Hippocampal place cells encode an animal's past, current, and future location through sequences of action potentials generated within each cycle of the network theta rhythm. These sequential representations have been suggested to result from temporally coordinated synaptic interactions within and between cell assemblies. Instead, we find through simulations and analysis of experimental data that rate and phase coding in independent neurons is sufficient to explain the organization of CA1 population activity during theta states. We show that CA1 population activity can be described as an evolving traveling wave that exhibits phase coding, rate coding, spike sequences and that generates an emergent population theta rhythm. We identify measures of global remapping and intracellular theta dynamics as critical for distinguishing mechanisms for pacemaking and coordination of sequential population activity. Our analysis suggests that, unlike synaptically coupled assemblies, independent neurons flexibly generate sequential population activity within the duration of a single theta cycle.
DOI:http://dx.doi.org/10.7554/eLife.03542.001
10.7554/eLife.03542.002eLife digest
When we explore a new place, we naturally create a mental map of the location as we go. This mental map is stored in a region of the brain called the hippocampus, which contains cells called place cells. These cells can carry information about our past, present, and future location in the form of electrical signals. They connect to each other to form networks and it has been proposed that these connections can store the information needed for the mental maps.
Real-time maps are represented in the information carried by the electrical signals themselves. A physical location is specified by the individual place cell that is activated, and by the timing of the electrical signal it produces relative to a ‘brain wave’ called the theta rhythm. Brain waves are patterns of electrical signals activated in sets of brain cells and the theta rhythm is produced in the hippocampus of an animal as it explores its surroundings.
Previous experiments suggested that when a rat explores an area, several sets of brain cells in the hippocampus are activated in sequence within each cycle of the theta rhythm. As the rat moves forward, the sequence shifts to different sets of cells to reflect the upcoming locations ahead of the rat. It has been thought that these sequences are triggered by the individual connections between the place cells.
Here, Chadwick et al. developed mathematical models of the electrical activity in the brains of rats as they explored. They used these models to analyze data from previous experiments and found that the sequences of electrical activity arise from the timing of each cell's activity relative to the theta rhythm, rather than from the connections between the cells.
Chadwick et al.'s findings suggest that the mental map may be highly flexible, allowing vast numbers of distinct memories to be stored within the same network of place cells without interference. Future studies will involve investigating the role of brain waves in the forming new mental maps and creating new memories.
DOI:http://dx.doi.org/10.7554/eLife.03542.002
Author keywordsmemorynavigationneural codecell assemblybrain oscillationneural computationResearch organismnonehttp://dx.doi.org/10.13039/501100000266Engineering and Physical Sciences Research Council (EPSRC)ChadwickAngushttp://dx.doi.org/10.13039/501100000268Biotechnology and Biological Sciences Research Council (BBSRC)NolanMatthew FThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementIndependent coding without synaptic coordination explains complex sequences of population activity observed during theta states and maximizes the number of distinct environments that can be encoded through population theta sequences.Introduction
Cognitive processes are thought to involve the organization of neuronal activity into phase sequences, reflecting sequential activation of different cell assemblies (Hebb, 1949; Harris, 2005; Buzsáki, 2010; Wallace and Kerr, 2010; Palm et al., 2014). During navigation, populations of place cells in the CA1 region of the hippocampus generate phase sequences structured around the theta rhythm (e.g., Skaggs et al., 1996; Dragoi and Buzsáki, 2006; Foster and Wilson, 2007). As an animal moves through the firing field of a single CA1 neuron, there is an advance in the phase of its action potentials relative to the extracellular theta cycle (O'Keefe and Recce, 1993). Thus, populations of CA1 neurons active at a particular phase of theta encode the animal's recent, current, or future positions (Figure 1A,B). One explanation for these observations is that synaptic output from an active cell assembly ensures its other members are synchronously activated and in addition drives subsequent activation of different assemblies to generate a phase sequence (Figure 1C) (Harris, 2005). We refer to this as the coordinated assembly hypothesis. An alternative possibility is that independent single cell coding is sufficient to account for population activity. According to this hypothesis, currently active assemblies do not determine the identity of future assemblies (Figure 1D). We refer to this as the independent coding hypothesis.10.7554/eLife.03542.003
Phase sequences in a place cell population.
(A) During navigation, place cells are sequentially activated along a route. (B) Within each theta cycle, this slow behavioral sequence of place cell activations is played out on a compressed timescale as a theta sequence. Theta sequences involve both rate and phase modulation of individual cells, but it remains unclear whether additional coordination between cells is present. (C) Internal coordination may bind CA1 cells into assemblies, and sequential assemblies may be chained together synaptically. This would require specific inter- and intra-assembly patterns of synaptic connectivity within the network. (D) Alternatively, according to the independent coding hypothesis, each cell is governed by theta phase precession without additional coordination.
DOI:http://dx.doi.org/10.7554/eLife.03542.003
Since these coding schemes lead to different views on the nature of the information transferred from hippocampus to neocortex and on the role of CA1 during theta states, it is important to distinguish between them. While considerable experimental evidence has been suggested to support the coordinated assembly hypothesis (e.g., Harris et al., 2003; Dragoi and Buzsáki, 2006; Foster and Wilson, 2007; Maurer et al., 2012; Gupta et al., 2012), the extent to which complex sequences of activity in large neuronal populations can be accounted for by independent coding is not clear. To address this we developed phenomenological models of independent and coordinated place cell activity during navigation. In the independent coding model, the spiking activity of each cell is generated by rate coding across its place field and phase precession against a fixed theta rhythm. We show that in this model phase coding generates a traveling wave which propagates through the population to form spike sequences. This wave is constrained by a slower moving modulatory envelope which generates spatially localized place fields. In the coordinated assembly model, the spikes generated by each cell are also influenced by the activity of other cells in the population. As a result, population spike patterns are further entrained by population interactions which counter the effects of single cell spike time variability and increase the robustness of theta sequences.
The independent coding hypothesis predicts that a population of independent cells will be sufficient to explain the spatiotemporal dynamics of cell assemblies in CA1. In contrast, the coordinated assembly hypothesis predicts that groups of cells show additional coordination beyond that imposed by a fixed firing rate and phase code (Harris et al., 2003; Harris, 2005). We show that the independent coding model is sufficient to replicate experimental data previously interpreted as evidence for the coordinated assembly hypothesis (Harris et al., 2003; Dragoi and Buzsáki, 2006; Foster and Wilson, 2007; Maurer et al., 2012; Gupta et al., 2012), despite the absence of coordination within or between assemblies. Moreover, novel analyses of experimental data support the hypothesis that place cells in CA1 code independently. Independent coding leads to new and experimentally testable predictions for membrane potential oscillations and place field remapping that distinguish circuit mechanisms underlying theta sequences. In addition we show that, despite the apparent advantage of coordinated coding in generating robust sequential activity patterns, it suffers from an inability to maintain these patterns in a novel environment. Thus, a key advantage of sequence generation through independent coding is to allow flexible global remapping of population activity while maintaining the ability to generate coherent theta sequences in multiple environments.
ResultsSingle cell coding model
To test the independent coding hypothesis, we developed a phenomenological model which generates activity patterns for place cell populations during navigation. While a phenomenological model of CA1 phase precession has previously been developed (Geisler et al., 2010), several features of this model limit its utility for investigation of coordination across neuronal populations. First, the previous model addresses only the temporal dynamics of single unit activity and population average activity, without addressing the spatiotemporal patterns of spiking activity within the population, the nature of which is a central question in the present study. Second, the previous model assumes coordination between cells in the form of fixed temporal delays and is formulated for a fixed running speed. In contrast, we wish to understand in detail the temporal relationships between cells arising in populations with no direct coordination and how these temporal relationships might depend on factors such as running speed. We therefore develop a model of a single cell with a given place field and phase code and proceed to derive the patterns of population activity under the independent coding hypothesis. To do this, we modeled the firing rate field for each neuron using a Gaussian tuning curve:rx(x)=A exp(−(x−xc)22σ2),where rx describes firing rate when the animal is at location x within a place field with center xc, width σ, and maximum rate A (Figure 2A, top panel). Simultaneously, we modeled the firing phase using a circular Gaussian:rϕ(ϕ(x),θ(t))=exp(k cos(ϕ(x)−θ(t))),where rϕ describes the firing probability of the neuron at each theta phase at a given location (Figure 2B). Here, θ(t) = 2πfθt is the local field potential (LFP) theta phase at time t and ϕ(x) is the preferred firing phase associated with the animal's location x, termed the encoded phase. The encoded phase ϕ(x) is defined to precess linearly across the place field (Figure 2A, bottom panel; Supplementary file 1, Appendix: A1). The phase locking parameter k determines the precision at which the encoded phase is represented in the spike output (Figure 2B). The instantaneous firing rate of the cell is given by the product of these two components r = rxrϕ. The phase locking can be set so that the cell exhibits only rate coding (at k = 0, where r = rx), only phase coding (as k → ∞, where all spikes occur at exactly the encoded phase ϕ(x)) or anywhere in between (Figure 2C).10.7554/eLife.03542.004
Single cell coding model.
(A) Firing rate and phase at different locations within a cell's place field are determined by a Gaussian tuning curve rx and linearly precessing encoded phase ϕ, respectively. (B) The dependence of single cell activity on the LFP theta phase θ is modeled by a second tuning curve rϕ which depends on the angle between the LFP theta phase θ and encoded phase ϕ at the animal's location. The phase locking parameter k controls the precision of the phase code. (C) The combined dependence of single cell activity on location and LFP theta phase. (D) Temporal evolution of the rate and phase tuning curves for a single cell as a rat passes through the place field at constant speed. (E) The total firing rate corresponding to (D), and spiking activity on 1000 identical runs.
DOI:http://dx.doi.org/10.7554/eLife.03542.004
10.7554/eLife.03542.005
Effect of normalization factor (<italic>N</italic><sub>spikes</sub>).
Firing rate vs time for runs with v = 50 cm/s, k = 0.7, and three different values of Nspikes.
DOI:http://dx.doi.org/10.7554/eLife.03542.005
To model place cell activity during navigation on a linear track, we set x(t) = vt, where v is the running speed (Figure 2D,E). This causes the encoded phase ϕ(t) to precess linearly in time at a rate fϕ which is directly proportional to running speed and inversely proportional to place field size, as in experimental data (Huxter et al., 2003; Geisler et al., 2007). To generate spikes we used an inhomogeneous Poisson process with an instantaneous rate r = rxrϕ. We normalized the firing rate such that the average number of spikes fired on a pass through a place field is independent of running speed (see Supplementary file 1, Appendix: A2) (Huxter et al., 2003). If the phase ϕ(x) at each location in the place field is fixed, the full rate and phase coding properties of a cell are encompassed by three independent parameters—the width of the spatial tuning curve σ, the degree of phase locking k, and the average number of spikes per pass Nspikes. Phase precession (Figure 2C) and firing rate modulation as a function of time in this model (Figure 2E) closely resemble experimental observations (e.g., Skaggs et al., 1996; Mizuseki and Buzsaki, 2013).
Place cells often show variations in firing rate in response to nonspatial factors relevant to a particular task (e.g., Wood et al., 2000; Fyhn et al., 2007; Griffin et al., 2007; Allen et al., 2012). In our model, such multiplexing of additional rate coded information can be achieved by varying the number of spikes per pass Nspikes without interfering with the other parameters ϕ(x), σ, and k (Figure 2—figure supplement 1).
It has been shown that the trial to trial properties of phase precession in individual cells are more variable than would be expected based on the pooled phase precession data (Schmidt et al., 2009). While it is possible that such trial to trial variability could reflect coordination between cell assemblies, such variability is equally consistent with an independent population code, and our model can be readily extended to incorporate such properties (Supplementary file 1, Appendix: A2).
Given this single cell model and assuming an independent population code, we next investigated the spatially distributed patterns of spiking activity generated in a CA1 population. To map the spatiotemporal dynamics of the population activity onto the physical space navigated by the animal, we analyzed the distributions of the rate components rx and phase components rϕ of activity in cell populations sorted according to the location xc of each place field (Supplementary file 1, Appendix: A3).
Our model naturally generates population activity at two different timescales: the slow behavioral timescale at which the rat navigates through space and a fast theta timescale at which trajectories are compressed into theta sequences. While the rat moves through the environment, the spatial tuning curves rx(x) generate a slow moving ‘bump’ of activity which, by definition, is comoving with the rat (Figure 3A, top, black). Simultaneously, the phasic component rϕ(ϕ(x),θ(t)) instantiates a traveling wave (Figure 3A, top, red). Due to the precession of ϕ(t), the wave propagates forward through the network at a speed faster than the bump, resulting in sequential activation of cells along a trajectory on a compressed timescale. The slower bump of activity acts as an envelope for the traveling wave, limiting its spatial extent to one place field (Figure 3A, bottom). The continuous forward movement of the traveling wave is translated into discrete, repeating theta sequences via a shifting phase relationship to the slow moving component (Figure 3B–D, Video 1). Moreover, this shifting phase relationship generates global theta oscillations at exactly the LFP frequency that cells were defined to precess against (Figure 3B, top panel). Thus, our model can be recast in terms of the dynamics of a propagating wavepacket comprising two components, with network theta resulting from their interaction. While we define single cells to precess against a reference theta rhythm (i.e., the LFP), we now see that this same reference oscillation emerges from the population, despite the higher frequencies of individual cells.10.7554/eLife.03542.006
Spatiotemporal dynamics of CA1 populations governed by independent coding.
(A) Top: Population dynamics during a single theta cycle on a linear track after ordering cells according to their place field center xc in physical space. The two components of the population activity are shown—the slow moving envelope (black) and the fast moving traveling wave (red), which give rise to rate coding and phase coding, respectively (cf. Figure 2). Bottom: Resulting firing rates across the population. When the traveling wave and envelope are aligned, the population activity is highest (middle panel). The dashed line shows the location of the rat at each instant. (B) Firing rate in the population over seven consecutive theta cycles. The fast and slow slopes are shown (solid and dashed lines, respectively), corresponding to the speeds of the traveling wave and envelope as shown in part (A). The top panel shows the LFP theta oscillations and emergent population theta oscillations, which are generated by the changing population activity as the traveling wave shifts in phase relative to the slower envelope (see Video 1). (C and D) The spiking activity for a population of 180 cells. All panels used v = 50 cm/s, so that vp = 350 cm/s and c = 7.
DOI:http://dx.doi.org/10.7554/eLife.03542.006
10.7554/eLife.03542.007
CA1 population activity governed by coordinated assemblies.
(A) The simulated place cells interact via a combination of asymmetric excitation and feedback inhibition. The weights plotted here govern how the spikes emitted by a given cell will influence the spiking activity of its peers depending on their relative place field locations. (B) Population firing rate on a single run along a linear track (180 cells with v = 50 cm/s and k = 0.5). The firing rate in each cell is a product of the animal's location, the LFP theta phase and the influence of recent peer spiking activity. (C) The spiking activity, generated using an inhomogeneous Poisson process. (D) Comparison of the global population firing rate for an independent coding population (black) and a coordinated population (red), with identical single cell properties. Interactions between cells amplify theta oscillations and introduce a shift in firing phase.
DOI:http://dx.doi.org/10.7554/eLife.03542.007
10.7554/eLife.03542.008
Traveling wave dynamics in populations of CA1 place cells.
Top: Distribution of the rate (black) and phasic (red) tuning curves for a population of linear phase coding place cells during constant speed locomotion on a linear track (cf. Figure 3A). The evolution in the population over 7 consecutive theta cycles is shown, slowed by a factor of approximately 16×. Bottom: The evolution of the overall firing rate distribution in the population, generated by multiplying the two tuning curves shown in the top panel. Note that the population firing rate undergoes oscillations at LFP theta frequency and the center of mass of the population activity shifts from behind the animal to ahead of the animal in each theta cycle.
DOI:http://dx.doi.org/10.7554/eLife.03542.008
Our model's prediction of global theta oscillations emerging in networks of faster oscillating place cells is consistent with a previous phenomenological model which assumed a fixed running speed and fixed, experimentally determined temporal delays between cells (Geisler et al., 2010). However, in contrast to previous models, our model based on single cell coding principles allows an analysis in which only place field configurations and navigational trajectories are required to fully predict at any running speed both the global theta oscillation and the detailed population dynamics. Experimental data show that the frequency of LFP theta oscillations is relatively insensitive to the running speed of the animal, showing a mild increase with running speed compared to a larger single unit increase (Geisler et al., 2007). We therefore investigated the relationship between the running speed of the animal, the temporal delays between cells and the frequency of population theta oscillations in the independent coding model.
The spiking delays between cells in our model are determined by speed of the fast moving traveling wave vp, which is related to the rat's running speed v by:vp=cv,where c is called the compression factor. This factor is equivalent to the ratio of the rat's actual velocity and the velocity of the representation within a theta cycle and has been quantified in previous experimental work (Skaggs et al., 1996; Dragoi and Buzsáki, 2006; Geisler et al., 2007; Maurer et al., 2012), although the relationship to the traveling wave model developed here was not previously identified (see Supplementary file 1, Appendix: A2 for derivation).
Analysis of our model demonstrates that for an independent population code the compression factor naturally depends on running speed. This change in compression factor with running speed ensures that the network maintains a fixed population theta frequency while running speed and single unit frequency vary:vp−v=λfθ,where the constant λ is the wavelength of the traveling wave (equal to the size of a place field, measured as the distance over which a full cycle of phase is precessed [Maurer et al., 2006]) and vp − v stays constant across running speeds due to the changing compression factor.
Hence, independent coding predicts temporal delays which are dependent on running speed. Conversely, our analysis shows that models incorporating fixed temporal delays between cells (e.g., Diba and Buzsáki, 2008; Geisler et al., 2010) cannot maintain an invariant relationship between spike phase and location without producing a population theta oscillation whose frequency decreases rapidly with running speed, in conflict with experimental observations (Geisler et al., 2007).
In order to compare activity patterns predicted by independent coding schemes with those predicted when interactions between cell assemblies are present, we developed a second model in which the spiking activity of each place cell influences the spiking activity of peer cells within the population. While single cell rate and phase tuning curves in this coordinated assembly model are identical to those in the independent coding model, a peer weight function also modulates the probability of a spike occurring in each cell depending on the spikes of its peers (Figure 3—figure supplement 1A, Supplementary file 1, Appendix: A4). In this model, asymmetric excitation stabilizes the temporal relationship between sequentially activated assemblies, while feedback inhibition between place cells normalizes firing rates (cf. Tsodyks et al., 1996). The resulting sequences are considerably more robust than those generated by independent coding with the same single cell properties (Figure 3—figure supplement 1B–C). Assembly interactions also amplify theta oscillations in the network (Figure 3—figure supplement 1D) (Stark et al., 2013). Hence, assembly coordination provides a potential mechanism for stabilizing the sequential activity patterns generated by noisy neurons, as interactions entrain cells in the population into coherent activation patterns within each theta cycle.
While alternative forms of assembly coordination might also be considered, we choose the present model for two key reasons. First, this model is simple, containing relatively few adjustable parameters while capturing the essential features of sequence generation via assembly coordination. Second, as we will show below, the coordination between cells under this model is sufficient to evaluate statistical tests of independence, allowing a systematic framework with which to interpret the results of such tests on experimental data.
Independent coding accounts for apparent peer-dependence of CA1 activity
We next investigated the extent to which models for population activity based on independent coding and coordinated assemblies can account for observations previously suggested to imply coordination within and between assemblies (Harris et al., 2003; Dragoi and Buzsáki, 2006; Foster and Wilson, 2007; Maurer et al., 2012; Gupta et al., 2012). We show below that, although these observations at first appear to imply assembly coordination, they can be accounted for by the independent coding model. We go on to establish the power of several tests to distinguish spike patterns generated by independent and coordinated coding models. By applying these tests to experimental data, we provide further evidence that CA1 population activity is generated through independent coding.
We first assessed whether independent coding accounts for membership of cell assemblies. A useful measure of the coding properties of place cell populations is to test how accurately single unit activity can be predicted from different variables. If, after accounting for all known single cell coding properties, predictions of the activity of individual place cells can be further improved by information about firing by their peer cells, it is likely that such cells are interacting through cell assemblies (Harris, 2005). Initial analysis of CA1 place cell firing suggested this is the case, with coordination between cells at the gamma timescale being implicated (Harris et al., 2003). Because this improved predictability directly implies interactions between CA1 neurons, it would constitute strong evidence against the independent coding hypothesis. However, in accounting for single cell phase coding properties, the prediction analysis of Harris et al. (2003) assumed that firing phase is independent of movement direction in an open environment. In contrast, more recent experimental data show that in open environments firing phase always precesses from late to early phases of theta, so that firing phase at a specific location depends on the direction of travel (Huxter et al., 2008; Climer et al., 2013; Jeewajee et al., 2014). Therefore, to test if the apparent peer-dependence of place cell activity is in fact consistent with independent coding, the directionality of phase fields must be accounted for.
To address this we first considered whether the assumption of a nondirectional phase field would lead to an erroneous conclusion of coordinated coding when analyzing spike patterns generated by the independent coding model. To do this, we extended the traveling wave model to account for phase precession in open environments (Supplementary file 1, Appendix: A6). We then constructed phase fields from simulated spiking data following the approach of Harris et al. (2003), in which firing phase is averaged over all running directions, and separately constructed directional phase fields consistent with recent experimental observations (Huxter et al., 2008; Climer et al., 2013; Jeewajee et al., 2014). We then calculated the predictability of neuronal firing patterns generated by the independent coding model using each of these phase fields. For simplicity, we considered the problem in one dimension, treating separately passes from right to left, left to right, and the combined data in order to generate the directional and nondirectional phase fields (Figure 4A,B, respectively). We ignored any shifts in place field centers for different running directions (e.g., Battaglia et al., 2004; Huxter et al., 2008) and assumed that the place cells did not engage in multiple reference frames (Jackson and Redish, 2007; Fenton et al., 2010).10.7554/eLife.03542.009
Peer prediction analysis for an independent population code.
(A) Combined place and phase fields constructed from simulated data using only runs with a single direction. (B) Place/phase field constructed from a combination of both running directions, as used by Harris et al. (2003). (C) Predictability analysis, using various combinations of place, phase, and peer activity. When using the nondirectional phase field of Harris et al. (2003), an additional peer predictability emerges (black vs green and purple). However, this additional predictability is seen to be erroneous if the directional phase field is used to predict activity (red). (D) Dependence of peer predictability on the peer prediction timescale and phase locking of individual cells, for an independent population code. The heat map shows the predictability of a cell's activity from peer activity (cf. part C, green line). The optimal peer prediction timescale depends on the amount of phase locking. The 20 ms characteristic timescale of peer correlations reflects independent phase precession of single cells rather than transient gamma synchronization of cell assemblies.
DOI:http://dx.doi.org/10.7554/eLife.03542.009
10.7554/eLife.03542.010
Change in information after addition of peer activity to prediction metrics.
Distributions of information gain/loss in individual cells after including peer activity in addition to all other prediction metrics. For independent coding and experimental data, peer prediction causes a decrease in information on average (p = 3.9 × 10−17 and p = 1.4 × 10−6, respectively). For coordinated coding, peer prediction causes an increase in information on average (p = 9 × 10−83). The decrease in information observed for independent coding simulations when peer activity is included occurs due to overfitting on a dataset of finite size. Due to statistical fluctuations in the data, peer weights are generally estimated as non-zero. Both the peer weights and the change in information when peers are included would be expected to approach zero as the amount of data increases for independent coding simulations, but not for coordinated coding simulations.
DOI:http://dx.doi.org/10.7554/eLife.03542.010
10.7554/eLife.03542.011
Results of prediction analysis on individual sessions.
Top: Number of cells for which prediction improved with peers after place fields, velocity modulation factors and directional phase fields had been fitted, shown for each session/running direction in the experimental dataset. Middle: The results when the same analysis was applied to data simulated with independent coding (twice as many sessions were simulated for comparison). Bottom: The results when data were simulated with coordinated assemblies.
DOI:http://dx.doi.org/10.7554/eLife.03542.011
For the independent coding model, we find that peer prediction provides a higher level of information about a neuron's firing than predictions based on place and nondirectional phase fields, despite the absence of intra-assembly coordination in our simulated data (Figure 4C, green and purple). However, prediction based on place fields and directional phase fields outperforms both of these metrics (Figure 4C, red). Therefore, previous evidence for intra-assembly coordination can be explained by a failure to account for the phase dependence of CA1 firing. Instead, our analysis indicates that independent phase precession of CA1 neurons is sufficient to account for observations concerning membership of CA1 assemblies. We also find that nondirectional phase fields (Figure 4B), as assumed by (Harris et al., 2003), yield little improvement in predictability of a neuron's firing compared with predictions based on the place field alone, and for high phase locking are detrimental (Figure 4C, blue vs black). While Harris et al. (2003) found that nondirectional phase fields generally do improve prediction, this discrepancy may arise from more complex details of experimental data in open exploration, for example a nonuniform distribution of running directions through the place field, which would cause the information in nondirectional phase fields to increase.
Because peers share a relationship to a common theta activity and implement similar rules for generation of firing, a cell's activity in the independent coding model can nevertheless be predicted from that of its peers in the absence of information about location or theta phase (Figure 4C, green). The quality of this prediction is dependent on the timescale at which peer activity is included in the analysis, so that the optimal timescale for peer prediction provides a measure of the temporal resolution of assembly formation. In experimental data the optimal timescale for peer prediction is approximately 20 ms, which corresponds to the gamma rhythm and the membrane time constant of CA1 neurons (Harris et al., 2003). We find that in the independent coding model the optimal peer prediction timescale depends strongly on phase locking (Figure 4D). Even though the model does not incorporate gamma oscillations or neuronal membrane properties, high values of phase locking also show a striking peak in peer predictability around the 20 ms range (Figure 4D). We show below that for running speeds in the range 35–75 cm/s phase locking is likely to lie within the range at which the observed 20 ms prediction timescale dominates. Thus, the 20 ms timescales found both here and experimentally are explainable as a signature of the common, independent phase locking of place cells to the theta rhythm, rather than transient gamma coordination or intrinsic properties of CA1 neurons.
While the above analysis demonstrates that independent coding is consistent with previous experimental results, it does not exclude the presence of coordinated assemblies. In particular, it is not clear whether, when applied to experimental data, including information about peer activity would continue to improve prediction compared to place and directional phase fields alone. We therefore applied the prediction analysis based on directional phase fields to experimental datasets recorded from CA1 place cells (Mizuseki et al., 2014). To provide benchmarks for the interpretation of experimental results, we also analyzed simulated datasets generated with either independent coding or coordinated assemblies. We simulated datasets with the same number of sessions and recorded cells per session as the experimental dataset in order to obtain measures of peer prediction performance expected under each hypothesis (see ‘Materials and methods’). In simulations of independent cells, we found that information about peer activity continues to improve predictability compared to prediction from place and directional phase fields alone. The source of this predictability was found to be the common modulation of firing rate in each cell with the running speed of the animal, which is a further single cell coding feature not previously accounted for in prediction analyses (McNaughton et al., 1984; Czurko et al., 1999; Huxter et al., 2003; Ahmed and Mehta, 2012). We therefore included in our analysis an additional prediction factor, termed the velocity modulation factor (see ‘Materials and methods’).
After accounting for rate fields, directional phase fields and velocity modulation factors, inclusion of peer information increased the predictability of 84% of place cells simulated through coordinated coding, but only 38% of cells simulated through independent coding (see Table 1 for a summary of all prediction metrics). On average, information decreased by 0.047 bits/s for each cell simulated by independent coding and increased by 0.24 bits/s for coordinated coding when peer information was added (Wilcoxon signed rank test, p = 3.9 × 10−17 and p = 9 × 10−83, respectively, Figure 4—figure supplement 1). Thus, this new prediction analysis which accounts for directional phase fields and velocity modulation can effectively distinguish between independent and coordinated coding.10.7554/eLife.03542.012
Performance of prediction metrics on experimental and simulated data
DOI:http://dx.doi.org/10.7554/eLife.03542.012
Prediction metric
Independent coding
Coordinated coding
Experimental data
Location
100%
100%
44.6% (SEM 5.8%)
Running speed
99.3%
99.7%
77.8% (SEM 3.7%)
Phase field
99.3%
100%
75.7% (SEM 5.7%)
Peer activity
38%
84.3%
32.5% (SEM 11%)
The percentage of cells for which prediction performance increased with the addition of each metric. Percentages refer to the number of cells for which information increased when the specified metric was included in addition to those listed in rows above. Note that for velocity, phase and peer prediction, only those cells for which prediction performance improved with information about location were considered. Simulations demonstrate that, after taking into account place fields, velocity modulation factors and phase fields, information about peer activity improves prediction for the majority of cells when coordination is present, but not when cells are independent. Experimental data are consistent with independent coding.
When we applied this prediction analysis to experimental data, prediction performance improved for 75.7% (±5.7%, SEM, n = 10 sessions) of experimentally observed place cells when phase fields were included and 77.8% (±3.7%) of place cells when velocity modulation factors were included. In contrast, prediction performance improved for only 32% (±11%) of the experimentally observed place cells when peer information was included after accounting for single cell coding properties (Figure 4—figure supplement 2 shows the results for individual experimental sessions). On average, addition of peer information decreased the predictability of each cell by 0.049 bits/s (±0.013, SEM, n = 270 cells, Wilcoxon signed rank test, p = 1.4 × 10−6), in agreement with independent coding simulations and in contrast to coordinated coding simulations. Hence, after fully accounting for the directional properties of phase fields and the dependence of firing rate on running speed, peer prediction analysis supports independent coding as the basis of experimentally observed place cells in CA1. Therefore, based on comparison of simulated with experimental datasets, coordinated assemblies appear unlikely to account for the observed activity in CA1.
Independent coding accounts for phase sequences
While the above analysis demonstrates that intra-assembly interactions are not required to account for membership of CA1 assemblies, several studies support a role for inter-assembly coordination in the generation of theta sequences (Dragoi and Buzsáki, 2006; Foster and Wilson, 2007; Maurer et al., 2012; Gupta et al., 2012). We therefore investigated whether the independent coding or coordinated assembly model would better account for the results of these studies. We focus initially on the path length encoded by spike sequences, which we define as the length of trajectory represented by the sequence of spikes within a single theta cycle. Experimental data show that this path length varies with running speed (Maurer et al., 2012; Gupta et al., 2012), but it is not clear whether this phenomenon is a feature of independent coding or instead results from coordination between assemblies. To address this we first derived analytical approximations to the sequence path length for strong phase coding, where k → ∞ (Supplementary file 1, Appendix: A2). This analysis predicts a linear increase in sequence path length with running speed, but with a lower gradient than that found experimentally (Maurer et al., 2012). Hence, independent coding with strong phase locking does not quantitatively explain the changes in sequence properties with running speed.
We reasoned that independent coding might still explain observed sequence path lengths if a more realistic tradeoff between rate and phase coding is taken into account. To test this, we varied phase locking k and decoded the path length following the method of Maurer et al. (2012), which decodes the location represented by the population at each time bin in a theta cycle to estimate the encoded trajectory. We found that a good match to the data of Maurer et al. (2012) can be obtained by assuming that the degree of phase locking increases with running speed (Figure 5A). This is due to the dependence of the decoded path length on the strength of phase locking (Figure 5—figure supplement 1A).10.7554/eLife.03542.013
Decoded sequence path lengths and population activity propagation speeds.
(A) With constant phase locking, the decoded path length increases linearly with running speed, but to account for experimental data a dependence of phase locking on running speed is required. The shaded regions show lower and upper bounds (k = 0 and k = ∞). (B) Dependence of decoded fast slope on running speed (cf. our Figure 3B; Figure 3 of Maurer et al. (2012)). Again, a match to the data requires a velocity dependent phase locking. (C) The decoded slow slope matches the analytical value, where the population travels at the running speed v. Bounds show LFP theta frequencies below 4 Hz (upper bound) and above 12 Hz (lower bound) at each given running speed.
DOI:http://dx.doi.org/10.7554/eLife.03542.013
10.7554/eLife.03542.014
Dependence of decoded sequence path lengths, fast slopes, and slow slopes on phase locking.
(A) The decoded path length depends on the phase locking of individual cells. For zero phase locking, the decoded path length is the distance traveled by the rat in a theta cycle. This is because the decoded location in each time bin is simply the location of the rat. As phase locking is increased the path length increases asymptotically towards our analytical result, which is the distance traveled by the rat plus one full place field. This effect arises due to the gradual separation of cells representing different locations into separate theta phases, as seen explicitly in Figure 3C,D. Phases within a single theta cycle represent past, present, and future locations along the track. Dashed lines show the phase locking values plotted in Figures 2, 3. (B) Dependence of decoded fast slope on phase locking. While the analytical result for vp is independent of phase locking, the decoded value shown here is consistent with the intuitive notion that the sequence path length D is equal to the distance traveled by the fast moving wave in a theta cycle. (C) The decoded slow slope does not depend on phase locking, which is expected given the separation of timescales involved.
DOI:http://dx.doi.org/10.7554/eLife.03542.014
10.7554/eLife.03542.015
Results of shuffling analysis.
(A–D) The analysis of Foster and Wilson (2007) and (E–F) a corrected analysis. (A) Spike phases were initially calculated by interpolation between theta peaks, shown as dotted lines. (B) After shuffling the phases of spikes, a new spike time is calculated by interpolation between the nearest two theta troughs (dotted lines) to the original spike, which often generates erroneous spike times. The shuffled spike in this case acquires a small phase jitter, but a large temporal jitter. (C) The unshuffled sequence correlations between cell rank order and spike times. The red line shows the mean correlation. (D) Shuffled sequence correlations remained greater than zero, but were significantly reduced relative to the unshuffled case as in experimental data (Foster and Wilson, 2007). (E) Results of a corrected shuffling procedure applied to simulated independent coding datasets and an experimental dataset (height magnified for comparison). Displayed are the average changes in sequence correlations caused by shuffling for each simulated dataset. In 74% of simulated datasets, there was no significant difference between the original and shuffled distributions. (F) Results of the corrected shuffling procedure when applied to datasets simulated with coordinated assemblies. In 81% of simulated coordinated coding datasets, shuffling significantly changed the distribution of sequence correlations. The experimental dataset was not significantly affected by shuffling (p = 0.28, t-test, 2436 putative sequences).
DOI:http://dx.doi.org/10.7554/eLife.03542.015
Maurer et al. (2012) found that the compression factor c, which measures the compression of an encoded trajectory into a single theta cycle, also depends on running speed. To test whether independent coding might account for this observation, we investigated the behavior of the fast and slow slopes of population activity (as shown in Figure 3B), representing assembly propagation at theta timescales and behavioral timescales, respectively (i.e., vp and v). In the analysis of Maurer et al. (2012), the compression factor was estimated as the ratio of these two quantities. Following again the methods used by Maurer et al. (2012) to decode the fast and slow slopes from spiking data, we found that the dependence of the decoded fast slope on running speed in our simulated data matches experimental data provided that phase locking is again made dependent on running speed (Figure 5B, Figure 5—figure supplement 1B). However, the slower behavioral timescale dynamics did not match those reported by Maurer et al. (2012). Our decoded values for the slow slope closely matched the true value based on the rat's running speed. In contrast, the values reported by Maurer et al. (2012) are considerably lower (Figure 5C) which, if correct, would suggest that the population consistently moved more slowly than the rat, even moving backwards while the animal remained still. Because of this discrepancy we could not reproduce the compression factor reported by Maurer et al. (2012). Nevertheless, the independent coding model accurately reproduces the theta timescale activity reported by Maurer et al. (2012).
The above analysis has two important implications. First, both the decoded sequence path length and theta-compressed propagation speed in the independent coding model match experimental data provided the degree of theta modulation of spike output increases linearly with running speed. This dependence of phase locking on running speed is consistent with the observed increase in LFP theta amplitude (McFarland et al., 1975; Maurer et al., 2005; Patel et al., 2012), and is a novel prediction made by our model. Second, since the temporal delays between cells are determined by the propagation speed vp, the close match of this quantity to experimental data confirms the dependence of temporal delays on running speed predicted by our model, and argues against models based on fixed delays (Diba and Buzsáki, 2008; Geisler et al., 2010).
Further experimental support for the notion of inter-assembly coordination has come from an analysis suggesting that single cell phase precession is less precise than observed theta sequences (Foster and Wilson, 2007). This conclusion relies on a shuffling procedure which preserves the statistics of single cell phase precession yet reduces intra-sequence correlations. However, performing the same shuffling analysis on data generated by our independent coding model also reduced sequence correlations (t-test, p < 10−46) (Figure 5—figure supplement 2). The effect arises because the shuffling procedure does not preserve the temporal structure of single cell phase precession, despite preserving the phasic structure (Figure 5—figure supplement 2A,B). Hence, the phase–position correlations are unaffected, while the time–position correlations and hence sequence correlations are disrupted (Figure 5—figure supplement 2C,D). Thus, inter-assembly coordination is not required to account for these findings.
Nevertheless, although these results are reproducible by the independent coding model, it remains possible that coordinated assemblies underly the observed theta sequences. In particular, it is unclear whether this shuffling procedure could be modified to obtain a test for assembly coordination with greater statistical specificity and if so, whether it would reveal assembly coordination within experimental datasets. To address these questions, we analyzed experimental data along with data generated by independent coding and coordinated assembly models, using a modified version of this shuffling procedure (see ‘Materials and methods’). We found that the new shuffling procedure successfully detected assembly coordination with a statistical power of 81% (calculated for datasets containing the same number of sessions, cells, and sequences as our experimental dataset). When applied to experimental data from CA1, the shuffling test failed to detect any significant effect of shuffling (t-test, p = 0.28, 2436 events), as in most (74%) of the simulated independent coding datasets (Figure 5—figure supplement 2E,F). This failure to detect evidence of assembly coordination gives further support to the independent coding hypothesis.
In additional support for the coordinated assembly hypothesis, Dragoi and Buzsáki (2006) performed an analysis suggesting that, during continuous locomotion around a rectangular track, some cell pairs show a lap by lap covariation of firing rates (termed the dependent pairs). These cell pairs were found to spike with a more reliable temporal lag within theta cycles than cell pairs whose firing rates are independent, which was interpreted as evidence for direct interactions between dependent neurons. To test whether these results are instead consistent with independent coding, we applied the analysis of lap by lap firing rate covariations to data from simulations of the independent coding model. We found a similar fraction of apparently dependent cell pairs to that reported by Dragoi and Buzsáki (2006), despite the absence of any true dependencies between cells in the model (see ‘Materials and methods’). Hence, this analysis artificially separates homogeneous populations of place cells into apparently dependent and independent cell pairs. Moreover, these dependent and independent cell groups displayed different spatial distributions of place fields, with dependent cell pairs generally occuring closer together on the track (Wilcoxon rank sum test, p = 1.8 × 10−16). By separating a homogeneous population of cells into dependent and independent groups, the analysis therefore introduces a sampling bias, leading to dependent cell pairs having different properties. While we were unable to reproduce the analysis of the temporal lags in each group due to a lack of information provided within the original study (see ‘Materials and methods’), the emergence of dependent cell pairs with measurably different properties in independent coding simulations nevertheless demonstrates that these results are not indicative of interactions between neurons.
Finally, precise coordination of theta sequences has been suggested on the basis that theta sequence properties vary according to environmental features such as landmarks and behavioral factors such as acceleration, with sequences sometimes representing locations further ahead or behind the animal (Gupta et al., 2012). To establish whether independent coding could also account for these results, we generated data from our model and applied the sequence identification and decoding analysis reported by Gupta et al. (2012). We found that, even for simulated data based on pure rate coding with no theta modulation (k = 0), large numbers of significant sequences were detected at high running speeds (Figure 6A). Therefore, to test the performance of the full sequence detection and Bayesian decoding protocol used by (Gupta et al., 2012), we analyzed two simulated datasets—one with a realistic value of phase locking (k = 0.5, Figure 6B–D, solid lines) and another with zero phase locking (i.e., no theta related activity, Figure 6B–D, dashed lines). In both cases, applying the reported Bayesian decoding analysis yielded similar decoded path lengths to those found experimentally (Figure 6C,D). Importantly, there was an inverse relationship between the ahead and behind lengths decoded from the simulated data, which reproduces the apparent shift in sequences ahead or behind the animal observed in experimental data (cf. Figure 4c of Gupta et al. (2012)). This effect was dependent on the density of recorded place fields on the track and the threshold for the minimum number of cells in a theta cycle required for sequence selection (Figure 6—figure supplement 1). As these results were obtained both in the case with realistic phase coding and in the case with only rate coding (and therefore no theta sequences), the properties of the decoded trajectories are not related to theta activity within the population. Hence, these data do not constrain models of theta activity in CA1.10.7554/eLife.03542.016
Analysis of individual sequence statistics.
(A) The fraction of theta cycles which are classified as ‘significant sequences’ according to the Gupta et al. (2012) analysis, as a function of running speed and phase locking (for simulated data generated under the independent coding model). Large fractions of significant sequences are generated even without phase coding or theta sequences within the population (i.e., at k = 0). The black line shows the fraction reported experimentally. (B) The distribution of significant sequences over running speed and decoded path length for simulated data with phase locking k = 0.5, as calculated by Gupta et al. (2012) (cf. their Figure 1c). (C) The relationship between decoded path length and decoded ahead and behind lengths for significant sequences, calculated for a dataset with no theta activity (k = 0) and a dataset with realistic theta activity (k = 0.5). (D) The relationship between the ahead length of the sequence and the behind length of the sequence for these two datasets. Note that the properties of the decoded trajectories do not depend on the theta activity in the data. This replicates the experimental data (cf. Figure 4a-c of Gupta et al. (2012)), showing that similar trajectories are decoded by this algorithm regardless of the presence of theta sequences.
DOI:http://dx.doi.org/10.7554/eLife.03542.016
10.7554/eLife.03542.017
Dependence of decoded trajectories on the number of cells in a sequence.
(A–C) Distributions of the number of cells which spike in a theta cycle, for simulations of the independent coding model with different densities of place fields on the track (i.e., different numbers of place fields on a track of fixed length). (A) The cell density used to reproduce the results of Gupta et al. (2012). (B and C) Simulations with higher place field densities in which more active cells are recorded in each theta cycle on average. (D–F) Relationship between decoded ahead and behind length, calculated as in Gupta et al. (2012), shown for simulations with different place field densities and for different thresholds of the minimum number of cells required for a sequence to be included for analysis. (D) Simulations with 12 cells on the track and a threshold of three cells generate results similar to Gupta et al. (2012). (E–F) The density of place fields on the track and the threshold for sequence selection affect the decoded trajectories, with higher values for either resulting in a smaller change in behind length as a function of ahead length. (G–H) Spearman's rank correlation between ahead length and behind length for different place field densities plotted as a function of the threshold for the minimum number active of cells. Although the magnitude of the effect shown in (D–F) is diminished as these quantities increase, the correlation between ahead and behind length stays constant. Moreover, this correlation remains significant despite the decreasing effect size. Only when the number of selected sequences becomes too low to maintain a reliable measure does the effect become insignificant.
DOI:http://dx.doi.org/10.7554/eLife.03542.017
In total, our analysis demonstrates that a traveling wave model based on independent phase coding for CA1 theta states is consistent with existing experimental data. Thus, neither intra- nor inter-assembly interactions are required to explain spike sequences observed in CA1 during theta states. Our analyses of experimental data along with simulations from each hypothesis render it unlikely that assembly coordination significantly shapes the structure of theta sequences or CA1 cell assemblies. Below, we investigate some functional consequences of the independent coding and coordinated assembly hypotheses and show that, despite the advantage of assembly coordination in generating robust sequential activity patterns, it suffers from severe limitations in remapping and storage of multiple spatial maps. Independent coding offers a solution to this problem, allowing flexible generation of sequential activity over multiple spatial representations.
Linear phase coding constrains global remapping
What are the advantages of independent coding compared to sequence generation through interactions between cell assemblies? When an animal is moved between environments, the relative locations at which place cells in CA1 fire remap independently of one another (e.g., O'Keefe and Conway, 1978; Wilson and McNaughton, 1993). This global remapping of spatial representations poses a challenge for generation of theta sequences through coordinated assemblies as synaptic interactions that promote formation of sequences in one environment would be expected to interfere with sequences in a second environment. Indeed, in the coordinated assembly model, simulations of remapping reduced single cell phase precession to below the level of independent cells (i.e., of an identical simulation with interactions between cells removed). Remapping in the coordinated coding model also substantially reduced firing rate and population oscillations (Figure 7—figure supplement 1). This decrease in firing rate following remapping contradicts experimental data showing an increase in firing rate in novel environments (Karlsson and Frank, 2008). It is not immediately clear whether the independent coding model faces similar constraints on sequence generation across different spatial representations. We therefore addressed the feasibility of maintaining theta sequences following remapping given the assumptions that underpin our independent coding model.
We first consider the possibility that following remapping the phase lags between cell pairs remain fixed—that is, while two cells may be assigned new firing rate fields, their relative spike timing within a theta cycle does not change. This scenario would occur if the phase lags associated with phase precession were generated by intrinsic network architectures (e.g., Diba and Buzsáki, 2008; Geisler et al., 2010; Dragoi and Tonegawa, 2011, 2013) or upstream pacemaker inputs. For fixed phase lags, place cells display linear phase coding, whereby a cell continues to precess in phase outside of its rate coded firing field at a constant rate (Figure 7A). In this scenario, the phase lag between two neurons depends linearly on the distance between their place field centers, while cells separated by multiples of a place field width share the same phase (Figure 7A). Each cell pair therefore has a fixed phase lag in all environments and all cells can in principle be mapped onto a single chart describing their phase ordering (Figure 7A). If this mechanism for determining phase ordering is hardwired, then following arbitrary global remapping, cells with nearby place field locations will in most cases no longer share similar phases (Figure 7B). As a result, theta sequences and the global population theta will in general be abolished (Figure 7B). However, there exist a limited set of remappings which in this scenario do not disrupt the sequential structure of the population (e.g., Figure 7C). On a linear track, these remappings are: translation of all place fields by a fixed amount, scaling of all place fields by a fixed amount, and permuting the place field locations of any cell pair with zero phase lag.10.7554/eLife.03542.018
Properties of CA1 populations governed by linear phase coding.
(A) On a linear track, cells which precess linearly in phase maintain fixed theta phase lags. This is illustrated as a phase ordering (colored bar), which describes the relative phase of each cell (arrows show locations of cells at each phase). Each cell has a constant, running speed dependent frequency and a fixed phase offset to each other cell. (B) A complete global remapping with phase lags between cells held fixed. Theta sequences and population oscillations are abolished. (C) In a constrained place field remapping, theta sequences are preserved. (D) In open environments, phase lags depend on running direction. The set of population phase lag configurations needed to generate sequences in each direction is called a phase chart. (E) If a population has a fixed phase chart, the possible remappings are restricted to affine transformations.
DOI:http://dx.doi.org/10.7554/eLife.03542.018
10.7554/eLife.03542.019
Remapping with coordinated assemblies.
(A) Comparison of single cell phase precession generated by coordinated assemblies (before and after remapping) and independent coding. For this simulation, single cell phase and rate fields were assumed to be perfectly remapped, so that any changes are purely due to assembly interactions. Note that, while assembly interactions improve phase coding in single cells in the initial environment, after remapping these same interactions disrupt phase precession and cause a lower (circular-linear) correlation between spike phase and animal location than that generated by independent cells. (B) Population firing rate on a single trial along a linear track. While assembly interactions initially entrain and amplify theta oscillations in the population compared to independent cells, after remapping these interactions disrupt theta activity and cause a lower overall activity level.
DOI:http://dx.doi.org/10.7554/eLife.03542.019
When considering global remapping in an open environment similar constraints apply. Because the phase lag between any two cells depends on running direction (e.g., Huxter et al., 2008), the population phase ordering must always be aligned with the direction of movement (Figure 7D). Hence, in open environments, the notion of a phase chart must be extended to include a fixed phase ordering for each running direction. Given such a fixed phase chart, a set of remappings known as affine transformations preserve the correct theta dynamics (see Supplementary file 1, Appendix: A7). Such remappings consist of combinations of linear transformations (scaling, shear, rotation, and reflection) and translations (Figure 7E). Remappings based on permutation of place field locations of synchronous cells, which are permissible in one dimensional environments, are no longer tenable in the two dimensional case due to constraints over each running direction.
Since place cell ensembles support statistically complete (i.e., non-affine) remappings (e.g., O'Keefe and Conway, 1978) while maintaining phase precession, CA1 network dynamics are not consistent with the model outlined above. Moreover, this analysis demonstrates that previous models based on fixed temporal delays between cells (e.g., Diba and Buzsáki, 2008; Geisler et al., 2010) cannot maintain theta sequences following global remapping. Nevertheless, it remains possible that CA1 theta dynamics are based on fixed phase charts, provided that multiple such phase charts are available to the network, similar to the multiple attractor charts which have been suggested to support remapping of firing rate (Samsonovich and McNaughton, 1997). In this case, each complete remapping recruits a different phase chart, fixing a new set of phase lags in the population. The number of possible global remappings that maintain theta sequences is then determined by the number of available phase charts. Such a possibility is consistent with recent suggestions of fixed sequential architectures (Dragoi and Tonegawa, 2011, 2013) and has not been ruled out in CA1. It is also of interest that affine transformations are consistent with the observed remapping properties in grid modules (Fyhn et al., 2007), suggesting that a single phase chart might be associated with each grid module.
Sigmoidal phase coding enables theta sequence generation and flexible global remapping
The above analysis demonstrates that both coordination of assemblies and independent, linear phase coding pose severe restrictions on global remapping which appear at odds with experimental observations. We asked if it is possible to overcome these constraints so that phase sequences can be flexibly generated across multiple environments. We reasoned that experimental data on phase precession only imply that phase precesses within a cell's firing field and need not constrain a cell's phase outside of its firing field. We therefore implemented a version of the independent coding model in which firing phase has a sigmoidal relationship with location (Figure 8A–B, solid line; Supplementary file 1, Appendix: A5), such that phase precesses within the firing field but not outside of the field. In this case, each cell's intrinsic frequency increases as the animal enters the spatial firing field and drops back to LFP frequency when the animal exits the firing field (Figure 8C, solid line). This is in contrast to the linear phase model and previous work with fixed delays (Geisler et al., 2010) in which each cell's intrinsic frequency is always faster than the population oscillation, both inside and outside of the place field (Figure 8C, dashed line). In a given environment, spike phase precession and sequence generation in a population of cells with sigmoidal phase coding (Figure 8D–F) are similar to models in which cells have linear phase coding. However, in addition, sigmoidal phase coding enables theta sequences to be generated after any arbitrary global remapping (Figure 8G). This flexible global remapping is in contrast to the scrambling of theta sequences following global remapping when cells have linear phase coding (Figure 8G). Thus, independent sigmoidal coding is able to account for CA1 population activity before and after global remapping.10.7554/eLife.03542.020
Properties of CA1 populations governed by sigmoidal phase coding.
(A–C) Firing rate and intracellular phase and frequency in the linear (dashed lines) and sigmoidal models (solid lines) during the crossing of a place field. In the sigmoidal model, phase precession is initiated inside the place field by an elevation of intracellular frequency from baseline. (D–F) Firing rate and intracellular phase and frequency for a place cell population on a linear track. In the sigmoidal model, an intracellular theta phase lag between cell pairs develops as the animal moves through their place fields. Outside their place fields, cell pairs are synchronized. (G) Global remapping in the linear and sigmoidal models. The sigmoidal model allows arbitrary remapping without disrupting population sequences.
DOI:http://dx.doi.org/10.7554/eLife.03542.020
Linear and sigmoidal models of phase coding lead to distinct experimentally testable predictions. Recordings of the membrane potential of CA1 neurons in behaving animals show that although spikes precess against the LFP, they always occur around the peak of a cell's intrinsic membrane potential oscillation (MPO) (Harvey et al., 2009). Therefore the intrinsic phase ϕ of each cell in our model (Figure 2D,E) can be interpreted as MPO phase. While data concerning the MPO phase outside of the firing field are limited, such data would likely distinguish generation of theta sequences based on a linear and sigmoidal phase coding. If CA1 implements linear phase coding, then the MPO of each cell should precess linearly in time against LFP theta at a fixed (velocity dependent) frequency, both when the animal is inside the place field and when the animal is at locations where the cell is silent (Figure 8A–C, dashed line). Alternatively, sigmoidal phase coding predicts that precession of the MPO against the LFP occurs only inside the firing rate field (Figure 8A,B, solid line) and that the MPO drops back to the LFP frequency outside of the place field (Figure 8C, solid line) as reported by Harvey et al. (2009). A further prediction of sigmoidal coding is that, in contrast to models based on fixed delays (Diba and Buzsáki, 2008; Geisler et al., 2010), the phase lag between any two cells changes when the animal moves through their place fields. Outside their place fields the cells are synchronized with each other and with the LFP, whereas a dynamically shifting phase lag develops as the animal crosses the place fields (Video 2). Finally, phase precession under the sigmoidal model behaves differently to the linear model in open environments. In the linear model, the phase chart fixes a different population phase ordering for each running direction, so that spike phase depends on the location of the animal and the instantaneous direction of movement. In the sigmoidal model, however, each cell has a location dependent frequency, so that the spike phase depends on the complete trajectory through the place field and no explicit directional information is required (see Supplementary file 1, Appendix: A6). Rather, the directional property of the sequence arises purely through a location dependent oscillation frequency in each cell combined with the trajectory of the animal through each place field. In summary, our analysis demonstrates how evaluation of theta sequences following global remapping and of theta phase within and outside of a cell's firing field will be critical for distinguishing models of CA1 assemblies and theta generation.10.7554/eLife.03542.021
Population dynamics with sigmoidal phase coding.
Top: Distribution of the rate (black) and phasic (red) tuning curves for a population of sigmoidal phase coding place cells during constant speed locomotion on a linear track. Bottom: The evolution of the overall firing rate distribution in the population. Again, the population firing rate undergoes oscillations at LFP theta frequency and the center of mass of the population activity shifts from behind the animal to ahead of the animal in each theta cycle. However, in this case cells with place field centers distant from the animal's current location are synchronized with zero phase lag.
DOI:http://dx.doi.org/10.7554/eLife.03542.021
Discussion
Our analysis demonstrates how complex and highly structured population sequences can be generated without coordination between neurons. In contrast to previous suggestions (Harris et al., 2003; Dragoi and Buzsáki, 2006; Foster and Wilson, 2007; Maurer et al., 2012; Gupta et al., 2012), we find that the theta-scale population activity observed in CA1 is consistent with phase precession in independent cells, without interactions within or between cell assemblies. We demonstrate that independent coding enables flexible remapping of CA1 population activity while maintaining the ability to generate theta sequences. These properties are consistent with maximization of the capacity of CA1 for representation of distinct spatial experiences.
The independent coding hypothesis leads to a novel view of the CA1 population as a fast moving traveling wave with a slower modulatory envelope. This model implements an invariant phase code via a change in the frequency and temporal delay between cells with running speed. Amplitude modulation of the envelope provides a mechanism for multiplexing spatial with nonspatial information, such as task specific memory items (Wood et al., 2000) and sensory inputs (Rennó-Costa et al., 2010). The independence of each neuron naturally explains the robustness of phase precession against intrahippocampal perturbations (Zugaro et al., 2005), an observation which is difficult to reconcile with models based on assembly interactions. Depending on the exact nature of the single cell phase code, independent phase coding can enable theta sequences to be maintained with arbitrary global remapping. This flexibility may maximize the number and diversity of spatial representations that CA1 can provide to downstream structures, offering a strong functional advantage over mechanisms based on interactions between cell assemblies.
Independent phase coding leads to new and experimentally testable predictions that distinguish mechanisms of CA1 function during theta states. First, an absence of coordination within or between assemblies has the advantage that neural interactions do not interfere with sequence generation after global remapping. Rather, for independent coding models the constraints on sequence generation following remapping arise from the nature of the phase code. With linear phase coding the set of sequences available to the network is fixed, resulting in a limited set of place field configurations with a particular mathematical structure (Figure 7). Interestingly, the remappings observed in grid modules (Fyhn et al., 2007), but not CA1, are consistent with those predicted for networks with a single fixed set of theta phase lags called a phase chart. These findings, together with the fact that the temporal delays between cells depend on running speed, argue against previous models based on fixed delays within CA1 populations (Diba and Buzsáki, 2008; Geisler et al., 2010). Nevertheless, more complex scenarios with multiple phase charts could explain CA1 population activity during theta oscillations and ‘preplay’, which suggests a limited remapping capacity for CA1 (Dragoi and Tonegawa, 2011, 2013). Alternatively, sigmoidal phase coding massively increases the flexibility for global remapping as cells can remap arbitrarily while maintaining coherent theta sequences within each spatial representation (Figure 8). Second, linear and sigmoidal phase coding predict distinct MPO dynamics. With linear phase coding, the temporal frequency of each MPO is independent of the animal's location. With sigmoidal phase coding, the MPO frequency increases inside the place field, so that phase precession occurs inside but not outside the place field. In this case, only the spiking assembly behaves as a traveling wave, whereas the MPOs of cells with place fields distant from the animal are phase locked to the LFP. Sigmoidal phase precession could emerge due to inputs from upstream structures (Chance, 2012) or be generated intrinsically in CA1 place cells (Leung, 2011). Finally, in contrast to linear phase coding populations, sigmoidal phase coding populations do not require additional information from head direction or velocity cells to generate directed theta sequences in open environments. Instead, sigmoidal theta sequences are determined solely by the recent trajectory of the rat through the set of place fields together with a location dependent oscillation frequency, consistent with recent observations of reversed theta sequences during backwards travel (Cei et al., 2014; Maurer et al., 2014). In summary therefore, these two models may be distinguished experimentally on the basis of observations of the number of non-affine remappings in CA1, the intracellular frequency and delay between place cells as a function of location and of the dependence of firing phase on the trajectory through a place field in open environments.
While theta sequences of CA1 activity are most commonly observed during spatial navigation, similar activity patterns associated with short term memory have been observed during wheel running (Pastalkova et al., 2008). In this situation each cell's activity depends on the phase of the LFP theta rhythm and on the temporal location within an ‘episode field’ rather than a place field. Our model can be applied equally well to these internally generated sequences if the rate coded episode field is assumed to have a similar temporal structure to a place field. An entirely different class of sequences, however, is observed during non-theta states such as sharp wave ripples (SWR) (Buzsáki et al., 1992; Diba and Buzsáki, 2007). In contrast to theta sequences, SWR sequences are generally observed during states of immobility and are believed to arise from synchronous discharge in CA3 (Buzsáki et al., 1983). Because SWR sequences are generated without co-occurence of longer time-scale firing fields or theta oscillations, they cannot be accounted for by the independent coding schemes that we investigate here, in which rate and phase information determine the activity of each cell. Instead, the nature of cell assemblies in CA1 may be highly state dependent, operating in at least two modes. During theta states, sequences are generated by independently precessing neurons, whereas during SWRs sequences may result from interactions between consecutively activated cell assemblies.
Can independent coding account for manipulations that modify place cell dynamics? Administration of cannabinoids disrupts phase precession by CA1 neurons and impairs spatial memory, but does not appear to affect the rate coded place firing fields of CA1 neurons (Robbe and Buzsáki, 2009). This dissociation between rate and phase coding can be accounted for in our model by assuming that rate fields are maintained while phase fields are disrupted (Figure 2A) or the degree of phase locking (k) is substantially reduced (Figure 2B). In contrast, increased in-field firing of place cells following optogenetic inactivation of hippocampal interneurons (Royer et al., 2012) can be accounted for in our model by increased Nspikes, while altered phase of place cell firing following inactivation of parvalbumin interneurons can be accounted for in our model by modifying the phase fields (Figure 2A) of the place cells. Important future tests of the independent coding model will include comparison of its predictions of sequence activity, remapping and intracellular dynamics to experimental measures made during these kinds of manipulations.
Our independent coding model offers a comprehensive account of population activity in CA1 during theta states and makes new predictions for coordination of network dynamics and remapping at the population level, but it does not aim to distinguish cellular mechanisms for phase precession. Nevertheless, by demonstrating that existing observations of population sequences can be explained by independent coding our model argues against mechanisms for phase precession that rely on synaptic coordination at theta time scales (e.g., Tsodyks et al., 1996; Maurer and McNaughton, 2007; Lisman and Redish, 2009). In contrast, our model does not distinguish between specific single cell mechanisms for phase precession such as dual oscillators (Lengyel et al., 2003; Burgess et al., 2007), depolarizing ramps (Mehta et al., 2002), intrinsic membrane currents (Leung, 2011) or dual inputs from CA3 and entorhinal cortex (Chance, 2012). Our model is also consistent with inheritance of phase precession in CA1 from upstream circuits in CA3 and entorhinal cortex (Jaramillo et al., 2014). However, it argues against the possibility that CA1 inherits coordinated sequences from CA3 (Jaramillo et al., 2014). It is possible that CA3 nevertheless generates sequences by synaptic coordination. Because CA3 neurons are connected by dense recurrent collaterals (Miles and Wong, 1986; Le Duigou et al., 2014), there are likely to be substantial correlations in their output to CA1, which could induce deviations from the independent population code outlined here. However, feedback inhibition motifs such as those found in CA1 may counteract such correlations (Renart et al., 2010; Tetzlaff et al., 2012; Bernacchia and Wang, 2013; King et al., 2013; Sippy and Yuste, 2013). Hence, the local inhibitory circuitry in CA1 may actively remove correlations present in its input in order to generate an independent population code (Ecker et al., 2010).
A major advantage of independently precessing cell populations is that they provide a highly readable, robust, and information rich code for working and episodic memory in downstream neocortex. In particular, a downstream decoder with access to an independent population code need only extract the stereotyped correlational patterns associated with traveling waves under a given place field mapping. In this way it can flexibly decode activity across a large number of spatial representations. Decoding in the presence of additional correlations would likely lead to a loss of information (Zohary et al., 1994). While this loss can to some extent be limited by including knowledge of these additional correlations (Nirenberg and Latham, 2003; Eyherabide and Samengo, 2013), this likely requires a high level of specificity and therefore a lack of flexibility in the decoder. The flexibility afforded by an independent population code may therefore provide an optimal format for the representation and storage of the vast number of spatial experiences and associations required to inform decision making and guide behavior.
Materials and methodsSimulations of CA1 population activity
In the independent coding model, we simulated data from a population of place cells with place field centers xc and width σ which precess linearly through a phase range of Δϕ over a distance 2R on a linear track using Equation (A3.6) in Supplementary File 1. The initial phase ψs was either taken as 0, or a uniform random variable ψs ∈ [0,2π] set at the beginning of each run. In all simulations, parameters were set as: 2R = 37.5 cm (Maurer et al., 2006), Δϕ = 2π, σ = 9 cm, fθ = 8 Hz, Nspikes = 15. Finite numbers of place cells were simulated with place field centers xc which were either uniformly distributed along a linear track with equal spacing or randomly sampled from a uniform distribution over the track. All cells were therefore identical up to a shift in place field center xc. Simulations were performed using Matlab 2010b and 2013b.
Simulations of population activity generated through coordinated assemblies used equations (A4.1–4.5) in Supplementary File 1, with the single cell properties simulated as for the independent coding model. The peer interaction timescale was set to τ = 25 ms, and the interaction length for asymmetric excitation was set to ℓ = 10 cm with an excitatory amplitude of wE = 1/4. The amplitude of the inhibitory weights was varied until the same number of spikes were generated as in the independent coding simulation (for the parameters used in these simulations, the inhibitory amplitude was wI = 1/18).
Experimental datasets
We used data recorded from CA1 during navigation along a linear track. For details of experimental data see Mizuseki et al. (2014). For the analysis performed in this study, simultaneous recordings of a large number of coactive cells in CA1 are required, which restricted the number of suitable datasets. In particular, we used datasets ec016.233, ec016.234, ec016.269, ec014.468, ec014.639.
Prediction analyses
To replicate the results of Harris et al. (2003), we simulated constant speed movement along a linear track, with lap by lap running speeds drawn from a normal distribution with mean 35 cm/s and standard deviation of 15 cm/s. We simulated motion in each direction, using the same set of place fields in each case. We estimated the preferred firing phase at each location from the simulated data using the methods stated in Harris et al. (2003), using either single-direction data or data consisting of runs in both directions to generate nondirectional or directional phase fields. The prediction analysis was performed according to the methods given in Harris et al. (2003). For these initial simulations (Figure 4), we used the simulated value of phase locking rather than estimating it from the data. To display the peer prediction performance shown in Figure 4C, the optimal prediction timescale for each phase locking value was chosen. This was done separately for the peer only case and the peer plus phase field case.
We then performed additional, more detailed simulations to test the performance of simulated and experimental data when using the new directional phase fields. We separated datasets according to the running direction along a linear track, analyzing each direction individually. In addition to fitting the place field, phase field, and peer factor used by Harris et al. (2003), we also fitted a velocity modulation factor given by:A(v)=∑tntw(|v−vt|)∑tr0(xt)dtw(|v−vt|),which estimates the changes in firing rate of a cell according to running speed. In the above equation, the notation follows that of Harris et al. (2003) (their Supplementary Information), that is, w is a Gaussian smoothing kernel of width 3.5 cm/s, nt is the number of spikes fired by the cell in time bin t, r0 is the estimated firing rate field at location x, xt is the animal's location in time bin t, and vt is its velocity. Our simulations showed that, using the methods of Harris et al. (2003), the phase locking parameter k was underestimated outside of the place field center. Misestimation of phase field parameters introduces false peer predictability in simulated datasets. We therefore replaced their location dependent estimation with a fixed value equal to the phase locking estimated in regions where the place field is over 2/3 its maximum value. We also found that the 5 cm spatial smoothing kernel used by Harris et al. (2003) resulted in a high level of spurious peer prediction in simulations based on independent coding, since it extended the boundaries of place fields, allowing non-overlapping peer cells to compensate via inhibitory weights. A smaller kernel of 3.5 cm reduced the rate of false positive for peer prediction and was therefore used instead. We simulated 300 cells in each session of which we randomly sampled 15 for analysis in order to match the number of place cells typically recorded experimentally. 28 laps were simulated for each session and 10 sessions were simulated in total (representing the two running directions over the five experimental sessions we analyzed). Peer prediction was performed at a timescale of 25 ms (the optimal timescale in Harris et al. (2003)).
Changes in sequence properties with running speed
To compare the sequence path length in spiking data generated from the independent coding model to experimental data, we followed the decoding methods outlined in Maurer et al. (2012). Briefly, this involves constructing trial averaged time by space population activity matrices in order to decode the location represented by the population in each time bin over an average theta cycle. The decoded path length is measured as the largest distance between decoded locations within the theta cycle. To test the influence of phase locking in this analysis, k was varied incrementally from 0 to 6 and the sequence path length for the resulting data was calculated in each case. We used the same spatial and temporal bins (0.7 cm and 20° of LFP θ) as the original study.
To calculate the fast and slow slopes, we generated the contour density plots described by Maurer et al. (2012) using the same parameters as the sequence path length analysis. We simulated 100 trials for each running speed. We then divided these 100 trials into 10 subsets of 10 and applied the contour analysis to each subset. We fitted the fast slope to the 95% contour of the central theta peak, and measured the slow slope as the line joining the maximum of the top and bottom peaks of the central 3. We averaged over the results from each subset to obtain the final value. The analytical value for the fast slope in the limit of high phase locking is FS = vp/(360fθ), where the denominator arises due to the normalization to cm/deg in the analysis of Maurer et al. (2012). Similarly for zero phase locking, FS = v/(360fθ). The analytical value for the slow slope is independent of phase locking, SS = v/(360fθ). Upper and lower bounds for the slow slope were therefore fitted assuming the reported running speed is accurate, and that the LFP theta frequency is in the range 4 Hz < fθ < 12 Hz.
Shuffling analyses
To reproduce the results of Foster and Wilson (2007), we generated data from 1000 theta cycles, each with a running speed drawn from the same distribution as for the prediction analysis. Following the protocol outlined by Foster and Wilson (2007), we found the set of all spike phases for each cell when the rat was at each position and analyzed events defined as 40 ms windows around firing rate peaks. Spike phases were calculated by interpolation between LFP theta peaks. For the shuffling analysis, each spike in an event was replaced by another spike taken from the same cell while the animal was at the same location. The new spike time was then calculated from its phase by interpolation between the closest two LFP theta troughs of the original spike, as reported in the original study. 100 such shuffles were performed for each event, and the correlation between cell rank order and spike times was calculated in each case.
For the corrected shuffling procedure, we followed the methods of Foster and Wilson (2007) but with the following adjustments: the correlations between spike times and place field rank order within an event calculated in the original study were replaced with circular-linear correlations between spike phase and place field peaks in order to remove issues arising from conversion between spike time and spike phase (Kempter et al., 2012); a minimum running speed of 20 cm/s and a maximum running speed of 100 cm/s were imposed; the LFP phase was measured using a Hilbert transform rather than a linear interpolation between theta peaks; spikes that occured more than 50 cm away from the place field peak were discarded from the analysis. The circular-linear correlation requires a mild restriction of the range of possible regression slopes between the circular and linear variables, which in this case describes the distance traveled by a theta sequence within a theta cycle (Kempter et al., 2012). We set this range as 25–80 cm/cycle, that is, around the size of a place field. For simulations using this shuffling procedure, we simulated 300 cells in each session on a linear track and randomly sampled 15 of these for further analysis. We again simulated 10 sessions with 28 laps each, for which the number of detected events was similar to that of the experimental dataset. We generated a large number of such datasets in order to obtain a distribution of shuffling test results to compare against the experimental dataset.
Dependent and independent cells
To reproduce the results of Dragoi and Buzsáki (2006), we simulated population activity on a linear track. To recreate the experimental conditions of Dragoi and Buzsáki (2006), we set the track length as 250 cm and simulated 8 sessions (i.e., four animals by two running directions), each with 25 place cells. As the original experiment consisted of continuous locomotion around a rectangular track, we wrapped the boundaries of the linear track and simulated continuous sessions rather than single laps. Place fields were randomly distributed over the track following a uniform distribution. Running speed on each lap was drawn from the same distribution as the prediction and shuffling analyses. Phase locking was set to 0.5. We calculated the dependent and independent cell pairs following the methods of Dragoi and Buzsáki (2006), which uses temporal bins of 2 s to calculate firing rate correlations and a shuffling procedure to find significantly correlated cells.
Dragoi and Buzsáki (2006) did not state the number of dependent and independent cell pairs obtained from their analysis. Therefore, to compare the results of our simulations to their experimental data, we estimated the number of points in their CCG-lag plot for dependent and independent cell pairs (their Figure 3B) and compared the result to the same measure in our simulations. CCG plots were calculated using the methods described in Dragoi and Buzsáki (2006). Using this method, we found that 33% of cell pairs were dependent compared to an estimated 30–35% in Dragoi and Buzsáki (2006).
To calculate the reliability of temporal lags between dependent and independent pairs, Dragoi and Buzsáki (2006) took the central cloud of the CCG-lag vs place field distance scatter plot (their Figure 2A) and calculated the correlation between these two variables. However, the method for isolating the central cloud from the surrounding clusters was not disclosed. Without this information, we were unable to reproduce this analysis.
To test for differences between place field separations of dependent and independent cell pairs, we again considered only cell pairs whose CCG lags passed the inclusion criteria (as described in Dragoi and Buzsáki (2006)). We compared the vectors of cell pair separations for each group.
Decoding individual sequences
To reproduce the results of Gupta et al. (2012), we used the significant sequence testing protocol and Bayesian decoding algorithm described therein, with spatial binning set as 3.5 cm, as in the original study. Briefly, the significant sequence testing analysis tests if population activity within a theta cycle has significant sequential structure, whereas the Bayesian decoding algorithm generates a time by space probability distribution which is used to decode the ahead and behind lengths represented by the theta sequence. For Figure 6A, we varied phase locking and running speed independently and generated spiking data for each pair of values. In the simulations used to generate Figure 6, we assumed that the number of spikes fired per theta cycle does not vary with running speed, as such a dependence introduces an additional change of the decoded sequence path length with running speed. In order to best match the fraction of theta cycles with three or more cells active reported by Gupta et al. (2012), each simulated theta cycle contained 12 place cells with place fields randomly distributed over a region of space 94.5 cm ahead or behind the rat. We then applied the significant sequence detection methods for each resulting data set to obtain the fraction of significant sequences in each case. For Figure 6B, we used k = 0.5 and generated 1000000 theta cycles, each with a running speed drawn from a normal distribution with mean 30 cm/s and standard deviation 10 cm/s. Running speeds less than 10 cm/s were discarded and the remaining theta cycles were tested for significant sequential structure. For Figure 6C,D, we applied the Bayesian decoding algorithm to these significant sequences in order to calculate the path length, ahead length, and behind length. In addition, we applied the same analysis to another dataset simulated with k = 0.
Remapping simulations
To simulate remapping in the coordinated assembly model, we simulated spiking activity for a population of 300 cells on a linear track with weights as described in Supplementary file 1, Appendix: A4. To simulate the remapped population, we left this set of weights intact but randomly reassigned the place and phase fields of each cell, such that phase coding and rate coding were perfectly remapped but peer interactions were preserved between environments.
To simulate remapping in the linear phase coding model, we assumed that phase lags were preserved between environments. The remapped population was simulated by randomly permuting the place field centers of cells while leaving the phase fields of each cell intact.
To simulate remapping in the sigmoidal phase coding model, we assumed that the field of elevated frequency is locked to the place field before and after remapping. Hence, place fields were randomly permuted and the single cell frequency was defined to increase within the new place field.
Acknowledgements
This work was supported by the EPSRC, BBSRC, and MRC. We thank Gyuri Buzsaki, Kamran Diba, and Iris Oren for helpful comments on the manuscript. We are grateful for the provision of experimental data, made freely available at crcns.org (Mizuseki et al., 2014).
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
AC, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
MCWR, Conception and design, Analysis and interpretation of data, Drafting or revising the article
MFN, Conception and design, Analysis and interpretation of data, Drafting or revising the article
Additional files10.7554/eLife.03542.022
Mathematical appendices. This file includes all mathematical methods and derivations pertaining to the models described in the main text.
DOI:http://dx.doi.org/10.7554/eLife.03542.022
Major dataset
The following previously published dataset was used:
MizusekiK, DibaK, PastalkovaE, TeetersJ, SirotaA, BuzsakiG, 2014, Multiple single unit recordings from different rat hippocampal and entorhinal regions while the animals were performing multiple behavioral tasks, http://dx.doi.org/10.6080/K09G5JRZ, Publicly available at Collaborative Research in Computational Neuroscience (http://crcns.org).
ReferencesAhmedOJMehtaMR2012Running speed alters the frequency of hippocampal gamma oscillationsThe Journal of Neuroscience327373738310.1523/JNEUROSCI.5110-11.2012AllenKRawlinsJNPBannermanDMCsicsvariJ2012Hippocampal place cells can encode multiple trial-dependent features through rate remappingThe Journal of Neuroscience32147521476610.1523/JNEUROSCI.6175-11.2012BattagliaFPSutherlandGRMcNaughtonBL2004Local sensory cues and place cell directionality: additional evidence of prospective coding in the hippocampusThe Journal of Neuroscience244541455010.1523/JNEUROSCI.4896-03.2004BernacchiaAWangX2013Decorrelation by recurrent inhibition in heterogeneous neural circuitsNeural Computation251732176710.1162/NECO_a_00451BurgessNBarryCO'KeefeJ2007An oscillatory interference model of grid cell firingHippocampus1780181210.1002/hipo.20327BuzsákiG2010Neural syntax: cell assemblies, synapsembles, and readersNeuron6836238510.1016/j.neuron.2010.09.023BuzsákiGHorváthZUriosteRHetkeJWiseK1992High-frequency network oscillation in the hippocampusScience2561025102710.1126/science.1589772BuzsákiGLeungLWSVanderwolfCH1983Cellular bases of hippocampal EEG in the behaving ratBrain Research Reviews613917110.1016/0165-0173(83)90037-1CeiAGirardeauGDrieuCKanbiKEZugaroM2014Reversed theta sequences of hippocampal cell assemblies during backward travelNature Neuroscience1771972410.1038/nn.3698ChanceFS2012Hippocampal phase precession from dual input componentsThe Journal of Neuroscience321669316703a10.1523/JNEUROSCI.2786-12.2012ClimerJRNewmanELHasselmoME2013Phase coding by grid cells in unconstrained environments: two-dimensional phase precessionThe European Journal of Neuroscience382526254110.1111/ejn.12256CzurkóAHiraseHCsicsvariJBuzsákiG1999Sustained activation of hippocampal pyramidal cells by ‘space clamping’ in a running wheelEuropean Journal of Neuroscience1134435210.1046/j.1460-9568.1999.00446.xDibaKBuzsákiG2007Forward and reverse hippocampal place-cell sequences during ripplesNature Neuroscience101241124210.1038/nn1961DibaKBuzsákiG2008Hippocampal network dynamics constrain the time lag between pyramidal cells across modified environmentsThe Journal of Neuroscience28134481345610.1523/JNEUROSCI.3824-08.2008DragoiGBuzsákiG2006Temporal encoding of place sequences by hippocampal cell assembliesNeuron5014515710.1016/j.neuron.2006.02.023DragoiGTonegawaS2011Preplay of future place cell sequences by hippocampal cellular assembliesNature46939740110.1038/nature09633DragoiGTonegawaS2013Distinct preplay of multiple novel spatial experiences in the ratProceedings of the National Academy of Sciences of USA1109100910510.1073/pnas.1306031110EckerASBerensPKelirisGABethgeMLogothetisNKToliasAS2010Decorrelated neuronal firing in cortical microcircuitsScience32758458710.1126/science.1179867EyherabideHGSamengoI2013When and why noise correlations are important in neural decodingThe Journal of Neuroscience33179211793610.1523/JNEUROSCI.0357-13.2013FentonAALyttonWWBarryJMLenck-SantiniPPZinyukLEKubíkSBuresJPoucetBMullerRUOlypherAV2010Attention-like modulation of hippocampus place cell dischargeThe Journal of Neuroscience304613462510.1523/JNEUROSCI.5576-09.2010FosterDJWilsonMA2007Hippocampal theta sequencesHippocampus171093109910.1002/hipo.20345FyhnMHaftingTTrevesAMoserMBMoserEI2007Hippocampal remapping and grid realignment in entorhinal cortexNature44619019410.1038/nature05601GeislerCDibaKPastalkovaEMizusekiKRoyerSBuzsákiG2010Temporal delays among place cells determine the frequency of population theta oscillations in the hippocampusProceedings of the National Academy of Sciences of USA1077957796210.1073/pnas.0912478107GeislerCRobbeDZugaroMSirotaABuzsákiG2007Hippocampal place cell assemblies are speed-controlled oscillatorsProceedings of the National Academy of Sciences of USA1048149815410.1073/pnas.0610121104GriffinALEichenbaumHHasselmoME2007Spatial representations of hippocampal CA1 neurons are modulated by behavioral context in a hippocampus-dependent memory taskThe Journal of Neuroscience272416242310.1523/JNEUROSCI.4083-06.2007GuptaASvan der MeerMATouretzkyDSRedishAD2012Segmentation of spatial experience by hippocampal theta sequencesNature Neuroscience151032103910.1038/nn.3138HarrisKD2005Neural signatures of cell assembly organizationNature Reviews Neuroscience639940710.1038/nrn1669HarrisKDCsicsvariJHiraseHDragoiGBuzsákiG2003Organization of cell assemblies in the hippocampusNature42455255610.1038/nature01834HarveyCDCollmanFDombeckDATankDW2009Intracellular dynamics of hippocampal place cells during virtual navigationNature46194194610.1038/nature08499HebbDO1949The organization of behavior: a neuropsychological theoryVolume 44 of A Wiley book in clinical psychologyWileyHuxterJBurgessNO'KeefeJ2003Independent rate and temporal coding in hippocampal pyramidal cellsNature42582883210.1038/nature02058HuxterJRSeniorTJAllenKCsicsvariJ2008Theta phase-specific codes for two-dimensional position, trajectory and heading in the hippocampusNature Neuroscience1158759410.1038/nn.2106JacksonJRedishAD2007Network dynamics of hippocampal cell-assemblies resemble multiple spatial maps within single tasksHippocampus171209122910.1002/hipo.20359JaramilloJSchmidtRKempterR2014Modeling inheritance of phase precession in the hippocampal formationThe Journal of Neuroscience347715773110.1523/JNEUROSCI.5136-13.2014JeewajeeABarryCDouchampsVMansonDLeverCBurgessN2014Theta phase precession of grid and place cell firing in open environmentsPhilosophical Transactions of the Royal Society of London3692012053210.1098/rstb.2012.0532KarlssonMPFrankLM2008Network dynamics underlying the formation of sparse, informative representations in the hippocampusThe Journal of Neuroscience28142711428110.1523/JNEUROSCI.4261-08.2008KempterRLeiboldCBuzsákiGDibaKSchmidtR2012Quantifying circular-linear associations: hippocampal phase precessionJournal of Neuroscience Methods20711312410.1016/j.jneumeth.2012.03.007KingPDZylberbergJDeWeeseMR2013Inhibitory interneurons decorrelate excitatory cells to drive sparse code formation in a spiking model of V1The Journal of Neuroscience335475548510.1523/JNEUROSCI.4188-12.2013Le DuigouCSimonnetJTeleñczukMTFrickerDMilesR2014Recurrent synapses and circuits in the CA3 region of the hippocampus: an associative networkFrontiers in Cellular Neuroscience726210.3389/fncel.2013.00262LengyelMSzatmáryZÉrdiP2003Dynamically detuned oscillations account for the coupled rate and temporal code of place cell firingHippocampus1370071410.1002/hipo.10116LeungLS2011A model of intracellular theta phase precession dependent on intrinsic subthreshold membrane currentsThe Journal of Neuroscience31122821229610.1523/JNEUROSCI.0586-11.2011LismanJRedishAD2009Prediction, sequences and the hippocampusPhilosophical Transactions of the Royal Society of London Series B, Biological Sciences3641193120110.1098/rstb.2008.0316MaurerAPBurkeSNLipaPSkaggsWEBarnesCA2012Greater running speeds result in altered hippocampal phase sequence dynamicsHippocampus74773774710.1002/hipo.20936MaurerAPCowenSLBurkeSNBarnesCAMcNaughtonBL2006Organization of hippocampal cell assemblies based on theta phase precessionHippocampus1678579410.1002/hipo.20202MaurerAPLesterAWBurkeSNFerngJJBarnesCA2014Back to the future: preserved hippocampal network activity during reverse ambulationThe Journal of Neuroscience34150221503110.1523/JNEUROSCI.1129-14.2014MaurerAPMcNaughtonBL2007Network and intrinsic cellular mechanisms underlying theta phase precession of hippocampal neuronsTrends in Neurosciences3032533310.1016/j.tins.2007.05.002MaurerAPVanrhoadsSRSutherlandGRLipaPMcNaughtonBL2005Self-motion and the origin of differential spatial scaling along the septo-temporal axis of the hippocampusHippocampus1584185210.1002/hipo.20114McFarlandWLTeitelbaumHHedgesEK1975Relationship between hippocampal theta activity and running speed in the ratJournal of Comparative and Physiological Psychology8832432810.1037/h0076177McNaughtonBLBarnesCAO'KeefeJ1984The contributions of position, direction, and velocity to single unit activity in the hippocampus of freely-moving ratsExperimental Brain Research5419510.1007/BF00235832MehtaMRLeeAKWilsonMA2002Role of experience and oscillations in transforming a rate code into a temporal codeNature41774174610.1038/nature00807MilesRWongRK1986Excitatory synaptic interactions between CA3 neurones in the guinea-pig hippocampusThe Journal of Physiology37339741810.1113/jphysiol.1986.sp016055MizusekiKBuzsakiG2013Theta oscillations decrease spike synchrony in the hippocampus and entorhinal cortexPhilosophical Transactions of the Royal Society of London Series B, Biological Sciences3692012053010.1098/rstb.2012.0530MizusekiKDibaKPastalkovaETeetersJSirotaABuzsákiG2014Neurosharing: large-scale data sets (spike, LFP) recorded from the hippocampal-entorhinal system in behaving ratsF1000Research39810.12688/f1000research.3895.1NirenbergSLathamPE2003Decoding neuronal spike trains: how important are correlations?Proceedings of the National Academy of Sciences of USA1007348735310.1073/pnas.1131895100O'KeefeJConwayDH1978Hippocampal place units in the freely moving rat: why they fire where they fireExperimental Brain Research3157359010.1007/BF00239813O'KeefeJRecceML1993Phase relationship between hippocampal place units and the EEG theta rhythmHippocampus331733010.1002/hipo.450030307PalmGKnoblauchAHauserFSchüzA2014Cell assemblies in the cerebral cortexBiological Cybernetics10855957210.1007/s00422-014-0596-4PastalkovaEItskovVAmarasinghamABuzsákiG2008Internally generated cell assembly sequences in the rat hippocampusScience3211322132710.1126/science.1159775PatelJFujisawaSBerényiARoyerSBuzsákiG2012Traveling theta waves along the entire septotemporal axis of the hippocampusNeuron7541041710.1016/j.neuron.2012.07.015RenartAde la RochaJBarthoPHollenderLPargaNReyesAHarrisKD2010The asynchronous state in cortical circuitsScience32758759010.1126/science.1179850Rennó-CostaCLismanJEVerschurePF2010The mechanism of rate remapping in the dentate gyrusNeuron681051105810.1016/j.neuron.2010.11.024RobbeDBuzsákiG2009Alteration of theta timescale dynamics of hippocampal place cells by a cannabinoid is associated with memory impairmentThe Journal of Neuroscience29125971260510.1523/JNEUROSCI.2407-09.2009RoyerSZemelmanBVLosonczyAKimJChanceFMageeJCBuzsákiG2012Control of timing, rate and bursts of hippocampal place cells by dendritic and somatic inhibitionNature Neuroscience1576977510.1038/nn.3077SamsonovichAMcNaughtonBL1997Path integration and cognitive mapping in a continuous attractor neural network modelThe Journal of Neuroscience1759005920SchmidtRDibaKLeiboldCSchmitzDBuzsákiGKempterR2009Single-trial phase precession in the hippocampusThe Journal of Neuroscience29132321324110.1523/JNEUROSCI.2270-09.2009SippyTYusteR2013Decorrelating action of inhibition in neocortical networksThe Journal of Neuroscience339813983010.1523/JNEUROSCI.4579-12.2013SkaggsWEMcNaughtonBLWilsonMABarnesCA1996Theta phase precession in hippocampal neuronal populations and the compression of temporal sequencesHippocampus614917210.1002/(SICI)1098-1063(1996)6:23.0.CO;2-KStarkEEichlerRRouxLFujisawaSRotsteinHGBuzsákiG2013Inhibition-Induced theta resonance in cortical circuitsNeuron801263127610.1016/j.neuron.2013.09.033TetzlaffTHeliasMEinevollGTDiesmannM2012Decorrelation of neural-network activity by inhibitory feedbackPLOS Computational Biology8e100259610.1371/journal.pcbi.1002596TsodyksMVSkaggsWESejnowskiTJMcNaughtonBL1996Population dynamics and theta rhythm phase precession of hippocampal place cell firing: a spiking neuron modelHippocampus627128010.1002/(SICI)1098-1063(1996)6:33.3.CO;2-QWallaceDJKerrJND2010Chasing the cell assemblyCurrent Opinion in Neurobiology2029630510.1016/j.conb.2010.05.003WilsonMAMcNaughtonBL1993Dynamics of the hippocampal ensemble code for spaceScience2611055105810.1126/science.8351520WoodERDudchenkoPARobitsekRJEichenbaumH2000Hippocampal neurons encode information about different types of memory episodes occurring in the same locationNeuron2762363310.1016/S0896-6273(00)00071-4ZoharyEShadlenMNNewsomeWT1994Correlated neuronal discharge rate and its implications for psychophysical performanceNature37014014310.1038/370140a0ZugaroMBMonconduitLBuzsákiG2005Spike phase precession persists after transient intrahippocampal perturbationNature Neuroscience8677110.1038/nn136910.7554/eLife.03542.023Decision letterSkinnerFrances KReviewing editorUniversity Health Network, and University of Toronto, Canada
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
[Editors’ note: this article was originally rejected after discussions between the reviewers, but the authors were invited to resubmit after an appeal against the decision.]
Thank you for choosing to send your work entitled “Independent Theta Phase Coding Accounts for CA1 Population Sequences and Enables Flexible Remapping” for consideration at eLife. Your full submission has been evaluated by Eve Marder (Senior editor) and three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the decision was reached after long discussions between the reviewers. We regret to inform you that your work will not be considered further for publication.
In summary, although there was some enthusiasm among the reviewers for the work, this was tempered by concerns regarding insufficient support provided for the strong claims being made. Specifically, it was felt that this work does not represent a significant advance in the field without a more substantial analysis to support the authors’ claim that the experimental evidence is consistent with independent generation. Details of major substantive concerns raised by the reviewers are provided below for your consideration.
Reviewer #1:
The authors propose an independent coding hypothesis (as compared to a coordinated assembly hypothesis), in which an essential difference lies in currently active assemblies not determining future ones in the independent hypothesis case. An addition to existing phenomenological models (Geisler et al.) is the addition of spike generation (via an inhomogeneous Poisson process).
1) Given that a 'pacemaker' theta rhythm is included, it does not seem quite appropriate to talk about the generation of an emergent population theta rhythm. And, how should this be interpreted in light of the mentioned alternative of “populations… generate their own theta frequency…”, as stated in the seventh paragraph of the subsection “Independent coding accounts for phase sequences”, in the Results section?
2) It is not clear to me what the new and experimentally testable predictions are? Given that this is mentioned in the Introduction and Discussion, it should be clearly delineated. When it appears in the Discussion, subsequent sentences are more about the differences between the different hypotheses and its consequences, and how the independent hypothesis is better etc., but not what is 'experimentally testable' (and feasible?). In earlier parts of the manuscript, there is mention of optimal peer prediction timescale depending on phase locking, running speed dependencies etc.
So, what sorts of (feasible) experimental tests are being suggested? And what explicit predictions (differences from other hypotheses, coordinated assembly) should one look for? It would help the reader if this was more clearly set down, rather than the statement of “Important future tests of the independent coding…”, in the fourth paragraph of the Discussion section.
Since several experimental studies exist, the work as presented does not make it clear what 'new' experiments need to be done to support and distinguish this proposed independent hypothesis.
Reviewer #2:
This modeling study explores the population-level implications of the assumption that CA1 cells phase precess independently.
The main claim that the model “is sufficient to explain the organization of CA1 population activity during theta states” is a bold one, because several experimental studies argue strongly against independence. The authors re-examine some of these findings, providing useful new insights in the process. However, I do not think this main claim is adequately supported. In particular:
1) Some key pieces of evidence against independence are not considered: a) Dragoi and Buzsaki (2006) show that cells with correlated firing rates across laps have more reliable phase lags within theta cycles. If the authors model correlated firing rates using the Nspikes parameter, it is not obvious that an increased sequence compression index would result.
b) Schmidt et al. (J Neurosci, 2009) found that on a single pass through a place field, CA1 cells tend to precess through only about 180 degrees, and that the full 360-degree range is only obtained after averaging across passes. Thus, a single theta sequence does not randomly sample the 360 degree range as I believe would be the prediction from the authors' model. In addition, single trial sequences drawn from the average place-phase fields were a poor match to the observed sequences, a direct and strong test of independence.
c) Gupta et al. (2012) show that forward and backward sequence length are anticorrelated; this is not the distribution that would result from the authors' model. In addition, the relationship between velocity and sequence length reported here in Figure 6B is the opposite relationship from that shown in the Gupta paper (their Figure 1c).
2) The authors argue that some pertinent pieces of evidence, although originally advanced as evidence against independence, are in fact consistent with results from their model. However, in both cases (Harris et al., 2003; Foster and Wilson, 2007) the authors' argument hinges on a technical point in the original reports. It thus remains an open, empirical issue whether or not in those papers the main result would hold if the analyses were performed following the revised method the authors suggest.
To expand: in the case of Harris et al. it is shown that including more accurate (direction-dependent) phase fields into the analysis of data generated by the model improves peer prediction. However, this leaves open the empirical issue of when tested on actual data, adding peers would further improve the prediction. In the case of Foster and Wilson, the authors find that shuffling their model data as reported in that paper reduces correlations indicative of theta sequences, demonstrating that this result can in fact be obtained from independent neurons. Interestingly, the authors note that the method reported in the original paper may be modified to be a stronger test of non-independence. As reported it remains to be seen whether if, with this modified shuffling procedure, correlations would be reduced.
Given the above issues, I think the authors should align their interpretation with what they have shown (and not shown!). Overall I am enthusiastic about several aspects of the work: the model is compact and intuitive, providing not only satisfying insight, but also novel applications to experimental data sets with arbitrary speed profiles. It is thus a useful tool that sharpens the interpretation of such data sets, and suggests new analyses and experiments moving forward. The exploration of the relationship between theta sequences and remapping is thoughtful and generates useful testable predictions.
Reviewer #3:
This manuscript describes a model of hippocampal cell activity in which phase precession arises due to “traveling waves” within CA1. The authors first define their model, and then demonstrate that this model matches a number of data sets from experiments examining phase precession.
I think this model is clearly described and easy to understand. It should be easily generalized to other areas that demonstrate phase precession (for example CA3). However, the model is mainly descriptive, meaning it describes and characterizes the phenomenon of phase precession, but does not provide any new insights into the underlying mechanisms of phase precession, nor any new insights into how information is encoded in the hippocampus. The equations used to predict the firing rates of CA1 cells are very similar to equations used in previous studies of phase precession (the authors even cite some of these references). While I do think that having a clear description of phase precession and the implications that the existence of this phenomenon has on activity patterns within CA1 would be useful to the field, such a discussion is almost more appropriate for a review paper rather than a research article. Also, the manuscript is rather dense with concepts that are specific to temporal coding in the hippocampus. I am not sure how understandable this paper would be to those outside of this field.
Major comments:
1) Through their comparisons of the model with experimental data, the authors have provided an excellent review of the literature concerning hippocampal phase precession. I suspect, however, that sections of this paper may be incomprehensible to those not specifically working on phase precession. For example, I would have rather seen a more in-depth review of peer-predictions and its implications, rather than what felt like a cursory explanation and a citation of the Harris paper.
2) There is not much discussion of previous models of phase precession, although many of the experimental results used by the authors to support their model is predicted/explained by other models as well. For example, Geisler et al (2010) already demonstrated that the LFP theta rhythm can from a population of neurons oscillating faster than the theta-frequency. Although this paper is cited, I feel more attention should be paid to the analytical results of that paper rather than just focusing on the experimental data.
3) Along the same lines, it would have been nice to see at least some discussion of previous models of phase precession. While an exhaustive comparison is perhaps beyond the scope of this manuscript, O'Keefe's dual oscillator model is a classic and should at least be discussed.
10.7554/eLife.03542.024Author response
[Editors’ note: this article was originally rejected after discussions between the reviewers, but the authors were invited to resubmit after an appeal against the decision.]
In summary, although there was some enthusiasm among the reviewers for the work, this was tempered by concerns regarding insufficient support provided for the strong claims being made. Specifically, it was felt that this work does not represent a significant advance in the field without a more substantial analysis to support the authors’ claim that the experimental evidence is consistent with independent generation. Details of major substantive concerns raised by the reviewers are provided below for your consideration.
We thank the reviewers and editor for their helpful comments. In recognizing the initial concerns identified by the reviewers we have carried out substantial additional analyses: a) we now compare predictions from independent coding with a coordinated coding model, and b) we now compare analysis of experimental data with model predictions. Our new analysis provides further support to our initial conclusions. We outline these and further changes in the response below.
Reviewer #1:
The authors propose an independent coding hypothesis (as compared to a coordinated assembly hypothesis), in which an essential difference lies in currently active assemblies not determining future ones in the independent hypothesis case. An addition to existing phenomenological models (Geisler et al.) is the addition of spike generation (via an inhomogeneous Poisson process).
We note that while our model, like that of Geisler et al., addresses phase precessing assemblies at a phenomenological level, it differs conceptually in a number of important ways. First, our model allows an analysis of the spatiotemporal patterns of population activity, whereas Geisler et al. only investigated the temporal dynamics of single unit and population average activity. This is important because it allows analysis of theta sequences at the population level, which is central to the new advances made by our study. Second, our model generates realistic activity at arbitrary running speeds, while the fixed phase lags assumed by Geisler et al. are inconsistent with experimental data if running speeds are allowed to vary. Third, our model allows systematic variation of the phase locking of cells against the theta rhythm, leading to novel predictions for sequence properties, including a dependence of the decoded sequence path length and propagation speed on phase locking (Figures 5 and Figure 5–figure supplement 1 in our original and revised manuscripts). We appreciate we may not have made these important conceptual differences clear in our initial manuscript and have addressed this in the revised submission (across the subsection headed “Independent phase coding generates traveling waves”, in the Results section).
1) Given that a 'pacemaker' theta rhythm is included, it does not seem quite appropriate to talk about the generation of an emergent population theta rhythm. And, how should this be interpreted in light of the mentioned alternative of “populations… generate their own theta frequency…”, as stated in the seventh paragraph of the subsection “Independent coding accounts for phase sequences”, in the Results section?
We appreciate the reviewer's point and believe it perhaps reflects a lack of clarity on our part in the original manuscript. Thus, while the origin of the theta frequency oscillation is not central to our main conclusions, the reviewer identifies an element of our model that, because it is conceptually similar to that of Geisler et al., we perhaps did not explain sufficiently clearly. In our single cell model, we define neurons to precess in phase against a reference theta rhythm. As a result, each neuron oscillates at a velocity-dependent frequency which is always higher than that of the reference theta. Regardless of velocity, however, we find that the global population activity oscillates at the same frequency as the reference theta, i.e. at a lower frequency than each individual cell. Our use of the word “generate” is restricted to this scenario, where the network theta is “generated” from the sum of the higher frequency oscillations in each neuron. In the revised manuscript we have clarified this point (second paragraph of the subsection “Independent phase coding generates traveling waves”, in the Results).
2) It is not clear to me what the new and experimentally testable predictions are? Given that this is mentioned in the Introduction and Discussion, it should be clearly delineated. When it appears in the Discussion, subsequent sentences are more about the differences between the different hypotheses and its consequences, and how the independent hypothesis is better etc., but not what is 'experimentally testable' (and feasible?). In earlier parts of the manuscript, there is mention of optimal peer prediction timescale depending on phase locking, running speed dependencies etc.
So, what sorts of (feasible) experimental tests are being suggested? And what explicit predictions (differences from other hypotheses, coordinated assembly) should one look for? It would help the reader if this was more clearly set down, rather than the statement of “Important future tests of the independent coding… ”, in the fourth paragraph of the Discussion section.
Since several experimental studies exist, the work as presented does not make it clear what 'new' experiments need to be done to support and distinguish this proposed independent hypothesis.
We appreciate this was a major weakness of the previous manuscript and have carried out substantial new simulations and analysis of experimental data to address the point at length. We previously identified predictions that distinguish different scenarios for independent coding, but we did not make explicit predictions for analyses that would distinguish independent from coordinated coding. We also did not compare predictions from independent and coordinated coding models directly with experimental data and, as the reviewer notes, we did not distinguish predictions that require new experiments. We have addressed these issues as follows:
a) To address the question of how the independent coding hypothesis might be empirically distinguished from the coordinated coding hypothesis, we have developed an additional model and performed extensive additional simulations and analyses. The additional model includes interactions between cells within the population in order to simulate data under the coordinated assembly hypothesis (Figure 3–figure supplement 1 in the revised manuscript), while the additional simulations compare the behavior of the independent coding and coordinated assembly models when subjected to statistical tests of independence. In particular, we compared the performance of a shuffling analysis (adapted from Foster and Wilson, 2007; see Figure 5–figure supplement 2E, F in the revised manuscript) and a prediction analysis (adapted from Harris et al. 2003; please see Table 1 and Figure 4–figure supplement 1 and 2 in the revised manuscript). We find in both cases that spike patterns generated by independent coding and coordinated assembly models can be distinguished by the shuffling analysis and by the prediction analysis. We are able to estimate the statistical power of each analysis method given assumptions about the effect size and the number of recorded neurons.
We include these new results in the Results section of the revised manuscript (please see the subsections: “Assembly coordination stabilizes sequential activation patterns”, “Independent coding accounts for apparent peer-dependence of CA1 activity”, and “Independent coding accounts for phase sequences”).
We also clarify novel predictions requiring new data, including predictions involving membrane potential oscillations and place field remapping (subsection “Linear phase coding constrains global remapping” in the Results and paragraph three in the Discussion). We note here that, in addition to our previous predictions, our new coordinated assembly model has allowed the additional prediction that phase precession would be severely disrupted following remapping if CA1 assemblies were generated by coordinated coding (Figure 7–figure supplement 1 and subsection “Linear phase coding constrains global remapping”, in the Results of the revised manuscript).
b) Having demonstrated that these new analyses have the statistical power to distinguish independent from coordinated data, we applied these analyses to experimental data (for details of these data, see Mizuseki et al., 2014). For both tests, the experimental data favor the independent coding hypothesis (please see the subsections “Independent coding accounts for apparent peer-dependence of CA1 activity” and “Independent coding accounts for phase sequences” of the Results, Table 1, Figure 4–figure supplement 1, 2 and Figure 5–figure supplement 2E, F in the revised manuscript). We believe this new analysis provides very substantial new evidence which supports our original conclusions.
Reviewer #2:
This modeling study explores the population-level implications of the assumption that CA1 cells phase precess independently.
The main claim that the model “is sufficient to explain the organization of CA1 population activity during theta states” is a bold one, because several experimental studies argue strongly against independence. The authors re-examine some of these findings, providing useful new insights in the process. However, I do not think this main claim is adequately supported. In particular:
1) Some key pieces of evidence against independence are not considered: a)Dragoi and Buzsaki (2006)show that cells with correlated firing rates across laps have more reliable phase lags within theta cycles. If the authors model correlated firing rates using the Nspikes parameter, it is not obvious that an increased sequence compression index would result.
We appreciate the suggestion, but unfortunately we find that certain steps of the analysis reported by Dragoi et al. were not exactly reproducible due to a lack of information in their study. For example, the method for isolating the central cloud (their Figure 2B) from the overall CCG plot (their Figure 2A) was not disclosed. Nevertheless, in attempting to reproduce their methods as closely as possible, we found that their key results could be accounted for by independent coding. First, when analyzing the correlations in lap by lap firingrates using the methods of Dragoi et al., we found a similar number of apparently dependent cell pairs as the original study, despite the absence of true firingrate correlations within the simulated data. Hence, the analysis used by Dragoi et al. artificially separates homogeneous populations of place cells into apparently dependent and independent cell pairs. Second, these dependent and independent cell groups displayed different spatial distributions of firing rate fields. This suggests that the effects reported by Dragoi et al. might result from a sampling bias introduced by the separation of a homogeneous population of cells into dependent and independent groups. Thus, as far as we can tell, the results reported by Dragoi et al. are consistent with the independent coding hypothesis. We report these new analyses in the revised manuscript (paragraph seven of subsection “Independent coding accounts for phase sequences”, in the Results). We hope the reviewers and editors will also recognize the difficulty in making comparisons to previous work where that work has not been documented to a level where it can be reproduced.
b) Schmidt et al. (J Neurosci, 2009) found that on a single pass through a place field, CA1 cells tend to precess through only about 180 degrees, and that the full 360-degree range is only obtained after averaging across passes. Thus, a single theta sequence does not randomly sample the 360 degree range as I believe would be the prediction from the authors' model. In addition, single trial sequences drawn from the average place-phase fields were a poor match to the observed sequences, a direct and strong test of independence.
We note that the data of Schmidt et al. do not provide evidence for or against independent coding. This is because Schmidt et al. did not analyze sequences, but only single unit phase precession. Hence, while their results suggest a more complex single cell phase code than that included in our model, they cannot reveal coordination between cells as this would require an analysis of ensemble activity. Our single cell model can be readily extended to incorporate more complex single cell phase codes while maintaining independence between cells in the population. We now discuss this issue in the manuscript and include an additional appendix detailing a model which includes trial-by-trial single cell coding properties resembling those described by Schmidt et al. while maintaining independence between cells (final paragraph of subsection headed “Single Cell Coding Model”, in the Results section, and Appendix: A2).
c)Gupta et al. (2012)show that forward and backward sequence length are anticorrelated; this is not the distribution that would result from the authors' model. In addition, the relationship between velocity and sequence length reported here inFigure 6Bis the opposite relationship from that shown in the Gupta paper (theirFigure 1c).
While we agree with the reviewer that our previous presentation of the independent coding model did suggest a difference to the observations in Gupta et al., it does not follow that the Gupta et al data are inconsistent with independent coding. We have performed additional simulations using the Gupta protocol which show that, if the number of cells simulated is sufficiently small as to match the number of cells per theta cycle reported by Gupta et al., the anticorrelation between ahead and behind length arises naturally due to the sequence selection criteria. Importantly, these results are fully reproducible for realistic values of phase locking and also for zero phase locking, where no theta sequences exist in the data (revised manuscript Figure 6C, D). We further note that the relationship between velocity and sequence path length that we presented in our original manuscript was a consequence of our assumed change in firingrate as a function of running speed. Further simulations in which the number of spikes per theta cycle does not vary with running speed produce a relationship similar to that reported by Gupta et al. (revised manuscript Figure 6B). Thus, the Gupta et al. data are fully consistent with the independent coding model. We make these issues clear in the revised manuscript (paragraph eight of the subsection “Independent coding accounts for phase sequences”, in the Results section).
2) The authors argue that some pertinent pieces of evidence, although originally advanced as evidence against independence, are in fact consistent with results from their model. However, in both cases (Harris et al., 2003;Foster and Wilson, 2007) the authors' argument hinges on a technical point in the original reports. It thus remains an open, empirical issue whether or not in those papers the main result would hold if the analyses were performed following the revised method the authors suggest.
To expand: in the case of Harris et al. it is shown that including more accurate (direction-dependent) phase fields into the analysis of data generated by the model improves peer prediction. However, this leaves open the empirical issue of when tested on actual data, adding peers would further improve the prediction. In the case of Foster and Wilson, the authors find that shuffling their model data as reported in that paper reduces correlations indicative of theta sequences, demonstrating that this result can in fact be obtained from independent neurons. Interestingly, the authors note that the method reported in the original paper may be modified to be a stronger test of non-independence. As reported it remains to be seen whether if, with this modified shuffling procedure, correlations would be reduced.
Given the above issues, I think the authors should align their interpretation with what they have shown (and not shown!). Overall I am enthusiastic about several aspects of the work: the model is compact and intuitive, providing not only satisfying insight, but also novel applications to experimental data sets with arbitrary speed profiles. It is thus a useful tool that sharpens the interpretation of such data sets, and suggests new analyses and experiments moving forward. The exploration of the relationship between theta sequences and remapping is thoughtful and generates useful testable predictions.
We appreciate these points reflected substantial weaknesses in the previous manuscript. As detailed in our response to Reviewer 1, we have now performed extensive additional simulations and analyses which address these issues directly and in full. In particular, we show through simulations that our improved tests can successfully distinguish between coordinated and independent coding (please see the subsections “Independent coding accounts for apparent peer-dependence of CA1 activity” and “Independent coding accounts for phase sequences”, in the Results section of the revised manuscript), and we show that the results of these tests when applied to experimental data suggest independent coding rather than coordinated assemblies (see revised manuscript Table 1, Figure 4–figure supplement 1, 2 and Figure 5–figure supplement 2E, F). Our new simulations and analysis therefore provide further and we believe very substantial support to the independent coding hypothesis.
Reviewer #3:
This manuscript describes a model of hippocampal cell activity in which phase precession arises due to “traveling waves” within CA1. The authors first define their model, and then demonstrate that this model matches a number of data sets from experiments examining phase precession.
I think this model is clearly described and easy to understand. It should be easily generalized to other areas that demonstrate phase precession (for example CA3). However, the model is mainly descriptive, meaning it describes and characterizes the phenomenon of phase precession, but does not provide any new insights into the underlying mechanisms of phase precession, nor any new insights into how information is encoded in the hippocampus. The equations used to predict the firing rates of CA1 cells are very similar to equations used in previous studies of phase precession (the authors even cite some of these references). While I do think that having a clear description of phase precession and the implications that the existence of this phenomenon has on activity patterns within CA1 would be useful to the field, such a discussion is almost more appropriate for a review paper rather than a research article. Also, the manuscript is rather dense with concepts that are specific to temporal coding in the hippocampus. I am not sure how understandable this paper would be to those outside of this field.
We disagree with the Reviewer 3's suggestion that the model is descriptive and does not provide new insights. The reviewer's comments focus on our phenomenological model of phase precession in single cells. While this is in fact novel, as we make clear in our response to Reviewer 1 above, the conceptual importance of our work comes from our investigation of the population level activity predicted by this model. In this respect, it is only necessary that our model provides a good account of phase precession in single cells. We do not make any claims about mechanisms of phase precession. Given this misconception, we highlight again the key conceptual advances made by our study.
While considerable previous work has argued that population activity in CA1 during theta states involves coordinated coding, our model demonstrates that experimental evidence to support this conclusion can be fully accounted for by independent coding. We then use the model to develop novel insights into the implications of independent coding for place cell remapping. Thus, our manuscript provides a fundamentally different conception of population activity in CA1 to previous studies. Because theta activity in CA1 is coordinated with other circuits including prefrontal cortex and entorhinal cortex, our results have wide reaching implications for neural coding in general.
Our new simulations identify experimentally testable predictions that distinguish population activity under coordinated and independent coding scenarios. By comparison of these predictions to experimental data we now provide strong evidence in support of independent coding. We provide novel predictions for the consequences of different independent coding models for remapping of place representations. We have now extended this analysis to show that coordinated and independent coding fundamentally differ in their capabilities and limitations. Independent coding offers a massively increased ability to encode multiple environments, while coordinated coding provides a mechanism by which robust sequential activity can be generated despite the noisy intrinsic properties of individual place cells.
Thus, the models and analysis introduced by our study offer fundamental insights into both information processing and coordination of spike timing in hippocampal populations. In our revised manuscript we take care to make these novel insights much clearer to the reader.
Major comments:
1) Through their comparisons of the model with experimental data, the authors have provided an excellent review of the literature concerning hippocampal phase precession. I suspect, however, that sections of this paper may be incomprehensible to those not specifically working on phase precession. For example, I would have rather seen a more in-depth review of peer-predictions and its implications, rather than what felt like a cursory explanation and a citation of the Harris paper.
We have performed extensive additional peer prediction simulations on both simulated and experimental data (see comments above). Accordingly, this section of the manuscript has been expanded and a more in-depth explanation is included (subsection “Independent coding accounts for apparent peer-dependence of CA1 activity”, in the Results section).
2) There is not much discussion of previous models of phase precession, although many of the experimental results used by the authors to support their model is predicted/explained by other models as well. For example,Geisler et al. (2010)already demonstrated that the LFP theta rhythm can from a population of neurons oscillating faster than the theta-frequency. Although this paper is cited, I feel more attention should be paid to the analytical results of that paper rather than just focusing on the experimental data.
While we agree that several previous models of phase precession can account for the same phenomenological results as our model at the single-cell level, the purpose of our study is not to explain the mechanisms of single-cell phase precession but rather to understand the emergence of population activity. We now carefully explain the similarities and differences from the Geisler model, which we detailed above in our response to Reviewer 1.
3) Along the same lines, it would have been nice to see at least some discussion of previous models of phase precession. While an exhaustive comparison is perhaps beyond the scope of this manuscript, O'Keefe's dual oscillator model is a classic and should at least be discussed.
In the Discussion we outline which previous models of phase precession could provide a mechanistic basis for our single cell coding model and which models would instead imply coordinated assemblies. We now pay specifically mention the dual oscillator and other models in the Discussion section of the updated manuscript. Since the cellular mechanisms of phase precession are not a focus of our study we do not discuss these at length.
Many retroviral Gag proteins contain PPXY late assembly domain motifs that recruit proteins of the NEDD4 E3 ubiquitin ligase family to facilitate virus release. Overexpression of NEDD4L can also stimulate HIV-1 release but in this case the Gag protein lacks a PPXY motif, suggesting that NEDD4L may function through an adaptor protein. Here, we demonstrate that the cellular protein Angiomotin (AMOT) can bind both NEDD4L and HIV-1 Gag. HIV-1 release and infectivity are stimulated by AMOT overexpression and inhibited by AMOT depletion, whereas AMOT mutants that cannot bind NEDD4L cannot function in virus release. Electron microscopic analyses revealed that in the absence of AMOT assembling Gag molecules fail to form a fully spherical enveloped particle. Our experiments indicate that AMOT and other motin family members function together with NEDD4L to help complete immature virion assembly prior to ESCRT-mediated virus budding.
DOI:http://dx.doi.org/10.7554/eLife.03778.001
10.7554/eLife.03778.002eLife digest
To multiply and spread infections, viruses must enter and exit cells. Once inside a cell, many viruses conscript the cell's machinery to produce new viral particles and release them into the surroundings. Some viruses—like HIV-1—exit the cell in a way that leads to them being wrapped (or ‘enveloped’) in membrane from the host cell.
A virus protein called Gag is required for the release of HIV-1 and other enveloped viruses. In some cases, Gag proteins bind directly to members of the NEDD4 protein family to enable the viruses to be released. However, the Gag protein from HIV-1 does not appear to be able to interact directly with NEDD4 proteins, so it was not clear how Gag works in this case.
Mercenne et al. studied how HIV-1 is released from human cells grown in the laboratory. The experiments show that members of a human protein family called the Angiomotins bind to both Gag and NEDD4L (a member of the NEDD4 family) and are required for the efficient release of viruses. Using a technique called electron microscopy, Mercenne et al. observed that when Angiomotins are present, Gag proteins assemble in spheres at the cell membrane and viruses are able to exit the cell. However, when Angiomotins are depleted or absent, incomplete spheres of Gag proteins accumulate on the inner membrane surface and viruses are not released.
These findings show that Angiomotins act as a link between Gag and NEDD4L to promote the release of HIV-1 from human cells. The next step will be to learn precisely how this works. There are indications that the Angiomotins may also be involved in the release of other enveloped viruses, so the findings may be useful for the development of treatments for a variety of viral infections.
DOI:http://dx.doi.org/10.7554/eLife.03778.002
Author keywordsHIV Gag proteinNEDD4Lmotin protein familyESCRT pathwayvirus-host interactionsResearch organismhumanviruseshttp://dx.doi.org/10.13039/100000060National Institute of Allergy and Infectious Diseases (NIAID)AI051174SundquistWesley Ihttp://dx.doi.org/10.13039/100000057National Institute of General Medical Sciences (NIGMS)GM082545SundquistWesley Ihttp://dx.doi.org/10.13039/501100001691Japan Society for the Promotion of Science (JSPS)AriiJunThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementAngiomotin (AMOT) is a cellular host factor that links HIV-1 to the ubiquitin E3 ligase, NEDD4L, and promotes virion assembly and envelopment.Introduction
As obligate parasites, viruses must harness host cellular pathways to complete many different steps in viral replication. For example, a large number of enveloped viruses, including HIV-1, usurp the cellular ESCRT (Endosomal Sorting Complexes Required for Transport) pathway to bud from cells (for recent reviews, see Meng and Lever, 2013; Votteler and Sundquist, 2013; Weissenhorn et al., 2013). Viral structural proteins typically use one or more short peptide motifs, termed ‘late assembly domains’, to bind and recruit early-acting ESCRT factors or ESCRT-associated E3 ubiquitin ligases to sites of viral assembly and budding. The three best characterized viral late assembly domains are: (1) the ‘PTAP’ motif (Pro-Thr/Ser-Ala-Pro), which binds the UEV domain of the TSG101 subunit of the ESCRT-I complex, (2) the ‘YPXnL’ motif (Tyr-Pro-Xn-Leu, where X can vary in identity and sequence length), which binds the V domain of the ESCRT factor ALIX, and (3) the ‘PPXY’ motif (Pro-Pro-X-Tyr), which binds the WW domains of NEDD4 family E3 ubiquitin ligases. All three of these viral late assembly domains mimic cellular protein–protein interactions that facilitate ESCRT-dependent vesiculation processes such as multi-vesicular body (MVB) biogenesis and ectosome formation.
In addition to the three well-characterized late assembly domains described above, there are indications that enveloped viruses can also connect to the ESCRT pathway via alternative interactions that are not yet well-understood. For example, efficient release of the model paramyxovirus, PIV5, requires a non-canonical ‘FPIV’ late assembly domain motif, ubiquitylation of the matrix (M) protein, and host Angiomotin-like protein 1 (AMOTL1) (Schmitt et al., 2005; Pei et al., 2010; Harrison et al., 2012). There is no evidence that AMOTL1 can bind the FPIV late assembly domain, however, and any relationships between ubiquitin, AMOTL1 and the FPIV late assembly domain activity are not yet understood (Pei et al., 2010).
Owing to its medical importance, HIV-1 has become a leading model for understanding ESCRT pathway recruitment and viral budding. The C-terminal p6 region of the viral Gag structural protein contains canonical PTAP and YPXnL motifs that serve as the major functional late assembly domains in most cell types. Nevertheless, the release of ‘crippled’ HIV-1 Gag constructs that lack both of these late domains can still be stimulated by overexpression of the HECT E3 ubiquitin ligase, NEDD4L (also known as NEDD4-2) (Chung et al., 2008; Usami et al., 2008). Humans express nine different NEDD4 protein family members (reviewed in Ingham et al., 2004; Bernassola et al., 2008), but NEDD4L stimulates virus budding more potently than any other NEDD4 family member, implying that HIV-1 Gag preferentially engages this E3 ubiquitin ligase (Chung et al., 2008; Usami et al., 2008).
Previous studies have defined many of the requirements for NEDD4L-stimulated HIV-1 release. NEDD4L has many different isoforms (Chen et al., 2001; Dunn et al., 2002; Itani et al., 2005), but isoform 2 (also called NEDD4-2s, and here termed NEDD4L) potently stimulates the budding of crippled HIV-1 constructs, and uniquely rescues the Gag processing defects that accompany defective budding (Chung et al., 2008; Usami et al., 2008). This NEDD4L isoform contains only the final 31 residues of the N-terminal C2 membrane binding domain, a central linker region that contains four WW domains, and a C-terminal HECT E3 ubiquitin ligase domain that forms Lys-63-linked poly-ubiquitin chains (see Figure 1A) (Kim and Huibregtse, 2009; Weiss et al., 2010). Stimulation of HIV-1 budding requires NEDD4L E3 ubiquitin ligase activity, indicating that Lys-63-linked poly-ubiquitin chains perform an essential function (Chung et al., 2008; Usami et al., 2008). The functional target(s) of Lys-63 poly-ubiquitylation has not been established definitively, but HIV-1 Gag itself is the leading candidate because functional virus budding correlates with Lys-63-linked poly-ubiquitylation of Gag, and because NEDD4L truncation constructs that include the HECT domain can be functionally rescued by fusion to the HIV-1 Gag-targeting cyclophilin A domain (Weiss et al., 2010; Sette et al., 2013). HIV-1 Gag lacks identifiable PPXY motifs, however, and there is no evidence that Gag can bind NEDD4L directly.10.7554/eLife.03778.003
AMOT p130 is a NEDD4L binding partner.
(A) Schematic illustrations of the domain structures and motifs in NEDD4L, AMOT p130 and p80, AMOTL1, AMOTL2 and HIV-1 Gag. Here and throughout, NEDD4L refers to a naturally occurring protein isoform (isoform 2) that contains only the C-terminal 32 residues of the C2 domain (Genebank accession number AAM46201.1, denoted NEDD4Ls in other publications and NEDD4LΔC2 in our previous publication (Chung et al., 2008)). To maintain consistency with that publication, the NEDD4L numbering scheme used here corresponds to NEDD4L isoform 1, which is 121 residues longer at the N-terminus and contains the entire C2 domain (denoted NEDD4LWT in reference Chung et al., 2008). (B) Affinity co-purification and identification of AMOT p130 as a binding partner of OSF (One-STrEP-FLAG)-tagged NEDD4L. SDS-PAGE/Coomassie-stained gel showing STrEP-Tactin matrix affinity purified proteins from 293T cells transfected with an OSF-NEDD4L expression construct (lane 2) or an empty vector control (lane 1). Labels denote OSF-NEDD4L (bait) and AMOT p130 (prey). Trypsin-digested AMOT p130 peptides identified by mass spectrometric analyses are summarized in Figure 1—figure supplement 1. A representative image from three independent repetitions is shown. (C) NEDD4L WW domains are required for AMOT p130 binding. Western blots showing co-immunoprecipitations of endogenous 293T cell AMOT proteins (prey) with OSF-NEDD4L (bait). Panel 1: endogenous AMOT proteins (anti-AMOT blot) co-immunoprecipitated with OSF-NEDD4L baits from cells that lacked exogenous OSF-NEDD4L (lane 1, control), expressed wild type OSF-NEDD4L (lane 2), expressed OSF-NEDD4LΔWW (lane 3, construct has inactivating point mutations in all four WW domains), expressed OSF-NEDD4LC942A (lane 4, construct has an inactivating point mutation in the HECT E3 domain), or expressed OSF-NEDD4LC942A/ΔWW (lane 5, construct has inactivating point mutations in the WW and HECT E3 domains). Panel 2: same co-immunoprecipitation experiment blotted with anti-FLAG antibodies to detect OSF-NEDD4L proteins. Panel 3: input levels of endogenous AMOT (anti-AMOT blot, 10% of total). Panel 4: input levels of exogenous OSF-NEDD4L (anti-FLAG, 10% of total). Note that endogenous AMOT p80 was present in the input lysate, but did not co-immunoprecipitate with OSF-NEDD4L. Representative image from three independent repetitions.
DOI:http://dx.doi.org/10.7554/eLife.03778.003
10.7554/eLife.03778.004
Identification of AMOT p130 as a NEDD4L binding partner.
The protein band denoted ‘AMOT p130’ in Figure 1B, lane 2 was excised, digested with trypsin and the eluted peptides were identified by mass spectrometry as described in ‘Materials and methods’. This analysis was performed three times and the identified peptides are mapped onto the AMOT p130 primary sequence, with peptides identified in experiments 1–3 color coded in black, blue and red, respectively.
DOI:http://dx.doi.org/10.7554/eLife.03778.004
10.7554/eLife.03778.005
The AMOT p130 PPXY motifs contribute to NEDD4L binding.
Western blots show co-immunoprecipitations of OSF-NEDD4L (bait, captured on STrEP-Tactin affinity matrices) with endogenous 293T cells AMOT and exogenously expressed HA-AMOT p130 proteins (prey). Co-immunoprecipitations with OSF-NEDD4L were from cells that lacked exogenous AMOT p130 (lane 1), expressed wild type HA-AMOT p130 (lane 2) or expressed a mutant HA-AMOT p130 protein carrying PP/AA mutations in the three N-terminal ‘PPXY’ motifs (Wang et al., 2012; Yi et al., 2013). Panel 1: AMOT protein co-immunoprecipitations (anti-AMOT blot) with OSF-NEDD4L. Note that OSF-NEDD4L co-immunoprecipitated low levels of endogenous AMOT p130 (lanes 1 and 3, denoted p130) but not endogenous AMOT p80 (denoted p80 with a dashed line). Panel 2: The same co-immunoprecipitation experiment blotted with anti-HA antibodies to detect exogenously expressed HA-AMOT p130 proteins. Panel 3: The same co-immunoprecipitation experiment blotted with anti-FLAG antibodies to detect exogenously expressed OSF-NEDD4L. Panels 4–6 show the input quantities (10%) of endogenous AMOT and HA-AMOT p130 (panel 4, anti-AMOT), exogenous HA-AMOT p130 (panel 5, anti-HA), and exogenous OSF-NEDD4L (panel 6, anti-FLAG). A representative image from three independent repetitions is shown.
DOI:http://dx.doi.org/10.7554/eLife.03778.005
These observations imply that NEDD4L may be recruited to sites of HIV-1 assembly through an alternative mechanism. We undertook the present study with the goal of identifying co-factors that might help link NEDD4L to HIV-1 Gag and stimulate virus budding. We identified Angiomotin (AMOT) as a protein that binds both NEDD4L and HIV-1 Gag, and functions in HIV-1 assembly and release. AMOT is required to complete virion assembly and envelopment, and is therefore a new host factor that acts in a previously uncharacterized step of HIV-1 morphogenesis.
ResultsNEDD4L binds Angiomotin
Affinity purification/mass spectrometry experiments were performed to identify candidate NEDD4L binding partners. A number of cellular proteins co-purified with OSF (One-STrEP-FLAG)-tagged NEDD4L, as visualized by SDS-PAGE (Figure 1B, compare lanes 1 and 2). Prominent co-purifying proteins were excised from the gel, digested with trypsin, and identified by mass spectrometry. The p130 isoform of human Angiomotin was unambiguously identified as a protein that co-purified with OSF-NEDD4L in three independent repetitions of this experiment, (lane 2, labeled AMOT p130, peptide coverage data are summarized in Figure 1—figure supplement 1). Smaller AMOT p130 fragments were also identified, as was the related AMOT family protein, Angiomotin-like protein 1 (AMOTL1, data not shown).
Cells express two AMOT isoforms, a longer p130 isoform, and a shorter C-terminal p80 isoform that is expressed from an alternatively spliced message (Figure 1A) (Moreau et al., 2005). AMOT p130 binds NEDD4 protein family members via interactions between the four central NEDD4 WW domains and three PPXY motifs located within the N-terminal extension that is unique to the AMOT p130 isoform (Figure 1A) (Wang et al., 2012). In good agreement with this report and with our affinity purification/mass spectrometry results, we found that AMOT p130 co-immunoprecipitated with exogenously expressed OSF-NEDD4L (Figure 1C, panel 1, lane 2). In contrast, AMOT p130 did not co-immunoprecipitate with a mutant OSF-NEDD4L protein that contained inactivating Trp to Ala substitutions in the key PPXY-binding ‘W39’ residues from each of the four WW domains (denoted OSF-NEDD4LΔWW, panel 1, compare lanes 2 and 3) (Otte et al., 2003).
A previous study has demonstrated that NEDD4 proteins can ubiquitylate and reduce AMOT p130 levels (Wang et al., 2012), and we also observed that OSF-NEDD4L overexpression modestly reduced AMOT p130 levels (Figure 1C, panel 3, compare lanes 1 and 2). We therefore tested whether AMOT p130 also co-precipitated with OSF-NEDD4L proteins that carried inactivating Cys942Ala point mutations in the active site cysteine of the HECT E3 ubiquitin ligase domain (OSF-NEDD4LC942A). This mutation increased OSF-NEDD4L levels in the lysate and immunoprecipitate (panels 2 and 4 compare lanes 4 and 5 to lanes 2 and 3) and restored normal AMOT protein levels (panel 3, compare lanes 4, 5 and 1 to lane 2). Once again, AMOT p130 (and breakdown products) co-immunoprecipitated with OSF-NEDD4LC942A (panel 1, lane 4), but not with a protein that also contained inactivating WW domain mutations (OSF-NEDD4LΔWW/C942A panel 1, lane 5). AMOT p130 co-precipitation levels were higher with OSF-NEDD4LC942A than with wild type OSF-NEDD4L, presumably owing to the higher cellular levels of both OSF-NEDD4L and AMOT p130 (compare lanes 2 and 4).
Three additional observations confirmed that the NEDD4L WW domains and AMOT p130 PPXY motifs are critical for the NEDD4L-AMOT p130 interaction. Firstly, mutating the three N-terminal AMOT p130 PPXY motifs strongly inhibited co-immunoprecipitation of HA-AMOT p130 with OSF-NEDD4L (Figure 1—figure supplement 2, panel 1, compare lanes 2 and 3). Secondly, endogenous AMOT p80, which lacks the N-terminal PPXY motifs, failed to co-immunoprecipitate with either OSF-NEDD4L or OSF-NEDD4LC942A (Figure 1C, panel 1, lanes 2 and 4), even though AMOT p80 was present in the 293T cell extracts (panel 3, lanes 1–5). Thirdly, we observed that exogenously expressed, epitope-tagged NEDD4L and AMOT p130 co-localized well in HeLa-M cells, and this co-localization required both the AMOT PPXY and NEDD4L WW motifs (data not shown), in good agreement with previous reports that motins co-localize with other NEDD4 family members (Skouloudaki and Walz, 2012; Wang et al., 2012). Taken together, these experiments demonstrate that NEDD4L interacts specifically with AMOT p130 in cells and that the interaction requires the WW domains of NEDD4L and the PPXY motifs of AMOT.
AMOT binds directly to NEDD4L and to HIV-1 Gag
Pull-down experiments with purified recombinant proteins were performed to test whether AMOT p130 binds NEDD4L directly, and whether AMOT p130 can also bind HIV-1 Gag and thereby link NEDD4L to the viral Gag protein. Wild type and mutant NEDD4L proteins and an HIV-1 Gag protein that lacked the flexible p6 domain (GagΔp6) were expressed in Escherichia coli and purified to near homogeneity (Figure 2A, lanes 1 and 2, and Figure 2B, lane 1, respectively). OSF-AMOT p130 was overexpressed in SF-9 insect cells and substantially enriched by binding to STrEP-Tactin affinity resin (Figure 2A, lane 6 and Figure 2B, lane 7), although we were unable to purify the protein to homogeneity. As shown in Figure 2A, NEDD4L did not bind an immobilized control OSF-GFP protein, but bound with at least 1:1 stoichiometry to immobilized OSF-AMOT p130 (compare lanes 4 and 7). In contrast, binding of the NEDD4LΔWW mutant protein was nearly undetectable (Figure 2A, compare lanes 7 and 8, particularly in the western blot in panel 2). These experiments confirm that NEDD4L and AMOT p130 bind one another directly, with the WW domains of NEDD4L binding the PPXY motifs of AMOT p130.10.7554/eLife.03778.006
AMOT p130 binds directly and specifically to NEDD4L and HIV-1 Gag<sub>∆p6</sub>.
(A) OSF-AMOT p130 binds NEDD4L. Recombinant OSF-GFP (specificity control, lanes 3–5) or OSF-AMOT p130 (lanes 6–8) were expressed, captured on STrEP-Tactin affinity matrices, and incubated with either a buffer control (lanes 3 and 6) or buffers containing 0.5 µM wild type NEDD4L (lanes 4 and 7) or NEDD4LΔWW (lanes 5 and 8, inactivating point mutations in all four WW domains). Matrix-bound proteins were released by boiling in denaturing buffer and detected by SDS-PAGE with Coomassie blue staining (panel 1) or by western blotting (panel 2, anti-NEDD4L) to confirm the identities of the bound NEDD4L and help to distinguish background proteins from low-level binding in panel 1. Input levels of NEDD4L (lane 1, 1.5% of total) and NEDD4LΔWW (lane 2, 1.5% of total) are shown for reference. A representative image from three independent repetitions is shown. (B) OSF-AMOT p130 binds HIV-1 Gag∆p6. Recombinant OSF-GFP (control, lanes 3–6) or OSF-AMOT p130 (lanes 7–10) were expressed, captured on STrEP-Tactin affinity matrices and incubated with a buffer control (lanes 3 and 7) or with buffers containing 1.0 µM HIV-1 Gag∆p6 (lanes 3 and 8), 0.5 µM wild type NEDD4L (lanes 5 and 9) or both proteins (lanes 6 and 10). Matrix-bound proteins were released by boiling in denaturing buffer and detected by SDS-PAGE with Coomassie blue staining (panel 1) or by western blotting to confirm the identities of the bound Gag (panel 2, anti-MA and anti-CA) and NEDD4L (panel 3, anti-NEDD4L) proteins and help to distinguish background proteins from low-level binding in panel 1. Input Gag∆p6 (lane 1, 2% of total) and NEDD4L (lane 2, 1.5% of total) are shown for reference. A representative image from three independent repetitions is shown.
DOI:http://dx.doi.org/10.7554/eLife.03778.006
10.7554/eLife.03778.007
AMOT p130, AMOTL1 and AMOTL2 bind directly and specifically to NEDD4L and HIV-1 Gag<sub>∆p6</sub>.
(A) OSF-AMOT p130, OSF-AMOTL1, and OSF-AMOTL2 bind NEDD4L. Recombinant OSF-GFP (specificity control, lanes 2 and 6), OSF-AMOT p130 (lanes 3 and 7), OSF-AMOTL1 (lanes 4 and 8), and OSF-AMOTL2 (lanes 5 and 9) were expressed, captured on STrEP-Tactin affinity matrices, and incubated with either a buffer control (lanes 2–5) or with a buffer containing 0.5 µM wild type NEDD4L (lanes 6–9). Matrix-bound proteins were released by boiling in denaturing buffer and detected by SDS-PAGE with Coomassie blue staining. Input NEDD4L (lane 1, 1.5% of total) is shown for reference. Asterisk denotes an OSF-AMOTL2 breakdown product that co-migrates with NEDD4L. A representative image from two independent repetitions is shown. (B) OSF-AMOT p130, OSF-AMOTL1, and OSF-AMOTL2 bind Gag∆p6. Recombinant OSF-GFP (specificity control, lanes 2 and 6), OSF-AMOT p130 (lanes 3 and 7), OSF-AMOTL1 (lanes 4 and 8), and OSF-AMOTL2 (lanes 5 and 9) were expressed, captured on STrEP-Tactin affinity matrices, and incubated with either a buffer control (lanes 2–5) or buffers containing 1 µM Gag∆p6 (lanes 6–9). Matrix-bound proteins were released by boiling in denaturing buffer and detected by SDS-PAGE with Coomassie blue staining. Input Gag∆p6 (lane 1, 2% of total) is shown for reference. A representative image from two independent repetitions is shown.
DOI:http://dx.doi.org/10.7554/eLife.03778.007
Analogous pull-down experiments were performed to test whether AMOT binds HIV-1 GagΔp6. Removing the p6 region facilitates Gag protein expression and purification, and is not expected to interfere with any biologically relevant interactions because NEDD4L potently stimulates HIV-1 GagΔp6 release from cells (Chung et al., 2008; Usami et al., 2008). As shown in Figure 2B, HIV-1 GagΔp6 bound immobilized OSF-AMOT p130, but did not bind an immobilized OSF-GFP control (compare lanes 4 and 8). This interaction did not appear to be mediated by nucleic acids because the affinity-purified OSF-AMOT p130 protein was pre-washed with high salt solution to remove any bound nucleic acid (see ‘Materials and methods’), and because the interaction was maintained, albeit at slightly lower levels, even when immobilized OSF-AMOT p130 was incubated with high concentrations of RNase that were sufficient to eliminate other RNA-mediated Gag-protein interactions (not shown). GagΔp6 and NEDD4L appeared to bind OSF-AMOT p130 independently because their binding levels did not change significantly when all three proteins were incubated together (compare lanes 8 and 9 to lane 10).
AMOT is the founding member of the human motin family, which also includes Angiomotin-like protein 1 (AMOTL1) and Angiomotin-like protein 2 (AMOTL2) (Moleirinho et al., 2014). AMOTL1 and AMOTL2 share 52% and 45% pair-wise sequence identity with AMOT p130, and both contain the N-terminal extension and PPXY motifs that are present in AMOT p130 but absent in AMOT p80 (Figure 1A). Pull-down experiments demonstrated that AMOTL1 and AMOTL2 also bind both NEDD4L and HIV-1 GagΔp6 (Figure 2—figure supplement 1). Thus, AMOT p130 binds directly and specifically to both NEDD4L and HIV-1 Gag, and these binding activities are shared by the other two human motins.
AMOT p130 overexpression enhances NEDD4L stimulation of HIV-1 budding
Co-expression experiments were performed to test whether AMOT and NEDD4L can cooperate in stimulating HIV-1 release (Figure 3). These experiments employed the crippled HIV-1ΔPTAP, ΔYP viral expression construct, which lacks functional PTAP and YPXnL late domains and is therefore particularly responsive to cellular NEDD4L activity (Chung et al., 2008; Usami et al., 2008). As expected, HIV-1ΔPTAP, ΔYP was poorly released from control 293T cells, as measured by western blot that detected virion-associated CA and MA proteins (Figure 3, upper right panel, lane 1), and low viral titers (lower right panel, lane 1). Overexpression of HA-AMOT p130 increased the levels of the full-length protein (and presumptive breakdown products were also evident, left panel 1, compare lanes 1 and 2). AMOT p130 overexpression modestly increased HIV-1ΔPTAP.ΔYP release and infectivity (right panels compare lanes 1 and 2, 1.7-fold increases in both release and infectivity), without significantly altering the intracellular levels of Gag (left, panel 4, compare lanes 1 and 2) or a control cellular protein (GAPDH, left panel 3, compare lanes 1 and 2). Thus, AMOT p130 overexpression stimulates HIV-1ΔPTAP, ΔYP release and infectivity, but the effect is modest.10.7554/eLife.03778.008
AMOT p130 stimulates NEDD4L-dependent release of HIV-1<sub>ΔPTAP, ΔYP</sub>.
Left panels are western blots showing 293T cellular levels of endogenous AMOT and exogenous HA-AMOT p130 proteins (panel 1, anti-AMOT), exogenous FLAG-NEDD4L (panel 2, anti-FLAG), endogenous GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 GagΔPTAP, ΔYP proteins (panel 4, anti-MA and anti-CA). Cells were co-transfected with expression vectors for HIV-1ΔPTAP, ΔYP (lanes 1–6), FLAG-NEDD4L proteins (lanes 3–6, with wild type (WT) FLAG-NEDD4L in lanes 3 and 4 and FLAG-NEDD4LΔWW, in lanes 5 and 6), and wild type HA-AMOT p130 (lanes 2, 4 and 6) or an empty vector control (lanes 1, 3, and 5). Right panels show corresponding levels of extracellular, virion-associated CAGag and MAGag proteins (panel 1, anti-MA and anti-CA) and viral titers (panel 2, IU denotes ‘infectious units’). Here and in subsequent figures: (1) error bars denote standard deviations between independent repetitions of the experiment, n = 5 in this case, and (2) numbers within the blots show integrated intensities of the CA band intensities (relative to the value in the control experiment, set to 1.0). Here and throughout significance is denoted by: NS, not significant (p > 0.05); *0.05 > p > 0.01; **p < 0.01.
DOI:http://dx.doi.org/10.7554/eLife.03778.008
10.7554/eLife.03778.009
Dose-dependent AMOT p130 stimulation of NEDD4L-dependent release of HIV-1<sub>ΔPTAP, ΔYP</sub>.
Left panels are western blots showing 293T cellular levels of endogenous AMOT and exogenous HA-AMOT p130 proteins (panel 1, anti-AMOT), exogenous FLAG-NEDD4L (panel 2, anti-FLAG), GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 GagΔPTAP, ΔYP proteins (panel 4, anti-MA and anti-CA). Cells were co-transfected with expression vectors for HIV-1ΔPTAP, ΔYP (lanes 1–8), FLAG-NEDD4L (lanes 2–8), and increasing quantities of an AMOT p130 expression construct (lanes 3–8, 0.25–1.5 μg DNA). Cells were also co-transfected with empty vectors as necessary to keep total DNA levels constant (3 μg DNA total). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, anti-MA and anti-CA) and viral titers (lower panel), n = 3.
DOI:http://dx.doi.org/10.7554/eLife.03778.009
10.7554/eLife.03778.010
The AMOT p130 isoform specifically stimulates NEDD4L-dependent release of HIV-1<sub>ΔPTAP, ΔYP</sub>.
Left panels are western blots showing 293T cellular levels of endogenous AMOT, exogenous p80 and exogenous HA-AMOT p130 proteins (panel 1, anti-AMOT), exogenous FLAG-NEDD4L (panel 2, anti-FLAG), GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 GagΔPTAP, ΔYP proteins (panel 4, anti-MA and anti-CA). Cells were co-transfected with expression vectors for HIV-1ΔPTAP, ΔYP (lanes 1–6), AMOT p80 (lanes 2 and 5), AMOT p130 (lanes 3 and 6) and FLAG-NEDD4L (lanes 4–6). Cells were also co-transfected with empty vectors as necessary to keep total DNA levels constant (2.5 μg DNA total). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, anti-MA and anti-CA) and viral titers (lower panel), n = 4.
DOI:http://dx.doi.org/10.7554/eLife.03778.010
As seen previously, FLAG-NEDD4L overexpression also stimulated HIV-1ΔPTAP, ΔYP release (Figure 3, right panel 1, compare lanes 1 and 3, 3.2-fold stimulation) and infectivity (right panel 2, compare lanes 1 and 3, 5.3-fold stimulation), and also largely rescued the HIV-1ΔPTAP, ΔYP Gag processing defects (left panel 4, compare lane 3 to lanes 1 and 2, and note the reduction in p41 and p25 Gag processing intermediates) (Chung et al., 2008; Usami et al., 2008). Co-overexpression of both HA-AMOT p130 and FLAG-NEDD4L further increased HIV-1ΔPTAP, ΔYP release (right panel 1, lane 4, sixfold increase over basal levels) and infectivity (right panel 2, lane 4, 14-fold increase over basal levels). As shown in Figure 3—figure supplement 1, the degree of hyper-stimulation correlated positively with AMOT p130 expression levels, until an inhibitory level was reached. In contrast to wild type NEDD4L, the NEDD4LΔWW mutant stimulated HIV-1ΔPTAP, ΔYP release and infectivity only very modestly (Figure 3, right panels, compare lanes 5 and 3) and did not synergize with AMOT p130 (right panels, compare lanes 6 and 5). Similarly, the shorter AMOT p80 isoform, which lacks the PPXY motifs found in the N-terminal domain of AMOT p130, failed to synergize with NEDD4L in stimulating virus release and infectivity, and instead modestly reduced NEDD4L stimulation, probably by dominantly inhibiting endogenous AMOT p130 activity (Figure 3—figure supplement 2, compare lanes 4–6). These data indicate that AMOT p130 and NEDD4L cooperate in stimulating HIV-1ΔPTAP, ΔYP release and infectivity.
AMOT p130 is required for NEDD4L stimulation of HIV-1 budding
siRNA depletion experiments were performed to determine whether AMOT was also necessary for NEDD4L stimulation of HIV-1ΔPTAP, ΔYP release and infectivity. We used an siRNA that targeted both AMOT isoforms and reduced cellular AMOT p130 and p80 levels by at least fivefold (Figure 4, left panel 1, compare lanes 1 and 2 and lanes 3 and 4). AMOT depletion reduced the already low levels of virus release and infectivity even further, to near background levels (Figure 4, right panels, compare lanes 1 and 2, 7.5-fold infectivity reduction). Cellular levels of Gag and a GAPDH control were unaffected by this treatment (left panels 3 and 4, compare lanes 1 and 2), although AMOT depletion reduced cellular protein levels slightly in some repetitions (not shown).10.7554/eLife.03778.011
AMOT p130 is required for NEDD4L-stimulated release of HIV-1<sub>ΔPTAP, ΔYP</sub>.
Left panels are western blots showing 293T cellular levels of endogenous AMOT and exogenous HA-AMOT p130 proteins (panel 1, anti-AMOT), exogenous FLAG-NEDD4L (panel 2, anti-FLAG), endogenous GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 GagΔPTAP, ΔYP proteins (panel 4, anti-MA and anti-CA). Cells were co-transfected with a control siRNA (lanes 1 and 3) or with an siRNA that depleted both AMOT p130 and p80 (lanes 2 and 4–6), with expression vectors for HIV-1ΔPTAP, ΔYP (lanes 1–6), and with FLAG-NEDD4L (lanes 3–6) or an empty vector control (lanes 1 and 2), and with siRNA-resistant expression constructs for HA-AMOT p130 proteins (with wild type HA-AMOT p130 in lane 5 and an HA-AMOT p130ΔPPXY protein carrying inactivating point mutations in the three PPXY motifs in lane 6). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel), n = 6.
DOI:http://dx.doi.org/10.7554/eLife.03778.011
10.7554/eLife.03778.012
Dose-dependent AMOT p130 rescue of HIV-1 release from cells depleted of endogenous AMOT.
Left panels are western blots showing 293T cellular levels of endogenous AMOT and exogenous HA-AMOT p130 proteins (panel 1, anti-AMOT), exogenous FLAG-NEDD4L (panel 2, anti-FLAG), GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 Gag proteins (panel 4, anti-MA and anti-CA). Cells were co-transfected with a control siRNA (lanes 1 and 3) or with an siRNA that depleted both AMOT p130 and p80 (lanes 2 and 4–10), with expression vectors for HIV-1ΔPTAP, ΔYP (lanes 1–10), for FLAG-NEDD4L (lanes 3–10), and with increasing quantities of an siRNA-resistant AMOT p130 construct (lanes 5–10, 0.25–1.5 μg DNA). Cells were also co-transfected with empty vectors as necessary to keep total DNA levels constant (3 μg DNA total). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel), n = 3.
DOI:http://dx.doi.org/10.7554/eLife.03778.012
10.7554/eLife.03778.013
The AMOT p130 isoform specifically stimulates NEDD4L-dependent release of HIV-1<sub>ΔPTAP, ΔYP</sub>.
Left panels are western blots showing 293T cellular levels of endogenous AMOT and exogenous HA-AMOT p130 or AMOT p80 proteins (panel 1, anti-AMOT), exogenous FLAG-NEDD4L (panel 2, anti-FLAG), endogenous GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 GagΔPTAP, ΔYP proteins (panel 4, anti-MA and anti-CA). Cells were transfected with a control siRNA (lanes 1 and 3) or with an siRNA that depleted both AMOT p130 and p80 (lanes 2 and 4–7). Cells were co-transfected with expression vectors for HIV-1ΔPTAP, ΔYP (lanes 1–7), FLAG-NEDD4L (lanes 3–7), and siRNA-resistant expression constructs for AMOT p80 (lane 5), HA-AMOT p130 (lane 6) or both (lane 7). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel), n = 6.
DOI:http://dx.doi.org/10.7554/eLife.03778.013
AMOT depletion blocked the ability of NEDD4L to stimulate HIV-1ΔPTAP, ΔYP release and infectivity (Figure 4 and Figure 4—figure supplements 1, 2). In the control case, NEDD4L overexpression stimulated HIV-1ΔPTAP, ΔYP release (Figure 4, right panels, compare lanes 1 and 3, 4.1-fold infectivity increase). However, virus release was reduced to the levels seen in untreated control cells when endogenous AMOT p80 and p130 proteins were simultaneously depleted (compare lanes 2–4). Similar results were observed for a second siRNA that targeted both AMOT isoforms (not shown). Thus, AMOT is required for NEDD4L stimulation of HIV-1ΔPTAP, ΔYP release.
Rescue experiments with exogenous siRNA-resistant AMOT expression constructs were performed to confirm the specificity of the AMOT depletion and to test the requirements for AMOT function. Re-expression of AMOT p130 completely rescued the block to virus release and infectivity imposed by AMOT depletion (Figure 4, right panels, compare lanes 5 and 4), restoring viral infectivity to a level that was actually slightly higher than the control (compare lanes 3 and 5). The degree of rescue generally correlated positively with AMOT p130 re-expression levels, and peaked at levels that approximated those of the native endogenous protein (Figure 4—figure supplement 1, lane 8). In contrast, an AMOT p130 protein that lacked the three N-terminal PPXY motifs failed to rescue NEDD4L-dependent HIV-1ΔPTAP, ΔYP release (Figure 4, AMOT p130ΔPPXY, right panels, compare lanes 6 and 5). Similarly, re-expression of an siRNA-resistant AMOT p80 construct also failed to rescue virus release and infectivity (Figure 4—figure supplement 2, right panels, compare lanes 4 and 5), unless exogenous AMOT p130 was present (Figure 4—figure supplement 2, right panels, lane 7). Hence, NEDD4L stimulation of HIV-1ΔPTAP, ΔYP release and infectivity requires the presence of an AMOT p130 protein that can bind NEDD4L, but is not significantly affected by AMOT proteins that cannot bind NEDD4L, including AMOT p80.
AMOT p130 is required for efficient release of wild type HIV-1
siRNA depletion and rescue experiments were also performed to test whether AMOT p130 is required for release and infectivity of wild type HIV-1. In control experiments, depletion of either of the well-characterized HIV-1 budding factors, ALIX and TSG101, reduced virus release and infectivity, although the effects were much greater for TSG101 depletion, as reported previously (Figure 5, right panels, compare lanes 2 and 3 to lane 1) (Fujii et al., 2009). AMOT depletion also significantly reduced both HIV-1 release and infectivity (Figure 5, right panels, compare lane 4 to lane 1, eightfold reduction in infectivity and 2.5-fold reduction in release). The magnitudes of these effects were intermediate between those seen for depletion of TSG101 (20-fold reduction in infectivity and threefold reduction in virion release) and ALIX (1.5-fold reduction in infectivity and insignificant reduction in virion release). Importantly, the detrimental effects of AMOT p130 depletion were almost completely rescued by re-expression of wild type AMOT p130 (compare lanes 1, 4 and 5) but not by the mutant AMOT p130ΔPPXY protein (lane 6). These experiments demonstrate that AMOT p130 is required for efficient release of wild type HIV-1 from 293T cells and mutations that abolish NEDD4L binding block the ability of AMOT p130 to function in virus release.10.7554/eLife.03778.014
AMOT p130 is required for efficient release of wild type HIV-1.
Left panels are western blots showing 293T cellular levels of endogenous ALIX (panel 1, anti-ALIX), TSG101 (panel 2, anti-TSG101), AMOT (panel 3, anti-AMOT), GAPDH (panel 4, anti-GAPDH, loading control) and levels of HIV-1 Gag proteins (panel 5, anti-MA and anti-CA). Cells were co-transfected with a control siRNA (lane 1) or with an siRNA that depleted ALIX (lane 2), TSG101 (lane 3) or AMOT p130 and p80 (lanes 4–6), with expression vectors for wild type HIV-1 (lanes 1–6) and with an siRNA-resistant expression constructs for HA-AMOT p130 proteins (with wild type HA-AMOT p130 in lane 5 and an HA-AMOT p130ΔPPXY protein with inactivating point mutations in all three PPXY motifs in lane 6). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel), n = 4.
DOI:http://dx.doi.org/10.7554/eLife.03778.014
AMOT stimulates HIV-1 release from HeLa cells
AMOT mRNA is expressed in 293T cells, as well as in white blood cells and lymphoid tissues that are the natural hosts for HIV-1 infection (Moleirinho et al., 2014). In contrast, HeLa cells reportedly express little or no AMOT mRNA, although they do express AMOTL2 mRNA (Moleirinho et al., 2014). We confirmed that endogenous AMOT protein levels are at least 10-fold lower in HeLa cells than in 293T cells (Figure 6—figure supplement 1). This observation begs the question of how HIV-1 can be released from a cell type that lacks AMOT, and this issue is particularly relevant because HeLa cells are commonly used to study HIV-1 assembly. We therefore tested the effects of AMOT p130 expression on HIV-1 release and infectivity in HeLa cells. As shown in Figure 6A, AMOT p130 expression increased HIV-1 release from HeLa cells in a dose-dependent fashion, up to 15-fold at the highest levels of AMOT tested (right panel 1, compare lanes 1 and 5). Viral titers similarly increased with AMOT expression in a dose-dependent fashion, although the overall effects were less dramatic (right panel 2, compare lanes 1–5, threefold infectivity increase). Thus, HIV-1 release from wild type HeLa cells is suboptimal, and can be enhanced by expression of AMOT p130.10.7554/eLife.03778.015
AMOT p130 stimulates release of HIV-1 from HeLa and Jurkat T cells.
(A) Left panels are western blots showing HeLa cellular levels of endogenous AMOT p130 and exogenous HA-AMOT p130 (panel 1, anti-AMOT), endogenous GAPDH (panel 2, anti-GAPDH, loading control) and HIV-1 Gag proteins (panel 3, anti-MA and anti-CA). HeLa cells were co-transfected with expression vectors for HIV-1 (lanes 1–5), and with increasing concentrations of an expression construct for HA-AMOT p130 (0–4 μg DNA in lanes 1–5). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel), n = 4. (B) Levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel). Jurkat T cells were co-transfected with expression vectors for HIV-1 (lanes 1–5), together with a control vector (lane 1) or expression vectors for wild type HA-AMOT p130 (lanes 2 and 3, 5 and 10 µg DNA respectively) or an HA-AMOT p130ΔPPXY protein with inactivating point mutations in the three PPXY motifs (lanes 4 and 5, 5 and 10 µg DNA respectively), n = 3.
DOI:http://dx.doi.org/10.7554/eLife.03778.015
10.7554/eLife.03778.016
HeLa cells express little or no AMOT.
Western blots showing endogenous AMOT p130 and p80 expression levels in 293T cells (lanes 1–4) and in HeLa cells (lanes 5–8). Increasing numbers of cell equivalents were added as indicated (ranging from 1.5 × 105 cells in lanes 1 and 5 to 1.2 × 106 cells in lanes 4 and 8). Upper and lower panels show lighter and darker exposures of the same blot. Image is representative of two independent repetitions of this experiment.
DOI:http://dx.doi.org/10.7554/eLife.03778.016
AMOT also stimulates HIV-1 release from Jurkat T cells
T cells are natural hosts for HIV-1 infection, and AMOT mRNA is present in these cells, including Jurkat T cells (Yeung et al., 2009; Sheynkman et al., 2013). To determine whether AMOT p130 can function in virus release from Jurkat T cells, we tested whether expressing exogenous AMOT p130 expression stimulated HIV-1 release and infectivity. As shown in Figure 6B, AMOT p130 expression increased HIV-1 release from Jurkat T cells in a dose-dependent fashion, with up to 16-fold stimulation observed at the highest levels of AMOT p130 tested (panel 1, compare lanes 1–3). Viral titers similarly increased in a dose-dependent fashion (panel 2, compare lanes 1–3, sevenfold infectivity increase). Western blots confirmed that virus expression levels were the same in all cases (not shown), but AMOT p130 expression levels were below our detection limits in this cell type. We therefore also tested the effects of overexpressing the mutant AMOT p130ΔPPXY protein. Unlike wild type AMOT p130, AMOT p130ΔPPXY failed to stimulate virus release, and actually reproducibly decreased viral titers as compared to the untreated control (compare lanes 4 and 5 to lane 1). We speculate that this effect again reflects dominant inhibition of endogenous AMOT p130. Regardless, our data indicate that AMOT p130 also stimulates virus release and infectivity from T cells, which are natural targets of HIV-1 infection. Consistent with this conclusion, a global shRNA screen for host cofactors revealed that Jurkat T cells transduced with an shRNA that targeted AMOT were protected against HIV cytotoxicity (Yeung et al., 2009). This observation implies that AMOT contributes to HIV-1 replication in cultured Jurkat T cells, although the role of AMOT in HIV-1 replication was not characterized further in that study. (Yeung et al., 2009).
Other human motins can also stimulate HIV-1 release
siRNA depletion/rescue experiments were performed to test whether other motin family members can substitute for AMOT p130 in supporting release of wild type HIV-1 (Figure 7). In control experiments, HIV-1 release and infectivity were again decreased by depletion of endogenous AMOT from 293T cells (right panels, compare lanes 1 and 2, fivefold infectivity reduction). As expected, these effects were largely rescued by expression of exogenous siRNA-resistant AMOT p130 (right panels, compare lanes 2 and 3). Expression of either AMOTL1 or AMOTL2 also significantly rescued HIV-1 release and infectivity (right panels, compare lanes 4 and 5 to lanes 2 and 3). Hence, both AMOT-like proteins can facilitate HIV-1 release in the absence of endogenous AMOT p130. Neither AMOTL1 nor AMOTL2 was quite as effective as AMOT p130 in rescuing HIV-1 infectivity in this assay, possibly because these proteins are intrinsically less active than AMOT p130 and/or because they were expressed at lower levels (left panel 2, compare lanes 3–5).10.7554/eLife.03778.017
AMOTL1 and AMOTL2 can substitute for AMOT p130 in HIV-1 release.
Left panels are western blots showing 293T cellular levels of endogenous AMOT and exogenous HA-AMOT p130 proteins (panel 1, anti-AMOT), exogenous HA-AMOT p130, HA-AMOTL1 or HA-AMOTL2 proteins (panel 2, anti-HA), endogenous GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 Gag proteins (panel 4, anti-MA and anti-CA). Cells were co-transfected with a control siRNA (lane 1) or with an siRNA that depleted AMOT p130 and AMOT p80 (lanes 2–5), and with expression vectors for HIV-1 (lanes 1–5), with siRNA-resistant expression constructs for AMOT p130 (lane 3), AMOTL1 (lane 4) or AMOTL2 (lane 5). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel), n = 4.
DOI:http://dx.doi.org/10.7554/eLife.03778.017
10.7554/eLife.03778.018
AMOTL2 can stimulate NEDD4L-dependent release of HIV-1<sub>ΔPTAP, ΔYP</sub> and rescue HIV-1<sub>ΔPTAP, ΔYP</sub> release from cells depleted of endogenous AMOT.
(A) AMOTL2 stimulates NEDD4L-dependent release of HIV-1ΔPTAP, ΔYP. Left panels are western blots showing 293T cellular levels of exogenous HA-AMOT p130, HA-AMOTL1 or HA-AMOTL2 proteins (panel 1, anti-HA), exogenous FLAG-NEDD4L (panel 2, anti-FLAG), GAPDH (panel 3, anti-GAPDH, loading control) and HIV-1 Gag proteins (panel 4, anti-MA and anti-CA). Cells were co-transfected with expression vectors for HIV-1ΔPTAP/ΔYP (lanes 1–8), FLAG-NEDD4L (lanes 5–8), and either HA-AMOT p130 (lanes 2 and 6), HA-AMOTL1 (lanes 3 and 7) or HA-AMOTL2 (lanes 4 and 8). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel) and viral titers (lower panel), n = 3. (B) AMOTL2 can rescue HIV-1ΔPTAP, ΔYP release from 293T cells depleted of endogenous AMOT. Left panels are western blots showing cellular levels of endogenous AMOT and exogenous HA-AMOT proteins (panel 1, anti-AMOT), exogenous HA-AMOT p130, HA-AMOTL1 or HA-AMOTL2 (panel 2, anti-HA), exogenous FLAG-NEDD4L (panel 3, anti-FLAG), GAPDH (panel 4, anti-GAPDH, loading control) and HIV-1 Gag proteins (panel 5, anti-MA and anti-CA). Cells were co-transfected with a control siRNA (lanes 1 and 3) or with an siRNA that depleted both AMOT p130 and p80 (lanes 2 and 4–7), with expression vectors for HIV-1 (lanes 1–7), FLAG-NEDD4L (lanes 3–7), and either an siRNA-resistant HA-AMOT p130 expression construct (lane 5), or expression constructs for HA-AMOTL1 (lane 6), or HA-AMOTL2 (lane 7). Right panels show corresponding levels of extracellular virion-associated Gag proteins (upper panel, western blot, anti-MA and anti-CA) and viral titers (lower panel), n = 4.
DOI:http://dx.doi.org/10.7554/eLife.03778.018
We also tested whether the other AMOT family members could cooperate with NEDD4L in stimulating release of the crippled HIV-1ΔPTAP, ΔYP virus (Figure 7—figure supplement 1A) and whether AMOTL1 and AMOTL2 could rescue the block to NEDD4L stimulation of HIV-1ΔPTAP, ΔYP release imposed by depletion of endogenous AMOT (Figure 7—figure supplement 1B). AMOTL2 strongly stimulated/rescued HIV-1ΔPTAP, ΔYP release in both of these assays, albeit slightly less efficiently than AMOT p130 (Figure 7—figure supplement 1A, compare lane 8 to lanes 5 and 6 and Figure 7—figure supplement 1B, compare lane 7 to lanes 4 and 5). In contrast, AMOTL1 exhibited little or no activity in either assay (Figure 7—figure supplement 1A, compare lane 7 to lanes 5 and 6 and Figure 7—figure supplement 1B, compare lane 6 to lanes 4 and 5). Thus, AMOTL2 could substitute for AMOT p130 in three different virus release assays, whereas AMOTL1 was slightly less active than AMOT in the release of wild type HIV-1, and failed to stimulate release of the ‘crippled’ HIV-1ΔPTAP, ΔYP construct.
AMOT depletion inhibits an early stage of HIV-1 budding
Scanning and transmission electron microscopy (SEM and TEM, respectively) were used to visualize the stage at which AMOT depletion inhibited HIV-1 assembly or release (Figures 8, 9). SEM imaging was initially performed because this method can survey the entire cell surface and thereby identify global changes in virion assembly and budding profiles. SEM analyses were performed on 293T cells that expressed HIV-1 and were treated with either: (1) a control siRNA, (2) an siRNA that depleted TSG101 (positive control), or (3) an siRNA that depleted AMOT. As expected, viruses budding from control siRNA-treated cells were detectable but rare (e.g., Figure 8A–C). In contrast, large numbers of assembling virions were visible on the surfaces of cells that lacked TSG101 (Figure 8D–F) or that lacked AMOT (Figure 8G–I). We also observed that the arrested virions tended to cluster in the center of cells lacking AMOT, whereas they were more evenly distributed across the surfaces of cells lacking TSG101. The dramatic increase in the numbers of arrested budding virions implies that AMOT, like TSG101, must function following the initial stages of Gag assembly, but prior to viral membrane fission.10.7554/eLife.03778.019
AMOT p130 is required for HIV-1 assembly/budding.
Scanning electron microscopic (SEM) images of HIV virions budding from 293T cells depleted with a control siRNA (A–C), depleted of TSG101 (D–F) or depleted of AMOT (G–I). Successive images across each row show expansions of the adjacent boxed regions, and scale bars are 500 nm in all cases. Arrows in panel (C) highlight budding wild type HIV-1 virions.
DOI:http://dx.doi.org/10.7554/eLife.03778.019
10.7554/eLife.03778.020
AMOT p130 is required for an early stage of HIV-1 assembly/budding.
(A) Transmission electron microscopic (TEM) images of HIV virions budding from 293T cells depleted of TSG101. Panel 1: wide view. Panel 2: expanded view. Scale bars here and in part B are 100 nm. (B) TEM images of HIV virions budding from 293T cells depleted of AMOT p130 and p80. Panel 1: wide view. Panel 2: expanded view. (C) Quantification of the release and maturation status of HIV-1 virions associated with cells that were treated with a control siRNA (black bars), depleted of TSG101 (red), depleted of AMOT (teal), or depleted of AMOT and re-expressing AMOT p130 from an siRNA-resistant construct (light blue). This experiment was repeated twice with similar results, and error bars were derived by quantifying three separate sets of >100 virions each from one of the experiments. (D) Quantification of the completeness of virions budding from cells treated with a control siRNA (black triangles), depleted of TSG101 (red squares), or depleted of AMOT (teal diamonds). The extent of the Gag shell arc (in degrees) was measured from TEM images of 500 budding virions of each type, the measurements were binned into the intervals shown below the x-axis, and the virion numbers in each bin are shown in the plot.
DOI:http://dx.doi.org/10.7554/eLife.03778.020
TEM experiments were performed to characterize how virus assembly/budding arrested in the absence of AMOT. These experiments again visualized HIV-1 budding from 293T cells treated with a control siRNA, an siRNA that depleted TSG101 (positive control), an siRNA that depleted AMOT, or an siRNA that depleted endogenous AMOT together with an siRNA-resistant construct that re-expressed wild type AMOT p130 (rescue control). Culture supernatants were removed from the virus producing cells and the adherent cells were fixed while still attached to the culture dish in order to maximize the numbers of cell-associated virions. Fixed cells were harvested, embedded in resin, thin sectioned, stained, and visualized by TEM. Cell-associated virions observed in these experiments were scored as being either: (1) mature (discernable conical or condensed cores), (2) immature (discernable immature Gag shells), or (3) budding (contiguous cellular and viral membranes).
HIV-1 virions associated with wild type cells were usually mature (76 ± 5%, see Figure 9C), but immature and budding virions were also observed (9 ± 3% and 15 ± 5%, respectively). As expected, TSG101 depletion induced a strong virus budding defect, with the majority of detected virions arrested at the budding stage (67 ± 7%, shown in Figure 9A and quantified in Figure 9C). In addition, the ratio of mature (10 ± 6%) to immature (23 ± 2%) virions (1:2.3) was much lower than in the control case (1:0.1), consistent with previous reports that TSG101 is also required for efficient Gag processing and maturation (Göttlinger et al., 1991; Garrus et al., 2001; Martin-Serrano et al., 2003).
AMOT depletion also inhibited virus release, as reflected in a large increase in the percentage of virions detected at the budding stage (72 ± 6%, shown in Figure 9B and quantified in Figure 9C). In this case, however, the virions that were released exhibited a normal ratio of mature:immature virions (25 ± 5% vs 3 ± 2%; 1:0.1). Importantly, the budding defect seen in cells lacking endogenous AMOT was almost completely rescued by re-expression of the siRNA-resistant AMOT p130 (Figure 9C), so that the majority of virions were now mature (70 ± 5% vs 76 ± 5% in the control), and budding virion levels were nearly normal (26 ± 9% vs 15 ± 5% in the control). Thus, AMOT depletion blocks virus release at the assembly/budding stage, but does not inhibit Gag processing or virion maturation.
It was also striking that AMOT depletion caused virion assembly/budding to arrest at an earlier stage than TSG101 depletion. This difference can be seen in the EM images shown in Figure 9A,B, where most of the virions in TSG101-depleted cells exhibit the characteristic ‘lollipop’ morphology associated with a late assembly/membrane fission defect (Figure 9A), whereas most of the virions budding from AMOT-depleted cells exhibit a ‘half-moon’ morphology in which the nascent viral membrane and Gag shell are incomplete (Figure 9B). This observation was quantified by imaging 500 budding HIV-1 virions from: (1) wild type cells (control), (2) cells depleted of TSG101, and (3) cells depleted of AMOT. The arc formed by the Gag shell was measured manually for each budding virion, and the 500 measurements for each condition were then binned in 15° intervals and displayed graphically (Figure 9D).
The Gag arc distribution of control virions budding from wild type cells exhibited two similarly sized peaks; one in which the Gag shells covered an arc of ∼155°, and another in which the Gag shells covered an arc of ∼320°. These measurements indicate that virion assembly intermediates (or possibly dead-end products) appreciably accumulate at two different stages, suggesting that either stage can be rate limiting. In cells lacking TSG101, budding-arrested virions were much more prevalent and they typically arrested at very late stages of assembly, with the maximal population exhibiting a Gag shell arc of ∼320°. In cells lacking AMOT, arrested virions were again prevalent but they arrested at a much earlier stage, with the maximal population exhibiting a Gag shell arc of ∼165°. The dramatic increases in numbers of budding virions induced by depletion of TSG101 or AMOT can explain how the assembly phenotypes could distribute as single peaks in these cases, whereas virions budding from wild type cells (which were much rarer) distributed into two peaks. Our experiments reveal that HIV budding arrests at an earlier stage in the absence of AMOT than in the absence of TSG101, implying that AMOT acts at an earlier stage of virion morphogenesis.
Discussion
We have identified Angiomotin (AMOT) as a host protein required for efficient HIV-1 release. In the absence of AMOT, most virions arrested at a ‘half-moon’ stage in which a hemisphere of assembled Gag molecules distended the membrane, but did not form a spherical particle (Figure 9). The AMOT depletion phenotype was distinct from the defect induced by the absence of ESCRT factors, where virus budding arrested at a ‘lollipop’ stage in which nearly complete immature virions remained tethered to the host cell via thin membrane stalks. Thus, cellular factors facilitate (at least) two distinct steps in virion morphogenesis, with AMOT required to complete assembly/envelopment and the ESCRT machinery required for membrane fission. HIV-1 virions budding from control cells accumulated at both stages, suggesting that both steps can be kinetic bottlenecks (see Figure 9D and reference Ku et al., 2013).
The Gag lattice organizes the virion and undoubtedly helps to drive membrane envelopment because pure recombinant Gag proteins can form spherical particles in vitro (Campbell et al., 2001). In cells, however, Gag appears to require additional activities to bend the membrane, extrude the particle, and sever the bud neck. The well-studied process of endocytic vesicle formation provides precedence for these additional requirements because clathrin alone can form spherical assemblies but nevertheless requires assistance from BAR domain-containing proteins to help bend membranes, and assistance from actin and dynamin to release the vesicle (Weinberg and Drubin, 2012). It is intriguing that AMOT p130 also contains a candidate BAR-like domain (Heller et al., 2010) and can associate with F-actin (Ernkvist et al., 2006, 2008; Chan et al., 2013). The AMOT BAR-like domain can remodel membranes in vitro (Heller et al., 2010), although it is not yet clear whether it functions as a classical or an inverse BAR domain. This distinction is mechanistically significant because classical BAR domains stabilize positive membrane curvature whereas inverse BAR domains stabilize negative curvature (Mim and Unger, 2012), both of which are generated during enveloped virus assembly. In principle, the BAR domain could mediate AMOT p130 recruitment at the half-moon stage, once the nascent viral membrane has reached the proper degree of curvature. The BAR domain could also help drive the additional membrane curvature required to proceed to the lollipop stage. Similarly, F-actin recruitment and polymerization could help ‘push’ the virion toward the lollipop stage, although several recent studies argue strongly against an essential role for F-actin in HIV-1 assembly (Bharat et al., 2014; Rahman et al., 2014).
Our experiments indicate that another important AMOT p130 function is to recruit NEDD4L to sites of virus budding. In support of this idea, we have shown that: (1) the two proteins bind one another directly through interactions between the NEDD4L WW domains and PPXY elements present in the unique N-terminal region of the AMOT p130 isoform (Figures 1, 2 and references Wang et al., 2012; Moleirinho et al., 2014), (2) AMOT p130 can recruit NEDD4L to cellular membranes, and this activity requires the NEDD4L WW-AMOT p130 PPXY interactions ([Heller et al., 2010] and our data not shown), (3) NEDD4L cannot stimulate release of ‘crippled’ HIV-1 constructs in the absence of AMOT (Figure 4), (4) NEDD4L mutants that cannot bind AMOT are also impaired in their ability to stimulate virus budding (Figure 5), (5) AMOT p80 and AMOT p130 PPXY mutants that cannot bind NEDD4L do not support virus budding (Figures 3–6) and (6) recombinant AMOT p130 and HIV-1 Gag proteins can bind one another directly, albeit weakly in vitro (Figure 2B), and could bind more robustly in cells owing to avidity and membrane topology effects.
Previous studies indicate that other retroviruses also employ distinct host factors to promote membrane envelopment and fission. The clearest example is the D-type Betaretrovirus Mason-Pfizer Monkey Virus (M-PMV), which forms spherical particles in the cytoplasm and then obtains its outer membrane and exits the cell by crossing the plasma membrane. M-PMV Gag contains two distinct late assembly domains that function non-redundantly: a PPPY motif that binds NEDD4 (and perhaps other NEDD4 family members), and a PSAP motif that binds TSG101/ESCRT-I (Gottwein et al., 2003). PPPY and PSAP mutations induce strikingly different phenotypes: mutation of the PPPY motif causes virions to stack up against the plasma membrane where they fail to initiate membrane envelopment whereas virions lacking the PSAP motif acquire an envelope, but accumulate at the lollipop stage or are released as chains of membrane-tethered particles. These observations imply that Gag PPPY-NEDD4 complexes are required for early stages of membrane bending and envelopment, whereas Gag PSAP-ESCRT-I complexes recruit the ESCRT machinery to mediate membrane fission. Like other family members, NEDD4 binds AMOT (Moleirinho et al., 2014), and we therefore speculate that M-PMV may also recruit AMOT to facilitate membrane bending and particle envelopment. This would imply that viruses can recruit NEDD4-motin complexes by binding either member of the complex. The deltaretrovirus HTLV-I is another case where a PPXY motif appears to be required for an early step in particle envelopment and a PTAP motif may function at a later membrane fission step ([Le Blanc et al., 2002; Blot et al., 2004; Heidecker et al., 2004], but also see reference [Wang et al., 2004]). Distinct roles for PPXY and PTAP motifs have not been delineated in other well-studied viruses that use both motifs, such as the gammaretrovirus Murine Leukemia Virus (Yuan et al., 2000) or the filovirus Ebola Virus (Timmins et al., 2003).
Although many details of HIV-1 assembly and budding remain to be elucidated, our data are consistent with a stepwise pathway in which: (1) AMOT p130 is recruited at the half-moon stage of HIV-1 assembly, perhaps by binding cooperatively to Gag and to appropriately curved membranes, (2) AMOT p130 promotes further membrane curvature and recruits NEDD4L via PPXY-WW domain interactions, (3) NEDD4L builds Lys-63-linked ubiquitin chains on the viral Gag protein (and/or other nearby substrates), (4) the juxtaposition of Gag late assembly domains and Lys-63-linked Ub chains creates high affinity binding sites for the early-acting ESCRT factors, TSG101 and ALIX, and (5) these factors then recruit the downstream ESCRT-III and VPS4 components required to effect membrane scission. This model is attractive because: (1) our work shows that AMOT functions upstream of the ESCRT pathway, (2) AMOT appears to function as a NEDD4L adaptor (see above), (3) ubiquitin transfer is required for efficient HIV-1 release (Martin-Serrano, 2007), (4) NEDD4L builds Lys-63-linked ubiquitin chains and can use HIV-1 Gag as a substrate (Weiss et al., 2010; Sette et al., 2013), and (5) both TSG101/ESCRT-I and ALIX have ubiquitin binding activities. The ALIX activity is specific for Lys-63-linked ubiquitin chains (Dowlatshahi et al., 2012; Keren-Kaplan et al., 2013; Pashkova et al., 2013), and ESCRT-I complexes also have multiple ubiquitin binding sites, although linkage specificity has not been observed (Stefani et al., 2011; Agromayor et al., 2012; McCullough et al., 2013). Cooperative binding to the late domains and ubiquitin could explain how ALIX and TSG101/ESCRT-I can be efficiently recruited, despite making intrinsically weak interactions with the p6Gag late assembly domains. The proposed sequence of events could also provide a ‘timing’ mechanism for recruiting (or activating) these early-acting ESCRT proteins to function at a late stage of virion assembly, although the precise timing of early-acting ESCRT recruitment remains to be clarified because published reports differ on this issue (Jouvenet et al., 2011; Ku et al., 2014).
All enveloped viruses must bend and remodel membranes as they bud, and AMOT requirements may therefore be quite general. Indeed, AMOTL1 has been shown to bind the M protein of PIV5 and to be required for the efficient, ubiquitin-dependent release of this virus (Pei et al., 2010). Motin family members therefore contribute to the assembly of at least two highly diverged enveloped viruses; the paramyxovirus PIV5 and the lentivirus HIV-1. This suggests that AMOT proteins, like the ESCRT machinery, may provide general activities required for the assembly and release of a variety of different enveloped viruses.
Materials and methodsPlasmids, siRNA constructs and antibodies
Expression constructs, siRNA sequences, and antibodies are described in detail in Supplementary file 1. Constructs for expressing wild type and mutant NEDD4L and AMOT proteins in E. coli, insect SF-9 cells, and human 293T cells were created by standard cloning and mutagenesis methods, with detailed methods available upon request. All of the new expression constructs used in our studies have been deposited in the public DNASU plasmid repository (http://dnasu.org/DNASU/Home.jsp).
Affinity purification and mass spectrometric identification of AMOT
For the experiment shown in Figure 1B, an empty vector control or a pCAG-OSF-NEDD4L expression vector (Morita et al., 2007) expressing NEDD4L with an N-terminal One-STrEP-FLAG (OSF) affinity tag was transfected into 293T cells (calcium phosphate method, 2 × 10 cm plates each, 25 μg DNA/plate). After 48 hr, cells from each sample were collected by scraping, pooled, washed three times with PBS (1.5 mM KH2PO4, 155 mM NaCl, 2.7 mM Na2HPO4, pH 7.4), and lysed with 500 μl Triton lysis buffer (1% Triton X-100, 50 mM Tris pH 8.0, 150 mM NaCl, 150 μg/ml PMSF). This, and all subsequent preparation steps, were performed at 4°C. Soluble extracts were pre-incubated for 30 min with 5 μl Protein-A resin slurry (Millipore, Temecula, CA) to remove non-specific binding proteins. Supernatants were collected and incubated for 2 hr with 10 μl STrEP-Tactin resin (IBA-Lifesciences, Göttingen, Germany). The resins were washed three times with 500 μl washing buffer (0.1% Triton-X100, 50 mM Tris, 150 mM NaCl, pH 8.0), resuspended in 35 μl SDS-PAGE loading buffer (2% SDS, 10% glycerol, 2% 2-mercaptoethanol (BME), 0.002% bromophenol blue and 62.5 mM Tris, pH 6.8), and boiled to release the bound proteins. 12 μl samples of each solution were electrophoresed (10% SDS-PAGE) and proteins were visualized by staining with Coomassie brilliant blue.
This procedure was repeated in three independent experiments, and in each case the protein band corresponding to AMOT p130 was visible in the Coomassie-stained gel (Figure 1B). This band was excised (and the equivalent region from the control lane was also excised in one of the repetitions), de-stained in 50% methanol with 50 mM ammonium bicarbonate (2 × 1 ml, 1 hr, 23°C, gentle vortexing), re-hydrated in 50 mM ammonium bicarbonate (1 ml, 30 min, 23°C), and cut into several pieces. Each piece was dehydrated in acetonitrile (1 ml, 30 min, 23°C, gentle shaking), and the excess acetonitrile was removed. Proteins were in-gel digested using 10–20 μl sequence-grade modified trypsin (Promega, Madison, WI, 20 ng/μl) in 50 mM ammonium bicarbonate (16 hr, 37°C), and the reaction was quenched by the addition of 20 μl 1% formic acid. The solution was removed and the gel washed twice with 50% acetonitrile in 1% formic acid (20 μl, 20 min, 37°C, with sonication) and once with 100% acetonitrile (20 min, 37°C, with sonication). The wash solutions were combined, dried, and reconstituted in 100 μl 5% acetonitrile with 0.1% formic acid for LC/MS/MS analysis.
Peptides were analyzed using a nano-LC/MS/MS system comprising a nano-LC pump (2D-ultra, Eksigent) and an LTQ-FT mass spectrometer (ThermoElectron Corporation, San Jose, CA), equipped with a nanospray ion source (ThermoElectron Corporation). 5–20 fmoles of each tryptic digest were injected onto a homemade C18 nanobore LC column (C18 [Atlantis, Waters Corporation, Milford, MA]); 3 μm particle; column (75 μm ID × 100 mm length), and eluted using a linear gradient of 5–70% solvent B in 78 min (solvent B: 80% acetonitrile with 0.1% formic acid; solvent A: 5% acetonitrile with 0.1% formic acid, 350 nl/min). The LTQ-FT mass spectrometer was operated in the data-dependent acquisition mode (Xcalibur 1.4 software) with the 10 most intense peaks in each FT primary scan determined on-the-fly and trapped for MS/MS analysis and CID peptide fragmentation/sequencing in the LTQ linear ion trap. Spectra in the FT-ICR were acquired from m/z 350 to 1400 at 50,000 resolution (∼2 ppm mass accuracy) with the following LTQ linear ion trap parameters: precursor activation time 30 ms and activation Q at 0.25; collision energy at 35%; dynamic exclusion width at 0.1 Da–2.1 Da, with one repeat count and 10 s duration.
Raw LTQ FT MS data files were converted to peak lists with BioworksBrowser 3.2 software (ThermoElectron Corporation) using the following processing parameters: precursor mass 351–5500 Da; grouping enabled, 5 intermediate MS/MS scans; precursor mass tolerance 5 ppm, minimum ion count in MS/MS = 15, and minimum group count = 1. Resulting DTA files were merged, and searched to identify peptides/proteins against NCBInr using the MASCOT search engine (Matrix Science Ltd.; version 2.2.1, Boston, MA). Searches were done with tryptic specificity, allowing two missed cleavages, or ‘non-specific cleavage’ and a mass error tolerance of 5 ppm in MS spectra (i.e., FT-ICR data) and 0.5 Da for MS/MS ions (i.e., LTQ linear ion trap). Identified peptides were accepted when the MASCOT ion score value exceeded 20. In all three repetitions, at least two peptides corresponding to AMOT p130 were identified in the tryptic digest, and the maximal coverage was 33% (see Figure 1—figure supplement 1).
Immunoprecipitations
To assay OSF-NEDD4L protein binding to endogenous AMOT p130 (Figure 1C), 293T cells (2 × 106 cells per 10 cm plate) were transfected using Lipofectamine 2000 (Invitrogen/Life Technologies, Carlsbad, CA) with either 3.5 μg of an empty pCAG-OSF expression vector (control) or with pCAG-OSF vectors that expressed the designated OSF-NEDD4L proteins, using 3.5 μg of expression constructs for wild type NEDD4L or NEDD4LΔWW, or 2 μg each of expression constructs for NEDD4LC942A or NEDD4LΔWW/C942A. In this case, and in all other experiments, empty vector was added to keep total DNA levels constant (i.e., 1.5 μg empty pCAG-OSF vector in the case of NEDD4LC942A or NEDD4LΔWW/C942A). 48 hr post transfection, cells were washed in PBS buffer and lysed (25 mM Tris, 150 mM NaCl, 1 mM EDTA, 0.5% CHAPS, pH 7.5, 10 min, 4°C). Soluble lysates were collected by centrifugation (5000×g, 5 min, 4°C) and incubated with 50 μl STrEP-Tactin resin (IBA-Lifesciences, Göttingen, Germany, 3 hr, 4°C) equilibrated with lysis buffer. The resin was washed (25 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 0.1% CHAPS, pH 7.5, 3 × 1 ml), resuspended in 50 μl 2× SDS-PAGE loading buffer, boiled, and the released proteins were analyzed by SDS-PAGE and western blotting. Separated proteins were transferred to PVDF membranes, blocked with 5% non-fat dry milk for 45 min at RT, and incubated overnight at 4°C with primary antibodies (see Supplementary file 1). Secondary antibodies were anti-rabbit IgG or anti-mouse IgG conjugated to IRdye800 or IRdye700, respectively (1:10,000 and 1:20,000 respectively, Rockland Immunochemicals Inc., Gilbertsville, PA). All western blots were visualized using an Odyssey scanner (Li-Cor Biosciences, Lincoln, NB).
To assay OSF-NEDD4L protein binding to exogenous HA-AMOT p130 proteins (Figure 1—figure supplement 2), 293T cells (2 × 106 cells per 10 cm plate) were co-transfected using Lipofectamine 2000 (Invitrogen/Life Technologies) with 3.5 μg of a pCAG-OSF vector expressing wild type OSF-NEDD4L and 3.5 μg of an empty pCAG vector (control), or pcDNA vectors expressing HA-AMOT p130 or HA-AMOT p130ΔPPXY. Thereafter, the co-immunoprecipitation protocol was identical to the one described above.
Bacterial protein expression and purification
Proteins were expressed in the Rosetta-pLysS bacterial strain grown in auto-induction media ZYP-5052 (Studier, 2005).
NEDD4L protein purification
E. coli expressing GST-NEDD4L proteins (2 l) were harvested, resuspended in 80 ml lysis buffer (50 mM Tris, 150 mM NaCl, 5% glycerol, 2 mM MgCl2, 1 mM TCEP, 0.5% NP-40, 0.5 mM EDTA, pH 8.0) supplemented with the protease inhibitors PMSF (20 μg/ml), pepstatin (0.4 μg/ml), leupetin (0.8 μg/ml), and aprotinin (1.6 μg/ml). This and all subsequent steps were performed at 4°C. Lysate viscosity was reduced by addition of DNase (12.5 μg/ml, Roche, Basel, Switzerland). The lysate was clarified by centrifugation at 39,000×g for 45 min, and loaded onto 10 ml glutathione-superose resin (GE Healthcare Bio-Sciences, Pittsburgh, PA). The resin was washed with high-salt wash buffer (20 mM Tris, 1 M LiCl, 0.01% NP-40, 1 mM TCEP, 0.5 mM EDTA, 5% glycerol, pH 8.0) followed by low-salt wash buffer (20 mM Tris, 150 mM NaCl, 1 mM TCEP, 0.5 mM EDTA, 5% glycerol, pH 8.0) prior to elution (50 mM Tris, 150 mM NaCl, 5% glycerol, 1 mM TCEP, 0.5 mM EDTA, 20 mM reduced glutathione pH 8.0). PreScission protease (4 nmol protease/120 nmol purified protein, GE Healthcare Bio-Sciences) was added to remove the GST affinity tag and the sample was dialyzed overnight against the same buffer without glutathione. The cleaved protein was further purified on a Q-column (10 HiTrap HP-Q, GE Healthcare Bio-Sciences) equilibrated in buffer A (20 mM Tris, 50 mM NaCl, 5% glycerol, 1 mM TCEP, 0.5 mM EDTA, pH 8.0), and eluted with a linear gradient to buffer B (Buffer A, but with 1 M NaCl). NEDD4L-containing fractions were pooled and dialyzed against storage buffer (20 mM Tris, 150 mM NaCl, 20% glycerol, 1 mM TCEP, pH 8.0) and stored at −80°C. Typical yields were 60 nmol/l culture, and the protein identities were confirmed by electrospray mass spectrometry (NEDD4L: MWcalc = 98,472 g/mol, MWexp = 98,474 g/mol; NEDD4LΔWW: MWcalc = 98,012 g/mol, MWexp = 98,014 g/mol).
Gag<sub>Δp6</sub> protein purification
E. coli expressing HIV-1HXB2 GagΔp6-GST (4 l) were harvested and resuspended in 150 ml of lysis buffer (50 mM HEPES, 500 mM NaCl, 5% glycerol, 2 mM MgCl2, 10 mM BME, 0.1% NP-40, 10 μM ZnCl2, pH 7.5) supplemented with protease inhibitors (PMSF, pepstatin, leupetin, and aprotinin). This and all subsequent steps were performed at 4°C. Lysate viscosity was reduced by addition of DNase (Roche) and the lysate was clarified by centrifugation (39,000×g, 45 min). Nucleic acids were removed by adding polyethylenamine (PEI at pH 8.0) to a final concentration of 1% (wt/vol), followed by centrifugation (27,000×g, 20 min). The fusion protein was precipitated by addition of 0.35 equivalents of saturated ammonium sulfate (to 26% saturation). The resulting pellets were resuspended in 80 ml of GST Loading Buffer (50 mM HEPES, 500 mM NaCl, 5% glycerol, 10 mM BME, 0.01% NP-40 and 10 μM ZnCl2, pH 7.5), and loaded onto 5 ml of glutathione-sepharose resin (GE Healthcare Bio-Sciences). The resin was washed with GST loading buffer, and the protein was cleaved overnight on-column by incubation in 80 ml cleavage buffer (50 mM Tris, 500 mM NaCl, 5% glycerol, 10 mM BME, and 10 μM ZnCl2, pH 7.5) supplemented with 0.5 mg PreScission protease (GE Healthcare Bio-Sciences). The flowthrough containing the cleaved protein was diluted to lower the salt concentration to 200 mM, and loaded onto an SP-column equilibrated in buffer A (50 mM Tris, 50 mM NaCl, 5% glycerol, 10 mM BME, 10 μM ZnCl2, pH 7.5). Gag∆p6 was eluted with a linear gradient to buffer B (buffer A, but with 1 M NaCl). Protein-containing fractions were pooled and dialyzed against storage buffer (20 mM Tris, 500 mM NaCl, 20% glycerol, 10 mM BME, 10 μM ZnCl2, pH 7.5) and stored at −80°C. Typical yields were 75 nmol/l, and the protein identity was confirmed by electrospray mass spectrometry (MWcalc, −Met1 = 50,762 g/mol, MWexp = 50,758 g/mol).
Streptactin pull-downs of Gag<sub>Δp6</sub> and NEDD4L with OSF-AMOT
For the experiments described in Figure 2 and Figure 2—figure supplement 1, SF9 cells (2 l) were infected at an MOI of 5 with Bac-to-Bac baculoviruses expressing OSF-tagged GFP (control), AMOT p130, AMOTL1 or AMOTL2. Pellets from 200 ml of culture were lysed in 10 ml of lysis buffer (1% NP-40, 150 mM NaCl, 20 mM Tris pH 8.0, 0.5 mM EDTA supplemented with protease inhibitors). All steps were performed at 4°C. Supernatants were collected by centrifugation for 30 min at 13,000 RPM in a microcentrifuge and 0.5 ml of each lysate was incubated with 30 μl of a slurry of STrEP-Tactin resin (1 hr). Resins were washed with high salt wash buffer (20 mM Tris, 1 M LiCl, 0.5 mM EDTA, 0.1% NP-40, pH 8.0), followed by low salt buffer (20 mM Tris, 150 mM NaCl, 0.5 mM EDTA, 0.1% NP-40, pH 8.0). For NEDD4L binding experiments, the OSF-GFP or OSF-AMOT-bound resins were washed with binding buffer (150 mM NaCl, 50 mM Tris, 1 mM TCEP, 10 μM ZnCl2, 0.1% NP-40, pH 7.5). Pure recombinant NEDD4L or NEDD4LΔWW proteins (0.5 μM in 500 μl binding buffer) were pre-incubated with washed resins for 1 hr, and then incubated for 1 hr with GFP- or AMOT-bound resins. Resins were washed three times with 250 μl binding buffer, resuspended in 30 μl SDS-PAGE loading buffer, boiled, separated by SDS-PAGE, and visualized by Coomassie staining or by Western blotting (with 1 μl samples used for Coomassie staining, and 1 μl samples of a 50-fold dilution used for Western blotting).
For GagΔp6 binding experiments, the OSF-GFP (control)- or OSF-AMOT-bound resins were washed with binding buffer. Pure recombinant HIV-1 GagΔp6 (1 μM) in 500 μl binding buffer was incubated with the washed resins for 1 hr. Resins were washed with binding buffer, and processed as described above.
NEDD4L and AMOT co-overexpression experiments
NEDD4L and AMOT protein family co-overexpression experiments shown in Figure 3 and Figure 3—figure supplement 1, Figure 3—figure supplement 2, and Figure 7—figure supplement 1A were performed following the protocol: 293T cells (4 × 105 cells/well in a 6-well plate) were co-transfected (Lipofectamine, 2000; Invitrogen/Life Technologies) with 1 μg of an R9-based expression vector for the NL4-3 strain of HIV-1ΔPTAP, ΔYP (Swingler et al., 1997), 0.5 μg of pCI-FLAG NEDD4L or NEDD4LΔWW (Chung et al., 2008) or pCI-FLAG empty vector in combination with either 0.5 μg of pCMV AMOT p80 (Figure 3—figure supplement 2) or 1 μg pcDNA3 HA-AMOT p130 (Figure 3 and Figure 3—figure supplement 2) or 0, 0.25, 0.5, 0.75, 1, 1.25 or 1.5 μg pcDNA3 HA-AMOT p130, (Figure 3—figure supplement 1) or 1 μg pcDNA3 HA AMOTL1 or 1 μg pcDNA3 HA AMOTL2 (Figure 7—figure supplement 1A). 48 hr post-transfection, media was harvested for titer measurements and virus purification, and cells were washed off the plate in PBS.
Viral titers were assayed in HeLa-TZM indicator cells using a β-galactosidase assay, and following the manufacturer's instructions (Promega, Madison, WI). Briefly, HeLa-TZM cells (5000 cells per well, 96-well plate) were infected with four different dilutions of virus-containing culture media, harvested after 48 hr, washed once (PBS, 4°C), and lysed in 25 mM Tris phosphate (pH 7.8), 2 mM DTT, 2 mM 1,2-diaminocyclohexane N,N,N′,N′-tetraacetic acid (DCTA), 10% glycerol and 1% Triton X-100 (10 min, 23°C). Cell lysates were incubated in 200 mM sodium phosphate buffer (pH 7.3), 2 mM MgCl2, 100 mM BME and 1.33 mg/ml ortho-Nitrophenyl-β-galactoside for 60 min at 37°C. Reactions were terminated by addition of 200 μl 1 M Tris base and absorbance at 420 nm was read in a plate reader.
For western blot analyses, virions were pelleted by centrifugation through a 20% sucrose cushion (90 min, 15,000×g, 4°C) and resuspended in SDS-PAGE loading buffer. Cells were washed in PBS, lysed in RIPA buffer (10 mM Tris, 1 mM EDTA, 0.5 mM EGTA, 1% Triton X-100, 0.1% sodium deoxycholate, 0.1% SDS, 140 mM NaCl, pH 8.0) supplemented with protease inhibitors (cOmplete, Roche) (10 min, 4°C), and the insoluble material was removed by centrifugation (10 min, 15,000×g, 4°C). Primary antibodies and dilutions used for western blots are given in Supplementary file 1.
AMOT siRNA depletion and re-expression experiments
Depletion/re-expression experiments (Figure 4, Figure 4—figure supplement 1, Figure 4—figure supplement 2, Figure 5, Figure 7, and Figure 7—figure supplement 1) were performed following the protocol: t = 0: 293T cells (2.5 × 105 cells/well, 6-well plate) were transfected with 10 nM siRNA using 7.5 μl Lipofectamine RNAiMax (Invitrogen/Life Technologies); t = 24 hr: media changed and second siRNA co-transfection with 10 nM siRNA and 0.5 μg wild type HIV-1 R9 expression vector or 1 μg HIV-1ΔPTAP, ΔYPXL expression vector and 0.5 μg of pCI-FLAG NEDD4L (10 μl Lipofectamine, 2000; Invitrogen/Life Technologies); t = 72 hr: media and cells harvested for titer measurements and western blot analyses as described above. For the rescue experiments, siRNA-resistant expression constructs expressing AMOT p80, AMOT p130, AMOT p130ΔPPXY, AMOTL1 or AMOTL2 were co-transfected at t = 24 hr.
AMOT expression and effects on HIV-1 release from HeLa and Jurkat T cells
AMOT mRNA is expressed in 293T cells and in white blood cells and lymphoid tissues, but cannot be detected in HeLa cells (Moleirinho et al., 2014). Western blots were performed to compare AMOT protein expression levels in 293T vs HeLa cells (Figure 6—figure supplement 1). 293T (2 x 106 cells/10 cm plate) and HeLa M (1.2 x 106 cells/10 cm plate) cells were seeded, harvested 72 hr later by trypsin treatment, washed twice in PBS, counted, lysed in RIPA buffer (1.2 × 105 cells/μl) for 15 min at 4°C, clarified by centrifugation (15 min, 15,000×g, 4°C) and diluted 1:1 with 2× SDS loading buffer. 2.5, 5, 10 and 20 μl of each cell lysate mixture was loaded on a 4–15% gradient gel and AMOT proteins were detected by western blotting.
Experiments to test whether AMOT overexpression can stimulate HIV-1 release from HeLa cells (Figure 6A) were performed as follows: HeLa M cells (2.5 × 105 cells/well in a 6-well plate) were co-transfected with 0.5 μg of wild-type R9 HIV-1 and 0.5, 1, 2 or 4 μg of pcDNA3-HA AMOT p130 expression vectors. Cells and viruses were harvested 48 hr post-transfection.
To test whether AMOT could stimulate virus release from physiologically relevant T cells, AMOT was overexpressed in Jurkat T cells as follows: t = 0, cells were harvested, washed in PBS and resuspended in Neon resuspension buffer R (Invitrogen/Life Technologies). 5 × 106 cells were combined with 3 μg of the wild type R9 HIV-1 expression vector and 0, 5 or 10 μg of pcDNA3-HA AMOT p130 or AMOT p130ΔPPXY expression vectors. Cells were then electroporated in a 100 µl tip with the Neon Transfection System using three 10 ms pulses of 1325 V. Following electroporation, each sample was immediately transferred to a 10 cm plate containing 10 ml pre-warmed RPMI (supplemented with 10% FBS and 2 mM glutamine). t = 96 hr: media and cells were harvested.
Protocols for analyzing protein expression and viral titers were identical to those described for 293T cells experiments.
Scanning electron microscopy (SEM)
293T cells (2.5 × 105 cells/well, 6-well plate) were seeded onto glass cover slips (24 hr prior transfection) and co-transfected with the designated siRNAs and the R9 HIV-1 expression construct as described above. 48 hr post-transfection, the media was removed and cells were fixed in 1 ml fixation buffer (2.5% glutaraldehyde/1% paraformaldehyde in sodium cacodylate buffer; 50 mM sodium cacodylate, 18 mM sucrose, 2 mM CaCl2, pH 7.4, 10 min). This and all subsequent steps were performed at 23°C. Fixed cells were washed three times in sodium cacodylate buffer (5 min), and stained with 2% OsO4 for 1 hr. Each sample was then dehydrated in a graded ethanol series, dried in HDMS (hexamethyldisilazane), and sputter coated with platinum. SEM images were collected on a FEI Quanta 600 FEG scanning electron microscope at a beam energy of 10 kV, a spot size of three, and magnifications of 10,000, 30,000 or 65,000 X.
Transmission electron microscopy (TEM)
293T cells (2.5 × 105 cells/well, 6-well plate) were co-transfected with the designated siRNAs and the R9 HIV-1 expression construct as described above. 48 hr post-transfection, the media was removed and cells were fixed in 1 ml fixation buffer (2.5% glutaraldehyde/1% paraformaldehyde in sodium cacodylate buffer [50 mM sodium cacodylate, 18 mM sucrose, 2 mM CaCl2, pH 7.4, 10 min]). This and all subsequent steps were performed at 23°C. Fixed cells were harvested, washed three times in sodium cacodylate buffer (5 min), and stained with 2% OsO4 for 1 hr. The pellet was rinsed once in sodium cacodylate buffer and twice in water (5 min), followed by incubation in 100 μl of a 4% uranyl acetate solution (30 min). Stained cells were dehydrated in a graded ethanol series followed by acetone, and embedded in epoxy resin EMBed-812 (Electron Microscopy Sciences, Hatfield, PA). Thin sections (80–100 nm) were cut, post-stained with saturated uranyl acetate (20 min), rinsed with water, dried, stained with Reynolds' lead citrate (10 min) and dried again. TEM images were collected on a Hitachi H-7100 transmission electron microscope at an accelerating voltage of 75 kV, equipped with a Gatan Orius sc1000 camera.
Imaged virions were scored as ‘mature’ if they lacked an observable cellular tether and a condensed core was evident, as ‘immature’ if they were untethered but lacked a condensed core, and as ‘budding’ if a Gag layer was clearly evident and the nascent virion was assembling or was clearly tethered to the cell. The degree of assembly was assessed by rendering a (hypothetical) spherical particle and measuring the arc angle of the nascent Gag layer. To image 500 virions budding from wild type cells, it was necessary to screen about eight-times as many thin slices of wild type control cells (∼2500 cells) vs TSG101- or AMOT-depleted cells (275 and 300, respesctively).
Acknowledgements
We thank Chad Nelson and the University of Utah Mass Spectrometry Core facility for mass spectrometry analyses, Chris Rodesch and the University of Utah Cell Imaging Core facility for assistance with fluorescence microscopy, Nancy Chandler in the University of Utah EM Core facility for expert thin sectioning, S Joshua Romney (Invitrogen/Life Technologies) for the loan of a Neon transfection system, and members of our laboratory for critical reading and feedback. This work made use of University of Utah shared facilities of the Micron Technology Foundation Inc. Microscopy Suite sponsored by the College of Engineering, Health Sciences Center, Office of the Vice President for Research, and the Utah Science Technology and Research (USTAR) initiative of the State of Utah. All new constructs reported in this manuscript have been submitted and are freely available from the DNASU repository (https://dnasu.org/DNASU/).
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
GM, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents
SLA, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents
JA, Conception and design, Acquisition of data, Drafting or revising the article
MSL, Conception and design, Drafting or revising the article, Contributed unpublished essential data or reagents
WIS, Conception and design, Analysis and interpretation of data, Drafting or revising the article
Additional files10.7554/eLife.03778.021
Expression plasmids, antibodies and siRNAs used in this study.
DOI:http://dx.doi.org/10.7554/eLife.03778.021
ReferencesAgromayorMSolerNCaballeAKueckTFreundSMAllenMDBycroftMPerisicOYeYMcDonaldBScheelHHofmannKNeilSJMartin-SerranoJWilliamsRL2012The UBAP1 subunit of ESCRT-I interacts with ubiquitin via a SOUBA domainStructure2041442810.1016/j.str.2011.12.013BernassolaFKarinMCiechanoverAMelinoG2008The HECT family of E3 ubiquitin ligases: multiple players in cancer developmentCancer Cell14102110.1016/j.ccr.2008.06.001BharatTACastillo MenendezLRHagenWJLuxVIgonetSSchorbMSchurFKKrausslichHGBriggsJA2014Cryo-electron microscopy of tubular arrays of HIV-1 Gag resolves structures essential for immature virus assemblyProceedings of the National Academy of Sciences of USA1118233823810.1073/pnas.1401455111BlotVPerugiFGayBPrevostMCBriantLTangyFAbrielHStaubODokhélarMCPiqueC2004Nedd4.1-mediated ubiquitination and subsequent recruitment of Tsg101 ensure HTLV-1 Gag trafficking towards the multivesicular body pathway prior to virus buddingJournal of Cell Science1172357236710.1242/jcs.01095CampbellSFisherRJTowlerEMFoxSIssaqHJWolfeTPhillipsLRReinA2001Modulation of HIV-like particle assembly in vitro by inositol phosphatesProceedings of the National Academy of Sciences of USA98108751087910.1073/pnas.191224698ChanSWLimCJGuoFTanILeungTHongW2013Actin-binding and cell proliferation activities of angiomotin family members are regulated by Hippo pathway-mediated phosphorylationThe Journal of Biological Chemistry288372963730710.1074/jbc.M113.527598ChenHRossCAWangNHuoYMacKinnonDFPotashJBSimpsonSGMcMahonFJDePauloJRJrMcInnisMG2001NEDD4L on human chromosome 18q21 has multiple forms of transcripts and is a homologue of the mouse Nedd4-2 geneEuropean Journal of Human Genetics992293010.1038/sj.ejhg.5200747ChungHYMoritaEvon SchwedlerUMullerBKrausslichHGSundquistWI2008NEDD4L overexpression rescues the release and infectivity of human immunodeficiency virus type 1 constructs lacking PTAP and YPXL late domainsJournal of Virology824884489710.1128/JVI.02667-07DowlatshahiDPSandrinVVivonaSShalerTAKaiserSEMelandriFSundquistWIKopitoRR2012ALIX is a Lys63-specific Polyubiquitin binding protein that functions in retrovirus buddingDevelopmental Cell231247125410.1016/j.devcel.2012.10.023DunnDMIshigamiTPankowJvon NiederhausernAAlderJHuntSCLeppertMFLalouelJMWeissRB2002Common variant of human NEDD4L activates a cryptic splice site to form a frameshifted transcriptJournal of Human Genetics4766567610.1007/s100380200102ErnkvistMBirotOSinhaIVeitonmakiNNyströmSAaseKHolmgrenL2008Differential roles of p80- and p130-angiomotin in the switch between migration and stabilization of endothelial cellsBiochimica Et Biophysica Acta178342943710.1016/j.bbamcr.2007.11.018ErnkvistMLuna PerssonNAudebertSLecinePSinhaILiuMSchlueterMHorowitzAAaseKWeideTBorgJPMajumdarAHolmgrenL2009The Amot/Patj/Syx signaling complex spatially controls RhoA GTPase activity in migrating endothelial cellsBlood11324425310.1182/blood-2008-04-153874FujiiKMunshiUMAblanSDDemirovDGSoheilianFNagashimaKStephenAGFisherRJFreedEO2009Functional role of Alix in HIV-1 replicationVirology39128429210.1016/j.virol.2009.06.016GarrusJEvon SchwedlerUKPornillosOWMorhamSGZavitzKHWangHEWettsteinDAStrayKMCoteMRichRLMyszkaDGSundquistWI2001Tsg101 and the vacuolar protein sorting pathway are essential for HIV-1 buddingCell107556510.1016/S0092-8674(01)00506-2GöttlingerHGDorfmanTSodroskiJGHaseltineWA1991Effect of mutations affecting the p6 gag protein on human immunodeficiency virus particle releaseProceedings of the National Academy of Sciences of USA883195319910.1073/pnas.88.8.3195GottweinEBodemJMüllerBSchmechelAZentgrafHKräusslichHG2003The Mason-Pfizer monkey virus PPPY and PSAP motifs both contribute to virus releaseJournal of Virology779474948510.1128/JVI.77.17.9474-9485.2003HarrisonMSSchmittPTPeiZSchmittAP2012Role of ubiquitin in parainfluenza virus 5 particle formationJournal of Virology863474348510.1128/JVI.06021-11HeideckerGLloydPAFoxKNagashimaKDerseD2004Late assembly motifs of human T-cell leukemia virus type 1 and their relative roles in particle releaseJournal of Virology786636664810.1128/JVI.78.12.6636-6648.2004HellerBAdu-GyamfiESmith-KinnamanWBabbeyCVoraMXueYBittmanRStahelinRVWellsCD2010Amot recognizes a juxtanuclear endocytic recycling compartment via a novel lipid binding domainThe Journal of Biological Chemistry285123081232010.1074/jbc.M109.096230InghamRJGishGPawsonT2004The Nedd4 family of E3 ubiquitin ligases: functional diversity within a common modular architectureOncogene231972198410.1038/sj.onc.1207436ItaniOAStokesJBThomasCP2005Nedd4-2 isoforms differentially associate with ENaC and regulate its activityAmerican Journal of Physiology Renal Physiology289F334F34610.1152/ajprenal.00394.2004JouvenetNZhadinaMBieniaszPDSimonSM2011Dynamics of ESCRT protein recruitment during retroviral assemblyNature Cell Biology1339440110.1038/ncb2207Keren-KaplanTAttaliIEstrinMKuoLSFarkashEJerabek-WillemsenMBlutraichNArtziSPeriAFreedEOWolfsonHJPragG2013Structure-based in silico identification of ubiquitin-binding domains provides insights into the ALIX-V:ubiquitin complex and retrovirus buddingThe EMBO Journal3253855110.1038/emboj.2013.4KimHCHuibregtseJM2009Polyubiquitination by HECT E3s and the determinants of chain type specificityMolecular and Cellular Biology293307331810.1128/MCB.00240-09KuPIBendjennatMBallewJLandesmanMBSaffarianS2014ALIX is recruited Temporarily into HIV-1 budding sites at the end of gag assemblyPLOS ONE9e9695010.1371/journal.pone.0096950KuPIMillerAKBallewJSandrinVAdlerFRSaffarianS2013Identification of pauses during formation of HIV-1 virus like particlesBiophysical Journal1052262227210.1016/j.bpj.2013.09.047Le BlancIPrévostMCDokhélarMCRosenbergAR2002The PPPY motif of human T-cell leukemia virus type 1 Gag protein is required early in the budding processJournal of Virology76100241002910.1128/JVI.76.19.10024-10029.2002Martin-SerranoJ2007The role of ubiquitin in retroviral egressTraffic81297130310.1111/j.1600-0854.2007.00609.xMartin-SerranoJYaravoyAPerez-CaballeroDBieniaszPD2003Divergent retroviral late-budding domains recruit vacuolar protein sorting factors by using alternative adaptor proteinsProceedings of the National Academy of Sciences of USA100124141241910.1073/pnas.2133846100McCulloughJColfLASundquistWI2013Membrane fission reactions of the mammalian ESCRT pathwayAnnual Review of Biochemistry8266369210.1146/annurev-biochem-072909-101058MengBLeverAM2013Wrapping up the bad news: HIV assembly and releaseRetrovirology10510.1186/1742-4690-10-5MimCUngerVM2012Membrane curvature and its generation by BAR proteinsTrends in Biochemical Sciences3752653310.1016/j.tibs.2012.09.001MoleirinhoSGuerrantWKissilJL2014The Angiomotins - from discovery to functionFEBS Letters5882693270310.1016/j.febslet.2014.02.006MoreauJLordMBoucherMBelleauPFernandesMJ2005Protein diversity is generated within the motin family of proteins by alternative pre-mRNA splicingGene35013714810.1016/j.gene.2005.02.001MoritaESandrinVAlamSLEckertDMGygiSPSundquistWI2007Identification of human MVB12 proteins as ESCRT-I subunits that function in HIV buddingCell Host & Microbe2415310.1016/j.chom.2007.06.003OtteLWiedemannUSchlegelBPiresJRBeyermannMSchmiederPKrauseGVolkmer-EngertRSchneider-MergenerJOschkinatH2003WW domain sequence activity relationships identified using ligand recognition propensities of 42 WW domainsProtein Science1249150010.1110/ps.0233203PashkovaNGakharLWinistorferSCSunshineABRichMDunhamMJYuLPiperRC2013The yeast Alix homolog Bro1 functions as a ubiquitin receptor for protein sorting into multivesicular endosomesDevelopmental Cell2552053310.1016/j.devcel.2013.04.007PeiZBaiYSchmittAP2010PIV5 M protein interaction with host protein angiomotin-like 1Virology39715516610.1016/j.virol.2009.11.002RahmanSAKochPWeichselJGodinezWJSchwarzURohrKLambDCKrausslichandHGMullerB2014Investigating the role of F-actin in human immunodeficiency virus assembly by live-cell microscopyJournal of Virology887904791410.1128/JVI.00431-14SchmittAPLeserGPMoritaESundquistWILambRA2005Evidence for a new viral late-domain core sequence, FPIV, necessary for budding of a paramyxovirusJournal of Virology792988299710.1128/JVI.79.5.2988-2997.2005SettePNagashimaKPiperRCBouamrF2013Ubiquitin conjugation to Gag is essential for ESCRT-mediated HIV-1 buddingRetrovirology107910.1186/1742-4690-10-79SheynkmanGMShortreedMRFreyBLSmithLM2013Discovery and mass spectrometric analysis of novel splice-junction peptides using RNA-SeqMolecular & Cellular Proteomics122341235310.1074/mcp.O113.028142SkouloudakiKWalzG2012YAP1 recruits c-Abl to protect angiomotin-like 1 from Nedd4-mediated degradationPLOS ONE7e3573510.1371/journal.pone.0035735StefaniFZhangLTaylorSDonovanJRollinsonSDoyotteABrownhillKBennionJPickering-BrownSWoodmanP2011UBAP1 is a component of an Endosome-specific ESCRT-i complex that is essential for MVB sortingCurrent Biology211245125010.1016/j.cub.2011.06.028StudierFW2005Protein production by auto-induction in high density shaking culturesProtein Expression and Purification4120723410.1016/j.pep.2005.01.016SwinglerSGallayPCamaurDSongJAboATronoD1997The Nef protein of human immunodeficiency virus type 1 enhances serine phosphorylation of the viral matrixJournal of Virology7143724377TimminsJSchoehnGRicard-BlumSScianimanicoSVernetTRuigrokRWWeissenhornW2003Ebola virus matrix protein VP40 interaction with human cellular factors Tsg101 and Nedd4Journal of Molecular Biology32649350210.1016/S0022-2836(02)01406-7UsamiYPopovSPopovaEGottlingerHG2008Efficient and specific rescue of human immunodeficiency virus type 1 budding defects by a Nedd4-like ubiquitin ligaseJournal of Virology824898490710.1128/JVI.02675-07VottelerJSundquistWI2013Virus Budding and the ESCRT PathwayCell Host Microbe1423224110.1016/j.chom.2013.08.012WangCAnJZhangPXuCGaoKWuDWangDYuHLiuJOYuL2012The Nedd4-like ubiquitin E3 ligases target angiomotin/p130 to ubiquitin-dependent degradationThe Biochemical Journal44427928910.1042/BJ20111983WangHMacheskyNJManskyLM2004Both the PPPY and PTAP motifs are involved in human T-cell leukemia virus type 1 particle releaseJournal of Virology781503151210.1128/JVI.78.3.1503-1512.2004WeinbergJDrubinDG2012Clathrin-mediated endocytosis in budding yeastTrends in Cell Biology2211310.1016/j.tcb.2011.09.001WeissERPopovaEYamanakaHKimHCHuibregtseJMGottlingerH2010Rescue of HIV-1 release by targeting widely divergent NEDD4-type ubiquitin ligases and isolated catalytic HECT domains to GagPLOS Pathogens6e100110710.1371/journal.ppat.1001107WeissenhornWPoudevigneEEffantinGBassereauP2013How to get out: ssRNA enveloped viruses and membrane fissionCurrent Opinion in Virology315916710.1016/j.coviro.2013.03.011YeungMLHouzetLYedavalliVSJeangKT2009A genome-wide short hairpin RNA screening of jurkat T-cells for human proteins contributing to productive HIV-1 replicationThe Journal of Biological Chemistry284194631947310.1074/jbc.M109.010033YiCShenZStemmer-RachamimovADawanyNTroutmanSShoweLCLiuQShimonoASudolMHolmgrenLStangerBZKissilJL2013The p130 isoform of angiomotin is required for Yap-mediated hepatic epithelial cell proliferation and tumorigenesisScience Signaling6ra7710.1126/scisignal.2004060YuanBCampbellSBacharachEReinAGoffSP2000Infectivity of moloney murine leukemia virus defective in late assembly events is restored by late assembly domains of other retrovirusesJournal of Virology747250726010.1128/JVI.74.16.7250-7260.200010.7554/eLife.03778.022Decision letterDikicIvanReviewing editorGoethe University Medical School, Germany
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Angiomotin Functions in HIV-1 Assembly and Budding” for consideration at eLife. Your article has been favorably evaluated by Vivek Malhotra (Senior editor) and 3 reviewers, one of whom is a member of our Board of Reviewing Editors.
The Reviewing editor and the other reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
It was shown in the manuscript that AMOT proteins can simultaneously interact with NEDD4l E3 ligase and HIV-Gag protein and thereby facilitate budding of newly formed HIV virions. The hypothesis is that Nedd4L could be indirectly recruited to Gag via these AMOT proteins. This is an attractive hypothesis because it would explain earlier findings that “crippled” HIV-1 Gag proteins lacking PTAP and YPXL late domains can be induced to function for budding by overexpression of Nedd4L, despite the lack of PPxY motifs in Gag, or any other obvious way to recruit Nedd4L to Gag protein. Functional budding assays demonstrate that AMOT and Nedd4L cooperate to stimulate budding of the crippled HIV-1 Gag, and that AMOT depletion reduces budding of wt HIV-1. Budding could be restored through re-expression of AMOT, AMOTL1, or AMOTL2 from plasmids. An extensive EM analysis of budding HIV-1 particles upon AMOT depletion was performed. AMOT depletion leads to a block in particle formation that is earlier (half-moon stage) than the one caused by Tsg101 depletion (lollipop stage).
The quality of the data presented here is high and the reviewers were enthusiastic about the overall importance of the presented data. However, the reviewers agreed that the following issues need to be addressed before the manuscript can be considered for publication in eLife:
More information on the physiological significance AMOT-Gag interactions should be provided. For example, AMOT does not exist in many cell types and most of the studies required a Gag missing the late domains. Maybe AMOT only plays a role in Gag budding missing the late domains – but never plays a role in normal physiological assembly. What is the importance of AMOT for virus budding in different cell types, and what is the status of HIV budding in HeLa cells that are naturally deficient for AMOT. Could AMOT be acting on something else which, in turn, facilitates assembly? Does AMOT play a role normally in assembly of HIV-1, or only when they have late domains missing? Is Amot functioning at the sites of assembly? What is the relative timing of AMOT and ESCRTs. What is the evidence that AMOT is a pre-requisite for recruitment of ESCRTs rather than being a parallel process? When AMOT is depleted in the HEK293 cells and the half-moon shapes are formed, is ESCRT recruited to these half-moon shapes?
The interaction of NEDD4 and AMOT has been already shown previously, where NEDD4 ubiquitination of AMOT leads to its degradation via UPS. Authors as well demonstrate in one figure that overexpression of NEDD4 leads to decrease in total level of AMOT. However, in contradiction to this, they used NEDD4 overexpression in all experiments. One would expect that presence of AMOT negatively affects HIV release, by blocking access of NEDD4 and that overexpression of NEDD4 hence neutralizes overexpression of AMOT, by targeting it for the UPS degradation. How is than degradation of AMOT regulation during the HIV budding? Does presence of Gag neutralizes NEDD4 mediated degradation of AMOT? Could authors test this in their assays, and correlate the levels of AMOT in setting where NEDD4 is not overexpressed, and HIV budding is followed? In such setting both wt and dPTAP Gag should be tested, and compared.
Does increased concentration of HIVGag in the binding assays can outcompete NEDD4 binding? Further, the domain/motif in AMOT that binds to HIVGag should be identified, and this mutant should be tested in all budding/infection assays.
Minor comments:
Reviewer #1:
1) Does HIVGag negatively influences NEDD4 mediated ubiqutination of AMOT? Could authors test this in cells or in vitro. In case is so, does that support the fact that AMOT is stabilized upon HIV infection and only than is not degraded but facilitates ubiquitination of Gag?
2) AMOT contains BAR domain, however authors do not have an attempt to test the requirement of BAR domain in proposed process, despite the fact that BAR domains are known to be able to remodel/recognize membrane curvature. Figure 3 shows clearly that overexpression of AMOT leads to formation of large aggregate-like structures that are rather artefacts of overexpression than functional co-localization of AMOT with NEDD4. Authors should look closely on localization of endogenous AMOT in cells, and demonstrate its co-localization with HIV-Gag. Picture of Hiv-Gag localization should be provided in both wt and AMOT deficient cells as well.
3) Figure 9 should contain pictures of wt and rescued cells as well, and not only TSG101 and AMOT depleted cells.
4) It is not clear how does presence of AMOT affect the assembly of ESCRT machinery via influencing NEDD4. Considering that AMOT deficient cells are not even able to form the buds one can speculate that AMOT might be important for the actin cytoskeleton regulation and membrane curvature around the bud. Could authors provide TIRF images, showing recruitment of AMOT to HIV buds, and subsequent recruitment of ESCRT components, as they propose in the Ddiscussion. Currently there is no evidence that AMOT acts upstream of ESCRT beside the fact that AMOT binds NEDD4. Does ESCRT get recruited at all in case of AMOT deficient cells? Does AMOT precede recruitment of ESCRT when budding events are followed by TIRF?
5) If AMOT is required for NEDD4 recruitment and presumably subsequent ubiquitination, could authors also provide western blots showing levels of HIVGag ubiquitination in case of AMOT depletion or overproduction, in setting where only endogenous NEDD4 is present?
6) Figure 8A, right panel: more representative blot should be provided, since this one is not convincing at all. MA protein is not detected at all in viral particles, and the differences in CA level are almost not visible, going against the whole proposed mechanism that AMOT is able to surpass the lack of PTAP in crippled HIVGag.
7) Would be beneficial for the reader to have domain/motif comparison of all AMOTs (including AMOTL1 and AMOTL2) in Figure 1, or somewhere else in the figures. Based on which characteristic are those proteins redundant in terms of HIVGag budding.
Reviewer #2:
1) The binding experiments for the Nedd4L-AMOT interaction are carried out using a variety of constructs to localize binding determinants. Less so for the experiments characterizing the new interaction between Gag and AMOT. Have the authors tested whether AMOT p80 binds to Gag? If so, it would be useful to mention it, for example, on the third paragraph of the subsection “AMOT p130 is required for NEDD4L stimulation of HIV-1 budding”, in the Results section, referring to Figure 5–figure supplement 2: “Hence, Nedd4L stimulation… requires the presence of an AMOT p130 protein that can bind NEDD4L…”. The authors seem to attribute the lack of Nedd4L stimulation solely to lack of binding between p80 and NEDD4L. This is a case where it would be helpful to know if p80 binds to HIV-1 Gag. If not, it would mean that there are multiple defects in p80 that prevent it from functioning for the stimulation of Gag budding.
2) The sentence in the fourth paragraph of the subsection “AMOT p130 is required for NEDD4L stimulation of HIV-1 budding”, in the Results section, referring to Figure 5: “However, this stimulation was completely abrogated when endogenous AMOT proteins were depleted”. This is a key experiment in the paper. It is clear that full stimulation in these experiments requires both AMOT (endogenously expressed) and Nedd4L (from plasmid), and I believe that is the key point being made here. However, the statement that stimulation was completely abrogated when AMOTs were depleted may be a bit misleading. It implies that in the absence of AMOT, Nedd4L expression has no effect. But it does (compare lanes 2 and 4 of Figure 5). Please rephrase. This result is, perhaps, to be expected given that AMOTL1 and AMOTL2 are still present in these cells.
3) The sentence in the third paragraph of the subsection “AMOT p130 is required for NEDD4L stimulation 242 of HIV-1 budding”, in the Results section, referring to Figure 5–figure supplement 1: “the degree of rescue correlated positively with AMOT p130 re-expression levels…”. Certainly this appears to be true when looking at the infectivity assays in Figure 5–figure supplement 1, but such definitive language is not supported as well by the Gag budding experiment shown in the western blot of that figure.
4) The authors may want to point out that the partial rescue by AMOTL1 and AMOTL2 seen in Figure 8 could be related to differences in expression levels of the 3 AMOT proteins in the cells, which would be consistent with the dose dependence effect observed in Figure 7.
Reviewer #3:
It would help in the text to indicate which cells were used when (e.g., for the EM the cell type is not mentioned in the text nor the figure legend. It would help to have more information on the constructs used. The text says that HIV-1 was expressed using R9 wildtype. From looking back at a previous article that is referenced (Chung, 2008), it appears to be a pro-viral plasmid. However, that article does not have the R9 listed in its supplemental table of plasmids and references back to a paper from the year before (Fisher, 2007) that does not have the R9 listed. For Figure 4 and 5, there are panels called “virus”. Would it be better to call these “supernatant”? There are virions in both the cells and the supernatent. Why is there such a dramatic effect on infectivity when the effect on viral protein in the supernatant is not as pronounced?
It is claimed in the first paragraph of the subsection headed “AMOT binds directly to NEDD4L and to HIV-1 Gag2 in the Results section that NEDD4L binds AMOTp130 with at least 1:1 stoichiometry. How was this conclusion reached from the gel?
10.7554/eLife.03778.023Author response
[Editors’ note: the authors asked for clarification about the revision requirements prior to resubmission.]
We have first listed your final set of requirements and our responses to them. In addition, we have also addressed many of the comments made in the original reviews, and we have also described those cases.
Test the role of AMOT directly in physiologically relevant cells. If experiments with the T-cells are difficult, use macrophages instead.
As the reviewers pointed out, our original manuscript did not test the role of AMOT in cells that are physiologically for HIV-1 infection. To address this shortcoming, we have now tested how altering AMOT p130 levels affects the release of infectious HIV-1 particles from Jurkat T cells. These experiments were performed by comparing the effects of overexpressing either wild type AMOT p130 or an inactive AMOT p130 mutant with point mutations in the three NEDD4-binding “PPXY” motifs (AMOT p130ΔPPXY). As shown in the new Figure 6B, we find that overexpression of wild type AMOT p130 strongly stimulates HIV release and infectivity in a dose-dependent fashion, with maximal stimulations of 16-fold (virion release) and 7-fold (viral infectivity) vs. a transfected vector control. In contrast, overexpression of AMOT p130ΔPPXY reproducibly reduced HIV-1 release and infectivity, presumably by dominantly inhibiting the functions of endogenous AMOT p130 proteins. We have confirmed that the levels of HIV-1 protein expression are comparable in all cases (not shown). AMOT expression levels in Jurkat T cells are beneath our detection limits, but we feel confident in concluding that AMOT stimulates HIV-1 release in this setting because the experiment is well controlled by the use of an inactive point mutant AMOT p130 construct, because the effects are dose dependent, and because we repeated the experiment three times with similar results.
In the process of investigating how best to perform this experiment, we also noticed that one of published the “global” siRNA screens for HIV cofactors was performed in Jurkat T cells (Yeung et al., 2009). Gratifyingly, the data from that screen also indicate that AMOT is important for HIV-1 replication in Jurkat T cells because an shRNA that targeted AMOT protected Jurkat cells against the cytotoxicity that occurs upon HIV infection (although the functional role of AMOT was not pursued further). Thus, there are now two independent lines of evidence that AMOT p130 is important for HIV-1 replication in Jurkat T cells (and we have, of course, now referenced Yeung et al, 2009).
Explain why Tsg101 knockdown is so much more effective than AMOT in HIV infectivity.
This is also a good point. We hypothesized that the strong, but incomplete effects that we were seeing might reflect incomplete silencing of endogenous AMOT p130. We therefore repeated the experiment with a new, freshly repurified batch of siRNA. Importantly, this new AMOT p130 siRNA depleted both AMOT p130 and p80 even more efficiently than our previous batch of siRNA (see new Figure 5, panel 3, lane 4). As shown in the new Figure 5, this siRNA treatment reduces viral titers by approximately 8-fold, and the inhibition of virion release is approaching the inhibition that we see for our best siRNA against TSG101 (right panels, compare lanes 3 and 4). We therefore conclude that most (or all) of the residual virus release and infectivity that we saw in previous experiments was due to incomplete silencing of AMOT p130 (and/or the residual presence of AMOT L1 and L2). These data indicate that AMOT p130 is (nearly) as important as TSG101 for HIV-1 release. Finally, we note that: 1) In addition to reducing viral budding, TSG101 depletion also reduces HIV-1 infectivity by inhibiting Gag processing and viral maturation, whereas AMOT depletion only appears to inhibit by blocking budding (see the data in Figure 9 and the discussion below), and 2) the level of infectivity seen upon depletion of either AMOT p130 or TSG101 is significantly greater than the infectivity reduction seen upon depletion of the third known HIV budding factor, ALIX (Figure 5).
Please note that we have not gone back and repeated all of our siRNA rescue experiments with the new batch of siRNA because this would have required repeating a very large number of experiments and because the new experiment did not change our original conclusions.
Test the effect of AMOT expression with HIV-1 R-9 in Hela cells.
This experiment, i.e., examining the release of wild type HIV-1NL4-3 from HeLa cells (expressed from the R9 proviral expression vector), was already shown in Figure 7 of the original paper (now Figure 6A). As we have discussed, the misunderstanding arose from an error that we made in mislabeling the HIV-1 construct in the figure caption (for which we apologize) and also from an apparent misunderstanding of the nomenclature that we used within the figure (and elsewhere). Viruses and proteins labeled “HIV-1” and “Gag” without any subscripts refer to the wild type virus and Gag protein, respectively. Viruses and Gag proteins that lack late domains are labeled with the subscripts “ΔPTAP, ΔYP”. We apologize for the confusion, and note that AMOT expression strongly stimulates release of wild type HIV-1 in HeLa cells.
More information on the physiological significance AMOT-Gag interactions should be provided. For example, AMOT does not exist in many cell types.
Firstly, we note that AMOT is expressed in white blood cells and lymphoid tissues—the normal host cells for HIV-1 infection—and we have provided this information and relevant references in the manuscript. AMOT is not expressed in HeLa cells (but AMOTL2 is), and we show that exogenous expression of AMOT stimulates virus release from HeLa cells (Figure 6A). AMOT is expressed in Jurkat T cells (and this point is now made explicitly and referenced in the text, but AMOT nevertheless appears to be limiting because exogenous AMOT expression also stimulates HIV release from Jurkat T cells (new Figure 6B).
Reviewer #1:
Figure 3shows clearly that overexpression of AMOT leads to formation of large aggregate-like structures that are rather artefacts of overexpression than functional co-localization of AMOT with NEDD4.
We agree (and had pointed that out ourselves), but the reviewer’s point is well taken and we have therefore removed this figure from the manuscript and simply mentioned the co-localization as “data not shown”. We think there is no doubt that AMOT p130 and NEDD4L bind one another in cells, and do so through interactions between the AMOT p130 PPXY motifs and the NEDD4L WW domains. We have shown this in several ways in our paper, including by co-IP (Figure 2). We have also referenced other relevant papers showing that motins bind NEDD4 family members.
Figure 9should contain pictures of wt and rescued cells as well, and not only TSG101 and AMOT depleted cells.
We, and others, have previously published EM images of thin-sectioned cells that show budding of wild type HIV-1 particles. Wild type virions bud efficiently, however, and it is therefore difficult to obtain images from single thin sections that contain numerous examples of budding wild type virions. To help address this limitation, we have turned to scanning electron microscopy (SEM). The advantage of SEM is that we can look at the entire surface of a cell and therefore capture many more budding events in a single image. We have now included such images for virions budding from control cells and from cells depleted of TSG101 and AMOT (new Figure 8). These images make it clear that there is an enormous difference between the efficiency of virus budding from wild type cells vs. cells that lack either TSG101 or AMOT.
Would be beneficial for the reader to have domain/motif comparison of all AMOTs (including AMOTL1 and AMOTL2) inFigure 1, or somewhere else in the figures. Based on which characteristic are those proteins redundant in terms of HIVGag budding.
We agree and have now included the requested domain comparison within the modified Figure 1.
Reviewer #2:
The binding experiments for the Nedd4L-AMOT interaction are carried out using a variety of constructs to localize binding determinants. Less so for the experiments characterizing the new interaction between Gag and AMOT. Have the authors tested whether AMOT p80 binds to Gag? If so, it would be useful to mention it, for example, on the third paragraph of the subection “AMOT p130 is required for NEDD4L stimulation of HIV-1 budding”, in the Results section, referring to Figure 5–figure supplement 2: “Hence, Nedd4L stimulation… requires the presence of an AMOT p130 protein that can bind NEDD4L…”. The authors seem to attribute the lack of Nedd4L stimulation solely to lack of binding between p80 and NEDD4L. This is a case where it would be helpful to know if p80 binds to HIV-1 Gag. If not, it would mean that there are multiple defects in p80 that prevent it from functioning for the stimulation of Gag budding.
We have not tested AMOT p80 binding to Gag because the p80 isoform does not express well and we have therefore been unable to make pure recombinant AMOT p80. The reviewer is therefore correct that there may be multiple reasons that the p80 isoform does not support virus release. We believe, however, that we have shown that the lack of NEDD4L binding is sufficient to explain the lack of function of AMOT p80. The paragraph that the reviewer refers to also describes an experiment in which we show that an AMOT p130 protein with substitution mutations in the three PPXY motifs also fails to support HIV budding (Figure 4, AMOT p130ΔPPXY, right panels, compare lanes 6 and 5) and we therefore feel justified in concluding that NEDD4L stimulation requires the presence of an AMOT p130 protein that can bind NEDD4L.
The sentence in the fourth paragraph of the subection “AMOT p130 is required for NEDD4L stimulation of HIV-1 budding”, in the Results section, referring toFigure 5: “However, this stimulation was completely abrogated when endogenous AMOT proteins were depleted”. This is a key experiment in the paper. It is clear that full stimulation in these experiments requires both AMOT (endogenously expressed) and Nedd4L (from plasmid), and I believe that is the key point being made here. However, the statement that stimulation was completely abrogated when AMOTs were depleted may be a bit misleading. It implies that in the absence of AMOT, Nedd4L expression has no effect. But it does (compare lanes 2 and 4 ofFigure 5). Please rephrase. This result is, perhaps, to be expected given that AMOTL1 and AMOTL2 are still present in these cells.
The reviewer makes a good point. We have rephrased the sentence to read: “However, virus release was reduced to the levels seen in untreated control cells when endogenous AMOT proteins were simultaneously depleted (compare lanes 2-4).
The sentence in the third paragraph of the subsection “AMOT p130 is required for NEDD4L stimulation 242 of HIV-1 budding”, in the Results section, referring to Figure 5–figure supplement 1: “the degree of rescue correlated positively with AMOT p130 re-expression levels…”. Certainly this appears to be true when looking at the infectivity assays in Figure 5–figure supplement 1, but such definitive language is not supported as well by the Gag budding experiment shown in the western blot of that figure.
We take the reviewer’s point, and have modified the sentence as follows: “the degree of rescue generally correlated positively with AMOT p130 re-expression levels.”
The authors may want to point out that the partial rescue by AMOTL1 and AMOTL2 seen inFigure 8could be related to differences in expression levels of the 3 AMOT proteins in the cells, which would be consistent with the dose dependence effect observed inFigure 7.
We agree, and have added the following sentence that addresses this point: “Neither AMOTL1 nor AMOTL2 was quite as effective as AMOT p130 in rescuing HIV-1 infectivity in this assay, possibly because these motins are intrinsically less active than AMOT p130 and/or because they were expressed at lower levels (left panel 2, compare lanes 3-5).”
Reviewer #3:
It would help in the text to indicate which cells were used when (e.g., for the EM the cell type is not mentioned in the text nor the figure legend.
We have now defined the cell type used in every experiment.
It would help to have more information on the constructs used. The text says that HIV-1 was expressed using R9 wildtype. From looking back at a previous article that is referenced (Chung, 2008), it appears to be a pro-viral plasmid. However, that article does not have the R9 listed in its supplemental table of plasmids and references back to a paper from the year before (Fisher, 2007) that does not have the R9 listed.
The reviewer is correct that R9 is a wild type proviral HIV-1 construct. In the previous version of the manuscript, there were two references for the relevant sentence of the experimental methods. One (Swingler et al., 1997) referred to the first report of the R9 proviral plasmid, and the other (Chung et al., 2008) referred to our previous report of the NEDD4L plasmids used in this study. We have now separated the two references within the sentence to remove any confusion.
ForFigure 4 and 5, there are panels called “virus”. Would it be better to call these “supernatant”? There are virions in both the cells and the supernatent.
We understand the point, but prefer to use the term “virus” to remain consistent with other papers in the field.
Why is there such a dramatic effect on infectivity when the effect on viral protein in the supernatant is not as pronounced?
This is a good question. We (and others) generally observe a correlation between virion release and infectivity measurements, but the correlation is not perfect. This is why we believe it is important to measure both release and infectivity. One cause of these discrepancies is that budding defects can also induce defects in viral maturation and in virion contents. This is well characterized in the case of TSG101 depletion, where defective Gag processing and viral maturation contribute to loss of infectivity, even when virions are released. This effect is evident in the control experiment shown in Figure 9. Another effect that we have documented for TSG101 is that the released virions have reduced levels of reverse transcriptase. We believe that this is because even when virus budding does occur it is slow, and the processed RT proteins “float away” during the prolonged budding step. We envision that a similar effect probably occurs when AMOT depletion slows virus release, though we have not tested this directly. The important point is that aberrant kinetics of virion release appears to reduce infectivity, whereas any virions that eventually escape the cell will contribute to the signal in the release assay. Thus, treatments that disrupt (or restore) wild type virus budding kinetics can have differential effects on virion release and infectivity.
It is claimed in the first paragraph of the subsection headed “AMOT binds directly to NEDD4L and to HIV-1 Gag2, in the Results section, that NEDD4L binds AMOTp130 with at least 1:1 stoichiometry. How was this conclusion reached from the gel?
NEDD4L is a smaller protein than OSF-AMOT p130, yet the intensities of the Coomassie-stained NEDD4L band are significantly darker than the intensities of the OSF-AMOT p130 bands in the pulled down complexes shown in Figure 2B, panel 1, lanes 9 and 10. Thus, there must be at least as many molecules of NEDD4L as AMOT p130 in the complexes.
diff --git a/elife04363.xml b/elife04363.xml
new file mode 100644
index 0000000..bfb7cfb
--- /dev/null
+++ b/elife04363.xml
@@ -0,0 +1 @@
+elifeeLifeeLifeeLife2050-084XeLife Sciences Publications, Ltd0436310.7554/eLife.04363Registered reportDevelopmental biology and stem cellsHuman biology and medicineRegistered report: Tumour vascularization via endothelial differentiation of glioblastoma stem-like cellsChroscinskiDenise1SampeyDarryl2MaheraliNimet3group-author-id1Reproducibility Project: Cancer Biology*Noble Life Sciences, Gaithersburg, Maryland, United StatesBioFactura, Frederick, Maryland, United StatesHarvard Stem Cell Institute, Cambridge, Massachusetts, United StatesOkanoHideyukiReviewing editorKeio University School of Medicine, Japangroup-author-id1IornsElizabethScience Exchange, Palo Alto, Californiagroup-author-id1GunnWilliamMendeley, London, United Kingdomgroup-author-id1TanFraserScience Exchange, Palo Alto, Californiagroup-author-id1LomaxJoelleScience Exchange, Palo Alto, Californiagroup-author-id1ErringtonTimothyCenter for Open Science, Charlottesville, VirginiaFor correspondence: joelle@scienceexchange.com2502201520154e043631408201427012015
Ricci-VitianiLPalliniRBiffoniMTodaroMInverniciGCenciTMairaGParatiEAStassiGLaroccaLMDe MariaR. 09122010. Tumour vascularization via endothelial differentiation of glioblastoma stem-like cells. Nature468:824–828. doi: 10.1038/nature09557.
The Reproducibility Project: Cancer Biology seeks to address growing concerns about reproducibility in scientific research by conducting replications of 50 papers in the field of cancer biology published between 2010 and 2012. This Registered report describes the proposed replication plan of key experiments from ‘Tumour vascularization via endothelial differentiation of glioblastoma stem-like cells’ by Ricci-Vitiani and colleagues, published in Nature in 2010 (Ricci-Vitiani et al., 2010). The experiments that will be replicated are those reported in Figure 4B and Supplementary Figure 10B (Ricci-Vitiani et al., 2010), which demonstrate that glioblastoma stem-like cells can derive into endothelial cells, and can be selectively ablated to reduce tumor progression in vivo, and Supplementary Figures S10C and S10D (Ricci-Vitiani et al., 2010), which demonstrate that fully differentiated glioblastoma cells cannot form functionally relevant endothelium. The Reproducibility Project: Cancer Biology is a collaboration between the Center for Open Science and Science Exchange, and the results of the replications will be published by eLife.
DOI:http://dx.doi.org/10.7554/eLife.04363.001
Author keywordsReproducibility Project: Cancer Biologymethodologyglioblastoma multiformecancer stem cellsendotheliumResearch organismmouseLaura and John Arnold FoundationReproducibility Project: Cancer BiologyThe funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Introduction
Glioblastoma multiforme (GBM) is a highly aggressive form of cancer characterized by an extensive network of vasculature that contributes to its invasiveness. However, the mechanisms of angiogenesis and the origin of the tumor vasculature remain poorly understood. While conventional theory suggests that GBM tumor vasculature derives from existing vessels or from bone marrow progenitor cells, recent studies have indicated that this is not the case (Purhonen et al., 2008). Indeed, the tumor endothelium may actually be derived from the tumor itself. GBM is maintained via a population of self-renewing, tumorigenic cancer stem cells (CSCs), which have been implicated in tumor invasion and metastasis for a large variety of cancers (Vescovi et al., 2006; Fan et al., 2013). The progeny of these CSCs are not confined to a neural lineage but rather can differentiate into functional endothelium. Based on the work of Ricci-Vitiani et al., it appears that part of the vasculature in GBM originates from tumor cells, bypassing the normal mechanisms of angiogenesis. These findings offer insights into tumor self-renewal, and offer new options for cancer treatment by targeting the differentiation of tumor cells into endothelial progeny (Ricci-Vitiani et al., 2010; Kaur and Bajwa, 2014).
In order to demonstrate the CSC lineage of GBM tumor vasculature, Ricci-Vitiani et al. first traced the genetic lineage of the tumor endothelium. They analyzed the vasculature in 15 human glioblastoma patient samples and found that a large subset of endothelial cells harbored the same mutations and chromosomal aberrations as the tumors themselves. They also showed that in culture, glioblastoma stem-like cells (GSCs) could be differentiated to express multiple endothelial markers, and showed substantial tube-forming ability, whereas fully differentiated glioblastoma cell lines could not. Ricci-Vitiani et al. also traced the lineage of tumor vasculature in vivo, confirming the presence of human endothelial cells expressing green fluorescent protein (GFP) in a mouse xenograft of human GSCs expressing GFP. These experiments, and others, demonstrated that tumor xenografts obtained by injection of human glioblastoma neurospheres developed an intrinsic vascular network composed of tumor-derived endothelial cells (Ricci-Vitiani et al., 2010).
The authors next sought to investigate whether the GSC-derived endothelial cells contributed to tumor growth. This key finding is the focus of this replication study. The authors transduced glioblastoma neurospheres with a lentiviral vector containing the herpes simplex virus thymidine kinase gene (tk) under the control of the transcription regulatory elements of the endothelial marker Tie2. In this way, the tumor-derived endothelial cells could be selectively killed by exposure to ganciclovir. As negative controls, the authors used neurospheres transduced with an empty viral vector, as well as the differentiated glioblastoma cell line U87MG. As a positive control, they used neurospheres and U87MG cells transduced with a vector conferring constitutive expression of tk (PGK-tk). Upon administration of ganciclovir, selective targeting of endothelial cells generated by GSCs in mouse xenografts resulted in tumor reduction and degeneration, indicating the functional relevance of the GSC-derived endothelial vessels.
Prior to subcutaneous injection of transduced glioblastoma neurospheres, Ricci-Vitiani et al. first confirmed the lack of endogenous Tie2 expression in both the GSCs and the U87MG cells. This quality control step will be replicated in Protocol 1, the results of which will be compared to Figure S11C. Next, the viral transduction of GSCs and U87MG cells with expression constructs for Tie2-tk and PGK-tk, as well as empty viral vector, will be replicated in Protocol 2. The generation and analysis of xenografts using various cell lines in immunocompromised mice will be replicated in Protocol 3. This protocol will generate data that can be compared to original data presented in Figures 4B, S10B, S10C, and S10D.
Recently, multiple other studies have explored the phenomenon of tumor-derived vasculature in GBM. Two recent reports found results very similar to those of Ricci-Vitiani et al., showing that oncogene-induced glioblastoma tumors gave rise to tumor-derived endothelial cells, as indicated by GFP expression. These studies also found that endothelial cells within tumors harbored the same genetic signature as the tumor itself (Wang et al., 2010; Soda et al., 2011). Similarly, Chiao et al. reported that GCSs formed vasculogenic mimicry in tumor xenografts and expressed pro-vascular molecules (Chiao et al., 2011). However, other groups have found that endothelial cells comprising GBM vasculature do not share the same genetic make-up as neoplastic tissues, and that GSCs routinely do not give rise to endothelial cells (Rodriguez et al., 2012; Cheng et al., 2013). Interestingly, Cheng et al. present an alternative hypothesis to that of Ricci-Vitiani et al. by showing that GSCs can give rise to vascular pericytes—which also express Tie2 (De Palma et al., 2005)—rather than endothelial cells. Targeting these GSC-derived pericytes disrupted vessel function and inhibited tumor size similarly as the results presented by Ricci-Vitiani et al. for targeting endothelial cells (Cheng et al., 2013). El Hallani et al. demonstrated that GSCs were capable of vasculogenesis in vitro, and that a fraction of GSCs could transdifferentiate into vascular smooth muscle-like cells. However, their later work suggested that rather than transdifferentiating, the GSCs were fusing with endothelial cells to create a hybrid tumor vasculature (El Hallani et al., 2010; El Hallani et al., 2014). Conversely, using a GSC mouse xenograft, Lathia et al. did not observe the integration of tumor-derived cells into the vascular wall; however, this observation was only reported in the text. No data were shown (Lathia et al., 2011). Ghanekar et al. tried analogous experiments using hepatocellular carcinoma CSCs and did not find any evidence that the tumor cells gave rise to the endothelium (Ghanekar et al., 2013).
Materials and methods
Unless otherwise noted, all protocol information was derived from the original paper, references from the original paper, or information obtained directly from the authors.
Protocol 1: Evaluation of <italic>Tie2</italic> expression in various cell lines using qPCR
This protocol evaluates the expression of the endothelial marker Tie2 in three cell lines using semi-quantitative PCR: patient-derived glioblastoma neurospheres (GSC83), human glioblastoma cell line U87MG, and normal human dermal microvascular endothelial cells (HMVEC-d). The expression of Tie2 will be normalized against the endogenous expression of 18S rRNA. Expression of Tie2 is expected to be very low in GSC83 and U87MG cells, and robust in the endothelial cell line HMVEC-d, as depicted in Figure S11C. This protocol serves as a quality control step to ensure the lack of Tie2 expression in the glioblastoma cell lines used later in the study.
Sampling
This experiment will be performed three times (biological replicates) with each run using two technical replicates, for a final power of at least 80%.
A. Test conditions:
i. qRT-PCR of Tie2 (and 18S rRNA) from GSC83 glioblastoma neurospheres.
ii. qRT-PCR of Tie2 (and 18S rRNA) from U87MG cells.
iii. qRT-PCR of Tie2 (and 18S rRNA) from HMVEC cells.
Materials and reagents
Reagent
Type
Manufacturer
Catalog #
Comments
GSC83 glioblastoma neurospheres
Human cell line
N/A
N/A
Reagent being provided by original authors
U87MG glioblastoma cells
Human cell line
N/A
N/A
Reagent being provided by original authors
Normal human dermal microvascular endothelial cells (HMVEC-d)
Human cell line
N/A
N/A
Reagent being provided by original authors
25 cm2 tissue culture flasks
Labware
Corning
3289
Original brand not specified
Dulbecco's Modified Eagle's medium (DMEM)/F-12, no glutamine
Cell culture reagent
Sigma-Aldrich
D6421
Replaces Gibco cat. no. 21331-020 used in original study
Human recombinant epidermal growth factor (EGF)
Cell culture reagent
Sigma–Aldrich
E5036
Replaces Peprotech cat no. AF-100-15 used in original study
Human recombinant basic fibroblast growth factor (FGF2)
Cell culture reagent
Sigma–Aldrich
F0291
Replaces Peprotech cat no. AF-100-18B used in original study
Glutamine
Cell culture reagent
Sigma–Aldrich
G7513
Replaces Gibco cat no. 25030-081 used in original study
Glucose
Cell culture reagent
Sigma–Aldrich
49163
Putrescine
Cell culture reagent
Sigma–Aldrich
P5780
Progesterone
Cell culture reagent
Sigma–Aldrich
P6149
Sodium selenite
Cell culture reagent
Sigma–Aldrich
S9133
Insulin
Cell culture reagent
Sigma–Aldrich
I9278
Transferrin
Cell culture reagent
Sigma–Aldrich
T8158
Dulbecco's Modified Eagle's medium (DMEM), high glucose
Replaces Invitrogen cat. no. 15596-018 (Trizol) used in original study
First-Strand cDNA Synthesis Kit
cDNA synthesis
GE Healthcare (Sigma-Aldrich)
GE27-9261-01
Replaces Invitrogen cat. no. 28025-013 used in original study
96-Well qPCR plates
qPCR
Specific brand information will be left up to the discretion of the replicating laboratory and recorded later
TaqMan Gene Expression Master Mix
qPCR
Applied Biosystems
4369016
TaqMan Gene Expression Assay Hs00945155_m1
TaqMan probe
Applied Biosystems
4331182
18S rRNA Endogenous Control 4319413E
TaqMan probe
Applied Biosystems
4319413E
StepOnePlus Real-Time PCR System
Equipment
Applied Biosystems
Procedure
Note: All cell lines will be sent for STR profiling and mycoplasma testing.
Culture GSC83 glioblastoma neurospheres, U87MG cells, and HMVEC cells in 25 cm2 tissue culture flasks at 37°C at 5% CO2.
A. GSC83 cells should be plated at 20,000 cell/ml and subcultured once every 7 days at the same plating number. One week is sufficient time for two doublings to occur.
i. Cells should be cultured in stem cell medium consisting of serum-free Dulbecco's Modified Eagle's Medium (DMEM)/F-12 containing:
a. 20 ng/ml human recombinant epidermal growth factor (EGF).
b. 10 ng/ml human recombinant basic fibroblast growth factor (FGF2).
c. 2 mM glutamine.
d. 0.6% glucose.
e. 9.6 µg/ml putrescine.
f. 6.3 ng/ml progesterone.
g. 5.2 ng/ml sodium selenite.
h. 0.025 mg/ml insulin.
i. 0.1 mg/ml transferrin.
B. U87MG cells should be cultured in 25 cm2 tissue culture flasks in DMEM with 10% FBS.
i. Subculture cells at a ratio of 1:2 to 1:5; renew medium 2–3 times per week.
C. HMVEC-d (normal human dermal microvascular endothelial cells) should be cultured in 25 cm2 tissue culture flasks in Endothelial Growth Medium-2 Microvascular (EGM-2MV).
i. Subculture cells when they are 70–80% confluent; change growth media every other day.
Split each cell line into three separate 25 cm2 flasks. These separate flasks constitute biological replicates for eventual downstream gene expression analysis.
A. Allow cells to grow to log phase.
Isolate total RNA from the cells in each 25 cm2 flask (nine flasks in total) according to the manufacturer's instructions for TRI reagent. For each sample, harvest the entire population of cells in the flask.
A. Report total concentration and purity of isolated total RNA.
Reverse transcribe mRNA to cDNA according to the manufacturer's protocol.
A. Use 500 ng total RNA for each 20 µl reaction.
B. Use oligo(dT)12–18 primer for first-strand synthesis.
C. Add ribonuclease inhibitor at suggested step in the protocol.
D. Perform RNase H digestion at suggested step in the protocol.
Perform qPCR to assess Tie2 expression levels across cell types using a StepOnePlus Real-Time PCR System. Use 18S rRNA as an endogenous control. Perform duplicate technical replicates for each biological replicate (3 biological × 2 technical × 2 genes = 12 wells per cell line).
A. Use 1 µl of undiluted cDNA mixture for each reaction.
B. Use TaqMan probes for Tie2 and 18S rRNA (see reagent table).
C. Use an initial denaturation at 95°C for 10 min, following by 40 cycles of 95°C for 15 s; 60°C for 1 min.
Analyze and compute ΔΔCT values.
Deliverables
Data to be collected:
A. Purity (A260/280 and A260/230 ratios) and concentration of isolated total RNA from cells.
B. Raw qRT-PCR values, as well as analyzed ΔΔCT values and bar graph of Tie2 mRNA normalized to control mRNA levels for each condition (compare to Figure S11C).
Confirmatory analysis plan
This replication attempt will perform the statistical analyses listed below, compute the effect sizes, compare them against the reported effect size in the original paper and use a meta-analytic approach to combine the original and replication effects, which will be presented as a Forest plot.
Statistical analysis of the replication data:
A. One-way ANOVA to analyze the means of GSC83, U87MG, and HMVEC.
i. We will then perform a Fisher's LSD test to perform multiple pairwise comparisons:
a. GSC83 compared to HMVEC.
b. U87MG compared to HMVEC.
c. GSC83 compared to U87MG (sensitivity).
Known differences from the original study
In the original study, multiple human glioblastoma neurospheres were screened for Tie2 expression. The human glioblastoma cell lines U251 and T98G were also analyzed, as well as the human endothelial cell line HUVEC. This replication study will be using a single established glioblastoma neurosphere cell line (GSC83) provided by the authors. The authors will also provide their U87MG and HMVEC cell lines. All known differences in reagents and supplies are listed in the ‘Materials and reagents’ section above, with the originally used item listed in the ‘Comments’ section. All differences have the same capabilities as the original and are not expected to alter the experimental design.
Provisions for quality control
The cell lines used in this experiment will undergo STR profiling to confirm their identity and will be sent for mycoplasma testing to ensure there is no contamination. The sample purity (A260/280 and A260/230 ratios) of the isolated RNA from each sample will be reported. All data obtained from the experiment—raw data, data analysis, control data, and quality control data—will be made publicly available, either in the published manuscript or as an open access dataset available on the Open Science Framework project page for this study (https://osf.io/mpyvx/).
Protocol 2: Lentiviral infection of glioblastoma cells and stable cell generation
This protocol describes the methods necessary to virally transduce GSC83 glioblastoma neurospheres, as well as U87MG cells, with thymidine kinase expression constructs. The protocol first details the production of three different lentivirus strains (PGK-tk, Tie2-tk, and an empty viral vector), and then explains the techniques necessary to transduce the two human glioblastoma cell lines. Finally, the protocol includes methodology associated with assessing the transduction efficiency of glioblastoma cell lines via flow cytometry analysis as a quality control check.
Materials and reagents
Reagent
Type
Manufacturer
Catalog #
Comments
GenElute Endotoxin-free Plasmid Maxiprep Kit
Reagent
Sigma–Aldrich
PLEX15-1KT
Original brand not specified
pCMV-dR8.74
Viral packaging vector
N/A
N/A
Reagent being provided by original authors
pMD2G
Viral packaging vector
N/A
N/A
Reagent being provided by original authors
pRRLsin.Tie2p.TK.PGKp.GFP.spre
DNA construct
N/A
N/A
Reagent being provided by original authors
pRRLsin.PGKp.TK.PGKp.GFP.spre
DNA construct
N/A
N/A
Reagent being provided by original authors
pRRLsin.PGKp.GFP.spre
DNA construct
N/A
N/A
Reagent being provided by original authors
BamHI
Restriction enzyme
Sigma–Aldrich
R0260
NdeI
Restriction enzyme
Sigma–Aldrich
R5509
SpeI
Restriction enzyme
Sigma–Aldrich
R5257
SmaI
Restriction enzyme
Promega
R6121
75 cm2 tissue culture flask
Labware
Corning
430641U
Original brand not specified
HEK293T cells
Cell line
ATCC
CRL-3216
Dulbecco's Modified Eagle's Medium (DMEM), high glucose
Cell culture reagent
Sigma–Aldrich
D6429
Original brand not specified
Fetal bovine serum (FBS)
Cell culture reagent
Sigma–Aldrich
F2442
Original brand not specified
GSC culture media
See reagent list from Protocol 1 for a complete list of medium components
2× HEPES buffered saline (HBS)
Reagent
Sigma–Aldrich
51558
Replaces laboratory-made buffer used in original study. pH of substituted buffer is 7.2; pH of original buffer was 7.05
Calcium chloride dihydrate
Reagent
Sigma–Aldrich
C7902
Original brand not specified
Hexadimethrine bromide (polybrene)
Cell culture reagent
Sigma–Aldrich
107689
Original brand not specified
6-Well tissue culture plates
Labware
Corning
3516
Original brand not specified
7-Amino actinomycin D (7-AAD)
Flow reagent
Life Technologies
A1310
Original brand not specified
Tubes used for flow cytometry
Specific brand information will be left up to the discretion of the replicating laboratory and recorded later
FACSCalibur flow cytometry instrument
Equipment
Becton Dickinson
Sampling
Outline of experimental endpoints:
A. At the end of this protocol, we will have generated GSC83 glioblastoma neurospheres and U87MG cells stably expressing:
i. Empty vector.
ii. PGK-tk transduced.
iii. Tie2-tk transduced.
B. Total: six stable cell lines.
Protocol
Grow and prepare endotoxin-free plasmid constructs according to the manufacturer's protocol for the GenElute Endotoxin-free Plasmid Maxiprep Kit.
A. Viral packaging vectors:
i. pCMV-dR8.74 (∼50 µg DNA needed for production of three viruses).
ii. pMD2G (∼30 µg DNA needed production of three viruses).
B. DNA construct expression vectors:
i. pRRLsin.Tie2p.TK.PGKp.GFP.spre (∼30 µg DNA needed for virus production).
ii. pRRLsin.PGKp.TK.PGKp.GFP.spre (∼30 µg DNA needed for virus production).
iii. pRRLsin.PGKp.GFP.spre (∼30 µg DNA needed for virus production).
Perform restriction digestions on an aliquot of purified plasmid to check vector integrity for Tie2-tk and PGK-tk plasmids. Following digestion, run digested bands on an agarose gel to visualize band pattern.
A. For Tie2-tk vector (12.13 kb), digest vector with NdeI, BamHI, and SmaI.
On Day 1 of viral production, plate 1.2 × 106 HEK293T cells in a 75 cm2 tissue culture flask.
A. HEK293T cells should be maintained in DMEM supplemented with 10% FBS at 37°C with 5% CO2.
On Day 2, replace the cell medium with 18 ml of fresh medium. Prepare a transfection master mix for each of the DNA construct vectors.
A. Assemble the following components in a 15 ml polypropylene tube in the following order:
i. 20 µg of DNA construct expression vector.
ii. 13 µg of pCMV-dR8.74 packaging vector.
iii. 7 µg of pMD2G envelope vector.
iv. 150 µl of 2M CaCl2.
v. Bring volume to 1 ml with ddH2O.
vi. Add 1 ml of 2× HEPES buffered saline (HBS) and aerate solution for 20–30 s with a 2 ml pipette.
Immediately add the 2 ml transfection master mix directly to HEK293T cells by dropping slowly and evenly into the media, covering as much of the flask as possible.
A. Do not mix.
B. Place flasks in a 37°C incubator for 14–16 hr.
Day 3: after 14–16 hr, change media to remove DNA precipitate.
Day 4: 48 hr after transfection, collect viral supernatant, filter through a 0.45 µm syringe filter, and freeze in liquid nitrogen. Store at −80°C until use.
Culture GSC83 glioblastoma neurospheres and U87MG cells as described in Protocol 1.
A. Plate 150,000–200,000 cells in a 6-well plate.
B. Cells should be exponentially growing at time of lentiviral infection.
Infect GSC83 neurospheres and U87MG cells with lentivirus:
A. Add viral supernatant (1 ml/50,000 cells) along with 4 µg/ml polybrene to each well.
B. Spin plate of cells at 1800 rpm in centrifuge for 45 min.
C. Incubate cells for 75 min in a 5% CO2/37°C incubator.
D. Wash cells twice with culture medium, then add fresh serum-free media.
E. Seed cells into a 25 cm2 tissue culture flask at 20,000 cells per ml.
F. Allow cells to grow for 48 hr.
Evaluate infection efficiency at 48 hr post infection by flow cytometry.
A. Remove an aliquot of cells (20,000–50,000 cells) from each flask. Untransduced cells (both GSC83 and U87MG) should also be prepared for use as a negative control.
i. Pull down the cells by centrifuging each flask at ≤1000 rpm.
ii. Remove the supernatant, leaving approximately 150–200 µl of media in the flask.
iii. Use a P200 pipette to gently dissociate the cells. Pipette up and down several times to obtain a single cell suspension.
iv. Save an aliquot for flow analysis, and passage the remaining cells into a new flask to expand them for further experiments.
B. Incubate the freshly dissociated cells for 5 min with 7-amino actinomycin D (7-AAD; final concentration 5 µg/ml).
C. Analyze cells for GFP expression using a FACSCalibur instrument. Apply the following sequential gates to the dot plots to select viable cells:
i. FSC area/SSC area.
ii. SSC width/SSC area to exclude aggregates.
iii. FSC area/7-AAD area to select viable cells.
iv. Untransduced cells serve as a negative control.
D. Plot fluorescent protein expression in gated cells using bivariate plots.
Determine the percentage of transduced cells positive for GFP reporter expression in each population of cells.
A. Exclusion criteria: Expression should be ≥80% positive in each transduced population in order for cells to be used for xenograft injection.
Continue to expand and culture cells until ready for injection into immunocompromised mice.
Deliverables
Data to be collected:
A. Purity (A260/280 and A260/230 ratios) and concentration of plasmid DNA.
B. Agarose gel image of restriction-digested Tie2-tk and PGK-tk plasmids with molecular weight marker.
C. FACS plots of virally-transduced GSC83 and U87MG cells.
D. Achieved transduction efficiency (%GFP+ cells) for GSC83 and U87MG cells.
Samples delivered for further analysis:
A. Infected neurosphere and U87MG clones (see Protocol 3).
Confirmatory analysis plan
Statistical analysis of the replication data:
Not applicable.
Known differences from the original study
All known differences in reagents and supplies are listed in the ‘Materials and reagents’ section above, with the originally used item listed in the ‘Comments’ section. All differences have the same capabilities as the original and are not predicted to alter experimental outcome.
Provisions for quality control
Endotoxin-free plasmid DNA for expression constructs will be analyzed for concentration and purity. In order to verify the construction of Tie2-tk and PGK-tk constructs, restriction digestion mapping will be performed. Banding pattern will be compared to expected band sizes based on plasmid maps received from the original authors. Flow cytometry data will be analyzed using the software package FlowJo and the achieved transduction efficiency (%GFP+ cells) will be calculated for each infected cell population. Untransduced, wild-type cells (both GSC83 and U87MG) will serve as a negative control for flow cytometry. 7-AAD will be used to exclude dead cells from flow analysis. All data obtained from the experiment—raw data, data analysis, control data, and quality control data—will be made publicly available, either in the published manuscript or as an open access dataset available on the Open Science Framework project page for this study (https://osf.io/mpyvx/).
Protocol 3: Monitoring xenograft tumor size after selective targeting of cells with ganciclovir
This protocol is designed to test whether GSC-derived endothelial cells can contribute to tumor growth in vivo. Virally transduced glioblastoma neurospheres expressing the herpes simplex virus thymidine kinase gene (tk) under the control of the endothelial marker Tie2 are subcutaneously injected into immunocompromised mice. Following tumor formation, mice are treated with ganciclovir, which selectively kills any cells expressing tk. Negative controls include neurospheres transduced with an empty viral vector, as well as the differentiated glioblastoma cell line U87MG, which should not give rise to endothelial cells. Positive controls include neurospheres and U87MG cells transduced with a vector conferring constitutive expression of tk (PGK-tk), which should target all tumor cells. Selective targeting of tk-expressing tumor cells should result in tumor reduction and degeneration, indicating the functional relevance of the GSC-derived endothelial vessels (as shown in Figures 4B and S10B).
Sampling
These experiments will utilize at least three mice per treatment group, for a minimum power of 80%.
A. See ‘Power calculations’ section for details.
B. As per Ricci-Vitiani et al., U87MG cells have a 100% tumor incidence rate, while GSC83 neurospheres have a ∼60% tumor incidence rate.
i. To ensure that we have enough animals at the end of the study to accurately power the effects for the GSC83 mouse cohort, we are including two extra mice per group beyond the estimated sample size of our power calculations.
Outline of experimental conditions:
A. Mouse Cohort 1 (xenograft of GSC83 glioblastoma neurospheres).
i. 28 female CD1 athymic nude mice, 5 weeks old.
a. Seven mice injected with untransduced GSC83 cells.
b. Seven mice injected with empty vector transduced GSC83 cells.
c. Seven mice injected with PGK-tk transduced GSC83 cells.
d. Seven mice injected with Tie2-tk transduced GSC83 cells.
B. Mouse Cohort 2 (xenograft of U87MG cells).
i. 12 female CD1 athymic nude mice, 5 weeks old.
a. Three mice injected with untransduced U87MG cells.
b. Three mice injected with empty vector transduced U87MG cells.
c. Three mice injected with PGK-tk transduced U87MG cells.
d. Three mice injected with Tie2-tk transduced U87MG cells.
Materials and reagents
Reagent
Type
Manufacturer
Catalog #
Comments
4–5 Week old female CD1 athymic nude mice
Mouse line
Charles River
086
Ganciclovir
Drug
Sigma–Aldrich
G2536
Matrigel Matrix High Concentration (HC), phenol red-free
Cell culture reagent
Corning
354262
Original catalog number not specified
Dulbecco's phosphate buffered saline (PBS)
Reagent
Sigma–Aldrich
D8537
1 ml insulin syringe with attached needle; 29G × 1/2 in.
Labware
BD Biosciences
329411
Original brand not specified
IsoFlo (isoflurane, USP)
Anesthetic
Abbott Animal Health
05260-05
Paraformaldehyde
Reagent
Sigma–Aldrich
158127
Original brand not specified
Paraffin
Reagent
Specific brand information will be left up to the discretion of the replicating laboratory and recorded later
Xylene
Reagent
Ethanol
Reagent
Carazzi's hematoxylin
IHC stain
Eosin
IHC stain
Permount
Mounting medium
Protocol
Prepare virally transduced cells for injection into mice. Perform all steps under sterile conditions.
A. Dissociate cells in flasks to form a single-cell suspension:
i. Pull down the cells by centrifuging each flask at ≤1000 rpm.
ii. Remove the supernatant, leaving approximately 150–200 µl of media in the flask.
iii. Use a P200 pipette to gently dissociate the cells. Pipette up and down several times to obtain a single cell suspension.
B. Resuspend 1 × 106 dissociated cells in 0.1 ml cold PBS, then mix with an equal volume of cold Matrigel. Total volume should be 0.2 ml.
Subcutaneously inject 4–5 week old female athymic nude mice into the rear flank. Each mouse should receive a single injection.
A. Mice should be microchipped prior to injection, so that they can be easily monitored throughout the duration of the study.
B. For each cell type, inject the 0.2 ml cell/Matrigel mixture subcutaneously into the flanks of the mice using a 29G insulin syringe.
Allow tumor nodules to form. The estimated time for tumor formation for U87MG cells is 3–4 weeks. The estimated time for tumor formation for GSC83 neurospheres is 4–6 months.
A. Check for tumors weekly and measure diameter and volume.
i. Calculate tumor volume as (length × width2)/2.
B. Note time for tumor nodules to form as well as tumor diameter and volume.
C. Once tumor size reaches ∼10 mm in diameter, proceed to ganciclovir treatment.
Inject mice with 50 mg/kg of ganciclovir (GCV) per day for 5 days in total.
A. Immediately prior to injection, record final tumor diameter and volume measurements. These data will serve as baseline controls for downstream analyses.
B. Prepare a 5 mg/ml solution of GCV using sterile water under sterile conditions. Filter solution through a 0.2 µm sterile filter.
C. Inject 0.2 ml of sterile solution (containing 1 mg of GCV) intraperitoneally (i.p.) into mice.
i. This injection amount assumes an approximate mouse weight of 20 g. For mice that are larger or smaller, the injection volume should be adjusted accordingly.
Measure tumor size twice weekly following GCV injection for 4 weeks using calipers.
A. Record both tumor diameter and volume. Measure tumor in two directions with calipers. Calculate tumor volume as (length × width2)/2.
Four weeks after last GCV injection, take final tumor measurements and euthanize mice. Harvest a random subset of tumors for histological analysis.
A. Randomly choose one mouse from each treatment group to harvest tumor tissue (eight mice in total; four mice from each cohort).
i. Prior to euthanasia, deeply anesthetize animals using isofluorane and transcardially perfuse mice with sterile saline, followed by 4% paraformaldehyde (PFA) in PBS.
ii. Excise tumor nodules under an operating microscope.
iii. Fix excised tumors in 4% PFA for 24 hr at 4°C.
Process, embed, and mount tissues on slides.
A. Dehydrate tissues through graded alcohols and clear in xylene.
B. Infiltrate with, and then embed, tissues in paraffin and section into 3 µm sections.
C. Mount sections onto positively charged glass slides.
i. Mount a total of two sections for each tumor onto a single slide.
Perform H&E staining by hand using the following procedure:
A. Deparaffinize sections twice in xylene, then rehydrate through graded alcohols (95%, 70%, 50% ETOH) to water.
B. Stain sections with Carazzi's hematoxylin, then rinse slides in water.
C. Stain sections with eosin.
D. Dehydrate sections through graded alcohols (50%, 70%, 90%), and then place in xylene.
E. Apply coverslips to slides with Permount and store slides at room temperature.
Blindly image stained sections and have images blindly analyzed by a board certified veterinary pathologist to verify the tumor composition of the tissue sections and analyze vascular structures for endothelial vacuolization and disruption.
Deliverables
Data to be collected:
A. Xenograft transplant records (mouse health records [age, weight], injection location, time for nodules to form, etc.).
B. Tumor size measurements (diameter and calculated volume) throughout course of study and prior to euthanasia (end-point).
C. Graph of percent tumor diameter change for each condition (compare to Figure 4B and S10C).
D. Images of H&E stained sections from a random selection of tumors (indicate neovessels and vascular degeneration) (compare to Figure S10B and S10D).
E. Pathologist's report detailing the evaluation of the stained tumor sections.
Confirmatory analysis plan
This replication attempt will perform the statistical analyses listed below, compute the effect sizes, compare them against the reported effect size in the original paper and use a meta-analytic approach to combine the original and replication effects, which will be presented as a Forest plot.
Statistical analysis of the replication data:
A. Comparison of percent diameter change in control versus tk-expressing tumors.
i. The diameters of tumors directly before GCV treatment will serve as individual baselines for each mouse. Tumor measurements made 4 weeks after GCV treatment will be subtracted from the baseline measurements for each tumor, and a percent change in diameter will be calculated. The mean percent change in each mouse cohort will be analyzed with a one-way ANOVA.
a. Following the one-way ANOVA, the following planned pairwise comparisons will be made using the Bonferroni correction to account for multiple comparisons:
For mice implanted with GSC83 neurospheres:
A. Tie2-tk versus PGK-tk.
B. Tie2-tk versus empty vector.
C. PGK-tk versus empty vector.
D. Untransduced (wild-type) versus empty vector (sensitivity).
b. Following the one-way ANOVA, the following planned pairwise comparisons will be made using the Bonferroni correction to account for multiple comparisons:
For mice implanted with U87MG cells:
A. Tie2-tk versus PGK-tk.
B. Tie2-tk versus empty vector (sensitivity).
C. PGK-tk versus empty vector.
D. Untransduced (wild-type) versus empty vector (sensitivity).
ii. The authors originally examined the percent changes in tumor diameter between mouse treatment groups using multiple uncorrected two-tailed t-tests. We will replicate their t-tests, but also use Bonferroni-corrected t-tests within the framework of the ANOVA.
B. Comparison of percent volume change in control versus tk-expressing tumors.
i. Differences in percent volume change of tumors before and after GCV treatment will be analyzed as described for percent diameter change above.
C. Comparison of tumor growth rates.
i. We will measure tumor growth rates across all mouse cohorts over the length of the study, both before and after GCV treatment. These data were not analyzed in the original study, so we consider them exploratory data. We will plot tumor growth curves for each animal and calculate the area under the curve (AUC) before and after GCV treatment. We will perform an ANCOVA on the different treatment groups to evaluate the AUC after GCV treatment, with the baseline (AUC before GCV treatment) included as the covariate. Further, we will perform Bonferroni corrected t-tests for pairwise comparisons between controls and tk-expressing tumors.
Known differences from the original study
The methods section in the original paper stated that mice were dual-injected with both control and Tie2-tk expressing neurospheres into the right and left flanks, respectively, and bilateral tumors were allowed to form. However, subsequent dialogue with the authors clarified that mice actually only received a single injection, as dual injections often led to problematic differences in tumor growth rates. Therefore, we will be using a single-injection model, where mice will be either injected with tk vectors or controls, but not both simultaneously. We will only be comparing GSC83-derived cell lines and U87MG-derived cell lines, excluding the other GSC lines used in the original study. Along with measuring differences in tumor diameter, we will also be measuring tumor volume throughout the course of the study. In the original study, it was not specified how many tumors were harvested and histologically analyzed. We have elected to harvest a random subset of tumors that represent all treatment groups. All known differences in reagents and supplies are listed in the ‘Materials and reagents’ section above, with the originally used item listed in the ‘Comments’ section. All differences have the same capabilities as the original and are not expected to alter the experimental design.
Provisions for quality control
The genetic integrity, mycoplasma-free and rodent pathogen-free purity, and efficient viral transduction of each cell line used in this experiment have been previously validated in Protocols 1 and 2. All mice will be handled and housed in accordance with the Institutional Animal Care and Use Committee (IACUC). All data obtained from the experiment—raw data, data analysis, control data, and quality control data—will be made publicly available, either in the published manuscript or as an open access dataset available on the Open Science Framework (https://osf.io/mpyvx/).
Power calculations
Unless otherwise stated, all data values are derived from the original paper, or were provided by the original authors.
Protocol 1
Summary of original data provided by Ricci-Vitiani et al.
Normalized Tie2 expression across cell lines (Figure S11C)
Mean
SD
n
GSC83
0.035362
0.012455
2
U87MG
0.018498
0.010397
2
HMVEC
1.317634
0.021
2
Test family
ANOVA: fixed effects, omnibus, one-way, with alpha error of 0.05.
A. ANOVA F-test statistic (performed with GraphPad Prism, version 6.0).
B. Partial η2 calculated from Lakens (2013).
Power calculations (performed with G*Power software, version 3.1.7 [Faul et al., 2007])
F (Dfn, Dfd)
Partial η2
Effect size f
A priori power
Total sample size
F (2, 3) = 4732
0.999683
56.15669
99.9%
6* (2 per group)
A minimum of three samples per group will be used, making the total sample size 9.
Test family
Two-tailed, unpaired t-test, with alpha error of 0.05 (Fisher's LSD).
A. Power calculations (performed with G*Power software, version 3.1.7 [Faul et al., 2007])
Group 1
Group 2
Effect size d
A priori power
Group 1 sample size
Group 2 sample size
GSC83
HMVEC
83.68295
>99.9%
2*
2*
U87MG
HMVEC
84.78352
>99.9%
2*
2*
GSC83
U87MG
3.070892#
80.0%
3
3
A minimum of three samples per group will be used.
This is a sensitivity calculation. The original effect size is 1.100569.
Protocol 2
Not applicable. Power calculations are not necessary for this protocol.
Protocol 3
Summary of original data provided by Ricci-Vitiani et al.
Impaired tumor growth of GSC83 xenograft following ganciclovir treatment (Figure 4B)
Mean (% diameter change)
SD
n
Wild-type
7.8
5.2
3
Empty vector
8.5
5.0
3
Tie2-tk
−13.2
5.9
3
PGK-tk
−28.7
3.4
4
Test family
ANOVA: fixed effects, omnibus, one-way, with alpha error of 0.05.
A. ANOVA F-test statistic (performed with GraphPad Prism, version 6.0).
B. Partial η2 calculated from Lakens (2013).
C. Power calculations (performed with G*Power software, version 3.1.7 [Faul et al., 2007])
F (Dfn, Dfd)
Partial η2
Effect size f
A priori power
Total sample size
F (3, 9) = 48.38
0.941612
4.015856
99.9%
8* (2 per group)
A total sample size of 20 will be used based on the planned comparisons.
Test family
Two-tailed, unpaired t-test, with alpha error of 0.0125 (Bonferroni's correction).
A. Power calculations (performed with G*Power software, version 3.1.7 [Faul et al., 2007])
Group 1
Group 2
Effect size d
A priori power
Group 1 sample size
Group 2 sample size
PGK-tk
Tie2-tk
3.221481
93.9%
5
5
Wild-type
Empty vector
2.640807*
80.0%
5
5
Empty vector
Tie2-tk
4.510074
98.1%#
4#
4#
Empty vector
PGK-tk
7.731555
99.8%§
3§
3§
This is a sensitivity calculation. The original effect size is 0.1454863.
Five per group will be used based on the PGK-tk versus Tie2-tk comparisons making the power 99.9%.
Five per group will be used based on the PGK-tk versus Tie2-tk comparisons making the power 99.9%.
Summary of original data
Values estimated from graph in Figure S10C
Impaired tumor growth of U87MG xenograft following ganciclovir treatment (Figure S10C)
Mean (% diameter change)
SD
n
Empty vector
38.1
7.7
3
Tie2-tk
36.1
11.4
3
PGK-tk
−49.0
7.9
3
Note: We are including an additional negative control in this experiment (wild-type untransduced cells). We performed these calculations with the assumption that wild-type untransduced cells will have similar values as empty vector control.
Test family
ANOVA: fixed effects, omnibus, one-way, with alpha error of 0.05.
A. ANOVA F-test statistic partial η2 (performed with R software, version 3.1.2 [R Development Core Team, 2014]).
B. Power calculations (performed with G*Power software, version 3.1.7 [Faul et al., 2007])
F (Dfn, Dfd)
Partial η2
Effect size f
A priori power
Total sample size
F (3, 8) = 72.111
0.964339
5.200177
99.9%
8* (2 per group)
A total sample size of 12 will be used based on the planned comparisons.
Test family
Two-tailed, unpaired t-test, with alpha error of 0.0125 (Bonferroni's correction).
A. Power calculations (performed with G*Power software, version 3.1.7 [Faul et al., 2007])
Group 1
Group 2
Effect size d
A priori power
Group 1 sample size
Group 2 sample size
PGK-tk
Tie2-tk
9.651957
99.9%
3
3
Wild-type
Empty vector
4.521980
80.0%*
3
3
Empty vector
Tie2-tk
4.521980
80.0%#
3
3
Empty vector
PGK-tk
9.878795
99.9%
3
3
This is a sensitivity calculation. There is no original effect size.
This is a sensitivity calculation. The original effect size is 0.226838.
Acknowledgements
The Reproducibility Project: Cancer Biology core team would like to thank the original authors, including Ruggero De Maria, Roberto Pallini, and most especially, Lucia Ricci-Vitiani, for generously sharing critical information as well as reagents to ensure the fidelity and quality of this replication attempt. We thank Courtney Soderberg at the Center for Open Science for assistance with statistical analyses. We would also like to thank the following companies for generously donating reagents to the Reproducibility Project: Cancer Biology: American Type Culture Collection (ATCC), BioLegend, Cell Signaling Technology (CST), Charles River Laboratories, Corning Incorporated, DDC Medical, EMD Millipore, Harlan Laboratories, LI-COR Biosciences, Mirus Bio, Novus Biologicals, System Biosciences, and Sigma–Aldrich.
Additional informationCompeting interests
DC: Noble Life Sciences is a Science Exchange-associated lab.
DS: BioFactura is a Science Exchange-associated lab.
RP:CB: We disclose that Elizabeth Iorns, Fraser Tan, and Joelle Lomax are employed by and hold shares in Science Exchange Inc. The experiments presented in this manuscript will be conducted by DC at Noble Life Sciences and DS at BioFactura, which are both Science Exchange-associated laboratories.
The other authors declare that no competing interests exist.
Author contributions
DC, Conception and design
DS, Conception and design
NM, Drafting or revising the article
RP:CB, Conception and design, Drafting or revising the article
ReferencesChengLHuangZZhouWWuQDonnolaSLiuJKFangXSloanAEMaoYLathiaJDMinWMcLendonRERichJNBaoS2013Glioblastoma stem cells generate vascular pericytes to support vessel function and tumor growthCell15313915210.1016/j.cell.2013.02.021ChiaoMTYangYCChengWYShenCCKoJL2011CD133+ glioblastoma stem-like cells induce vascular mimicry in vivoCurrent Neurovascular Research821021910.2174/156720211796558023De PalmaMVenneriMAGalliRSergi SergiLPolitiLSSampaolesiMNaldiniL2005Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitorsCancer Cell821122610.1016/j.ccr.2005.08.002El HallaniSBoisselierBPeglionFRousseauAColinCIdbaihAMarieYMokhtariKThomasJLEichmannADelattreJYManiotisAJSansonM2010A new alternative mechanism in glioblastoma vascularization: tubular vasculogenic mimicryBrain13397398210.1093/brain/awq044El HallaniSColinCEl HoufiYIdbaihABoisselierBMarieYRavassardPLabussièreMMokhtariKThomasJLDelattreJYEichmannASansonM2014Tumor and endothelial cell hybrids participate in glioblastoma vasculatureBioMed Research International201482732710.1155/2014/827327FanYLZhengMTangYLLiangXH2013A new perspective of vasculogenic mimicry: EMT and cancer stem cells (Review)Oncology Letters61174118010.3892/ol.2013.1555FaulFErdfelderELangAGBuchnerA2007G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciencesBehavior Research Methods3917519110.3758/BF03193146GhanekarAAhmedSChenKAdeyiO2013Endothelial cells do not arise from tumor-initiating cells in human hepatocellular carcinomaBMC Cancer1348510.1186/1471-2407-13-485KaurSBajwaP2014A ‘tête-à tête’ between cancer stem cells and endothelial progenitor cells in tumor angiogenesisClinical & Translational Oncology1611512110.1007/s12094-013-1103-4LakensD2013Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAsFrontiers in Psychology486310.3389/fpsyg.2013.00863LathiaJDGallagherJMyersJTLiMVasanjiAMcLendonREHjelmelandABHuangAYRichJN2011Direct in vivo evidence for tumor propagation by glioblastoma cancer stem cellsPLoS ONE6e2480710.1371/journal.pone.0024807PurhonenSPalmJRossiDKaskenpaaNRajantieIYla-HerttualaSAlitaloKWeissmanILSalvenP2008Bone marrow-derived circulating endothelial precursors do not contribute to vascular endothelium and are not needed for tumor growthProceedings of the National Academy of Sciences of USA1056620662510.1073/pnas.0710516105R Development Core Team2014R: A language and environment for statistical computingR Foundation for Statistical Computinghttp://www.R-project.org/Ricci-VitianiLPalliniRBiffoniMTodaroMInverniciGCenciTMairaGParatiEAStassiGLaroccaLMDe MariaR2010Tumour vascularization via endothelial differentiation of glioblastoma stem-like cellsNature46882482810.1038/nature09557RodriguezFJOrrBALigonKLEberhartCG2012Neoplastic cells are a rare component in human glioblastoma microvasculatureOncotarget398106SodaYMarumotoTFriedmann-MorvinskiDSodaMLiuFMichiueHPastorinoSYangMHoffmanRMKesariSVermaIM2011Transdifferentiation of glioblastoma cells into vascular endothelial cellsProceedings of the National Academy of Sciences of USA1084274428010.1073/pnas.1016030108VescoviALGalliRReynoldsBA2006Brain tumour stem cellsNature Reviews Cancer642543610.1038/nrc1889WangRChadalavadaKWilshireJKowalikUHovingaKEGeberAFligelmanBLevershaMBrennanCTabarV2010Glioblastoma stem-like cells give rise to tumour endotheliumNature46882983310.1038/nature0962410.7554/eLife.04363.002Decision letterOkanoHideyukiReviewing editorKeio University School of Medicine, Japan
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Registered report: Tumour vascularization via endothelial differentiation of glioblastoma stem-like cells” for consideration at eLife. Your article has been favorably evaluated by Sean Morrison (Senior editor), a Reviewing editor, four reviewers, and a biostatistician. Two of the reviewers, Yoshiaki Kubota and Ruggero De Maria, have agreed to share their identity.
The Reviewing editor and the reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
As part of The Reproducibility Project: Cancer Biology, Chroscinski et al. propose to replicate key experiments from “Tumour vascularization via endothelial differentiation of glioblastoma stem-like cells” by Ricci-Vitiani and colleagues, published in Nature in 2010. The original Nature manuscript demonstrated that endothelial cells bearing the same genomic alteration as cancer cells may show a different sensitivity to conventional anti-angiogenic treatments and suggests the possibility of targeting the process of GSC differentiation into endothelial cells as a new therapeutic option in cancer treatment.
The authors accurately summarize the Nature paper and the proposed experiment using GSCs transduced with a Tie2-tk construct sufficiently replicates the key finding in that Nature paper.
The following questions should be addressed in a resubmission:
1) A directly conflicting paper by Cheng et al. (Cell, 2013) discussed that the discrepancy is attributed to possible activation of Tie2 promoter in pericytes, which may eliminate pericytes as well. Do the authors have any discussion on this issue?
2) Within protocol 1, the authors plan to use HMVEC instead of HUVEC as previously outlined in the original study. Endothelial cells of different lineage vary greatly in their gene expression profile in particular between HUVEC and HMVEC. As HUVECs are commercially available, the authors should provide a clearer explanation in their change of reagents.
3) To maximize the quality of the replication, we would suggest that the authors establish more precise criteria for neurosphere passage. Protocol 1.1.a.ii states that “GSC83 cells should be subcultured when neurosphere size prevents the needed provision of nutrients”. This criteria is subjective, and viability, differentiation potential, infection efficiency etc of glioma stem cells can all be influenced by the abundance or lack of nutrients. Always plating a fixed cell number and subculturing after a fixed period of time could be one way to avoid variability in culture methods.
Statistical comments:
Overall, I have no major concerns over the statistics.
1) For protocols 1 and 3, the aim is to replicate the detection of the very large effects seen in the original report. The original report used very small numbers so both the reported effect size and the SD will have very poor precision. With a sample size of 2-4 per group it is impossible to judge whether the data points have been sampled from a normal distribution so the p-value strongly relies on the assumption of normality (which is not so important for large sample sizes). However, the researchers presumably have extensive prior experience of using such techniques in similar situations and hence expect to see very large effects with no overlap of the distributions of values observed in each group. So I judge that the very small sample sizes, provided no data are missing, will be sufficient to answer these questions.
2) It seems reasonable to use the Bonferroni adjusted significance levels as the same sample is used for several comparisons.
3) When it comes to presenting the results and comparing with the original report, a Forest plot is appropriate but this should not include an overall summary statistic (combining the original and replicate).
4) The method used to compare the growth rates seems rather convoluted. Why not put the repeated measurements into a full linear model with the baseline included as a co-variate?
10.7554/eLife.04363.003Author response
1) A directly conflicting paper by Cheng et al. (Cell, 2013) discussed that the discrepancy is attributed to possible activation of Tie2 promoter in pericytes, which may eliminate pericytes as well. Do the authors have any discussion on this issue?
As reviewed in the Introduction, many conflicting reports exist within the field of tumor-derived vasculature in GBM. We believe the data presented in Cheng et al. provide a very interesting alternative hypothesis that may help to explain the results achieved by Ricci-Vitiani and colleagues. We have expanded our discussion of this paper in the Introduction, to better outline the results of Cheng et al. We will be sure to highlight these data again in the Replication Study as a possible alternative mechanism to understand the results of the replication.
2) Within protocol 1, the authors plan to use HMVEC instead of HUVEC as previously outlined in the original study. Endothelial cells of different lineage vary greatly in their gene expression profile in particular between HUVEC and HMVEC. As HUVECs are commercially available the authors should provide a clearer explanation in their change of reagents.
As demonstrated in Supplementary Figure 11C (Ricci-Vitiani et al., 2010), Ricci-Vitiani and colleagues measured Tie2 expression in both HUVEC and HMVEC cell lines. Both cell lines served as a positive control, demonstrating high levels of Tie2 expression. According to our correspondence with Ricci-Vitiani and colleagues, they included HMVEC cells as an additional control at the behest of their own reviewers for the original Nature manuscript. These reviewers felt that including an adult endothelial cell line constituted a more accurate control for adult-derived glioblastoma neurospheres than did the fetal-derived HUVEC cells. Thus, we have opted to also include this more-relevant control cell line. We are obtaining the same adult HMVEC cell line as used by the original authors to enable a direct comparison of gene expression.
3) To maximize the quality of the replication, we would suggest that the authors establish more precise criteria for neurosphere passage. Protocol 1.1.a.ii states that “GSC83 cells should be subcultured when neurosphere size prevents the needed provision of nutrients”. This criteria is subjective, and viability, differentiation potential, infection efficiency etc of glioma stem cells can all be influenced by the abundance or lack of nutrients. Always plating a fixed cell number and subculturing after a fixed period of time could be one way to avoid variability in culture methods.
We thank the reviewers for these suggestions. Upon further correspondence with Ricci-Vitiani and colleagues, we have updated the Registered Report to include more details pertaining to the culture of GSC#83, including defined cell plating numbers and subculturing schedule. We will be obtaining the same culture that was initially derived by the authors, and using the same conditions they use for this culture.
Statistical comments:
[…]
3) When it comes to presenting the results and comparing with the original report, a forest plot is appropriate but this should not include an overall summary statistic (combining the original and replicate).
We disagree with the reviewer and as described in Valentine et al., 2011 and Bumming, 2012, combining multiple studies using a meta-analytic approach is a statistical option that can be employed to describe all of the available evidence about a given effect size. Specifically we will utilize a random effects meta-analysis because this approach assumes that the effects vary due to known and unknown characteristics of the studies.
4 ) The method used to compare the growth rates seems rather convoluted. Why not put the repeated measurements into a full linear model with the baseline included as a co-variate?
We thank the reviewer for this helpful suggestion. We have changed the method used to compare the growth rates of the different cohorts. We will perform an ANCOVA to analyze the differences in the tumor growth curves after GCV treatment using the tumor growth curves before GCV treatment (baseline) as the covariate. We will also use the area under the curve to quantify and compare the tumor growth curves.
diff --git a/elife04489.xml b/elife04489.xml
new file mode 100644
index 0000000..fbea560
--- /dev/null
+++ b/elife04489.xml
@@ -0,0 +1 @@
+elifeeLifeeLifeeLife2050-084XeLife Sciences Publications, Ltd0448910.7554/eLife.04489Research articleBiophysics and structural biologyCell biologyNon-catalytic motor domains enable processive movement and functional diversification of the kinesin-14 Kar3MieckChristine1†MolodtsovMaxim I124†DrzewickaKatarzyna1van der VaartBabet1LitosGabriele1SchmaussGerald13VaziriAlipasha124*WestermannStefan1*Research Institute of Molecular Pathology, Vienna, AustriaMax F Perutz Laboratories, University of Vienna, Vienna, AustriaInstitute of Molecular Biotechnology, Vienna, AustriaResearch Platform Quantum Phenomena and Nanoscale Biological Systems, University of Vienna, Vienna, AustriaHarrisonStephen CReviewing editorHoward Hughes Medical Institute, Harvard Medical School, United StatesFor correspondence: alipasha.vaziri@univie.ac.at (AV);westermann@imp.ac.at (SW)
Motor proteins of the conserved kinesin-14 family have important roles in mitotic spindle organization and chromosome segregation. Previous studies have indicated that kinesin-14 motors are non-processive enzymes, working in the context of multi-motor ensembles that collectively organize microtubule networks. In this study, we show that the yeast kinesin-14 Kar3 generates processive movement as a heterodimer with the non-motor proteins Cik1 or Vik1. By analyzing the single-molecule properties of engineered motors, we demonstrate that the non-catalytic domain has a key role in the motility mechanism by acting as a ‘foothold’ that allows Kar3 to bias translocation towards the minus end. This mechanism rivals the speed and run length of conventional motors, can support transport of the Ndc80 complex in vitro and is critical for Kar3 function in vivo. Our findings provide an example for a non-conventional translocation mechanism and can explain how Kar3 substitutes for key functions of Dynein in the yeast nucleus.
DOI:http://dx.doi.org/10.7554/eLife.04489.001
10.7554/eLife.04489.002eLife digest
Molecules can be transported around a cell by so-called motor proteins that move along a network of filaments called microtubules. Many motor proteins—including the kinesin family of these proteins—can only move in one direction along a microtubule. In most cells, kinesins tend to transport other molecules away from the center and towards the cell edge.
Kinesins can have different structures, but most are made up of two subunits that are joined and work together to create a walking-like movement. Each subunit has a region called a motor domain (also known as its ‘head’) that can bind to the microtubule and to a molecule called ATP, which provides the energy required for the motor to step forward.
Kinesins can be classed either as processive or non-processive motors. Processive motors can walk continuously along a microtubule for several hundred steps, whereas non-processive motors fall off after just a few steps. A motor protein called Kar3 belongs to a group of non-processive kinesins. Kar3 is unusual; unlike most of the motors in this group (which work together in pairs), Kar3 motor protein subunits each bind to and work with non-motor protein subunits, including one called Cik1. The head of the non-motor protein cannot bind to ATP, although it can bind to microtubules. This means that the non-motor protein subunits are not provided with the energy to make a stepping motion; this raises questions about how the Kar3 motor protein moves along the microtubule, and whether this affects the roles the motor performs.
Mieck et al. studied how a molecular motor made up of Kar3 and Cik1 moves along microtubules using sensitive microscopy that allows single molecules to be observed. This revealed that, contrary to what is expected from a non-processive motor, Kar3–Cik1 moves long distances on microtubules without detaching from them. Further investigation showed that Cik1 provides a ‘foothold’ for the motor, binding it to the microtubule in such a way that allows it to move along the microtubule in the opposite direction to most kinesins. In addition, Mieck et al. found that the Kar3–Cik1 motor binds to and transports a protein complex that is crucial for separating chromosomes during cell division.
A challenge for the future is to understand in even greater detail how the movement of such a motor occurs. If it doesn't ‘walk’ like other motors, then how can its motion be explained? The benefits for the cell that underlie why this mechanism evolved also remain to be discovered.
DOI:http://dx.doi.org/10.7554/eLife.04489.002
Author keywordskinesinsmicrotubulescytoskeletonResearch organismS. cerevisiaehttp://dx.doi.org/10.13039/501100000781European Research Council (ERC)FP2/2007-2013/no.203499WestermannStefanhttp://dx.doi.org/10.13039/501100002428Austrian Science Fund (FWF)SFB F34-B03WestermannStefanhttp://dx.doi.org/10.13039/501100001821Wiener Wissenschafts-, Forschungs- und TechnologiefondsVRG10-11VaziriAlipashahttp://dx.doi.org/10.13039/501100000854Human Frontier Science Program (HFSP)RGP004½012VaziriAlipashaThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementThe kinesin-14 motor Kar3 moves along microtubules using a previously undescribed mechanism that critically requires the presence of a non-catalytic head.Introduction
Motors of the kinesin family are ubiquitous enzymes essential for intracellular transport along microtubules in eukaryotes. The mechanism by which kinesin motor proteins convert the chemical energy of ATP hydrolysis into coordinated, long-range directional movement has fascinated cell biologists, biochemists, and engineers for many decades. Biophysical studies of kinesins have focused on conventional Kinesin-1 and established the ‘hand-over-hand’ model for the processive walking behavior of this type of motor (Asbury et al., 2003; Yildiz et al., 2004; Kaseda et al., 2003). In analogy to other enzymes, the term ‘processivity’ describes the ability of individual motor molecules to undergo multiple catalytic cycles—and therefore translocate—before releasing from the microtubule.
Kinesin-14 family members, exemplified by the Drosophila Ncd motor, are common examples for nonprocessive kinesins (Case et al., 1997; Foster and Gilbert, 2000). They generate motility through the minus-end-directed rotational movement of a coiled-coil mechanical element that occurs upon ATP binding (Endres et al., 2006). After each catalytic cycle, Ncd motors release from the microtubule lattice, meaning that to support microtubule sliding and crosslinking in the spindle, many Ncd motors must work together cooperatively in an ensemble (Braun et al., 2009; Fink et al., 2009). Budding yeast kinesin-14 Kar3 is distinct from other family members in its heterodimeric composition with either Cik1 or Vik1 (Manning et al., 1999) (Figure 1A). High-resolution structural analysis has shown that these accessory proteins contain a motor homology domain that harbors a microtubule binding site but lacks the structural elements required to bind and hydrolyze ATP (Allingham et al., 2007). Biochemical experiments have indicated that Cik1 and Vik1 modulate the interaction of Kar3 with microtubules (Allingham et al., 2007; Chen et al., 2011; Rank et al., 2012). In vivo Kar3's heterodimeric composition governs its subcellular localization and function: Kar3 in complex with Vik1 crosslinks parallel microtubules in proximity to spindle pole bodies during mitosis (Manning et al., 1999), whereas antiparallel microtubule sliding is powered by Cik1–Kar3 complexes that associate with the microtubule lattice and plus-ends during mitotic and meiotic events (Maddox et al., 2003; Gardner et al., 2008). In addition, Kar3 has been implicated in kinetochore capture and transport (Middleton and Carbon, 1994; Tanaka et al., 2005, 2007). The unusual composition of the Kar3 motor with the combination of a catalytic and a non-catalytic domain, as well as its key roles for diverse cellular processes in yeast, has made it a particularly interesting object of study both from a biophysical and cell biological point of view. The understanding of the mechanistic basis of Kar3 function, however, has remained incomplete, as biochemical experiments have been limited to ensemble assays using truncated or artificially dimerized proteins. On the basis of such experiments and the interpretation of in vivo phenotypes, it has been proposed that Cik1–Kar3 acts as a microtubule depolymerase (Chu et al., 2005; Sproul et al., 2005; Allingham et al., 2007). We hypothesized that the presence of a non-catalytic domain may allow functionalities fundamentally different from conventional kinesin-14 homodimers. As the activity of individual full-length Kar3 motors had not been observed directly, we developed assays to investigate motors at the single molecule level and analyze the contribution of the non-catalytic domain.10.7554/eLife.04489.003
Purification and characterization of Cik1–Kar3 kinesin motors.
(A) Schematic representation of conventional Kinesin-1 in comparison to the kinesin-14 Cik1–Kar3. (B) Purification of recombinant Cik1–Kar3 from yeast extracts. Motors are covalently labeled with Tetramethylrhodamine (TMR) via a HaloTag on the amino-terminus of Kar3. Coomassie-stained SDS-PAGE shows purity of the motor preparation and fluorescent labeling of the Kar3 subunit. (C) Size-exclusion chromatography of Cik1–Kar3-Halo motors on a Superose 6 column. The void volume of the column (V0) and the elution position of standard proteins with their respective stokes radii is indicated. (D) SDS-PAGE analysis of Superose 6 fractions from (C). (E) Sucrose gradient centrifugation of Cik1–Kar3 motors. Consecutive fractions from top to bottom of a 5–25 (wt/vol)% sucrose gradient were analyzed by SDS-PAGE and Coomassie staining. The gradient positions of standard proteins are indicated together with their sedimentation coefficients. (F) Low angle Pt/C rotary shadowing electron microscopy of Cik1–Kar3 motors obtained after size exclusion chromatography. Overview of Cik1–Kar3 motors, scale bar 100 nm. (G) Gallery view of selected Cik1–Kar3 motors, scale bar 50 nm.
DOI:http://dx.doi.org/10.7554/eLife.04489.003
ResultsCik1–Kar3 is a processive kinesin-14 motor with a single catalytic domain
To study Kar3 motors at the single molecule level, we developed a protocol to express and purify full-length kinesin-14 heterodimers from Saccharomyces cerevisiae using affinity tagged Cik1 and Kar3 fused COOH-terminally to a HaloTag that served as covalent attachment site for the fluorescent dye tetramethylrhodamine (TMR). Purification and labeling yielded a homogenous preparation containing a heterodimer of Cik1 and TMR-labeled Kar3 (Figure 1B). During size exclusion chromatography, Cik1–Kar3 motors eluted as a single major peak with a Stokes radius of ∼9.1 nm, well separated from the void volume of the column (Figure 1C,D). Sucrose-gradient centrifugation revealed the presence of a single major species with an apparent sedimentation coefficient of ∼5.6S (Figure 1E). Combining these hydrodynamic values yielded a native molecular weight of 214 kDa, close to the calculated molecular weight of a Halo-tagged Cik1–Kar3 heterodimer of 190 kDa. We further characterized the oligomeric state of full-length Cik1–Kar3 motors by performing low-angle Pt/C rotary shadowing electron microscopy on peak fractions from the gel filtration experiments. This analysis revealed individual well-defined, highly elongated molecules that were characterized by globular domains separated by a 61 ± 8 nm (mean ± SD, n = 100) long coiled-coil (Figure 1F). Typically, two closely spaced globular domains likely corresponding to catalytic and non-catalytic head domains were discernible at one end, while a single globular domain decorated the other (Figure 1G). The highly elongated shape of the Cik1–Kar3 molecules can explain their early elution from the gel filtration column. Overall, the hydrodynamic analysis and the direct visualization of motor molecules by EM support the presence of heterodimeric Cik1–Kar3 molecules.
We next observed the behavior of single motor molecules on surface-immobilized microtubules in vitro using time-lapse multi-color TIRF microscopy. Unexpectedly, and contrary to the classification of kinesin-14s as non-processive motors, Kar3 molecules displayed efficient ATP-dependent movement over several micrometers and accumulated at microtubule minus ends (Figure 2A, Video 1). Automated tracking of motors revealed a Gaussian velocity distribution of Cik1–Kar3 with a mean speed of 77 ± 23 nm/s (mean ± SD; Figure 2B) in range with reported microtubule gliding velocities for truncated Cik1–Kar3 molecules (Chu et al., 2005; Allingham et al., 2007). The motor is therefore approximately 10-fold slower than conventional Kinesin-1, but similar in speed to yeast cytoplasmic Dynein, the major minus-end directed-motor in eukaryotes (Reck-Peterson et al., 2006). The run-length histogram followed an exponential distribution and revealed that individual Cik1–Kar3 motors advanced processively for an average of 5.2 μm before detaching from the microtubule track (Figure 2C). The motile parameters were highly sensitive to the ionic strength of the imaging buffer: at the same ATP concentration higher salt concentration increased the mean squared displacement of the motors (Figure 2D), but decreased the on-rate, run length and minus-end dwell time (Figure 2E). We additionally noticed that Cik1–Kar3 oscillated back-and-forth when entering microtubule overlap zones. Because Cik1–Kar3 is exclusively moving towards the minus-end on single MT filaments, we concluded that reversing motors encountered an antiparallel-oriented MT bundle. Individual motors were able to switch the track microtubule multiple times leading to a prolonged association with antiparallel bundles (Figure 2F, Video 2).10.7554/eLife.04489.004
Cik1–Kar3 motors move processively with a single catalytic domain.
(A) Kymograph showing two-color time lapse TIRF microscopy of Cik1–Kar3 (red) moving along taxol-stabilized microtubules (blue). See Video 1 for example of Cik1–Kar3 motility. (B) Histogram of velocities of Cik1–Kar3 molecules moving along taxol-stabilized microtubules (fit with a Gaussian function, black line). The mean velocity is 77 ± 23 nm/s, n = 699. (C) Histogram of run lengths of Cik1–Kar3 molecules moving along taxol-stabilized microtubules, n = 209. (D) Mean-squared displacement analysis of wild-type Cik1–Kar3 at two different salt concentrations in the presence of ATP. (E) Influence of ionic strength on the motile properties of Cik1–Kar3 molecules. Experiments were performed in BRB80-based imaging buffer containing the indicated concentrations of KCl. (F) Behavior of Cik1–Kar3 in microtubule networks. Typical kymograph showing directional movement of Cik1–Kar3 on single microtubules vs repeated directionality switches of individual motors in antiparallel overlaps. The dashed line indicates the beginning of an overlap zone. See Video 2.
DOI:http://dx.doi.org/10.7554/eLife.04489.004
10.7554/eLife.04489.005
Characterization of Cik1–Kar3 motility.
(A) Photobleaching experiment in the absence of nucleotide. Kymograph representation showing that Cik-Kar3 motors do not exhibit displacement or diffusion in this state. (B) Quantification of photobleaching. (C) Example for single-step photobleaching event. (D) Example for two-step photobleaching event. (E) Mixing experiment combining TMR-labeled Cik1–Kar3 motors with Alexa488 labeled Cik1–Kar3 motors prior to imaging. Kymograph reveals individual traces for red-and green fluorescent motors, indicating that a single Cik1–Kar3 heterodimer is sufficient for movement. (F) Microtubule-gliding experiment with varying concentrations of Cik1–Kar3. In this assay, the motor is immobilized via anti-Halo antibodies to the coverglass and movement of microtubules is recorded by time-lapse TIRF microscopy. Kymographs indicate gliding velocity at different motor concentrations, the lower panel shows that gliding is ATP dependent. (G) Movement of Cik1–Kar3 motors on dynamic microtubules. Dynamic extensions were grown from GMPCPP-stabilized microtubule seeds and imaged together with Cik1–Kar3. Note movement of motors opposite to the directions of microtubule growth, indicating minus-end directed motility of Cik1–Kar3. (H) Quantitative analysis of MT shrinkage rate of taxol-stabilized MTs in the presence of 1 nM Cik1/Kar3 (error bars represent SEM, n = 30).
DOI:http://dx.doi.org/10.7554/eLife.04489.005
10.7554/eLife.04489.006
A single Cik1–Kar3 heterodimer is sufficient to form a processive complex.
(A) Kymograph showing movement of Cik1–Kar3. Position along the microtubule is depicted on the vertical axis while time changes along the horizontal axis. (B–C) Kymographs showing binding and movement of single Cik1–Kar3 complexes each composed of 1 (B) or 2 (C) heterodimers. (D–E) Records for background subtracted brightness vs times for complexes in B and C, respectively. (E) Distribution of background subtracted brightness for Cik1–Kar3 complexes. The fit is with four peak Gaussian. (F) Distribution of the initial size of processively moving complexes based on the brightness of a single fluorophore from E. (G) Histogram of moving motor complexes containing the indicated number of Cik1-Kar3 heterodimers. (H) Model of oligomerization of Cik1–Kar3 complexes. Oligomer of two complexes is shown. Blue arrows show velocities of movement of each independent heterodimer. Interconnected tails act as a spring-like linker in between two moving heads. (I, J) Velocity of the movement and run length as a function of size. Blue squares are experimental data, red circles are results of theoretical modeling (for parameters seeFigure 2—figure supplement 3). The bars show average value and standard deviation.
DOI:http://dx.doi.org/10.7554/eLife.04489.006
10.7554/eLife.04489.007
Mathematical model of the cooperative kinesin movement.
(A) Schematics of the kinesin team. In the depicted case, the team consists of three kinesins and two of them are shown as bound to the microtubule and move at the velocities ν→1 and ν→3, respectively. Since ν1>ν3 the elastic linkage between the dimers stretches generating the forces f1→ and f3→. (B) Representation of the force–velocity relation used in the model. (C) Representation of the dependence of the unbinding rate on the applied force. (D) Example of a simulation for a complex size with three dimers. The blue graph is the position of the center of mass of the complex as a function of time (left axis), red graph shows the number of dimers attached to the microtubule at each time point. (right axis). (E–F) Velocity, run length and their standard deviations as a function of the complex size for different values of kstiff. Other parameters of the simulation are: V = 50 nm/s, σν = 20 nm/s; Fstall = 1 pN; kON = 0.55 s−1; kOFFo = 0.08 s−1; kOFFMAX = 4 s−1. (G) Comparison between experimental data (blue) and best fit (see values of objective function in Table 5) of the simplified theoretical model without force dependence (red). (H–I) Velocity, run length and their standard deviations as a function of the complex size for best parameters of the complete model. Data are shown for red—Fstall = 0.2 pN; kON = 0.57 s−1; kOFFMAX = 3.7 s−1; magenta—Fstall= 1 pN; kON = 0.55 s−1; kOFFMAX = 4 s−1; green—Fstall = 4 pN; kON = 0.49 s−1; kOFFMAX = 5.3 s−1; Other parameters for all simulations: V = 50 nm/s, σν = 20 nm/s; kOFFo = 0.08 s−1; kstiff = 0.03 pN/nm.
DOI:http://dx.doi.org/10.7554/eLife.04489.007
10.7554/eLife.04489.008
Time-lapse two-color TIRF microscopy of Cik-Kar3-TMR motors (red) moving on taxol stabilized microtubules (blue). 100 frames were taken every 3 s. The video is played at 20 frames/s, scale bar: 5 μm. The video corresponds to <xref ref-type="fig" rid="fig2">Figure 2A</xref>.
DOI:http://dx.doi.org/10.7554/eLife.04489.008
10.7554/eLife.04489.009
Behavior of Cik1–Kar3 in microtubule overlap zones. Two-color time-lapse TIRF video of Cik1–Kar3-TMR (red) moving on taxol-stabilized microtubules (blue). Note back and forth movement of individual Cik1–Kar3 motors in microtubule overlap zones. 100 frames were taken every 3 s, the video is played at 20 frames/s, scale bar 5 μm. The video corresponds to <xref ref-type="fig" rid="fig2">Figure 2F</xref>.
DOI:http://dx.doi.org/10.7554/eLife.04489.009
Individual Cik1–Kar3 heterodimers are sufficient for processive movement
The processive movement of Kar3 may either be a property of individual heterodimers or alternatively require the formation of motor ensembles combining multiple catalytic domains that coordinate stepping. Photobleaching experiments in the absence of nucleotide, in which the motor is persistently bound to the microtubule, revealed that 43 out of 50 molecules lost fluorescence in a single step consistent with the presence of a single Cik1–Kar3 heterodimer (Figure 2—figure supplement 1A–D). Mixing experiments combining TMR- with Alexa488 labeled Cik1–Kar3 molecules prior to imaging showed that the majority of moving complexes displayed exclusively either red or green fluorescence (Figure 2—figure supplement 1E, Video 3). To discriminate between processive and non-processive motility modes by an alternative approach, we varied the motor concentration in a standard microtubule gliding assay. Titration of Cik1–Kar3 over a 50-fold concentration range revealed no decrease in velocity for microtubule gliding, a characteristic feature of processive motility (Hancock and Howard, 1998) (Figure 2—figure supplement 1F). Contrary to previous reports, we did not observe substantial depolymerization of taxol-stabilized microtubules in the presence of Cik1–Kar3. On dynamic microtubules motors moved towards the minus-ends, overall microtubule dynamics appeared unchanged in the presence of Cik1–Kar3 and catastrophes did not coincide with plus-end localization of the motor (Figure 2—figure supplement 1G).10.7554/eLife.04489.010
Mixing experiment to demonstrate processivity of individual Cik1–Kar3 heterodimers. Cik1–Kar3-TMR motors (red) were mixed with Cik1–Kar3-Alexa488 motors (green) and imaged by multi-color TIRF microscopy on taxol-stabilized microtubules. Red and green motors are seen moving in opposite directions because of closely spaced microtubules with opposite orientation. Frames were taken every 3 s for 100 frames, the video is played at 20 frames/s. Scale bar corresponds to 5 μm. The video corresponds to <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1E</xref>.
DOI:http://dx.doi.org/10.7554/eLife.04489.010
Quantification of TMR brightness of moving motors allowed us to compare the motile properties of Cik1–Kar3 heterodimers vs larger teams that consisted of two or more heterodimers (Figure 2—figure supplement 2A–G). We found that Cik1–Kar3 velocity was largely independent of motor team size (Figure 2—figure supplement 2H,I). The run length, however, increased with larger team size while the variance of the velocity decreased (Figure 2—figure supplement 2J). These motile behaviors of Cik1–Kar3 complexes of different sizes can be quantitatively explained by a biophysical model in which individual motors influence each other through mechanical coupling with spring-like properties (Figure 2—figure supplement 3 and Supplementary file 1). Importantly, a key feature of the biophysical model is the ability of an individual Cik1–Kar3 heterodimer to move processively.
Mutations in the non-catalytic head domain abolish directional movement
We next sought to establish the molecular requirements for Kar3 motility: processive movement could be a property intrinsic to the head domains or require secondary microtubule interaction sites in the motor tails (Gudimchuk et al., 2013; Su et al., 2013). To distinguish between these possibilities, truncation constructs were designed, systematically eliminating parts of the tail, coiled-coil, or globular domains of either the motor or the partner protein (Figure 3A). Truncation of the amino-terminal tail domain, either in Kar3 (up to aa 174) or in Cik1 (up to aa 250) allowed robust processive movement in the single molecule assay and also had little effect on multi-motor gliding velocity. This distinguishes Cik1–Kar3 from the homodimeric Drosophila kinesin-14, Ncd, which has been shown to be capable of a weakly processive motion at very low ionic strength depending on its tail region that acts as an electrostatic tether to microtubules (Furuta and Toyoshima, 2008). Further truncations of the coiled-coil interfered with heterodimer formation (not shown). In contrast, carboxyterminal truncations that either completely eliminated the predicted globular Cik1 motor homology domain (Cik11–360) or removed a portion of the carboxyterminus (Cik11–521) had severe effects and prevented directional movement. The specific nature of the defect was most apparent for the shorter truncation Cik11–521–Kar3, which was able to bind microtubules under our standard conditions, but instead of smooth translocation it displayed erratic forward and backward displacements that did not lead to directional movement as revealed by kymographs (Figure 3B, Videos 4 and 5). The defect imposed by the Cik11–521 mutation was also readily apparent in multi-motor gliding assays, where microtubules frequently switched their direction of movement and displayed little net transport (Figure 3C). The pronounced defect of the Cik11–521 mutant points to an essential role for the non-catalytic head in the motility mechanism. To corroborate this point, we also co-overexpressed Flag-tagged Kar3 with Kar3-Halo and purified fluorescently labeled Kar3 homodimers that can form in the absence of Cik1 (Chu et al., 2005). We failed to observe processive movement of Kar3-Kar3 combining two catalytic domains or of Kar3-Kar3rigor complexes combining catalytically active and inactive Kar3 heads (Figure 3A). These results point to a specific role of the non-catalytic Cik1 head that cannot be simply replaced by a second catalytically active or inactive Kar3 head.10.7554/eLife.04489.011
Molecular requirements for processivity and identification of a translocation-deficient Cik1 mutant.
(A) Schematic showing analyzed Cik1 and Kar3 truncation constructs with the corresponding results from TIRF assays and multi-motor gliding assays. All constructs contained the Halo-tag at the aminoterminus of Kar3 for fluorescent labeling with TMR. (B) Kymographs of TMR-labeled Kar3 complexes containing either full-length Cik1 (aa 1–594) or the carboxyterminal truncation mutant Cik11–521. See Videos 4 and 5. (C) Kymograph of microtubule-gliding by full-length Cik1–Kar3 and the Cik11–521-Kar3 mutant. Note the impairment of directional translocation by the Cik11–521–Kar3 mutant.
DOI:http://dx.doi.org/10.7554/eLife.04489.011
10.7554/eLife.04489.012
Wild-type Cik1–Kar3-TMR in the presence of 5 mM ATP moving on taxol-stabilized microtubules imaged at higher frame rate in single color with TIRF microscopy. Time between frames is 273 ms, the frames are numbered in the upper left corner, total length of the video is 54 s. The video is played at 20 frames/s. Scale bar is 3 μm. The video corresponds to <xref ref-type="fig" rid="fig3">Figure 3B</xref>.
DOI:http://dx.doi.org/10.7554/eLife.04489.012
10.7554/eLife.04489.013
Cik1–Kar3-TMR with truncation of the non-catalytic Cik1 globular domain (Cik1 1–521) interacting with taxol-stabilized microtubules in the presence of 5 mM ATP. Frame rate and imaging conditions as in <xref ref-type="other" rid="video5">Video 5</xref>. Note non-directional movement of Cik1<sup>1–521</sup>-Kar3. Scale bar 3 μm. The video corresponds to <xref ref-type="fig" rid="fig3">Figure 3B</xref>.
DOI:http://dx.doi.org/10.7554/eLife.04489.013
The non-catalytic domain restrains diffusion of Kar3 motors
To analyze the motile cycle in more detail, we imaged single Cik1–Kar3 motors in different nucleotide states by TIRF microscopy with high temporal resolution. In the absence of nucleotide (+Apyrase) and in the presence of the non-hydrolyzable ATP analog AMP-PNP Kar3 motors bound with high affinity to the microtubule but displayed no movement (Figure 4A). Binding under these conditions was sensitive to ionic strength with on-rate and lifetime of motor–microtubule interactions decreasing as the salt concentration was raised from 30 mM to 100 mM KCl (not shown). In the ADP state, motors displayed diffusive microtubule interactions (D = 0.061 ± 0.003 μm2/s) with short residence times (τ = 0.6 ± 0.1 s) (Figure 4B,C).10.7554/eLife.04489.014
Single-molecule analysis of Kar3 motors in different nucleotide states.
(A) Kymographs showing single molecule TIRF microscopy of Cik1–Kar3 motors in different nucleotide states. Videos were taken with high temporal resolution (35 frames per second). Note the different motor concentrations and the directional displacement of Cik1–Kar3 molecules in the presence of ATP. (B) Diffusive movement of individual Cik1–Kar3 molecules in the presence of ADP. (C) MSD analysis of the diffusive movement of Cik1–Kar3 motors in the presence of ADP. Molecules with lifetimes between 0.5 and 5 s were analyzed. (D) Typical kymographs of Cik11–360–Kar3 motors lacking the non-catalytic head domain in no-nucleotide (Apyrase), AMPPNP and ATP states. Note the diffusive interactions of the motor with the microtubule in comparison to 3A. (E) Mean-squared displacement (MSD) analysis of Cik1–Kar3 and Cik11–360–Kar3 motors in the presence of ATP. Data points were fitted to the formula <x2> = a·tn. n = 1 for Cik11–360–Kar3 indicates random diffusion without bias and constraints, while n = 2 for the wild-type motor indicates directional, processive movement. (F) Quantitative comparison of microtubule interactions of wild-type vs Cik11–360–Kar3 motors in no nucleotide and AMPPNP states by mean squared displacement analysis. (G) MSD analysis of wild-type vs Cik11–360–Kar3 motors in the ADP state. (H) Summary of the diffusion coefficients obtained for wild-type vs Cik11–360-Kar3 motors in three different nucleotide states. Note the logarithmic scale on the y-axis. (I) Typical kymographs of a monomeric Kar3 head construct (residues 353–729 fused to an N-terminal Halo tag) in the presence of ATP.
DOI:http://dx.doi.org/10.7554/eLife.04489.014
What is the contribution of the non-catalytic domain to the translocation mechanism? To gain insights into this question, we quantitatively compared the microtubule-binding properties of single wild-type Cik1–Kar3 molecules with a mutant that lacks the globular Cik1 motor homology domain (Cik11–360–Kar3) (Figure 4D). In contrast to wild-type motors, the Cik11–360–Kar3 mutant displayed short-lived diffusive interactions with microtubules (Figure 4D). MSD data obtained in ATP for wild-type and mutant were fitted to the equation <x2> = a·tn. n = 1 suggests standard diffusion and in this case, a = 2D, where D is the one-dimensional diffusion coefficient. For the mutant, an optimal fit was obtained with n = 1.01 ± 0.04, indicating unconstrained diffusion with no directional component. By contrast, n = 1.96 ± 0.03 was obtained for the wild-type motor, consistent with directional processive movement (Figure 4E). Further MSD analysis indicated that the mutant is about 20-fold more diffusive than the wild-type motor in the no nucleotide state (D = 0.01 ± 0.0006 μm2/s vs 0.00045 ± 0.00011 μm2/s) and about 30-fold more diffusive in the AMPPNP state (D = 0.0036 ± 0.00011 μm2/s vs 0.00011 ± 0.00001 μm2/s) (Figure 4F). Further analysis indicated that the Cik11–360 mutation also increased diffusion in the ADP state, but the difference was less pronounced in comparison to the other states (Figure 4G,H). We additionally constructed and purified a monomeric Kar3 head encompassing residues 353–729 with an N-terminal Halo-Tag. Under standard imaging conditions in the presence of ATP only very short-lived microtubule interactions without directional movement could be observed (Figure 4I). Taken together, these results indicate only a heterodimer containing a non-catalytic Cik1 head is able to move processively and a key contribution of the non-catalytic domain is to promote tight, non-diffusive binding of the motor in no nucleotide and AMPPNP states.
The non-catalytic head determines the velocity of the Kar3 motor
Having established that the non-catalytic partner is critical for achieving mechanically processive Kar3, we sought to gain further insights into the underlying mechanism by studying the alternative Vik1–Kar3 motor. The non-catalytic proteins Cik1 and Vik1 are paralogs that display about 45% sequence similarity between each other. We purified Vik1–Kar3 complexes from yeast extracts and imaged their interaction with microtubules under the same conditions as previously employed for Cik1–Kar3 (Figure 5A). Interestingly, Vik1–Kar3 also displayed highly processive movement, but with faster velocity compared to Cik1–Kar3. Under standard assay conditions, Vik1–Kar3 complexes moved at 234 ± 29 nm/s towards microtubule minus-ends where they strongly accumulated (Figure 5B, Video 6). To ask how these differences might be determined by the non-catalytic partner, we constructed a chimeric protein in which the globular motor homology domain and the neck of Vik1 (aa 351–647) were fused to the tail domain of Cik1 (aa 1–353) (Figure 5C). The Vik1–Cik1 chimera formed a stable heterodimer with Kar3-HALO and supported movement with 188 nm/s to the minus-end, significantly faster than Cik1–Kar3 and approaching the velocity of Vik1–Kar3 (Figure 5D,E). Thus, the globular non-catalytic domain is a key determinant for setting the velocity of the motor. The engineered chimeric protein was able to functionally replace Cik1 in cells as demonstrated by its ability to rescue the phenotypes of a Cik1 deletion and to support growth at wild type level under all tested conditions (Figure 5F).10.7554/eLife.04489.015
The non-catalytic head domain determines the velocity of the Kar3 motor.
(A) Schematic representation of the kinesin-14 Vik1–Kar3 and purification of recombinant Vik1–Kar3 from yeast extracts. Motors are covalently labeled with Tetramethylrhodamine (TMR) via a HaloTag on the amino-terminus of Kar3. Coomassie-stained SDS-PAGE shows homogeneity of the motor preparation and fluorescent labeling of the Kar3 subunit. (B) Kymographs of single color time lapse TIRF microscopy highlighting the velocity difference between Vik1–Kar3 and Cik1–Kar3 when moving along taxol-stabilized microtubules. (C) Construction and purification of a chimeric motor in which the non-catalytic heads were switched. An asterisk denotes a proteolytic degradation product. (D) Kymograph of typical single molecule runs by the chimeric Cik1Tail–Vik1Head motor. (E) Comparison of velocities of the different Kar3 constructs. (F) Serial dilution growth assay of the indicated yeast strains at different temperatures and conditions. Note that the integration of the chimeric Cik1Tail–Vik1Head fusion protein fully rescues the phenotypes of a Cik1 deletion.
DOI:http://dx.doi.org/10.7554/eLife.04489.015
10.7554/eLife.04489.016
Vik1–Kar3 motors display processive movement and pronounced minus-end dwelling. Two-color time lapse TIRF microscopy of Vik1-Kar3-TMR (red) on taxol-stabilized microtubules (blue). Note strong accumulation of motors at minus-ends at the end of the video. 100 frames were taken every 3 s, video is played at 20 frames/s, scale bar is 5 μm. The video corresponds to <xref ref-type="fig" rid="fig5">Figure 5B</xref>.
DOI:http://dx.doi.org/10.7554/eLife.04489.016
Differential binding partners determine the subcellular localization of Kar3 in vivo
In addition to directly controlling the motile characteristics of individual Kar3 motors, the non-catalytic partners may also provide functional diversification by allowing differential interactions with regulatory proteins or cargos. To identify such interactors, we performed affinity purifications of Cik1–FLAG at different cell cycle stages and analyzed associated proteins by mass spectrometry. In addition to Kar3, we reproducibly mapped two plus-end tracking proteins Bim1 and Bik1, the EB1 and CLIP-170 homologues of budding yeast, as well as the microtubule-binding Ndc80 kinetochore complex (Figure 6A). By performing pull-down assays with recombinant proteins, we established that Cik1–Kar3, but not Vik1–Kar3, directly interacted with Bim1 (Figure 6B). The binding interface involved the C-terminal EB homology domain of Bim1 and the aminoterminal tail domain of Cik1 (Figure 6—figure supplement 1A–D). A physical interaction with Bim1 may account for the previously observed microtubule plus-end localization of Cik1–Kar3 in vivo (Sproul et al., 2005; Gardner et al., 2008). Live cell imaging showed that the localization of Kar3 to distinct foci along the yeast spindle, as well as to the tips of shmoo-tip-directed microtubules in alpha-factor-arrested cells (Figure 6—figure supplement 1E), was abolished in a bim1Δ strain, pheno-copying a cik1 deletion (Manning et al., 1999) (Figure 6C,D,E). Consistent with our biochemical experiments, spindle localization was maintained in the Cik11–360 and Cik11–521 mutants. Thus, the Cik1 tail domain specifies the differential localization of Kar3 in vivo by allowing a direct interaction with Bim1.10.7554/eLife.04489.017
Cik1 specifies differential interactions of Kar3 motors to determine their subcellular localization.
(A) Affinity purification of the Cik1–Kar3 complex from different cell cycle states and identification of interaction partners by mass spectrometry. (B) Pull-down assay with GST-Bim1 and Cik1- or Vik1-Kar3. Only Cik1-Kar3 interacts with Bim1 in a dose-dependent manner. (C) Localization of Kar3 in different deletion mutants in yeast investigated by live cell microscopy. (D) Panel shows line scans of Kar3–GFP intensity along the spindle axis. Arrowheads point to Kar3 foci along the spindle. Scale bar: 3 μm. Both Cik1 and Bim1 are required for the localization to the anaphase spindle. (E) Quantification of Kar3 foci on anaphase spindles in the indicated yeast strains. Kar3 spindle signals were never detected in a bim1Δ or cik1Δ strain. (F) The Cik11–521 and Cik11–360 mutants elicit a temperature-sensitive phenotype in vivo. Growth assay was performed with serial dilutions of the indicated yeast strains at different temperature on rich medium (YPD). (G) Quantification of chromosome segregation in the indicated yeast strains at 25°C, n = 100 for all strains analyzed. Right panel shows representative fluorescent micrographs with segregation of fluorescently labeled Chromosome V.
DOI:http://dx.doi.org/10.7554/eLife.04489.017
10.7554/eLife.04489.018
Biochemical and genetic characterization of the Cik1-Kar3-Bim1 interaction.
(A) Analytical size exclusion chromatography of Cik1–Kar3 and Bim1 individually or in combination. Shift to earlier elution position indicates complex formation. (B) Sequence alignment of the amino-termini of selected Cik1 and Vik1 proteins from diverse yeasts. Note the presence of an SxIP motif (EB1 interaction motif) in Cik1 proteins that is absent from Vik1. (C) Mutation of the SxIP motif in Cik1 does not impair binding of the Cik1–Kar3 complex to Bim1 in GST-pull down assays. (D) The more extended amino-terminus of Cik1 is required for Bim1 binding. GST-Bim1 pull-down assays with indicated Cik1–Kar3 truncation mutants. Elimination of the first 250 amino acids in Cik1 impairs binding of the motor complex to Bim1. (E) Localization of Kar3 to the tips of shmoo-directed microtubules in alpha-factor arrested cells depends on Bim1. Live cell microscopy of Kar3-3xGFP (green) in either wild-type (upper panel) or bim1-deletion mutant (lower panel). Note the lack of microtubule plus-end localization of Kar3 in the bim1 deletion mutant. (F) Mutation of the conserved SxIP motif in Cik1 does not cause a growth phenotype in vivo. Spot assays were carried out with the indicated strains either on YPD plates or on YPD plates supplemented with 100 mM Hydroxyurea. (G) Binding of Cik1–Kar3 to Bim1 occurs via the cargo-binding domain. GST-pull down assays with full-length Bim1 or versions constituting the microtubule (Lampert et al., 2013) in the absence of the motor.
DOI:http://dx.doi.org/10.7554/eLife.04489.018
The insights into the localization determinants of the Kar3 motor allowed us to evaluate the key roles of Kar3 in the cell: yeast strains expressing Cik11–360 or Cik11–521 from the endogenous chromosomal locus displayed slow growth at 25°C and were inviable at 37°C, similar to a cik1 deletion (Figure 6F). By contrast neither a bim1 deletion, which eliminates spindle localization, nor a vik1 deletion, removing spindle pole localization, displayed a pronounced growth phenotype (data not shown). Together with our finding that the Vik1head–Cik1tail chimera fully supports viability, this implies that the key function of Kar3 in vegetatively growing yeast cells requires motility—supported by any of the two non-catalytic heads—and the Cik1 tail domain, but does not involve spindle localization via Bim1. Based on previously characterized mutants and on our finding that the Ndc80 complex co-purifies with Cik1, this supports a key role for Kar3 at the kinetochore. Consistent with this notion, we found that 80% of Cik11–521 cells arrested as large budded cells upon shift to the restrictive temperature, indicative of mitotic checkpoint activation. Analysis of chromosome segregation at the semi-permissive temperature of 25°C by fluorescent labeling of chromosome V showed that 50% of large budded Cik1 mutant cells attempted nuclear division without proper bi-orientation of sister chromatids (Figure 6G). We conclude that the key contribution of the non-conventional Cik1 motility mechanism for cell proliferation lies in the promotion of sister chromatid bi-orientation in mitosis.
Cik1–Kar3 motors can promote transport of the Ndc80 kinetochore complex in vitro
To ask whether Cik–Kar3 can act as a transport motor, we first performed bulk in vitro recruitment assays in the absence of nucleotide using purified motor proteins and recombinant, fluorescently labeled Ndc80 complex as the candidate kinetochore binding partner revealed in the mass spectrometry experiments. In the absence of motor, the Ndc80 complex only weakly associated with taxol-stabilized microtubules under standard conditions (Figure 7A,B). Vik1–Kar3 motors strongly decorated microtubules in the absence of ATP, but had no effect on the Ndc80 complex. By contrast, combining Cik1–Kar3 and the Ndc80 complex resulted in decoration of microtubules with both molecules. Upon lowering protein concentrations, Cik1–Kar3 and the Ndc80 complex co-localized to distinct spots on the microtubule where their intensities were correlated (Figure 7C). Inclusion of ATP initiated co-transport of the Ndc80 complex and Cik1–Kar3 towards the minus end (Figure 7D, Video 7). While transporting the Ndc80 complex, the average speed of the motor was 76 ± 15 nm/s (mean ± SD), indicating that cargo binding did not significantly impede movement. The value is also in range with the reported velocity for kinetochore transport in vivo (Tanaka et al., 2005). Microtubule binding by the Ndc80 complex was not required for efficient transport, as shown by effective recruitment and translocation of the microtubule-binding defective calponin-homology domain mutant K122E K204E (Lampert et al., 2013) (Figure 7—figure supplement 1). This result is in agreement with the observation that Kar3 localizes to unattached kinetochores prior to microtubule binding. We conclude that the Ndc80 complex can be transported by the Kar3 motor in vitro and that in addition to promoting processive motility a key function of Cik1 may lie in the differential binding of this complex.10.7554/eLife.04489.019
Processive transport of the Ndc80 kinetochore complex by Cik1–Kar3 motors in vitro.
(A) Cik1–Kar3, but not Vik1–Kar3 recruits the Ndc80 kinetochore complex to taxol-stabilized microtubules in vitro. The experiment was performed with TMR-labeled Kar3 motors and fluorescently labeled recombinant Ndc80 complex in the absence of nucleotide. (B) Quantification of the recruitment experiment by analyzing fluorescence intensity of microtubule-bound Ndc80-eGFP complex in the presence of different motor constructs. Error bars denote standard error of the mean (s.e.m.) (C) Co-localization of the Ndc80 kinetochore complex and Cik1–Kar3 motors to distinct signals on the microtubule lattice. Right panel shows line scans of fluorescent intensity of Kar3 motors and Ndc80 complex along the length of a microtubule. (D) Kymograph showing triple-color time-lapse TIRF microscopy demonstrating that in the presence of ATP, Ndc80 complexes are processively transported towards microtubule minus ends by Cik1–Kar3 motors.
DOI:http://dx.doi.org/10.7554/eLife.04489.019
10.7554/eLife.04489.020
Cik1–Kar3 transports a mutant Ndc80 kinetochore complex that is unable to bind MTs.
Schematic and kymographs of an effective recruitment and translocation of the microtubule-binding defective calponin-homology domain mutant K122E K204E Ndc80 complex by the motor Cik1–Kar3. The experiment was performed with TMR-labeled Kar3 motors, Alexa647-taxol stabilized MTs, and fluorescently labeled recombinant Ndc80 complex in the presence of ATP.
DOI:http://dx.doi.org/10.7554/eLife.04489.020
10.7554/eLife.04489.021
Transport of the Ndc80 kinetochore complex by Cik1–Kar3 motors along taxol-stabilized microtubules. Triple-color TIRF video showing Ndc80-Alexa488 complex (green), Cik1–Kar3-TMR motors (red), and Alexa647-Taxol stabilized microtubules (blue) in the presence of 5 mM ATP. Frames were taken every 3 s. Note co-transport of Ndc80 complexes and Cik1–Kar3 complexes towards the end of the microtubule. Video is played at 20 frames/s, scale bar is 5 μm. The video corresponds to <xref ref-type="fig" rid="fig7">Figure 7D</xref>.
DOI:http://dx.doi.org/10.7554/eLife.04489.021
DiscussionKar3 motors and non-conventional translocation along microtubules
Our study provides direct evidence for the processivity of Kinesin-14 motors by analyzing full-length yeast Kar3 motors on the single-molecule level. Contrary to other kinesin-14s, we demonstrate that individual Cik1–Kar3 motors can move processively towards the minus-end and show that the non-catalytic Cik1 head domain is functionally required for this activity. We further demonstrate that the combination of Kar3 with two different non-catalytic partners generates motors with different motile properties and also allows the differential binding of partner proteins such as Bim1 and the Ndc80 complex (Figure 8A).10.7554/eLife.04489.022
Model for Kar3 motility and function.
(A) Functional contributions of the different domains of heterodimeric Kar3 motors analyzed in this study. (B) Model for Cik1–Kar3 motility depending on a non-catalytic head domain. Upon ATP uptake a conformational change occurs with the stalk rotating towards the minus end. A key role for the non-catalytic domain is to prevent diffusion in this state but allow one-dimensional diffusion in the subsequent ADP state with a bias towards the minus end.
DOI:http://dx.doi.org/10.7554/eLife.04489.022
The demonstration of Cik1–Kar3 processivity presented here provides evidence that effective translocation along microtubules does not strictly require the conventional hand-over-hand walking mechanism that has been established for conventional Kinesin-1. So far this point has been most clearly demonstrated for the Dynein motor, as recent studies have shown that cytoplasmic dynein is capable of processive movement in vitro despite inactivating mutations in one of the two force-generating AAA ring domains (DeWitt et al., 2012; Qiu et al., 2012), or even as a combination of one active head with a second passive microtubule-binding domain (Cleary et al., 2014).
Also several kinesin motors have been reported to move in violation of the hand-over-hand model: monomeric constructs of the kinesin-3 KIF1A have been shown to move along microtubules in vitro using a biased diffusion mechanism (Okada and Hirokawa, 1999), but the in vivo function of these motors probably involves dimeric molecules for effective cargo transport (Tomishige et al., 2002; Endres et al., 2006). We could not detect evidence for processive movement of monomeric Kar3, making it unlikely that it works via a KIF1A-type mechanism. Among the kinesin-14 family, Ncd can exhibit processive motility but requires at least two homodimers coupled together (Furuta et al., 2013). A number of studies have shown that inactivating mutations in one of the two heads of conventional kinesin can still allow residual processivity (Subramanian and Gelles, 2007; Thoresen and Gelles, 2008; Kaseda et al., 2003). We note, however, that in these examples processive movement of the mutant heterodimeric kinesins was compromised compared to wild-type homodimers. By contrast, Kar3 does not simply tolerate the loss of a second active head but instead it has evolved a specific requirement for the non-catalytic domain in order to move processively. In other words, Kar3 can only walk with a combination of an active and an inactive ‘leg’.
A working model for the movement of Kar3 kinesins
A precise elucidation of the stepping mechanism of Kar3 motors will require further biophysical experiments. Based on the results obtained in this and in previous studies, we propose a working model that describes the non-catalytic head as a ‘foothold’ for Kar3. Based on recent structural data Kar3 and its partner non-catalytic head may be located on adjacent protofilaments (Rank et al., 2012), where the non-catalytic domain could contribute to hold the motor complex in place and prevent diffusion. During its ATP cycle, the motor goes through rounds of tightly and weakly bound states. When ADP is bound, the motor is in its weakly bound state which allows for one-dimensional diffusion along the microtubule lattice. On its own, movement in this state would lack directionality and an additional mechanism providing minus-end directed bias is required. We speculate that the presence of Cik1 in combination with the kinesin-14 typical stalk rotation upon ATP binding by Kar3 (Endres et al., 2006; Gonzalez et al., 2013) might provide such a mechanism either by shifting the overall position of the molecule or by allowing subsequent directed binding/unbinding (Figure 7C). Because of the combination of one foot on the track and a diffusive movement to the next binding site, we refer to this possible mechanism as the ‘skateboard’ model. This model can explain a number of our experimental observations: the effect of raising the ionic strength is that initially motor movement occurs with faster velocity, probably by increasing the diffusion term. At some point however, stable binding of the non-catalytic domain is prevented and directional displacement is disrupted, similar to the situation in which the motor lacks the motor homology domain. In addition, diffusive movement in the ADP state and the observed large variation in movement rates are compatible with the model. The biochemical properties of the non-catalytic head binding to the microtubule will have profound effects on the motile characteristics as evidenced by the different properties of Cik1–Kar3 and Vik1–Kar3 motors.
One of the open questions regarding the mechanism of processivity is how the motor homology domain can provide ‘footing’ in no nucleotide and AMPPNP states yet allow effective diffusion in the ADP state. In agreement with previous work, this suggests that similar to conventional kinesins, a form of head-to-head communication must occur in Kar3 motors such that the catalytic head influences the microtubule affinity of the motor homology domain (Allingham et al., 2007; Chen et al., 2011; Duan et al., 2012; Rank et al., 2012). Such coordination may involve intramolecular strain communicated by mechanical elements between the heads as has been reported for conventional kinesin (Yildiz et al., 2008).
Implications for Kar3 function in vivo
Based on our experiments and the previously reported rates of microtubule shortening in the presence of Cik1–Kar3, we consider it unlikely that Kar3 functions as a microtubule depolymerase in vivo. We note that none of the Kar3 functions poses a strict requirement for such an activity and indeed recent experiments analyzing Kar3 function during karyogamy support the view that its primary function is transport, not shortening (Gibeaux et al., 2013).
What are the implications of Kar3's processivity? The ability to move processively may be used in the cell to enrich Kar3 motors at minus-ends. In addition, Kar3 could work more effectively in small teams if it can undergo multiple catalytic cycles before releasing from the microtubule. This may be especially important during kinetochore transport, where Ndc80 complexes can provide only a limited number of binding sites for the motor. Kar3 heterodimers have probably evolved from conventional Kinesin14-homodimers that performed Ncd-like roles in spindle organization. We speculate that with the exclusion of dynein from the budding yeast nucleus additional functional requirements for minus-end directed motility in spindle assembly and kinetochore function arose that could be fulfilled with the ‘invention’ of the Cik1 and Vik1 non-catalytic domains. While commonly perceived as conferring a loss of functionality, the pseudo-motor domains have the ability to convert a kinesin-14 into a processive motor and additionally create functional diversity through allowing differential interactions with partners. Together this allows Kar3 to function as a processsive sliding and transport motor that substitutes for key roles of dynein in the yeast nucleus. Other examples for heterodimeric motors exist in the Kinesin-2 family with different subunits conferring distinct activities (Brunnbauer et al., 2010). Kar3 motors may therefore constitute an extreme example for such diversification strategies that could be more widely used in other motor families. More generally, we note that pseudoenzymes have emerging roles in other cellular contexts, for example in the form pseudo-GTPases during human kinetochore assembly (Basilico et al., 2014), or as pseudo-kinases and pseudo-phosphatases in cell signaling (Boudeau et al., 2006; Tonks, 2009). This may suggest that the use of catalytically inactive enzyme derivatives could be a more widespread strategy employed by the cell. The establishment of an in vitro assay for Ndc80 transport serves as a starting point for a detailed analysis of the function and regulation of kinetochore motility. Furthermore, the unique design principle of the Kar3 motor, which allows the same catalytic domain to be paired with different non-catalytic heads, generates functional diversity that may also be exploited in nanotechnological applications.
Materials and methodsProtein Biochemistry
The protein coding sequences of S. cerevisae (S288c) full-length Kar3, Cik1, and Vik1 were amplified by PCR and cloned into the overexpression pESC-TRP (Agilent Technologies, Santa Clara, CA) vector. The non-catalytic motor subunit was tagged C-terminally with 1xFLAG and the motor itself was fused to a HaloTag (DHA, Promega) at the 5′ end of the coding sequence, separated by a 13 amino acid linker. Mutated and truncated versions of Cik1 and Kar3 were generated by site-directed mutagenesis PCR (Phusion, Thermo Scientific). To determine the TMR-labeling efficiency, Halo-Kar3 was fused with enhanced green fluorescent protein (eGFP), creating Halo-eGFP-Kar3. All motor proteins were overexpressed in budding yeast using the pESC vectors (Agilent Technologies) following the manufacturer's instructions. In brief, yeast cells harboring the respective pESC plasmid were induced with 2% Galactose at OD600 = 1.0 for 7–9 hr, harvested by centrifugation, washed, and frozen as droplets in liquid nitrogen. Lysis was conducted in liquid nitrogen using a freezer mill (Biospec Inc.). The cell powder was resuspended in lysis buffer (25 mM Hepes [7.4], 300 mM NaCl, 1 mM MgCl2, 5% glycerole, 0.1 mM EDTA, 0.5 mM EGTA, 0.1% Tween-20, 0.01 mM ATP, 0.1 mM PMSF supplemented with PhosSTOP Phosphatase Inhibitor Cocktail [Roche]). The lysed cells were centrifuged twice, first at 43.146×g for 20 min and afterwards at 125.749×g for 1 hr. The resulting supernatant was incubated with M2 affinity agarose (Sigma–Aldrich) for 1 hr, gently rotating at 4°C. The agarose resin was washed five times with lysis buffer (adjusted to 150 mM NaCl, 0.09% Tween-20, omitting the PhosSTOP reagent). Elution of the kinesin heterodimeric constructs from M2 agarose was achieved by applying one resin volume of 3xFLAG peptide at final 2 mg/ml in lysis buffer (adjusted to 250 mM NaCl, 1 mM DTT, 0.09% Tween-20, omitting ATP and PhosSTOP). If needed, the elution was loaded onto a cation exchange chromatography (MonoS 5/50 GL, GE Healthcare) in running buffer (10 mM Hepes [pH 7.2], 150 mM NaCl, 1 mM MgCl2, 1 mM DTT, 1 mM EGTA) to remove the FLAG peptide. Afterwards a linear salt gradient eluted a single peak, pure motor fraction at 250 mM salt. Elution fractions were supplemented with glycerol (final concentration: 5%), snap-frozen in liquid nitrogen and stored at −80°C. The protein concentration was measured using the DC Assay kit (Bio-Rad). All proteins were pre-cleared by centrifugation using a 0.1-μm spin filter (Millipore) to remove aggregates before each experiment.
Labeling of the NH2-terminus of Kar3 with the HaloTag TMR ligand or HaloTag Alexa488 ligand (Promega, Madison, WI) was performed during the above-described purification before eluting the kinesin heterodimer from the M2 affinity agarose: the proteins were incubated with 10 μM HaloTag ligand for 3 hr. Extensive washing removed unbound TMR ligand and the kinesin was eluted as described before. In order to assess the labeling efficiency the Halo-eGFP-Kar3 construct was expressed, purified, and TMR-labeled. Observing the motor in our multi-color single-molecule imaging setup in the absence or presence of ATP revealed that >90% of eGFP-labeled kinesins also had a TMR-ligand covalently bound to the HaloTag.
To obtain a stoichiometric heterodimer of Cik1 and Kar3, the motor was subjected to analytical SEC. The purified motor was loaded onto a Superose 6 PC 3.2/30 column (GE Healthcare), and 100 μl fractions were collected and separated by SDS-PAGE. Proteins were stained with Coomassie Brilliant blue R250.
Expression and purification of the full-length and mutant Ndc80 complex (Ndc80p-6xHis/Nuf2p-EGFP and Spc24p/6xHis/Spc25p) was performed as described previously (Lampert et al., 2013; Lampert et al., 2010). Bim1 was expressed and purified as described previously (Zimniak et al., 2009).
Rotary shadowing electron microscopy of full-length Cik1–Kar3
Peak fractions of motor from the gel filtration experiments were diluted to a final concentration of 60 μg/ml in spraying buffer, containing 100 mM ammonium acatate and 30% (vol/vol) glycerol, pH adjusted to 7.4. After dilution, the samples were sprayed onto freshly cleaved mica chips and immediately transferred into a Bal-Tec MED020 high vacuum evaporator equipped with electron guns. After drying in the vacuum, the rotating samples were coated with 0.6 nm Platinum at an elevation angle between 5° and 6°, followed by 9.5 nm Carbon at 90°. The produced replicas were floated off from the mica chips and picked up on 400 mesh Cu/Pd grids. The grids were inspected in an FEI Morgagni 268D TEM operated at 80 kV. Electron micrographs were acquired using an 11 megapixel Morada CCD camera from Olympus-SIS. Images were examined and analyzed using ImageJ.
Sucrose gradient centrifugation
Single-step purified Cik1–Halo–Kar3 was loaded on top of a 4.4 ml 5–25% linear sucrose gradient and spun at 50,000 rpm for 16 hr using an Sorvall TH-660 rotor and a Sorvall Discovery 90SE centrifuge. Fractions (270 μl) were collected and analyzed by SDS-PAGE and Coomassie Brilliant Blue R250 staining. The sedimentation value for Cik1-Halo-Kar3 was defined by comaring the mobilities of the motor with linear plots of mobility standards.
Single molecule imaging assay
The single molecule motor assays were conceptually designed as described previously (Korten et al., 2011; Lampert et al., 2010; Bieling et al., 2010). Biotin-PEG-SVA- and mPEG-SVA-functionalized coverslips (Laysan Bio) were prepared as described (Lampert et al., 2010; Jain et al., 2012). Coverslips were assembled onto passivated glass slides using double-sided tape creating a flow chamber. First, a solution of 1 mg/ml avidin DN (Vector Laboratories) was applied to the chamber for 30 min and exchanged for 1% pluronic F-127 (Sigma–Aldrich) in BRB80 (Zimniak et al., 2009; Nitzsche et al., 2010). Porcine-derived HiLyte-647-labeled, biotinylated and taxol-stabilized microtubules (MTs) were immobilized (labeled and biotinylated tubulin source: Cytoskeleton Inc., unlabeled tubulin was purified from pig brains as described previously [Ashford and Hyman, 2006]) and excess of MTs was washed out with BRB80 buffer supplemented with 5 μM taxol, 0.5% (vol/vol) β-mercaptoethanol, 4.5 μg/ml glucose, 200 μg/ml glucose-oxidase, and 35 μg/ml catalase. Single molecule imaging was performed by introducing the TMR- or Alexa488-labeled kinesins at low nanomolar range in assay buffer (BRB80, 0.33 mg/ml casein, 16.6 μM taxol, 0.13% [vol/vol] methylcellulose, 0.5% [vol/vol] β-mercaptoethanol, 4.5 μg/ml glucose, 200 μg/ml glucose-oxidase and 35 μg/ml catalase, 0.06% [vol/vol] Tween-20, the indicated amount of KCl and the respective amount of nucleotide) into the flow cell. Time-lapse videos were recorded at 28°C in 3 s intervals between frames (if not annotated differently) using a TIRF microscopy setup described previously (Lampert et al., 2010). Multi-color imaging was achieved by the use of an external filterwheel (Ludl Electronic Products Ltd.). Each channel (excitation: 488 nm, 561 nm, 639 nm) was exposed for 100 ms at every time-interval, recorded by a Cascade II EMCCD camera and projected to two-dimensional images (software: Metamorph [Molecular Devices], ImageJ). For photobleaching analysis, the oxygen-scavenger mix was omitted (β-mercaptoethanol, glucose, glucose-oxidase, catalase), and images were recorded at maximum laser power. Videos are represented as kymographs (time-space plot) or as example single frame (software: MetaMorph [Molecular Devices]).
High temporal resolution recordings (as presented in Figure 2) have been obtained using custom-made TIRF microscope based on Olympus IX-71 body and Coherent CUBE lasers in a temperature stabilized room (21 ± 0.1°C). Images were acquired using a Andor iXon3 897 EMCCD camera and subsequently analyzed using custom-made software written in MATLAB (MathWorks, Inc).
Microtubule gliding assay
The microtubule gliding assay was designed as described before (Nitzsche et al., 2010). For this assay, hydrophobic coverslips were prepared according to the following scheme: sonication in acetone for 15 min was followed by sonication for 15 min in ethanol. Coverslips were incubated 1 hr in boiling Piranha solution and rinsed with a lot of water afterwards. Then rinsed with 0.1 M KOH, MilliQ and dried with nitrogen and immersed in 5% dichlorodimethylsilane in heptane for 1 hr at room temperature. Coverslips were rinsed again with MilliQ and sonicated for 5 min, sonicated in chloroform for 5 min and air-dried.
Motor proteins were attached to the hydrophobic coverslips via application of 0.2–20 μg/ml anti-Halo antibody (Promega) to the flow chamber. After the excess of antibody was washed out with BRB80 supplemented with 0.5 mg/ml casein, motors were introduced to the chamber and incubated for 5 min. Finally a microtubule-containing solution supplemented with 5 mM ATP and oxygen scavenger mix was perfused into the chamber. Microtubules used for this experiment were assembled from porcine tubulin mixed together with HiLyte-647-labeled tubulin (Cytoskeleton Inc.) in the presence of 10 μM taxol. Time-lapse videos were recorded at 28°C in 3-s intervals between frames using a TIRF microsopy setup described previously (Lampert et al., 2010). MetaMorph (Molecular Devices) software was used to compile images into videos and create kymographs out of individual moving microtubules. Velocity of gliding microtubules was calculated based upon the slope of the kymographs.
Data analysis
The tracking of the proteins was performed automatically using the Definiens Software Suite. Prior to the analysis, the image data were processed performing a shading correction and smoothing. This was done by applying an 11 × 11 pixel (1463 × 1463 nm) kernel median filter and dividing the original raw image by the filtered image data. The resulting image was smoothed, using a 3 × 3 (399 × 399 nm) kernel Gaussian filter. The processed image data were searched and segmented for fluorescent signal. The signal area was searched for local intensity maxima within a search range of 532 nm. Circular objects with a radius of 332 nm were created on the found maxima and used for tracking the identified proteins.
The tracking of the movement of the TMR-, Alexa-488-, or GFP-labeled proteins was done through linking the proteins frame by frame by direct overlap, using the best fitting overlap.
Bioinformatic methods
Cik1 protein sequences of Saccharomyces cerevisiae (accession number NP_013925.1), Saccharomyces kudriavzevii (EJT42048.1), Saccharomyces arboricola (EJS44163.1) and Vik1 sequences from S. cerevisiae (NP_015070.1), S. kudriavzevii (EJT43871.1), S. arboricola (EJS41496.1) were retrieved from the National Center for Biotechnology Information (NCBI). The Saccharomyces bayanus protein sequences of Cik1 (WashU_Sbay_Contig651.30) and Vik1 (WashU_Sbay_Contig637.18) were retrieved from the Saccharomyces Genome Database (SGD). The sequences were aligned with MAFFT (version 6, L-INS-I method) and further processed with Jalview and colored according to ClustalW.
In vitro binding assay
Varying amounts (0.1–1 μM) of recombinant bait protein (GST, GST-Bim1FL, GST-Bim10–185, GST-Bim1185–344) were immobilized on 30 μl glutathione sepharose (GE Healthcare) in 0.5 ml binding buffer (25 mM Hepes pH 7.2, 250 mM NaCl, 1 mM MgCl2, 1 mM EGTA, 0.5 mM DTT, 0.05% NP-40). The binding partner was added at a constant concentration between 0.5 and 1 μM and incubation lasted for 1 hr at 4°C. Afterwards beads were washed three times with 0.5 ml binding buffer and analyzed by SDS-PAGE and Coomassie Brilliant blue R250 staining.
Yeast strains and spot assay
All modifications were performed in the S288C background (Supplementary file 2). Genetic modifications were introduced by using standard procedures.
For the spot assay, the desired strains were grown overnight in YPD medium. The following day cells were diluted to OD600 = 0.6 which was the starting point of a 1:4 dilution series and spotted on YPD or 100 mM hydroxyurea (HU) plates. Plates were incubated at the indicated temperatures to 2–3 days.
Live cell imaging
Imaging strains were grown in synthetic medium containing 2% glucose and imaged on concanavalin A-coated culture dishes (Matek) at ambient temperature. Eight z stacks with planes 0.3 μm apart were acquired at 30 s intervals on an Axiovert 200M microscope (Carl Zeiss) using an UPlanSApo 100× NA 1.40 oil immersion objective lens (Olympus) and a Coolsnap HQ CCD camera (Photometrics). Images were projected to two-dimensional images (SoftWoRx software) and further analyzed by MetaMorph (Molecular Devices). The linescans showing the fluorescence intensity for Kar3-3xGFP on spindles were plotted using ImageJ.
Single-step affinity purification of native motors and mass spectrometry analysis
Desired strains were grown to OD600 = 1.2 in YPD, centrifuged, drop frozen in liquid nitrogen, and lysed by freezer mill treatment. 5 g of yeast powder was dissolved in 10 ml buffer A (25 mM Hepes pH 8.0, 2 mM MgCl2, 0.5 mM EGTA pH 8.0, 0.1 mM EDTA, 0.1% NP-40, 15% glycerole, 150 mM KCl, 0.01 mM ATP, 0.1 mM PMSF, 1× protease inhibitor cocktail set IV [Calbiochem]). The lysed cells were centrifuged twice, first at 43.146×g for 20 min and afterwards at 125.749×g for 1 hr. The resulting supernatant was incubated for 2–3 hr with 100 μl Dynabeads (Life Technologies) that were coupled to 50 μl anti-FLAG M2 antibody (Sigma Aldrich). Beads were washed three times with buffer A and four times with buffer B (25 mM Hepes pH 8.0, 150 mM KCl). Elution was achieved by using 1 beads volume 2 mg/ml 3xFLAG peptide in buffer B. The elution fractions were analyzed by SDS-PAGE and silver stain. For mass spectrometry analysis, an on-bead digest replaced the elution procedure: 500 ng LysC protease was added per 50 μl dynabeads in ammonium bicarbonate buffer and incubated at 37°C overnight. The supernatant was filtered and applied to mass spectrometry analysis, which was performed on three independent preparations.
Acknowledgements
The authors thank all members of the Westermann lab for discussions and Jan-Michael Peters and David Keays for critical reading of the manuscript. We thank Karin Aumayr, Gabriele Stengl, and Pawel Paserbiek for help with microscopy and image analysis and Martin Colombini for fabrication of custom mechanical components. We thank Marlene Brandstaetter and Alexander Schleiffer for excellent technical support. This work received funding from the European Research Council under the European Community's Seventh Framework Programme (SW FP7/2007-2013)/ERC grant agreement number [203499], from the Austrian Science Fund FWF (SW, SFB F34-B03) and the Austrian Research Promotion Agency (FFG). MIM acknowledges the VIPS Program of the Austrian Federal Ministry of Science and Research and the City of Vienna. The research leading to these results has received funding from the Vienna Science and Technology Fund (WWTF) project VRG10-11, Human Frontiers Science Program Project RGP0041/2012, Research Platform Quantum Phenomena and Nanoscale Biological Systems (QuNaBioS). The IMP is funded by Boehringer Ingelheim.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
CM, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
MIM, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
SW, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
KD, Helped with single molecule in vitro imaging and in vivo experiements
BV, Helped with single molecule in vitro imaging and in vivo experiements
GL, Helped with single molecule in vitro imaging and in vivo experiements
GS, Helped with automated image analysis
AV, Conception and design, Analysis and interpretation of data, Drafting or revising the article
Additional files10.7554/eLife.04489.023
This file contains a detailed description of the mathematical model for cooperative Cik1–Kar3 kinesin movement.
DOI:http://dx.doi.org/10.7554/eLife.04489.023
10.7554/eLife.04489.024
This file contains tables listing the yeast strains and plasmids used in this study.
DOI:http://dx.doi.org/10.7554/eLife.04489.024
ReferencesAllinghamJSSproulLRRaymentIGilbertSP2007Vik1 modulates microtubule-Kar3 interactions through a motor domain that lacks an active siteCell1281161117210.1016/j.cell.2006.12.046AsburyCLFehrANBlockSM2003Kinesin moves by an asymmetric hand-over-hand mechanismScience3022130213410.1126/science.1092985AshfordAJHymanAA2006Preparation of tubulin from porcine brainCell Biology215516010.1016/B978-012164730-8/50094-0BasilicoFMaffiniSWeirJRPrumbaumDRojasAMZimniakTDe AntoniAJeganathanSVossBvan GerwenSKrennVMassimilianoLValenciaAVetterIRHerzogFRaunserSPasqualatoSMusacchioA2014The pseudo GTPase CENP-M drives human kinetochore assemblyeLife3e0297810.7554/eLife.02978BielingPTelleyIAHentrichCPiehlerJSurreyT2010Fluorescence microscopy assays on chemically functionalized surfaces for quantitative imaging of microtubule, motor, and +TIP DynamicsMethods in Cell Biology9555558010.1016/S0091-679X(10)95028-0BoudeauJMiranda-SaavedraDBartonGJAlessiDR2006Emerging roles of pseudokinasesTrends in Cell Biology1644345210.1016/j.tcb.2006.07.003BraunMDrummondDRCrossRAMcAinshAD2009The kinesin-14 Klp2 organizes microtubules into parallel bundles by an ATP-dependent sorting mechanismNature Cell Biology1172473010.1038/ncb1878BrunnbauerMMueller-PlanitzFKösemSHoTHDombiRGebhardtJCRiefMÖktenZ2010Regulation of a heterodimeric kinesin-2 through an unprocessive motor domain that is turned pocessive by its partnerProceedings of the National Academy of Sciences of USA107104601046510.1073/pnas.1005177107CaseRBPierceDWHom-BooherNHartCLValeRD1997The directional preference of kinesin motors is specified by an element outside of the motor catalytic domainCell90959966ChenCJRaymentIGilbertSP2011Kinesin Kar3Cik1 ATPase pathway for microtubule cross-linkingJournal of Biological Chemistry286292612927210.1074/jbc.M111.255554ChuHMYunMAndersonDESageHParkHWEndowS2005Kar3 interaction with Cik1 alters motor structure and functionThe EMBO Journal243214322310.1038/sj.emboj.7600790ClearyFBDeWittMABilyardTHtetZMBelyyVChanDDChangAYYildizA2014Tension on the linker gates the ATP-dependent release of dynein from microtubulesNature Communications5458710.1038/ncomms5587DeWittMAChangAYCombsPAYildizA2012Cytoplasmic dynein moves through uncoordinated stepping of the AAA+ ring domainsScience33522122510.1126/science.1215804DuanDJiaZJoshiMBruntonJChanMDrewDDavisDAllinghamJS2012Neck rotation and neck mimic docking in the noncatalytic Kar3-associated protein Vik1Journal of Biological Chemistry287402924030110.1074/jbc.M112.416529EndresNFYoshiokaCMilliganRAValeRD2006A lever-arm rotation drives motility of the minus-end-directed kinesin NcdNature43987587810.1038/nature04320FinkGHajdoLSkowronekKJReutherCKasprzakAADiezS2009The mitotic kinesin-14 Ncd drives directional microtubule-microtubule slidingNature Cell Biology1171772310.1038/ncb1877FosterKAGilbertSP2000Kinetic studies of dimeric Ncd: evidence that Ncd is not processiveBiochemistry3917841791FurutaKFurutaAToyoshimaYYAminoMOiwaKKojimaH2013Measuring collective transport by defined numbers of processive and nonprocessive kinesin motorsProceedings of the National Academy of Sciences of USA11050150610.1073/pnas.1201390110FurutaKToyoshimaYY2008Minus-end-directed motor Ncd exhibits processive movement that is enhanced by microtubule bundling in vitroCurrent Biology1815215710.1016/j.cub.2007.12.056GardnerMKHaaseJMythreyeKMolkJNAndersonMJoglekarAPO'TooleETWineyMSalmonEDOddeDJBloomK2008The microtubule-based motor Kar3 and plus end binding protein Bim1 provide structural support for the anaphase spindleThe Journal of Cell Biology1809110010.1083/jcb.146.2.415GibeauxRPolitiAZNédélecFAntonyCKnopM2013Spindle pole body-anchored Kar3 drives the nucleus along microtubules from another nucleus in preparation for nuclear fusion during yeast karyogamyGenes & Development2733534910.1101/gad.206318.112GonzalezMACopeJRankKCChenCJTittmannPRaymentIGilbertSPHoengerA2013Common mechanistic themes for the powerstroke of kinesin-14 motorsJournal of Structural Biology18433534410.1016/j.jsb.2013.09.020GudimchukNVitreBKimYKiyatkinAClevelandDWAtaullakhanovFIGrishchukEL2013Kinetochore kinesin CENP-e is a processive bi-directional tracker of dynamic microtubule tipsNature Cell Biology151079108810.1038/ncb2831HancockWOHowardJ1998Processivity of the motor protein kinesin requires two headsThe Journal of Cell Biology1401395140510.1083/jcb.140.6.1395JainALiuRXiangYKHaT2012Single-molecule pull-down for studying protein interactionsNature Protocols744545210.1038/nprot.2011.452KasedaKHiguchiHHiroseK2003Alternate fast and slow stepping of a heterodimeric kinesin moleculeNature Cell Biology51079108210.1038/ncb1067KortenTNitzscheBGellCRuhnowFLeducCDiezS2011Fluorescence imaging of single kinesin motors on immobilized microtubulesMethods in Molecular Biology78312113710.1007/978-1-61779-282-3_7LampertFHornungPWestermannS2010The Dam1 complex confers microtubule plus end-tracking activity to the Ndc80 kinetochore complexThe Journal of Cell Biology18964164910.1083/jcb.200912021LampertFMieckCAlushinGMNogalesEWestermannS2013Molecular requirements for the formation of a kinetochore-microtubule interface by Dam1 and Ndc80 complexesThe Journal of Cell Biology200213010.1083/jcb.201210091MaddoxPSStempleJKSatterwhiteLSalmonEDBloomK2003The minus end-directed motor Kar3 is required for coupling dynamic microtubule plus ends to the cortical shmoo tip in budding yeastCurrent Biology131423142810.1016/S0960-9822(03)00547-5ManningBDBarrettJGWallaceJAGranokHSnyderM1999Differential regulation of the Kar3p kinesin-related protein by two associated proteins, Cik1p and Vik1pThe Journal of Cell Biology1441219123310.1083/jcb.144.6.1219MiddletonKCarbonJ1994KAR3-encoded kinesin is a minus-end-directed motor that functions with centromere binding proteins (CBF3) on an in vitro yeast kinetochoreProceedings of the National Academy of Sciences of USA917212721610.1073/pnas.91.15.7212NitzscheBBormuthVBräuerCHowardJIonovLKerssemakersJKortenTLeducCRuhnowFDiezS2010Studying kinesin motors by optical 3D-nanometry in gliding motility assaysMethods in Cell Biology9524727110.1016/S0091-679X(10)95014-0OkadaYHirokawaN1999A processive single-headed motor: kinesin superfamily protein KIF1AScience2831152115710.1126/science.283.5405.1152QiuWDerrNDGoodmanBSVillaEWuDShihWReck-PetersonSL2012Dynein achieves processive motion using both stochastic and coordinated steppingNature Structural & Molecular Biology1919320010.1038/nsmb.2205RankKCChenCJCopeJPorcheKHoengerAGilbertSPRaymentI2012Kar3Vik1, a member of the kinesin-14 superfamily, shows a novel kinesin microtubule binding patternThe Journal of Cell Biology19795797010.1083/jcb.201201132Reck-PetersonSLYildizACarterAPGennerichAZhangNValeRD2006Single-molecule analysis of dynein processivity and stepping behaviorCell12633534810.1016/j.cell.2006.05.046SproulLRAndersonDJMackeyATSaundersWSGilbertSP2005Cik1 targets the minus-end kinesin depolymerase kar3 to microtubule plus endsCurrent Biology151420142710.1016/j.cub.2005.06.066SuXArellano-SantoyoHPortranDGaillardJVantardMTheryMPellmanD2013Microtubule-sliding activity of a kinesin-8 promotes spindle assembly and spindle-length controlNature Cell Biology1594895710.1038/ncb2801SubramanianRGellesJ2007Two distinct modes of processive kinesin movement in mixtures of ATP and AMP-PNPThe Journal of General Physiology13044545510.1085/jgp.200709866TanakaKKitamuraEKitamuraYTanakaTU2007Molecular mechanisms of microtubule-dependent kinetochore transport toward spindle polesThe Journal of Cell Biology17826928110.1083/jcb.200702141TanakaKMukaeNDewarHvan BreugelMJamesEKPrescottARAntonyCTanakaTU2005Molecular mechanisms of kinetochore capture by spindle microtubulesNature43498799410.1038/nature03483ThoresenTGellesJ2008Processive movement by a kinesin heterodimer with an inactivating mutation in one headBiochemistry479514952110.1021/bi800747eTomishigeMKlopfensteinDRValeRD2002Conversion of Unc104/KIF1A kinesin into a processive motor after dimerizationScience2972263226710.1126/science.1073386TonksNK2009Pseudophosphatases: grab and hold onCell13946446510.1016/j.cell.2009.10.008YildizATomishigeMGennerichAValeRD2008Intramolecular strain coordinates kinesin stepping behavior along microtubulesCell1341030104110.1016/j.cell.2008.07.018YildizATomishigeMValeRDSelvinPR2004Kinesin walks hand-over-handScience30367667810.1126/science.1093753ZimniakTStenglKMechtlerKWestermannS2009Phosphoregulation of the budding yeast EB1 homologue Bim1p by Aurora/Ipl1pThe Journal of Cell Biology18637939110.1091/mbc.E07-06-053610.7554/eLife.04489.025Decision letterHarrisonStephen CReviewing editorHoward Hughes Medical Institute, Harvard Medical School, United States
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Non-catalytic motor domains enable processive movement and functional diversification of the kinesin Kar3” for consideration at eLife. Your article has been favorably evaluated by Randy Schekman (Senior editor), a Reviewing editor, and two reviewers.
The Reviewing editor and the reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
The reviewers agreed that in principle this paper would be suitable for publication in eLife, but to make it so, the authors will need to carry out additional experiments and to reduce the strength of their claims, which both reviewers agreed were exaggerated (one reviewer even used the word “reckless”). The key result that the reviewers agreed makes the paper suitable for eLife is that the heterodimer walks processively toward the minus end of a MT.
1) The main conclusion of this manuscript is that Kar3 needs a second non-catalytic microtubule binding domain for processive translocation. If monomeric kinesin Kif1A can move along microtubules in vitro using a biased diffusion mechanism (Okada and Hirokawa, 1999, but see reservations below), Kar3 might do the same, obviating (or at least reducing) the requirement for a second motor head. A key question is whether the Cik1(1-360)-Kar3 mutant (or the Cik1(1-521)-Kar3 mutant) moves directionally in the presence of ATP. Now it is clear from Figures 2B and 2C that the velocity is substantially lower, but the authors need to analyze these data in more detail. If Cik1(1-360)-Kar3 and Cik1(1-521)-Kar3 do have significant directionality, then the authors might need to modify their description of the “foot-hold” mechanism.
2) Examples of exaggerated claims or incorrect information, all of which need to be fixed in a revision:
Abstract:
(i) “Whether this (hand-over-hand) model is universal, or if alternative mechanisms can support highly processive motion along microtubules, is an open question.” This statement is wrong. It is no longer an open question. Dewitt et al., Science, 2012 and Qiu et al., NSMB, 2012 showed that dynein walks in an uncoordinated manner.
(ii) “The noncatalytic domain acts as a foothold that allows Kar3 to bias translocation towards the minus end via a diffusive ADP state.” The reported experiments do not prove this statement. That kinesins have lower MT affinity in the ADP state was already known. The authors did not demonstrate how this motor steps along MT. In order to make these claims, they must demonstrate that: a) Cik1/Vik1 biases translocation of Kar3 to the minus end, b) translocation of Kar3 occurs in the ADP bound state and c) diffusion of the Kar3 head to the next tubulin binding site is a critical step. The authors need to delete this statement and replace it with one that sticks to the facts.
(iii) “This mechanism… allows transport of the Ndc80 kinetochore complex in vitro and is critical for Kar 3 function in vivo.” The authors show that Kar3 can bind and transport the isolated Ndc80 complex, not the Ndc80 complex as part of an assembled kinetochore. It has yet to be shown whether this in vitro transport reflects a property critical for in vivo function of Kar3. This latter statement is what they propose, but do not show, in this manuscript. The authors need to delete this statement and replace it with one that sticks to the facts.
Introduction section:
“In general motors equipped with only one catalytically active head fail to achieve long-range processive motility”. This is not true. The authors need to rewrite this paragraph citing: (i) the Okada et al. 2003 Kif1A paper showing processive single-headed motility; (ii) the Kaseda, Higuchi and Hirose (NSMB, 2003) and Thoresen et al. (Biochemistry, 2008) papers showed that processive motility of kinesin heterodimers with mutant motor domains that have lower catalytic activity; (iii) the Clearly et al. 2014 paper on cytoplasmic dynein showing that processive motility can occur when one of the heads in the dimer is replaced by a non-motor domain; and (iv) reports of processive motility with axonemal dynein. In other words, the authors need to remove overblown claims, discuss earlier results in a balanced way, and clearly and accurately indicate how their results advance the field.
The manuscript lacks evidence that their preps do not include aggregates. The authors could rule out aggregation by running gel filtration experiments, which are standard in the field.
The authors do cite evidence in Figure 1–figure supplement 1, for mostly single step photobleaching of labeled motors, arguing that processive motility is a property of individual motors. While this may be true, the single step photobleaching is not definitive. First, the data clearly show that some motors show 2 or >2 steps of photobleaching, suggesting protein aggregation or transient interactions. Second, the labeling efficiency is not calculated. If it is low (e.g. 40%), the fact that 10-20% of motors have 2 or more steps of photobleaching would suggest most proteins are aggregates. Third, if the motors are aggregated (e.g. 10 motors in an aggregate) then almost all the aggregates will have TMR even if the coupling efficiency is low. Aggregation may be an even bigger problem with the delta K-loop construct, which is expected to be less soluble. So gel filtration on this construct is also necessary.
In Figure 1G, how do the authors know that the MTs are antiparallel and overlap? Was the polarity marked?
Third paragraph of the Results section:
“We noticed that due to their relatively large velocity distribution…” This whole paragraph does not make sense. How do they know that the motors run into each other (e.g. literally colliding with each other)? What is the evidence that motors travel together in teams? Are the proteins in contacts with each other (which cannot be seen by a fluorescence microscope due to diffraction limited resolution of the assay). In this section, they can also cite the Cin8 study by the Surrey lab (Roostalu et al., Science, 2011).
Eighth paragraph of the Results section:
The interpretation of the K-tail mutant experiments is counterintuitive. One would expect that removing positively charged residues from the noncatalytic head would reduce its affinity for MTs (which is not shown biochemically in this manuscript and needs to be shown in a revised version). Since the authors wish to refer to the function of this head as “a foothold”, these mutations would be expected to alter motility by reducing processivity (because it is no longer a very strong MT tether), with increased or unchanged velocity (because it might be easier for Kar3 to pull off the noncatalytic head from MT by force). The description of these results and their interpretation need reconsideration and revision.
The authors do not present direct evidence that Kar3 transports Ndc80 in vivo. Therefore, they need to change “We conclude that the Ndc80 complex is a direct cargo for Kar3 transport” to “We conclude that the Ndc80 complex can be transported by the Kar3/Cik1 motor.”
Discussion:
(i) In the first paragraph, the authors claim to have discovered that Kar3/Cik or Kar3/Vik1 heterodimers have a new mode of processive motility, and they assert that this is the first example of a motor that violates hand over hand motility. These statements are wrong (see comments above). Also note that without high resolution stepping studies their conclusions are too speculative.
(ii) The authors claim that “Kar3's motile characteristics share similarities to Kif1A”. How so? Kif1A was claimed to be processive on its own and not to require a foothold. The effect of the positive charges is different. Note that processive motility of single-headed Kif1 has not been replicated in another laboratory. The authors should analyze the motility of a Kar3 monomer and compare their results with those published by Hirokawa on Kif1A.
Figure 7:
The model shown does not relate to the main story, and there is no evidence provided for it. It should be removed from the manuscript. The entire discussion of diffusive states, anisotropy and powerstroke are far too speculative. It would be more appropriate for some future review. The problem is that without high-resolution stepping data many of the conclusions are in fact speculations. The authors have to be very careful to distinguish fact from fiction.
In summary, to be suitable for eLife, a revised manuscript will need to include at least the following:
1) More convincing and complete demonstration, by careful gel filtration experiments, that the motor proteins are not aggregated.
2) More thorough analysis of the data in Figure 2B, C: is there directionality in the motion detected for the mutants?
3) Comparison of the behavior of a Kar3 monomer with the reported behavior of Kif1A.
The revised manuscript must also avoid the exaggerated and overstated claims that mar the current version.
10.7554/eLife.04489.026Author response
In summary, to be suitable for eLife, a revised manuscript will need to include at least the following:
1) More convincing and complete demonstration, by careful gel filtration experiments, that the motor proteins are not aggregated.
2) More thorough analysis of the data inFigure 2B, C: is there directionality in the motion detected for the mutants?
3) Comparison of the behavior of a Kar3 monomer with the reported behavior of Kif1A.
The revised manuscript must also avoid the exaggerated and overstated claims that mar the current version.
1) In the revised manuscript we have included a biochemical characterization of Cik1-Kar3 motors by gel filtration and velocity gradient centrifugation analysis (new Figures 1C, D, E). To go further and more directly characterize the motors we have visualized the motor proteins used in the experiments by low angle Platinum/Carbon rotary shadowing EM (new Figures 1F, G). The hydrodynamic analysis shows that the motor proteins are biochemically well behaved, as they elute as a single species with no indication of aggregation. The EM data (the first visualization of full-length Cik1-Kar3 motors) reveals the overall organization of the motors and supports the conclusion that individual Cik1-Kar3 heterodimers are being analyzed.
2) As suggested we have more carefully analyzed the movement of wild-type and mutant motors in ATP in the new Figure 4E. A mean-squared displacement analysis indicates that the movement of the Cik1(1-360)-Kar3 mutant is fitted well with n=1 to the one-dimensional diffusion equation <x2>= a·tn, indicating random diffusion with no directional component. In contrast, data for mean squared displacement of the wild-type motor is fitted with n=2, indicating directed movement.
3) We have constructed a monomeric, Halo-tagged Kar3 construct and imaged it in the TIRF assay. In the presence of ATP and under low nanomolar motor concentrations, only very short-lived binding events can be detected on microtubules (new Figure 4I). Thus, there is no evidence that monomeric Kar3 can move processively.
We believe that taken together these three points have greatly strengthened the manuscript. The biochemical and EM characterization shows that well-behaved, non-aggregated motors were purified. The additional experiments show that indeed the non-catalytic head is strictly required for processive movement, as no directional movement of the heterodimer can be detected after its removal and in the reciprocal experiment; the Kar3 monomeric head alone does not display processive movement.
The reviewers agreed that in principle this paper would be suitable for publication in eLife, but to make it so, the authors will need to carry out additional experiments and to reduce the strength of their claims, which both reviewers agreed were exaggerated (one reviewer even used the word “reckless”). The key result that the reviewers agreed makes the paper suitable for eLife is that the heterodimer walks processively toward the minus end of a MT.
1) The main conclusion of this manuscript is that Kar3 needs a second non-catalytic microtubule binding domain for processive translocation. If monomeric kinesin Kif1A can move along microtubules in vitro using a biased diffusion mechanism (Okada and Hirokawa, 1999, but see reservations below), Kar3 might do the same, obviating (or at least reducing) the requirement for a second motor head. A key question is whether the Cik1(1-360)-Kar3 mutant (or the Cik1(1-521)-Kar3 mutant) moves directionally in the presence of ATP. Now it is clear fromFigures 2B and 2Cthat the velocity is substantially lower, but the authors need to analyze these data in more detail. If Cik1(1-360)-Kar3 and Cik1(1-521)-Kar3 do have significant directionality, then the authors might need to modify their description of the “foot-hold” mechanism.
Please see our response to this point above. In addition to the analysis for the Cik1 (1-360)-Kar3 mutant presented in the new Figure 4E, we have conducted the same analysis for the Cik1 (1-521)-Kar3 mutant (Author response image 1).
n=1 for the fit indicates that the movement of this mutant also does not contain a directional component.
2) Examples of exaggerated claims or incorrect information, all of which need to be fixed in a revision:
Abstract:
(i) “Whether this (hand-over-hand) model is universal, or if alternative mechanisms can support highly processive motion along microtubules, is an open question.”
This statement is wrong. It is no longer an open question. Dewitt et al., Science, 2012 and Qiu et al., NSMB, 2012 showed that dynein walks in an uncoordinated manner.
As suggested we have rewritten the Abstract. We would like to point out, however, that we explicitly referred to kinesin motors in the original Abstract. Indeed, Dynein movement does not require head-to-head coordination and we have cited the above mentioned papers in our Discussion.
(ii) “The noncatalytic domain acts as a foothold that allows Kar3 to bias translocation towards the minus end via a diffusive ADP state.”
The reported experiments do not prove this statement. That kinesins have lower MT affinity in the ADP state was already known. The authors did not demonstrate how this motor steps along MT. In order to make these claims, they must demonstrate that: a) Cik1/Vik1 biases translocation of Kar3 to the minus end, b) translocation of Kar3 occurs in the ADP bound state and c) diffusion of the Kar3 head to the next tubulin binding site is a critical step. The authors need to delete this statement and replace it with one that sticks to the facts.
We have deleted and replaced the corresponding statements in the revised Abstract.
(iii) “This mechanism… allows transport of the Ndc80 kinetochore complex in vitro and is critical for Kar 3 function in vivo.” The authors show that Kar3 can bind and transport the isolated Ndc80 complex, not the Ndc80 complex as part of an assembled kinetochore. It has yet to be shown whether this in vitro transport reflects a property critical for in vivo function of Kar3. This latter statement is what they propose, but do not show, in this manuscript. The authors need to delete this statement and replace it with one that sticks to the facts.
We have replaced the statement with “… can support transport of the Ndc80 complex in vitro…”.
Introduction section:
“In general motors equipped with only one catalytically active head fail to achieve long-range processive motility”. This is not true. The authors need to rewrite this paragraph citing: (i) the Okada et al. 2003 Kif1A paper showing processive single-headed motility; (ii) the Kaseda, Higuchi and Hirose (NSMB, 2003) and Thoresen et al. (Biochemistry, 2008) papers showed that processive motility of kinesin heterodimers with mutant motor domains that have lower catalytic activity; (iii) theClearly et al. 2014paper on cytoplasmic dynein showing that processive motility can occur when one of the heads in the dimer is replaced by a non-motor domain; and (iv) reports of processive motility with axonemal dynein. In other words, the authors need to remove overblown claims, discuss earlier results in a balanced way, and clearly and accurately indicate how their results advance the field.
We have re-written the Introduction as suggested. In general we have put less emphasis on a comparison of a conventional hand-over-hand mechanism with the observed motility of Kar3 heterodimers and instead have focused on a comparison of Kar3 to other Kinesin-14 family members such as Ncd.
The manuscript lacks evidence that their preps do not include aggregates. The authors could rule out aggregation by running gel filtration experiments, which are standard in the field.
We have included these experiments in the revised manuscript (new Figure 1C–G), please see our response to the three general points above.
The authors do cite evidence in Figure 1–figure supplement 1, for mostly single step photobleaching of labeled motors, arguing that processive motility is a property of individual motors. While this may be true, the single step photobleaching is not definitive. First, the data clearly show that some motors show 2 or >2 steps of photobleaching, suggesting protein aggregation or transient interactions. Second, the labeling efficiency is not calculated. If it is low (e.g. 40%), the fact that 10-20% of motors have 2 or more steps of photobleaching would suggest most proteins are aggregates. Third, if the motors are aggregated (e.g. 10 motors in an aggregate) then almost all the aggregates will have TMR even if the coupling efficiency is low. Aggregation may be an even bigger problem with the delta K-loop construct, which is expected to be less soluble. So gel filtration on this construct is also necessary.
We have estimated the TMR labeling efficiency using a construct that in addition to the Halo tag contains an eGFP fusion. By comparing eGFP to TMR fluorescence at low motor concentrations we estimate the labeling efficiency to exceed 90% (see fluorescence micrographs, Author response image 2). Together with the photobleaching steps, the biochemical characterization of the oligomeric state, and the two-color motor mixing experiment, there is strong evidence that heterodimers are imaged. To the last point: we have removed the experiments with the delta K-tail construct in the revised manuscript.
Fluorescence Micrographs.
InFigure 1G, how do the authors know that the MTs are antiparallel and overlap? Was the polarity marked?
Bundles of microtubules are identified by increased fluorescence in the tubulin channel. In this experiment the polarity of the microtubules was not marked but inferred from the behavior of the motors.
Third paragraph of the Results section:
“We noticed that due to their relatively large velocity distribution…” This whole paragraph does not make sense. How do they know that the motors run into each other (e.g. literally colliding with each other)? What is the evidence that motors travel together in teams? Are the proteins in contacts with each other (which cannot be seen by a fluorescence microscope due to diffraction limited resolution of the assay). In this section, they can also cite the Cin8 study by the Surrey lab (Roostalu et al., Science, 2011).
We have re-worded this section to state that we are simply comparing properties of motors with different brightness, i.e. heterodimers compared to teams that consist of two or more heterodimers
Eighth paragraph of the Results section:
The interpretation of the K-tail mutant experiments is counterintuitive. One would expect that removing positively charged residues from the noncatalytic head would reduce its affinity for MTs (which is not shown biochemically in this manuscript and needs to be shown in a revised version). Since the authors wish to refer to the function of this head as “a foothold”, these mutations would be expected to alter motility by reducing processivity (because it is no longer a very strong MT tether), with increased or unchanged velocity (because it might be easier for Kar3 to pull off the noncatalytic head from MT by force). The description of these results and their interpretation need reconsideration and revision.
While our experiments establish that the non-catalytic head in combination with Kar3 determines the velocity of the motor complex, we indeed at this point cannot give a mechanistic explanation for the role of the “K-tail”. In additional experiments we determined that simply adding the “K-tail” to the carboxyterminus of Cik1-Kar3 is not sufficient to alter the velocity of this motor. Thus, the K-tail’s role in the mechanochemical cycle might be more complex than simply altering the affinity for microtubules. This is in line with recent work from Yoshi et al., JBC, 2013. We have decided to remove the figure panels describing the K-tail experiments from the revised manuscript, as these points require further experimentation and do not contribute to the main message of the paper.
The authors do not present direct evidence that Kar3 transports Ndc80 in vivo. Therefore, they need to change “We conclude that the Ndc80 complex is a direct cargo for Kar3 transport” to “We conclude that the Ndc80 complex can be transported by the Kar3/Cik1 motor.”
We agree and have changed the wording accordingly.
Discussion:
(i) In the first paragraph, the authors claim to have discovered that Kar3/Cik or Kar3/Vik1 heterodimers have a new mode of processive motility, and they assert that this is the first example of a motor that violates hand over hand motility. These statements are wrong (see comments above). Also note that without high resolution stepping studies their conclusions are too speculative.
(ii) The authors claim that “Kar3's motile characteristics share similarities to Kif1A”. How so? Kif1A was claimed to be processive on its own and not to require a foothold. The effect of the positive charges is different. Note that processive motility of single-headed Kif1 has not been replicated in another laboratory. The authors should analyze the motility of a Kar3 monomer and compare their results with those published by Hirokawa on Kif1A.
As suggested, we have tested the motility of a monomeric Kar3 head construct (residues 353-729 fused to an N-terminal Halo tag). We have found no evidence for processive movement under conditions that allow robust motility of the heterodimer (new Figure 4I). We agree that the suggested similarities to Kif1A have been confusing in our Discussion. We have rewritten the Discussion to conclude that Kar3’s mechanism of movement must be different from Kif1A as wehave not detected motility of a monomeric Kar3 construct and no directional movement was observed after removing the non-catalytic Cik1 head.
Figure 7:
The model shown does not relate to the main story, and there is no evidence provided for it. It should be removed from the manuscript. The entire discussion of diffusive states, anisotropy and powerstroke are far too speculative. It would be more appropriate for some future review. The problem is that without high-resolution stepping data many of the conclusions are in fact speculations. The authors have to be very careful to distinguish fact from fiction.
As suggested, we have re-written the discussion section to remove the more speculative statements and focus on the experimentally supported insights. We have removed the original Figure 7B, as we can indeed not give conclusive insights about the stepping mechanism. We would like to maintain a simplified model (new Figure 8B) that illustrates the following experimentally addressed points: 1) the Kar3 motor can take multiple steps along the microtubule before dissociating; 2) the non-catalytic head is required for the movement by promoting tight binding in no-nucleotide and AMP-PNP state 3) the movement likely has a diffusional component, as suggested by the diffusive behavior of the motor in ADP and the pronounced salt-sensitivity of the velocity.
We agree that further biophysical experiments are required to establish a mechanism and we have greatly tightened the discussion to avoid too much speculation about the mechanism.
diff --git a/elife04525.xml b/elife04525.xml
new file mode 100644
index 0000000..b5971f6
--- /dev/null
+++ b/elife04525.xml
@@ -0,0 +1 @@
+elifeeLifeeLifeeLife2050-084XeLife Sciences Publications, Ltd0452510.7554/eLife.04525Research articleHuman biology and medicineMicrobiology and infectious diseaseMucosal effects of tenofovir 1% gelHladikFlorianhttp://orcid.org/0000-0002-0375-2764123*BurgenerAdam45†BallweberLamar3†GottardoRaphael367VojtechLucia1FouratiSlim8DaiJames Y67CameronMark J8StroblJohanna3HughesSean M1HoesleyCraig9AndrewPhilip10JohnsonSherri10PiperJeanna11FriendDavid R12BallT Blake45CranstonRoss D1314MayerKenneth H15McElrathM Juliana2316McGowanIan1314*Department of Obstetrics and Gynecology, University of Washington, Seattle, United StatesDepartment of Medicine, University of Washington, Seattle, United StatesVaccine and Infectious Diseases Division, Fred Hutchinson Cancer Research Center, Seattle, United StatesDepartment of Medical Microbiology, University of Manitoba, Winnipeg, CanadaNational HIV and Retrovirology Laboratories, Public Health Agency of Canada, Winnipeg, CanadaDepartment of Biostatistics, University of Washington, Seattle, United StatesPublic Health Sciences Division, Fred Hutchinson Cancer Research Center, Seattle, United StatesVaccine and Gene Therapy Institute of Florida, Port Saint Lucie, United StatesDepartment of Medicine, University of Alabama, Birmingham, United StatesFHI 360, Durham, United StatesDivision of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, United StatesCONRAD, Eastern Virginia Medical School, Arlington, United StatesUniversity of Pittsburgh School of Medicine, Pittsburgh, United StatesMicrobicide Trials Network, Magee-Women's Research Institute, Pittsburgh, United StatesFenway Health, Beth Israel Deaconess Hospital, Harvard Medical School, Boston, United StatesDepartment of Global Health, University of Washington, Seattle, United StatesSgaierSemaReviewing editorBill & Melinda Gates Foundation, IndiaFor correspondence: fhladik@fhcrc.org (FH);mcgowanim@mwri.magee.edu (IMG)
These authors contributed equally to this work
0302201520154e045252708201402022015This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.10.7554/eLife.04525.001
Tenofovir gel is being evaluated for vaginal and rectal pre-exposure prophylaxis against HIV transmission. Because this is a new prevention strategy, we broadly assessed its effects on the mucosa. In MTN-007, a phase-1, randomized, double-blinded rectal microbicide trial, we used systems genomics/proteomics to determine the effect of tenofovir 1% gel, nonoxynol-9 2% gel, placebo gel or no treatment on rectal biopsies (15 subjects/arm). We also treated primary vaginal epithelial cells from four healthy women with tenofovir in vitro. After seven days of administration, tenofovir 1% gel had broad-ranging effects on the rectal mucosa, which were more pronounced than, but different from, those of the detergent nonoxynol-9. Tenofovir suppressed anti-inflammatory mediators, increased T cell densities, caused mitochondrial dysfunction, altered regulatory pathways of cell differentiation and survival, and stimulated epithelial cell proliferation. The breadth of mucosal changes induced by tenofovir indicates that its safety over longer-term topical use should be carefully monitored.
DOI:http://dx.doi.org/10.7554/eLife.04525.001
10.7554/eLife.04525.002eLife digest
Tenofovir is a drug that can stop some viruses—including HIV—from multiplying. It is commonly used in multidrug therapies to control HIV infection. Clinical trials are underway to find out whether using the drug in the form of a gel applied to the vagina or rectum could be an effective way to prevent HIV transmission during sex.
Some of the clinical trials carried out so far have produced promising results. However, since the use of gels containing anti-viral drugs is a new strategy for HIV prevention, there are limited data available about the safety of these products. Previous studies have shown that the concentration of tenofovir in the vagina is much higher in individuals using the gel than in those taking the tablet form of the drug. These high concentrations could lead to unexpected effects on the health of the cells exposed to the gel.
Here, Hladik, Burgener, Ballweber et al. used a systems biology approach to look at the broad effects of tenofovir gel on tissue from the rectum. Tissue samples taken from the rectums of 15 patients who used tenofovir gel for seven days were compared with tissue samples taken from individuals who used a control gel that did not contain the drug or who did not use any gel.
Genes that regulate inflammation were suppressed in the rectal tissue from patients who used tenofovir, as were genes that help these tissues regenerate and produce energy. The tissue from these patients also contained more immune cells, suggesting that their local immune systems were more active. Additionally, Hladik, Burgener, Ballweber et al. observed changes that could potentially lead to the increased growth of cells.
Similar differences were also observed in vaginal cells that had been treated with tenofovir in the laboratory. These findings suggest that tenofovir delivered directly to the vagina or rectum may have unintentional local side effects. However, it is important to acknowledge that tenofovir gel has been evaluated in multiple studies that have not observed overt clinical adverse effects. Therefore, the implication of these findings is currently unclear and warrants further study.
DOI:http://dx.doi.org/10.7554/eLife.04525.002
Author keywordsHIV/AIDSpreventionmicrobicidesmucosaside effectsResearch organismhumanhttp://dx.doi.org/10.13039/100000002National Institutes of Health (NIH)U01AI068633McGowanIanhttp://dx.doi.org/10.13039/100000002National Institutes of Health (NIH)U19AI082637McGowanIanhttp://dx.doi.org/10.13039/100000002National Institutes of Health (NIH)R01HD51455HladikFlorianThe funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementMucosal application of the anti-retroviral drug tenofovir, which is intended to prevent HIV transmission, has many off-target effects on the mucosa itself.Introduction
The HIV prevention field has invested considerable resources in testing the phosphonated nucleoside reverse transcriptase inhibitor (NRTI) tenofovir as a mucosally applied topical microbicide to prevent sexual HIV transmission. In a phase 2B trial, CAPRISA 004, pericoital tenofovir 1% gel was 39% efficacious in preventing vaginal HIV acquisition (Abdool Karim et al., 2010). However, in another phase 2B trial, the VOICE study (MTN-003), the daily vaginal tenofovir 1% gel arm was discontinued for futility (Marrazzo et al., 2015). Adherence to product use was low in VOICE, likely explaining the differences in findings between the two studies. Currently, the CAPRISA 004 study is being repeated in a phase 3 trial (FACTS 001 study).
A reduced glycerin formulation of the vaginal tenofovir 1% gel for use as a rectal microbicide appears safe when evaluated by epithelial sloughing, fecal calprotectin, inflammatory cytokine mRNA/protein levels, and cellular immune activation markers (McGowan et al., 2013). However, because topical application of an antiretroviral drug to the mucosa is a novel prevention strategy without clinical precedent, we conducted a comprehensive systems biology assessment of tenofovir gel's effects on the mucosa.
Materials and methodsDesign of the clinical study
MTN-007 (ClinicalTrials.gov registration NCT01232803) was a phase 1, double blind, placebo-controlled trial in which participants were randomized to receive rectal reduced glycerin tenofovir 1%, nonoxynol-9 (N-9) 2% or hydroxyethylcellulose (HEC) gels, or no-treatment (1:1:1:1), at three clinical research sites (Pittsburgh, PA; Birmingham, AL; Boston, MA). The study protocol was approved by IRBs at all three sites. All participants gave written informed consent. The study details, and general safety and acceptability data, have been published elsewhere and showed that rectal tenofovir 1% gel was well tolerated and appeared safe by established safety parameters (McGowan et al., 2013). Each gel was administered as a single dose and then, after at least a 1-week recovery period, once daily for seven consecutive days. The first dose of study product was self-administered under supervision by the clinic staff at the Treatment 1 Visit. Subsequent administrations occurred at home, and study participants were instructed to insert one dose of gel into the rectum once daily throughout the 7-day period in the evening or before the longest period of rest.
Study participants and products
A total of 65 study participants were enrolled and randomized in the study, 62 of whom completed it (tenofovir, n = 15; N-9, n = 16; HEC, n = 15; and no treatment, n = 16). 43 (69%) were male. Microarray studies were performed on eight randomly selected male participants in each group, and confirmatory gene expression studies were done on the remaining participants. The study population consisted of healthy, HIV-uninfected adults aged 18 or older who were required to abstain from receptive anal intercourse during the course of the clinical trial. Female participants were required to use effective contraception. Individuals with abnormalities of the colorectal mucosa, significant gastrointestinal symptoms (such as a history of rectal bleeding or inflammatory bowel disease), evidence of anorectal Chlamydia trachomatis or Neisseria gonorrhea infection, hepatitis B infection, or who used anticoagulants were excluded from the study. Reduced glycerin tenofovir 1% gel and HEC gel, known as the ‘Universal Placebo Gel’ (Tien et al., 2005), were supplied by CONRAD (Arlington, VA, USA). 2% N-9 gel was provided as Gynol II (Johnson & Johnson). All study products were provided in identical opaque HTI polypropylene pre-filled applicators (HTI Plastics, Lincoln, NE) containing 4 ml of study product.
Mucosal biopsy procedures
Rectal biopsies for the microarray studies were obtained before treatment at enrollment (time point ‘0’), 30–60 min following application of the single gel dose (time point ‘I’; to test acute single-dose effects), and again on the day following the last dose of the seven once-daily gel applications (time point ‘VII’; to test multiple-dose effects). Following an enema with Normosol-R pH 7.4, a flexible sigmoidoscope was inserted into the rectum and biopsies were collected at 15 cm from the anal margin. Following the sigmoidoscopy, a disposable anoscope was inserted into the anal canal for collection of rectal biopsies at 9 cm from the anal margin. Immediately after harvest, biopsies were immersed in RNA later (Qiagen, Germany), stored at 4°C overnight, and transferred to a −80°C freezer for long-term storage until shipping to Seattle and processing.
Primary vaginal keratinocyte cultures
Tissues routinely discarded from vaginal repair surgeries were harvested from four otherwise healthy adult women, placed in ice-cooled calcium- and magnesium-free phosphate-buffered saline containing 100 U/ml penicillin, 100 μg/ml streptomycin, and 2.5 μg/ml Fungizone (Thermo Fisher Scientific, Waltham, MA), and transported to the laboratory within 1 hr of removal from the donor. Tissue harvesting and experimental procedures were approved by the Institutional Review Boards of the University of Washington and the Fred Hutchinson Cancer Research Center. The deep submucosa was removed with surgical scissors and the remaining vaginal mucosa was cut into 5 × 5 mm pieces, which were incubated at 4°C for 18 hr in 5 ml of a 25 U/ml dispase solution (354235; BD Biosciences, Franklin Lakes, NJ). The epithelial sheets were dissected off under a stereoscope and incubated for 10–12 min at 37°C in 2 ml 0.05% trypsin while gently shaking. The dispersed cells were poured through a 100-μm cell strainer into a 50-ml tube, pelleted by centrifugation, and resuspended in F medium (3:1 [vol/vol] F12 [Ham]-DMEM [Thermo Fisher Scientific], 5% fetal calf serum [Gemini Bio-Products, Calabasas, CA], 0.4 μg/ml hydrocortisone [H-4001; Sigma-Aldrich, St. Louis, MO], 5 μg/ml insulin [700-112P; Gemini Bio-Products], 8.4 ng/ml cholera toxin [227036; EMD Millipore, Billerica, MA], 10 ng/ml epidermal growth factor [PHG0311; Thermo Fisher Scientific], 24 μg/ml adenine [A-2786; Sigma], 100 U/ml penicillin, and 100 μg/ml streptomycin [Thermo Fisher Scientific]). The keratinocytes were plated into culture flasks in the presence of ∼12,500/cm2 irradiated (6000 Rad) 3T3-J2 feeder fibroblasts (a kind gift by Cary A Moody) and 10 μM of Rho kinase inhibitor Y27632 (1254; Enzo Life Sciences, Farmingdale, NY) was added (Rheinwald and Green, 1975; Chapman et al., 2010; Liu et al., 2012). Keratinocytes were fed every 2–3 days and passaged when around 80% confluent by 1 min treatment with 10 ml versene (Thermo Fisher Scientific) to remove the feeder cells, followed by 5 min treatment with trypsin/EDTA (Thermo Fisher Scientific). Dislodged keratinocytes were washed and re-plated at ∼2500 keratinocytes/cm2 with irradiated 3T3-J2 feeder fibroblasts.
Tenofovir treatment of primary vaginal keratinocytes
Tenofovir (CAS 147127-20-6; T018500, Toronto Research Chemicals, Canada) was dissolved in phosphate-buffered saline, 7% dimethyl sulfoxide, and 5% 5N sodium hydroxide to result in a 767 mM stock solution, and further diluted in culture media for addition to keratinocyte cultures in concentrations ranging from 0.05 to 32 mM. Based on initial titration experiments in which we measured the intracellular concentration of tenofovir diphosphate, the active cellular metabolite of tenofovir, by liquid chromatography-tandem mass spectrometry (performed in the laboratory of Dr Craig Hendrix, Johns Hopkins University), concentrations in the culture media at the lower end (0.05–0.5 mM range) were estimated to be equivalent to the active concentrations likely achieved by topical tenofovir 1% gel (35 mM) in mucosal epithelial cells in vivo (Hendrix et al., 2013; Louissaint et al., 2013). In all experiments, control keratinocytes were cultured in parallel without tenofovir. Keratinocytes were harvested at pre-determined time points, split into several aliquots, and pelleted. Pellets were frozen and stored at −80°C either as dry pellets for protein assays or after suspension in RNAprotect Cell Reagent (Qiagen) for RNA assays.
alamarBlue cell proliferation and cell viability assay
Primary vaginal keratinocytes were cultured in six-well plates (Corning) with or without various concentrations of tenofovir for up to 14 days. At day 1, day 4, day 7, or day 14, the alamarBlue reagent (BUF012A; AbD Serotec) was added. After 5 hr, absorbance was measured at 570 and 600 nm on a Varioskan Flash Multimode reader (Thermo Fisher Scientific). Percentage reduction of alamarBlue, corresponding to the level of cell proliferation, was calculated as specified in the manufacturer's technical datasheet. This was converted to percentage of maximal alamarBlue reduction by dividing each well by the well with the highest amount of alamarBlue reduction.
Interleukin-10 ELISA assay
The IL-10 concentration in primary vaginal keratinocytes was measured after lysis of frozen cell pellets in Cell Lysis buffer (R&D Systems, Minneapolis, MN) using a commercial human IL-10 immunoassay (Quantikine HS, HS100C; R&D Systems) according to the manufacturer's specifications.
RNA isolation, cRNA preparation, and whole genome BeadArray hybridization
RNA was isolated from the biopsies using the RNeasy Fibrous Tissue Mini Kit (Qiagen), and from the vaginal keratinocytes using the Direct-zol RNA MiniPrep kit (Zymo Research, Irvine, CA), according to the manufacturer's instructions, treated with 27 Kunitz units of DNAse (Qiagen) to remove genomic DNA contamination, and evaluated for integrity using the Agilent RNA 6000 Nano Kit (Agilent, Palo Alto, CA) on an Agilent 2100 Bioanalyzer. All samples had an RNA Integrity Number of 7 or greater. 500 ng of total RNA was amplified and labeled using the Illumina TotalPrep RNA Amplification kit (Thermo Fisher Scientific). cRNA from a total of 192 rectal biopsy samples (eight men per study arm, four study arms, three time points, and two biopsy sites [9 cm and 15 cm]), and from a total of 36 primary vaginal keratinocyte cultures (three tissue donors, three arms, and four time points), was hybridized to HumanHT12 v4 Expression BeadChips (Illumina, San Diego, CA) according to the manufacturer's protocols. Each chip contains 47,323 probes, corresponding to 30,557 genes.
Unbiased mass spectrometry of rectal sponge eluates
Sponges were obtained on the same visits as the biopsies, prior to the biopsy procedure, and processed as previously described (Anton et al., 2011). Protein concentration of sponge eluates was determined by BCA assay (EMD Millipore). Equal amounts of total protein from each sample (100 μg) were then denatured in pH 8.0 urea exchange buffer (8 M Urea [GE HealthCare, United Kingdom], 50 mM HEPES [Sigma-Aldrich]) for 20 min at room temperature, and then placed into 10 kDa Nanosep filter cartridges. After centrifugation, samples were treated with 25 mM dithiothreitol (Sigma-Aldrich) for 20 min, then 50 mM iodoacetamide (Sigma-Aldrich) for 20 min, followed by a wash with 50 mM HEPES buffer. Trypsin (Promega, Fitchburg, WI) was added (2 μg/100 μg protein) and samples were incubated at 37°C overnight in the cartridge. Peptides were eluted off the filter with 50 mM HEPES, and the digestion was stopped with 1% formic acid. Peptides were dried via vacuum centrifugation and cleaned of salts and detergents by reversed-phase liquid chromatography (high pH RP, Agilent 1200 series micro-flow pump, Water XBridge column) using a step-function gradient. The fractions were then dried via vacuum centrifugation. Equal amounts of peptides were re-suspended in 2% acetonitrile (Thermo Fisher Scientific), 0.1% formic acid (EMD, Canada) and injected into a nano-flow LC system (Easy nLC, Thermo Fisher Scientific) connected in-line to a LTQ Orbitrap Velos mass spectrometer (Thermo Fisher Scientific). Mass spectrometry instrument settings were the same as described previously (Burgener et al., 2013).
Quality control and processing of microarray and mass spectrometry dataMicroarray data
All chips used in the analysis passed standard quality control metrics assessed by GenomeStudio (Illumina) as well as visual inspection for anomalies and artifacts. GenomeStudio calculates a detection p value for each probe, which represents the confidence that a given transcript is expressed above background defined by negative control probes. Further processing and statistical analysis of data was done using the R/Bioconductor software suite (Gentleman et al., 2004). To aid in the comparison of gene expression data gathered across the series of hybridization chips used, data were normalized by variance stabilizing transformation and robust spline normalization as described in the Bioconductor lumi package (Du et al., 2008; Lin et al., 2008). The pre-processed data were filtered to include only probes with detection threshold p values of <0.05 in 100% of biopsies in at least one of the four study arms, or in 100% of vaginal epithelial cell cultures, and to remove probes that had a low standard deviation (≤0.5) across all arrays, using the Bioconductor genefilter package (Bourgon et al., 2010). Finally, probes without an Entrez ID were removed, leaving 1928 probes in the rectal biopsies, and 6699 probes in the vaginal epithelial cell cultures, for further statistical analysis.
Mass spectrometry data
All spectra were processed using Mascot Distiller v2.3.2 (Matrix Science, Boston, MA), against the UniProtKB/SwissProt (2012-05) Human (v3.87) database using the decoy database option (2% false discovery rate), along with the following parameters: carbamidomethylation (cysteine residues) (K and N-terminus) as fixed modifications, oxidations (methionine residues) as a variable modification, fragment ion mass tolerance of 0.5 Da, parent ion tolerance of 10 ppm, and trypsin enzyme with up to 1 missed cleavage. Mascot search results were imported into Scaffold (v4.21) (Proteome Software, Portland, OR) and filtered using 80% confidence for peptides, 99% confidence for proteins, and at least two peptides per protein. Label-free protein expression levels based on MS peak intensities were calculated using Progenesis LC-MS software V4.0 (Nonlinear Dynamics, Durham, NC). Feature detection, normalization, and quantification were all performed using default settings in the software. Retention time alignment was performed using automatic settings and manually reviewed for accuracy on each sample. Only charge states between 2+ and 10+ were included. Protein abundances were further normalized by total ion current.
Statistical analysis of microarray and mass spectrometry dataMicroarray data
A Bayesian probabilistic framework, Cyber-T (Baldi and Long, 2001; Long et al., 2001), was run from the CyberT/hdarray library in R to test whether differences in gene expression observed between before and after tenofovir treatment were significant. The effect of tenofovir 1% gel on the rectal mucosa in vivo was tested in paired comparisons between time point I or VII and time point 0. Each treatment arm was considered separately, and researchers were blinded to treatment arm. The effect of tenofovir on vaginal epithelial cell cultures in vitro was tested in paired comparisons between treated and untreated cultures. Each tenofovir dosage (50 and 500 μM) and time point (1, 4, 7, and 14 days) was considered separately. The Benjamini & Hochberg method for estimating false discovery rates (FDR) was used to control for multiple comparisons (Benjamini and Hochberg, 1995). Criteria for significance and relevance were an estimated FDR ≤ 0.05 and a log2 fold expression change of ≥ 0.5 (induction) or ≤ −0.5 (suppression), respectively. To confirm the results of the in vivo MTN-007 study, we reanalyzed the data using a double subtraction strategy, where the paired differences between time point 0 and time points I or VII within each treatment arm were first calculated for each probe and study participant. In a second step, the mean paired differences for each probe within each of the three treatment arms were compared to mean paired differences for each probe within the no-treatment arm. Significance testing for this alternative analysis of the in vivo MTN-007 data was performed via a linear fit model using the limma package (Smyth, 2005), with the same significance criteria used for Cyber-T. The resulting gene lists greatly overlapped with the Cyber-T results. For simplicity, the numbers of induced and suppressed genes reported in ‘Results’ are based on the Cyber-T analysis. Heat maps of differentially regulated genes were generated using MeV 4.8 within the TM4 Microarray Software Suite (Saeed et al., 2006), and hierarchically clustered according to selected gene ontologies found in the databases DAVID 6.7 (Huang da et al., 2009) and InnateDB (Lynn et al., 2008). All microarray data were deposited into the GEO database (accession numbers, GSE57025 and GSE57026).
Mass spectrometry data
For each protein, log2 protein intensity values were averaged across all five subjects separately for time points 0 and VII, fold change values between the two time points were calculated, and statistical analysis for differences between the two time points was done by paired t tests.
Pathway and network analysis of microarray and mass spectrometry dataMicroarray data
Entrez ID designations, assigned to the array probes by Illumina, were uploaded to the Innate DB database and a Gene Ontology (GO) over-representation analysis was performed for gene groups signifying a particular molecular function or biological process, or occurring in specific cellular compartments (Lynn et al., 2008). The following strategy was used to determine which gene groups were enriched in the data set: separately for suppressed and induced genes, ratios of the number of genes in a particular GO group to the total number of genes detected in our data set were compared to the ratios in the same GO group reported for the complete human genome using Fisher's exact test. Ingenuity Pathways Analysis (IPA) (Qiagen) was used to visualize direct and indirect relationships between individual gene products and map their cellular localizations.
Mass spectrometry data
Proteomic data were uploaded into the IPA software package (Qiagen). Associations between protein groups in the data set and canonical pathways were measured similarly to the over-representation analysis described for the microarray data above. IPA software was also used to compare protein expression patterns against the IPA biological function database. IPA software calculates a Benjamini & Hochberg-adjusted p-value for the association of protein abundance changes with each biological function, and a weighted z-score, which estimates whether the protein abundance changes found associated with a specific biological function are likely to enhance (positive z values) or inhibit (negative) the function.
Quantitative confirmation of microarray results by RT-ddPCR
A two-step reverse transcription (RT) droplet digital PCR (ddPCR) was used to confirm microarray transcriptome data for selected genes of interest (Hindson et al., 2011; Pinheiro et al., 2012). In a ddPCR assay, each sample is partitioned into ∼20,000 droplets representing as many individual PCR reactions. The number of target DNA copies present per sample can be quantified based on Poisson distribution statistics, because each individual droplet is categorized as positive or negative for a given gene. For rectal biopsies, cDNA was generated using the High Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific) and the ddPCR was carried out using the 2× ddPCR Supermix for Probes (BioRad, Hercules, CA). For epithelial cell lines, these steps were combined using the 2× One-Step RT-ddPCR Kit for Probes (BioRad). ddPCR was performed on the QX100 droplet digital PCR system (BioRad). Reactions were set up with 20× 6-carboxyfluorescein (FAM)-labeled target gene-specific qPCR assay (Integrated DNA Technologies, Coralville, IA; or Thermo Fisher Scientific) and 20× VIC-labeled housekeeping hemoglobin B (HBB) gene-specific Taqman gene expression assay (Thermo Fisher Scientific). Each assembled ddPCR reaction mixture was loaded in duplicate into the sample wells of an eight-channel disposable droplet generator cartridge (BioRad) and droplet generation oil (BioRad) was added. After droplet generation, the samples were amplified to the endpoint in 96-well PCR plates on a conventional thermal cycler using the following conditions: denaturation/enzyme activation for 10 min at 95°C, 40 cycles of 30 s denaturation at 94°C, and 60 s annealing/amplification at 60°C, followed by a final 10 min incubation step at 98°C. After PCR, the droplets were read on the QX100 Droplet Reader (BioRad). Analysis of the ddPCR data was performed with QuantaSoft analysis software version 1.3.1.0 (BioRad).
Immunohistochemistry of formalin-fixed rectal biopsies
Four-micron sections were cut on a Leica RM2255 Automated Rotary Microtome (Leica, Germany), mounted on positively charged EP-3000 slides (Creative Waste Solutions, Tualatin, OR), dried for 1 hr at 60°C, and stored until staining at 4°C. Slides were deparaffinized in xylene and rehydrated in graded dilutions of ethanol in water. Antigen retrieval was performed by heating the slides in Trilogy Pretreatment Solution (Sigma-Aldrich) for 20 min at boiling temperature in a conventional steamer (Black&Decker, Towson, MD). Slides were then cooled for 20 min, rinsed three times in Tris-buffered 0.15 M NaCl solution containing 0.05% Tween 20 (Wash Buffer; Dako, Santa Clara, CA), and stained at room temperature in an automated slide-processing system (Dako Autostainer Plus). Endogenous peroxidase activity was blocked using 3% H2O2 for 8 min, followed by 10 min in Serum-Free Protein Block (Dako). The slides were then stained for 1 hr with the primary antibodies. Staining was performed with the following primary antibodies: anti-CD3 (RM9107S, rabbit clone SP7; Thermo Fisher Scientific), anti-ubiquitin D (NBP2-13498, rabbit polyclonal anti-FAT10 antibody; Novus Biologicals, Littleton, CO), anti-CD7 (M7255, mouse clone CBC.37; Dako), and anti-interleukin-10 (sc-8438, mouse clone E−10; Santa Cruz Biotechnology, Dallas, TX). Negative control primary immunoglobulins were either whole rabbit IgG (011-000-003; Jackson ImmunoResearch Laboratories, West Grove, PA) or mouse IgG (I-2000; Vector Laboratories, Burlingame, CA).
For CD3 staining, the slides were incubated with anti-CD3 for 60 min at a dilution of 1:25, washed in Wash Buffer, incubated for 30 min with sheep-anti-rabbit Dylight 649 (611-643-122; Rockland Immunochemicals, Limerick, PA), washed, and incubated for 30 min with donkey-anti-sheep Dylight 649 (613-743-168; Rockland). Sections were counter-stained for 20 min with the nucleic acid-binding dye Sytox Orange (S11368; Thermo Fisher Scientific) at a dilution of 1:20,000, and coverslipped in ProLong Gold antifade reagent (P36930; Thermo Fisher Scientific). For ubiquitin D (UBD), CD7 and interleukin-10 (IL-10) staining, the slides were incubated with 0.8 μg/ml (UBD and CD7) or 4 μg/ml (IL-10) of the primary antibody for 60 min. After washing, the anti-UBD-stained slides were incubated for 30 min with Leica Power Vision HRP rabbit-specific antibody polymer (PV6119; Leica); and the anti-CD7 and anti-IL-10 stained slides were incubated for 30 min with LeicaPower Vision HRP mouse-specific antibody polymer (PV6114; Leica). After washing, staining was visualized with 3,3′-diaminobenzidine (Liquid DAB+ Substrate Chromogen System; Dako) for 7 min, and the sections were counter-stained for 2 min with hematoxylin (Biocare, Concord, CA). Controls for all antibodies were run with identical procedures but replacing the primary antibodies with either rabbit IgG or mouse IgG, as appropriate, at calculated matching concentrations.
Acquisition and analysis of stained tissue sections
Sequences of slightly overlapping 20× (for CD3, UBD or CD7) or 40× (for IL-10) images covering each stained tissue section in its entirety were acquired on a bright-field Aperio Scansope AT (for UBD, CD7 and IL-10) or a fluorescent Aperio Scansope FL (for CD3) (Aperio ePathology Solutions, Leica). Overlapping images were stitched together using the Aperio Image Analysis Suite, and the entire tissue sections on each slide were analyzed using Definiens Tissue Studio (Definiens AG, Germany). Individual cells were identified based on the nuclear counter-stain. For CD3 and CD7, all positive cells were automatically counted and reported as number of positive cells per mm2 of tissue section. For IL-10 and UBD, tight regions were manually drawn around all areas of the tissue sections containing columnar epithelial cells. For IL-10, all positive columnar epithelial cells were automatically counted and reported as number of positive cells per mm2. For UBD, which stained all epithelial cells, the mean staining intensity (MSI) of each columnar epithelial cell was measured in arbitrary units and overall UBD staining intensity for each section was reported as the average MSI of all epithelial cells.
Electron microscopy
Formalin-fixed paraffin-embedded rectal biopsies were de-paraffinized and fixed overnight in half-strength Karnovsky's fixative. Staining, embedding, cutting, and viewing on a JEOL 1400 SX transmission electron microscope were performed as previously described (Hladik et al., 1999, 2007). 20 images per sample were acquired at 5,000× magnification. Using ImageJ (Collins, 2007), the two-dimensional sizes in µm2 of all individual mitochondria with a circularity index of ≥0.9 were calculated (for standardization purposes, only mitochondria cut near perfectly along their minor axis were evaluated). 10 images per sample were also acquired at 2,000× magnification, always including the epithelial cell brush border. Using ImageJ, a grid of 1.32 µm2 squares of defined size was overlaid onto each image. All mitochondria in the images were counted, except those in the first row of squares falling on the brush border and in squares along the image rims (which only partially covered the tissue), and the mean numbers of mitochondria per µm2 were calculated for each of the acquired 2,000× images (range of counted squares per image: 23–50).
General statistics
All p values reported were adjusted for multiple testing as appropriate, except for the exploratory protein mass spectrometry data. Correlations of log2 fold gene expression changes between 9 cm and 15 cm biopsies in Figure 1C and Figure 1—figure supplement 1 were tested by Spearman's rank correlation coefficient. Gene and protein expression changes over baseline were compared between study arms by two-tailed Mann–Whitney test (Figure 1—figure supplement 2 and Figure 6). The combined expression data from the microarray and RT-ddPCR tests shown in Figure 2C were compared separately for log-fold induction or suppression over baseline using a one-tailed Wilcoxon signed-rank test with Bonferroni adjustment. Immunohistology measurements in Figure 2D were compared between baseline and after 7 days of treatment by two-tailed paired t tests. Due to high skewness of the CD7+, CD3+, and IL-10+ cell counts, these were tested after log10 transformation. Ratios of induced to suppressed genes in Figure 3A were tested for a difference between cellular compartments using Chi-square statistics. DSP, IL-10, KIAA0101, PNPT1, and ATP6 copy numbers or protein concentrations in Figures 4B,C, 5A,B,F were tested for significant change across three time points by repeated measures ANOVA, and exact Sidak's (Figures 4B,C, 5F) or Tukey (Figure 5A,B) post-tests were run for comparisons between two time points. Due to high skewness of the KIAA0101 copy numbers, these were tested after log10 transformation. The effect of increasing tenofovir concentrations on the proliferation of primary vaginal epithelial cells in Figure 4E was assessed for significance by a linear model accounting for days and donors. Copy numbers between 9 cm and 15 cm biopsies in Figure 5A were compared by two-tailed paired t test. Copy numbers between baseline tenofovir and N-9 in Figure 5B were compared by a two-tailed unpaired t test. Mitochondria counts and sizes in Figure 5C,D were compared between baseline and Day VII, separately for the two subjects evaluated by electron microscopy, by two-tailed unpaired t tests with Bonferroni adjustment. Microarray chip probe expression values were tested for correlation between the three primary vaginal cell cultures by computing pairwise Pearson correlation coefficients (Figure 3—source data 1). Statistics packages used were Prism 6 (Graphpad Software, La Jolla, CA) and Bioconductor/Lumi in R. Statistical results are summarized in Supplementary file 1.
ResultsTenofovir 1% gel induces broad and pronounced gene expression changes in the rectum
We measured mRNA expression changes across the complete human transcriptome by microarrays in rectal biopsies taken at 9 and 15 cm proximal to the anal margin. Biopsies were obtained before treatment, after a single and after seven consecutive once-daily applications of reduced glycerin tenofovir 1% gel, nonoxynol-9 (N-9) 2% gel, hydroxyethyl cellulose (HEC) placebo gel, or no treatment (eight participants per arm were tested by microarrays). The primary results of the clinical study, MTN-007, a phase 1, randomized, double-blind, placebo-controlled trial at three US sites were reported elsewhere (McGowan et al., 2013). Relative to enrollment biopsies, after 7 days of treatment, tenofovir 1% gel suppressed 505 genes and induced 137 genes in the 9 cm biopsies, whereas the detergent N-9, a transient mucosal toxin, suppressed 56 genes and induced 60 genes (log2 fold expression change ≥ 0.5 for induction or ≤ −0.5 for suppression, FDR ≤ 0.05) (Figure 1A,B and Supplementary file 2). 15 suppressed and 4 induced genes were common to tenofovir and N-9 (Figure 1B). In the HEC gel and no treatment arms, 16 and 23 genes changed (Figure 1B). Tenofovir 1% gel affected more genes after 7 days of treatment than after a single application and more genes in 9 cm than in 15 cm biopsies (Figure 1B), with significant correlations between expression changes at 9 cm and 15 cm (suppression: Spearman rho = 0.4775 [95% CI: 0.4051–0.5440]; p < 0.001, Figure 1C; induction: Spearman rho = 0.427 [95% CI: 0.2840–0.5514]; p < 0.001, Figure 1—figure supplement 1). By fold change of the individual genes, tenofovir suppressed genes more strongly than N-9 (median 0.505 vs 0.627-fold; p < 0.001), but induced genes less strongly (1.58 vs 1.69-fold; p < 0.001) (Figure 1—figure supplement 2).10.7554/eLife.04525.003
Tenofovir-induced gene expression changes in the human rectum.
(A) Heat map of differentially expressed genes in eight participants after a single (I) and after seven consecutive once-daily (VII) rectal applications of tenofovir 1% gel compared to baseline in biopsies taken at 9 cm and 15 cm proximal to the anal margin. Red and green bars signify strength of gene induction and suppression, respectively. 642 genes are shown, all of which exhibited an estimated FDR ≤ 0.05 and a log2 fold expression change of ≥ 0.5 (induction) or ≤ −0.5 (suppression) when evaluated jointly for all eight participants at time point VII in the 9 cm biopsies. (B) Numbers of significantly suppressed (green borders and numbers) and induced (red borders and numbers) genes after reduced glycerin tenofovir 1% gel (TFV) treatment and their overlap with nonoxynol-9 (N-9), hydroxyethyl cellulose (HEC) and no treatment (No tx). Circle area symbolizes the number of affected genes, overlap the number of genes independently affected by two or three conditions. (C) Correlation of log2 fold gene suppression from baseline to Day VII between 9 and 15 cm biopsies. All 505 genes significantly suppressed at 9 cm are included. Genes depicted as blue dots were significantly suppressed in both 9 and 15 cm biopsies (15 cm FDR < 0.05), genes depicted as black dots were only significant in 9 cm biopsies (15 cm FDR ≥ 0.05). Spearman rho correlation between the 9 and 15 cm biopsies expression and the corresponding p-value of a Spearman rank correlation test are indicated on the plot. Genes tested in Figure 2B by RT-ddPCR are specifically indicated.
DOI:http://dx.doi.org/10.7554/eLife.04525.003
10.7554/eLife.04525.004
Correlation of log<sub>2</sub> fold gene induction by tenofovir 1% gel from baseline to Day VII between 9 and 15 cm biopsies.
All 137 genes at 9 cm significantly induced (FDR < 0.05) and with log2 fold change > 0.5 are included. Genes depicted as blue dots were significantly induced in both 9 and 15 cm biopsies (15 cm FDR < 0.05), genes depicted as black dots were only significant in 9 cm biopsies (15 cm FDR ≥ 0.05).
DOI:http://dx.doi.org/10.7554/eLife.04525.004
10.7554/eLife.04525.005
Log<sub>2</sub> fold gene suppression (<bold>A</bold>) and induction (<bold>B</bold>) from baseline to Day VII caused by N-9 and tenofovir in 9 cm biopsies.
(A) All genes with an estimated FDR ≤ 0.05 and a log2 fold expression change of ≤ 0.5 are included (N-9: 56 genes; tenofovir: 505 genes). (B) All genes with an estimated FDR ≤ 0.05 and a log2 fold expression change of ≥ 0.5 are included (N-9: 60 genes; tenofovir: 137 genes). Differences in magnitude of gene expression changes were statistically compared between N-9 and tenofovir by unpaired Mann–Whitney tests. The box plots indicate medians and interquartile ranges and the whiskers indicate 10th–90th percentiles.
DOI:http://dx.doi.org/10.7554/eLife.04525.005
Confirmatory RT-ddPCR quantification and in situ immunostaining in additional study participants
To independently confirm the microarray results, we selected nine induced and six suppressed genes and performed reverse transcription digital droplet PCR (RT-ddPCR) assays with RNA from the 9 cm biopsies in the remaining seven individuals enrolled in the tenofovir 1% gel arm (two men and five women), whose biopsies had not been analyzed by microarray (Figure 2). The mRNA copy numbers of all 15 genes increased or decreased as predicted from the microarray data between baseline and time point VII (Figure 2A,B). Next, we combined the microarray and RT-ddPCR expression data, normalized fluorescence and copy number values over their respective baselines, and compared the fold change after 7 days of treatment with baseline (Figure 2C). Expression changes for all 15 genes assessed by microarray and RT-ddPCR were statistically significant (p < 0.01 for all genes except TRIM5 [p = 0.02]).10.7554/eLife.04525.006
Confirmation of microarray data.
(A and B) Quantification of mRNA copy numbers measured in 9 cm biopsies by reverse transcription (RT) droplet digital PCR (ddPCR) relative to the housekeeping gene hemoglobin beta (HBB) copy numbers in seven additional study participants. (A) Nine selected genes induced in the microarrays: CCL19, CCL21, CCL23, CXCL9, CCR7, CD7, CD19, matrix metallopeptidase 12 (MMP12), and serine peptidase inhibitor of the Kazal type 4 (SPINK4). Copy numbers at baseline (0), after a single tenofovir gel application (I) and after seven consecutive once-daily applications (VII) are shown. Line colors signify each of the seven study participants. (B) Six selected suppressed genes: p21-activated kinase (PAK2), nuclear factor of activated T cells 5 (NFAT5), desmoplakin (DSP), TGF-β receptor associated protein (TGFBRAP), interleukin 10 (IL-10), and tripartite motif-containing protein 5 (TRIM5). (C) Normalized fold changes of gene expression at Day VII over baseline in all 15 individuals treated with tenofovir 1% gel. Red dots depict fold changes measured by microarray, blue dots depict fold changes measured by RT-ddPCR. The boxes indicate median and 25th–75th percentiles and the whiskers indicate 10th–90th percentile. Asterisks indicate statistical significance level relative to baseline (one asterisk p < 0.05; two asterisks p < 0.01, by one-sided Wilcoxon tests, adjusted for multiple testing). (D) Immunostaining of formalin-fixed 9 cm rectal biopsies from 10 participants for the proteins CD7 (immunohistochemistry [IHC]), CD3 (immunofluorescence) and ubiquitin D (UBD; IHC), predicted to be induced by the microarrays, and for IL-10 (IHC), predicted to be suppressed. For CD7 and CD3, tissue sections were evaluated in their entirety and positive cells per mm2 are shown at baseline (0) and after seven consecutive once-daily applications (VII). Representative images are shown in Figure 2—figure supplement 1. For UBD and IL-10, only columnar epithelial cells were evaluated. For UBD, the average mean staining intensities (MSI) per cell are shown. Representative images are shown in Figure 2—figure supplement 2. Colors signify each of the 10 study participants. The boxes indicate median and 25th–75th percentiles and the whiskers indicate the range. Paired Wilcoxon signed-rank test p values for differences between 0 and VII are listed.
DOI:http://dx.doi.org/10.7554/eLife.04525.006
10.7554/eLife.04525.007
Immunohistochemistry for CD7, and immunofluorescence for CD3, in rectal biopsies before (0) and after 7 days (VII) of daily tenofovir 1% gel use.
Individual study participants are designated by letters. Blue indicates nuclei, red indicates CD3 or CD7.
DOI:http://dx.doi.org/10.7554/eLife.04525.007
10.7554/eLife.04525.008
Immunohistochemistry for IL-10 and ubiquitin D (UBD) in rectal biopsies before (0) and after 7 days (VII) of daily tenofovir 1% gel use.
Individual study participants are designated by letters. Blue indicates nuclei, brown indicates IL-10 or UBD.
DOI:http://dx.doi.org/10.7554/eLife.04525.008
For additional confirmation, we selected three induced (CD7, CD3 and ubiquitin D) and one suppressed gene (IL-10) for immunohistochemical staining of the respective proteins in 9 cm rectal biopsies from 10 subjects (Figure 2D and Figure 2—figure supplements 1, 2). Consistent with the gene expression studies, infiltrating T lymphocytes increased twofold to fivefold in the mucosa (mean fold change 5.0, 95% CI 2.46–10.1, p < 0.001 for CD7+; 2.44, 1.17–5.11, p = 0.023 for CD3+), whereas IL-10+ columnar epithelial cells decreased by more than half (0.36, 0.18–0.79, p = 0.017), between baseline and following 7 days of tenofovir 1% gel use. Ubiquitin D was widely expressed in all biopsies, but tenofovir treatment increased the intensity of its expression (p = 0.007), as predicted by the microarrays.
Gene expression patterns and functional pathways
Tenofovir 1% gel was more suppressive than stimulatory, with a ratio of induced to suppressed genes in the 9-cm rectal biopsies of 0.116 (17 genes up-regulated to 146 down) for nuclear products. However, genes encoding secreted proteins were more often induced than suppressed (Figure 3A), with a ratio of 2.33 (35 genes up-regulated to 15 down; χ2 p < 0.001). Noteworthy among induced genes for secreted products were the chemokines CCL2, CCL19, CCL21, CCL23, CXCL9, and CXCL13 (Figure 3A). Correspondingly, transcripts of a number of leukocyte-specific cell surface markers increased, specifically CD2, CD3D, CD7, CD8A, CD19, CD52, CD53, CCR6, and CCR7. The kinetics of gene induction is depicted in Figure 3—figure supplement 1.10.7554/eLife.04525.009
Expression pattern and functional pathway analysis.
(A) Ingenuity pathways analysis of tenofovir-induced effects in rectal biopsies, showing cellular localizations of and relationships between individual gene products. Red symbols indicate induction and green symbols suppression at Day VII relative to baseline in 9 cm biopsies. The diagram includes all significant genes identified as primarily located in the extracellular space and the cell nucleus. A few selected significant genes with products localizing to the plasma membrane (PM) or cytoplasm (CP) are also shown based on their putative functional roles. Direct (solid lines) and indirect (dashed) interactions between gene products are indicated. Line color is arbitrary and meant to indicate relationships between groups of genes. Yellow-shaded areas indicate zinc finger transcription factors. (B) Pathways analysis of tenofovir-induced effects in primary vaginal epithelial cells. Only genes that were suppressed or induced by tenofovir both in 9 cm rectum in vivo and in vaginal epithelial cells in vitro are shown. Primary vaginal epithelial cells derived from three healthy women were cultured with 50 or 500 μM tenofovir for 14 days. Global gene expression microarrays at 4, 7, and 14 days of culture were evaluated in comparison to untreated epithelial cells. Pre-processed microarray expression data were extremely consistent between the three vaginal cell cultures (mean Pearson correlation coefficient 0.9912; Figure 3—source data 1).
DOI:http://dx.doi.org/10.7554/eLife.04525.009
10.7554/eLife.04525.010
Pearson correlation coefficients of pre-processed microarray probe expression values between the three primary vaginal cell cultures.
DOI:http://dx.doi.org/10.7554/eLife.04525.010
10.7554/eLife.04525.011
Average strength of gene induction by tenofovir 1% gel across all 8 microarray study participants.
Heat map colors depict fold-change at Day I and Day VII over baseline. Baseline is depicted as the vertical bar labeled ‘0’ filled with the shade of blue corresponding to a fold-change of 1. Genes included in the list exhibited ≥1.6-fold average induction on Day VII or were ≥1.1-fold induced in at least six of eight study participants, and some knowledge about the gene products exists in the literature. The heat map divisions signify three different kinetic patterns of gene induction: flat at day I and then induced; induced at day I and then flat; or induced at day I and then more strongly induced at day VII.
DOI:http://dx.doi.org/10.7554/eLife.04525.011
10.7554/eLife.04525.012
Selected biological processes defined in the InnateDB database with significant enrichment of genes suppressed or induced by tenofovir 1% gel at Day VII in 9 cm biopsies.
Green bars depict the percentage of genes identified as suppressed in a particular process out of the total number of genes included by InnateDB in that process. Red bars depict corresponding percentages for gene induction. Numbers of suppressed and induced genes are indicated above the bars. Gene enrichment in each biological process was tested for statistical significance as described in the ‘Materials and methods’ and the computed p values are depicted by the stars. Not all processes with significant gene enrichment are shown.
DOI:http://dx.doi.org/10.7554/eLife.04525.012
Among suppressed genes with products localizing to the cell nucleus, we identified a large number of known or putative transcription factors and their co-factors, including CREB1 (CAMP responsive element-binding protein 1) and CREBBP (CREB-binding protein), both activators of IL-10 transcription (Woodgett and Ohashi, 2005; Alvarez et al., 2009), NFAT5 (nuclear factor of activated T cells 5) (Neuhofer, 2010), and many zinc finger proteins (Figure 3A). Tenofovir 1% gel also suppressed genes important for regulation of transcription and translation, as well as biological processes involving transforming growth factor beta (TGF-β), epithelial structure organization, regulation of cell proliferation, and apoptosis (Figure 3—figure supplement 2). Lastly, tenofovir 1% gel suppressed genes important for mitochondrial function, including PNPT1 (polyribonucleotide nucleotidyltransferase 1) (Wang et al., 2010) and OPA3 (optic atrophy 3) (Misaka et al., 2002). Among the few genes with nuclear products that were strongly induced, KIAA0101 and ubiquitin D were notable (Yu et al., 2001; Lee et al., 2003; Hosokawa et al., 2007).
Similarities between vaginal keratinocytes and rectal biopsies
Tenofovir 1% gel is most advanced as a candidate product for vaginal HIV prevention. We were therefore interested to extend our findings from the rectum to the vaginal mucosa. We isolated epithelial cells from vaginal tissues donated by three healthy women and tested their response to 50 and 500 μM tenofovir in vitro for up to 14 days of culture using RNA expression analysis. Preliminary tests showed that these dosages give roughly similar intracellular concentrations of the active drug (tenofovir diphosphate) as measured in the vagina after tenofovir 1% gel use (not shown) (Hendrix et al., 2013). Pre-processed microarray expression data were extremely consistent between the three vaginal cell cultures (mean Pearson correlation coefficient 0.9912; Figure 3—source data 1). Tenofovir's effects on the purified vaginal epithelial cells (Figure 3B) were similar to its effects on the rectal mucosa (Figure 3A), but, as expected, changes likely driven by leukocytes in the biopsies were not seen in the purified epithelial cells, or were differently regulated. In the epithelial cells, tenofovir did not significantly change chemokine, chemokine receptor, and cluster of differentiation (CD) genes, and it suppressed ISG 15 (interferon-stimulated gene 15) and MX1 (myxovirus resistance 1), which were induced in the biopsies.
The number of genes affected was initially higher with 500 μM than with 50 μM tenofovir, but equalized after 14 days of culture (Figure 4A). Next, we confirmed the expression changes of select genes by ddPCR with vaginal epithelial cells from four healthy women: mRNA copies of DSP (desmoplakin) and IL-10 significantly decreased, and KIAA0101 significantly increased during 7 days of tenofovir treatment, as seen in the microarray data (Figure 4B). In fact, IL-10 transcripts were virtually eliminated at 7 days (p = 0.002 and p = 0.003 for 50 and 500 μM, respectively), and KIAA0101 increased more than 10-fold at 500 μM tenofovir (p = 0.005). By ELISA of cell lysates, IL-10 protein also decreased significantly (n = 4 cell lines; p = 0.007 and p < 0.001 for 50 and 500 μM, respectively) (Figure 4C).10.7554/eLife.04525.013
Effects of tenofovir on primary vaginal epithelial cells.
(A) Number of suppressed (green) and induced (red) genes in response to treatment with 50 μM (open circles) or 500 μM (filled circles) tenofovir for 1, 4, 7, or 14 days (n = 3 cell lines from different women). (B) Quantification of mRNA copy numbers at days 1 and 7 of culture by RT-ddPCR assays for two selected genes identified as suppressed (DSP and IL-10) and one induced (KIAA0101) in the microarray data set (n = 4 cell lines). (C) Quantification of IL-10 protein concentrations in vaginal epithelial cells at days 1 and 7 of culture by ELISA. Mean (±standard deviation) IL-10 concentrations in the untreated cultures were 5.65 pg/ml (±0.25) at day 1 and 5.86 pg/ml (±0.32) at day 7 of culture. Boxes and error bars in (B) and (C) signify means and standard deviations with vaginal epithelial cell cultures derived from four healthy women. Asterisks indicate statistical significance level relative to untreated (*p < 0.05; **p < 0.01; ***p < 0.001). (D) Selected biological processes defined in the InnateDB and DAVID databases with significant enrichment of genes suppressed or induced by 50 μM tenofovir in vaginal epithelial cells after 7 days of culture. Green bars depict the percentage of genes identified as suppressed in a particular process out of the total number of genes included in that process. Red bars depict gene induction. Numbers of suppressed and induced genes are indicated above the bars. Gene enrichment in each biological process was tested for statistical significance as described in the ‘Materials and methods’ and the computed p values are depicted by the stars. Not all processes with significant gene enrichment are shown. (E) Proliferation of vaginal epithelial cells without or with various concentrations of tenofovir (n = 3 cell lines). Boxes depict mean percent reduction of the alamarBlue reagent in comparison to the maximum reduction. Error bars signify standard deviations.
DOI:http://dx.doi.org/10.7554/eLife.04525.013
Tenofovir treatment of primary vaginal epithelial cells in vitro mostly impacted the same biological processes as it did in the rectum in vivo (Figure 4D and Figure 3—figure supplement 2). Additionally, it suppressed genes important for keratinocyte differentiation and cellular innate immunity, and induced genes involved in DNA damage repair. Furthermore, tenofovir enhanced vaginal epithelial cell proliferation and/or cell survival in vitro (p = 0.02) (Figure 4E).
Signs of mitochondrial dysfunction
Our microarray results indicated that tenofovir suppresses PNPT1 (Figure 1C), which has been characterized as a master regulator of RNA import into mitochondria and whose deletion impairs mitochondrial function (Wang et al., 2010). To explore tenofovir's effects on mitochondria, we first confirmed its inhibition of PNPT1 in the 9 cm and 15 cm rectal biopsies of all 15 study participants by RT-ddPCR. PNPT1 copy numbers decreased more than 10-fold at 9 cm (mean fold change 0.09, 95% CI 0.009–0.172, p < 0.001) and by half at 15 cm (0.53, 0.34–0.73, p < 0.001) after 7 days of treatment (Figure 5A). To directly assess mitochondrial function, we picked one of the 13 genes encoded by mitochondrial DNA, ATP synthase F0 subunit 6 (ATP6), a key component of the proton channel (Houstek et al., 2006), and measured its transcription by RT-ddPCR in the 9 cm biopsies of all 15 study participants in the tenofovir arm (Figure 5B). Because mitochondrial genes are not included on the microarray chips, we had no preexisting information on its expression. ATP6 mRNA copy numbers decreased on average threefold (p = 0.003) after a single application and sixfold after 7 days (p < 0.001). In contrast, ATP6 copy numbers were stable in all 15 study participants treated with 2% N-9 gel (p = 0.491) (Figure 5B).10.7554/eLife.04525.014
Quantification of mitochondria-associated parameters.
(A) PNPT1 mRNA copy numbers measured in 9 and 15 cm biopsies at baseline (0), after a single tenofovir gel application (I) and after seven consecutive once-daily applications (VII) by RT-ddPCR assay. (B) Mitochondrial ATP6 mRNA copy numbers measured in 9 cm biopsies after tenofovir or N-9 treatment. Line colors in (A) and (B) signify the 15 participants in the tenofovir arm. Black lines signify the 15 participants in the N-9 arm. Baseline values were compared between 9 and 15 cm biopsies by paired t-test and between tenofovir and N-9 by unpaired t-test. Expression changes over time were tested for statistical significance by ANOVA with Bonferroni adjusted post-tests. (C) Assessment of mitochondrial density by electron microscopy of 9 cm biopsies in two study participants. Each dot indicates the mean number of mitochondria per µm2 in a separate 2000× image. (D) Assessment of mitochondrial sizes by electron microscopy in the same biopsies. Each dot depicts the size in µm2 of an individual mitochondrion measured at 5000×. Dot colors in (C) and (D) correspond to the line colors of the same two study participants in (A) and (B). Density and size changes were tested for statistical significance by unpaired t-tests. Horizontal lines and error bars depict means and standard deviations. (E) Representative electron microscopy images of normal mitochondria at baseline, and of enlarged and dysmorphic mitochondria at time point VII, in 9 cm biopsies of Subject Y. Fine structural detail is limited due to formalin fixation of biopsies. (F) PNPT1 and ATP6 gene expression in vaginal epithelial cell cultures in response to 1 and 7 days of 50 μM (blue boxes) or 500 μM (green boxes) tenofovir exposure in vitro. Boxes and error bars signify means and standard deviations across four independent experiments with epithelial cell cultures derived from the vaginal mucosa of four healthy women. Statistical significance levels in all figure panels are indicated by asterisks (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ‘ns’, not significant).
DOI:http://dx.doi.org/10.7554/eLife.04525.014
Next, we evaluated changes in mitochondrial number and size between baseline and after 7 days of treatment in two study participants chosen for exhibiting pronounced PNPT1 suppression by tenofovir. The number of mitochondria per µm2 decreased by more than half after 7 days of tenofovir treatment (p < 0.001 for both subjects) (Figure 5C). In the second participant, mitochondria also increased in size by 1.4-fold (p < 0.001) (Figure 5D) and developed dysmorphic cristae during treatment (Figure 5E). In parallel, tenofovir also caused statistically significant inhibition of PNPT1 and ATP6 mRNA expression in vaginal epithelial cells (Figure 5F).
Proteomics of rectal secretions corroborates enhancing effect of tenofovir on cell survival
Extensive protein studies were not an intended part of MTN-007 and samples were not preserved optimally for this purpose. However, 483 proteins were detected consistently at baseline and time point VII by mass spectrometry. 382 proteins increased in the tenofovir arm vs 112 in the no treatment arm (five participants/arm). Among these, significant increases in individual protein expression were only seen in the tenofovir arm (Figure 6). The top 100 proteins exhibited an average fold increase of 3.8 (median 3.2, range 1.6–128.7) in the tenofovir arm, listed by p value in Figure 6. Enrichment analysis primarily indicated enhancement of cell survival and induction of leukocyte migration by tenofovir (Figure 6), which is consistent with our other findings.10.7554/eLife.04525.015
Unbiased mass spectrometry proteomics in rectal secretions from five study participants each in the tenofovir and the no treatment arm.
(A) 483 proteins were consistently detected in all 10 study participants. Of these, 382 proteins in the tenofovir arm had log2 values > 0 for fold change between baseline and after seven daily gel applications, and 112 had log2 fold change values > 0 in the no treatment arm. Log2 fold changes (x axis) and p values signifying the likelihood of change (y axis) for these proteins are shown. Upper panel, tenofovir arm (382 proteins); lower panel, no treatment arm (112 proteins). Data for proteins with log2 fold values ≤ 0 were not interpretable and are not shown. (B) Log2 fold changes of the top 100 proteins upregulated between baseline and after 7 days of daily gel application in each of the five study participants in the tenofovir arm. (C) Selected biofunctional processes defined in the Ingenuity database with significant enrichment of proteins induced in rectal secretions by 7 days of daily tenofovir 1% gel use. The red bars with arrow heads depict the z scores for these biofunctions, indicating the strength of the directionality of the effect. Protein enrichment in each biofunction was tested for statistical significance as described in the ‘Materials and methods’ and the computed p values are depicted by the stars. Numbers of induced proteins are indicated above/below the bars.
DOI:http://dx.doi.org/10.7554/eLife.04525.015
Discussion
Our findings indicate that reduced glycerin rectal tenofovir 1% gel affects expression of a different and much broader range of genes than N-9 2% gel, potentially affecting mucosal immune homeostasis, mitochondrial function, and regulation of epithelial cell differentiation and survival. These results make biological sense given that tenofovir is a DNA chain terminator, with possible off-target effects in human cells (Lewis et al., 2003), and that topical application achieves at least 100-fold higher active drug concentrations in the mucosa than oral administration of 300 mg tenofovir disoproxil fumarate (Anton et al., 2012; Hendrix et al., 2013). Moreover, tenofovir caused similar changes in primary vaginal epithelial cells cultured from several healthy women.
We did not find evidence that tenofovir directly causes inflammation, which is in keeping with our prior report that rectal tenofovir gel did not cause overt histological inflammation or increased mRNA/protein levels of a select panel of pro-inflammatory cytokines (McGowan et al., 2013). Rather, tenofovir dampened anti-inflammatory factors. Most prominently, it strongly inhibited IL-10 gene and protein expression, likely via blocking of CREB1 and its coactivator CREBBP (Martin et al., 2005; Woodgett and Ohashi, 2005; Gee et al., 2006; Alvarez et al., 2009). In addition, it suppressed signaling pathways downstream of TGF-β, a central anti-inflammatory mediator in the gut (Konkel and Chen, 2011; Surh and Sprent, 2012). Consequently, a number of chemokines were induced, such as the B lymphocyte chemoattractant CXCL13 (Ansel et al., 2002), and CCL19 and CCL21, both ligands of CCR7 on T lymphocytes and dendritic cells (Forster et al., 2008). Correspondingly, CCR7, the B cell marker CD19, and the T cell markers CD2, CD3D and CD7 increased. In keeping with this, we observed higher densities of CD3+ and CD7+ T lymphocytes in the rectal mucosa following 7 days of tenofovir 1% gel use. In concert, these changes suggest that tenofovir creates a state of potential hyper-responsiveness to external inflammatory stimuli but does not itself cause inflammation. In populations who, unlike our MTN-007 study cohort, have a high incidence of mucosal infections and associated immune activation, this could potentially diminish the anti-viral protective effect of topical tenofovir prophylaxis (Naranbhai et al., 2012).
Mitochondrial toxicity of nucleotide/nucleoside reverse transcriptase inhibitors such as tenofovir is well described but the mechanism remains unclear (Lewis et al., 2003). We found that tenofovir consistently inhibited expression of PNPT1, which encodes polynucleotide phosphorylase (PNPASE). PNPASE regulates nucleus-encoded RNA import into mitochondria (Wang et al., 2010). In PNPT1 knock-out mice, mitochondrial morphology and respiratory capacity are disrupted in a manner quite similar to the disruption in renal proximal tubular cells in patients with tenofovir-induced nephrotoxicity (Perazella, 2010; Wang et al., 2010). In our study, just 1 week of daily tenofovir 1% gel application lowered transcription of mitochondrial ATP6 by sixfold and caused visible ultrastructural mitochondrial changes. These findings suggest that tenofovir's suppression of PNPT1 expression may underlie its reported, but heretofore unexplained, mitochondrial toxicity.
A number of changes in rectal biopsies and primary vaginal epithelial cells also suggested that tenofovir can cause increased epithelial proliferation. Furthermore, tenofovir's negative effect on mitochondrial function could lead to impairment of tumor progenitor cell apoptosis (Modica-Napolitano et al., 2007; Ni Chonghaile et al., 2011), as has been specifically reported for loss-of-function mutations of mtATP6, a mitochondrial gene strongly suppressed by tenofovir in our study (Shidara et al., 2005). Neoplastic pressure could also arise from the strong induction of KIAA0101 and UBD (ubiquitin D). KIAA0101 is important for regulation of DNA repair (Simpson et al., 2006), is increased in tumor tissues (Yu et al., 2001), and enhances cancer cell growth (Jain et al., 2011; Hosokawa et al., 2007). UBD appears to increase mitotic non-disjunction and chromosome instability (Ren et al., 2006, 2011) and is highly up-regulated in gastrointestinal cancers (Lee et al., 2003; Ren et al., 2006, 2011). Notably, though, these findings remain circumstantial, as there is no actual clinical evidence for carcinogenicity. Nevertheless, they raise the question of whether the relatively high concentrations of tenofovir achieved in the mucosa during topical use could potentially lead to neoplastic lesions with continuous and long-term use. According to Viread's Product Monograph, gastrointestinal tumorigenicity has been observed in mice after high oral dosing of tenofovir disoproxil fumarate. Vaginal tumorigenicity has been documented for azido-thymidine, an NRTI and DNA chain terminator like tenofovir, which induced vaginal hyperplasia and carcinomas when delivered to mice intravaginally as a 2% solution (∼25% carcinoma rate) (Ayers et al., 1996).
This is the first time that a systems biology approach has been applied to a clinical trial of mucosal pre-exposure prophylaxis, and our study shows the value of using these technologies for comprehensive mucosal safety assessment. Our findings raise concerns regarding the safety of topical tenofovir 1% gel in the rectum with long-term use. Tenofovir's effects on vaginal epithelial cells suggest similar activities in the vagina, which we are currently verifying in MTN-014, a phase I clinical trial comparing vaginal and rectal tenofovir 1% gel in a cross-over format. Further studies are required to gauge whether tenofovir, which has become a valuable cornerstone drug in treating HIV infection, can also be safely and effectively used as a vaginal or rectal microbicide.
Acknowledgements
We thank the study participants for their time and effort, and staff in the participating clinics for enrolling and following study participants. We thank Peter Wilkinson and Rafick-Pierre Sékaly from the Vaccine and Gene Therapy Institute of Florida and Sangsoon Woo from the Vaccine and Infectious Disease Division at the Fred Hutchinson Cancer Research Center for advice about microarray data analysis; Gustavo Doncel from CONRAD and the Eastern Virginia Medical School, Michael Boeckh and Stephen Voght from the Fred Hutchinson Cancer Research Center, and Anna Wald from the University of Washington, for critical reading and editing of the manuscript; Keith Jerome from the Vaccine and Infectious Disease Division at the Fred Hutchinson Cancer Research Center for assistance with ddPCR assays; Julie Randolph-Habecker and Kim R Melton from the Experimental Histopathology and Bobbie Schneider from the Electron Microscopy Shared Resources Core Facility at the Fred Hutchinson Cancer Research Center for assistance with immunohistochemistry and electron microscopy; Cary A Moody from the Department of Microbiology and Immunology, University of North Carolina–Chapel Hill, for providing the 3T3-H2 fibroblast feeder cell line; Gretchen Lentz and Michael Fialkow from the Department of Obstetrics and Gynecology at the University of Washington for procuring vaginal tissues for the isolation of vaginal keratinocytes; Aornrutai Promsong and Surada Satthakarn, currently at Prince of Songkla University, Thailand, for help with establishing the four primary vaginal keratinocyte cultures; Allison A McBride from the Laboratory of Viral Diseases at the National Institute of Allergy and Infectious Diseases for advice regarding the culture of primary vaginal keratinocytes; and Craig W Hendrix and Mark A Marzinke from the Division of Clinical Pharmacology, Department of Medicine, The Johns Hopkins University School of Medicine, for measuring the intracellular concentration of tenofovir diphosphate in primary vaginal keratinocytes treated with tenofovir in vitro; Kenzie Birse from the Department of Medical Microbiology, University of Manitoba, for support in proteomic data analysis; Max Abou, Garrett Westmacott, and Stuart McCorrister of the Proteomic Core of the National Microbiology Laboratory at the Public Health Agency of Canada for proteomic sample preparation and technical assistance in mass spectrometry.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
FH, Conception and design, Analysis and interpretation of data, Drafting or revising the article
IMG, Conception and design, Analysis and interpretation of data, Drafting or revising the article
AB, Acquisition of data, Analysis and interpretation of data
LV, Acquisition of data, Analysis and interpretation of data
JS, Acquisition of data, Analysis and interpretation of data
TBB, Acquisition of data, Analysis and interpretation of data
LB, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
RG, Analysed gene expression data, Analysis and interpretation of data
SF, Analysed gene expression data, Analysis and interpretation of data
JYD, Analysed gene expression data, Analysis and interpretation of data
MJC, Analysed gene expression data, Analysis and interpretation of data
SMH, Analysis and interpretation of data, Drafting or revising the article
CH, Recruitment of study participants, Acquisition of data
PA, Operational support for the clinical study, Acquisition of data
SJ, Operational support for the clinical study, Acquisition of data
JP, Advised on all aspects of the study, Conception and design
DRF, Provided study drug, Contributed unpublished essential data or reagents
RDC, Recruitment of study participants, Drafting or revising the article
KHM, Recruitment of study participants, Acquisition of data, Drafting or revising the article
MJME, Conception and design, Drafting or revising the article
Ethics
Clinical trial registration NCT01232803.
Human subjects: MTN-007 (ClinicalTrials.gov registration NCT01232803) was a phase 1, double blind, placebo-controlled trial in which participants were randomized to receive rectal reduced glycerin tenofovir 1%, nonoxynol-9 (N-9) 2% or hydroxyethylcellulose (HEC) gels, or no-treatment (1:1:1:1), at three clinical research sites (Pittsburgh, PA; Birmingham, AL; and Boston, MA). The study protocol was approved by IRBs at all three sites. All participants gave written informed consent. The study details, and general safety and acceptability data, have been published elsewhere (McGowan I et al. [2013] A phase 1 randomized, double blind, placebo controlled rectal safety and acceptability study of tenofovir 1% gel [MTN-007]. PLOS ONE 8: e60147).
Additional files10.7554/eLife.04525.016
Summary of effect sizes and statistical significance testing.
DOI:http://dx.doi.org/10.7554/eLife.04525.016
10.7554/eLife.04525.017
Lists of genes significantly up- or down-regulated by tenofovir 1% or nonoxynol-9 2% gels in 9 cm rectal biopsies over 7 days of treatment.
DOI:http://dx.doi.org/10.7554/eLife.04525.017
Major datasets
The following datasets were generated:
FlemingL, HladikF, 2014, MTN_007 Data, GSE57025; Publically available at the NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/).
FlemingL, HladikF, 2014, Ex vivo TFV data, GSE57026; Publically available at the NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/).
Standard used to collect data:
The clinical trial recruiting subjects for the current paper was published in McGowan I et al. (2013) A phase 1 randomized, double blind, placebo controlled rectal safety and acceptability study of tenofovir 1% gel (MTN-007). PLOS ONE 8: e60147.
ReferencesAbdool KarimQAbdool KarimSSFrohlichJAGroblerACBaxterCMansoorLEKharsanyABSibekoSMlisanaKPOmarZGengiahTNMaarschalkSArulappanNMlotshwaMMorrisLTaylorD; CAPRISA 004 Trial Group2010Effectiveness and safety of tenofovir gel, an antiretroviral microbicide, for the prevention of HIV infection in womenScience3291168117410.1126/science.1193748AlvarezYMunicioCAlonsoSSánchez CrespoMFernándezN2009The induction of IL-10 by zymosan in dendritic cells depends on CREB activation by the coactivators CREB-binding protein and TORC2 and autocrine PGE2The Journal of Immunology1831471147910.4049/jimmunol.0900312AnselKMHarrisRBCysterJG2002CXCL13 is required for B1 cell homing, natural antibody production, and body cavity immunityImmunity16677610.1016/S1074-7613(01)00257-6AntonPACranstonRDKashubaAHendrixCWBumpusNNRichardson-HarmanNElliottJJanockoLKhanukhovaEDennisRCumberlandWGJuCCarballo-DiéguezAMauckCMcGowanI2012RMP-02/MTN-006: a phase 1 rectal safety, acceptability, pharmacokinetic, and Pharmacodynamic study of tenofovir 1% gel compared with oral tenofovir disoproxil fumarateAIDS Research and Human Retroviruses281412142110.1089/aid.2012.0262AntonPASaundersTElliottJKhanukhovaEDennisRAdlerACortinaGTannerKBoscardinJCumberlandWGZhouYVentuneacACarballo-DiéguezARabeLMcCormickTGabelnickHMauckCMcGowanI2011First phase 1 double-blind, placebo-controlled, randomized rectal microbicide trial using UC781 gel with a novel index of ex vivo efficacyPLOS ONE6e2324310.1371/journal.pone.0023243AyersKMCliveDTuckerWEJrHajianGde MirandaP1996Nonclinical toxicology studies with zidovudine: genetic toxicity tests and carcinogenicity bioassays in mice and ratsFundamental and Applied Toxicology3214815810.1006/faat.1996.0118BaldiPLongAD2001A Bayesian framework for the analysis of microarray expression data: regularized t -test and statistical inferences of gene changesBioinformatics1750951910.1093/bioinformatics/17.6.509BenjaminiYHochbergY1995Controlling the false discovery rate: a practical and powerful approach to multiple testingJournal of the Royal Statistical Society, Series B57289300BourgonRGentlemanRHuberW2010Independent filtering increases detection power for high-throughput experimentsProceedings of the National Academy of Sciences of USA1079546955110.1073/pnas.0914005107BurgenerATjernlundAKaldensjoTAbouMMcCorristerSWestmacottGRMogkKAmbroseEBrolidenKBallB2013A systems biology examination of the human female genital tract shows compartmentalization of immune factor expressionJournal of Virology875141515010.1128/JVI.03347-12ChapmanSLiuXMeyersCSchlegelRMcBrideAA2010Human keratinocytes are efficiently immortalized by a Rho kinase inhibitorThe Journal of Clinical Investigation1202619262610.1172/JCI42297CollinsTJ2007ImageJ for microscopyBiotechniques43253010.2144/000112517DuPKibbeWALinSM2008lumi: a pipeline for processing Illumina microarrayBioinformatics241547154810.1093/bioinformatics/btn224ForsterRDavalos-MisslitzACRotA2008CCR7 and its ligands: balancing immunity and toleranceNature Reviews Immunology836237110.1038/nri2297GeeKAngelJBMaWMishraSGajanayakaNParatoKKumarA2006Intracellular HIV-Tat expression induces IL-10 synthesis by the CREB-1 transcription factor through Ser133 phosphorylation and its regulation by the ERK1/2 MAPK in human monocytic cellsThe Journal of Biological Chemistry281316473165810.1074/jbc.M512109200GentlemanRCCareyVJBatesDMBolstadBDettlingMDudoitSEllisBGautierLGeYGentryJHornikKHothornTHuberWIacusSIrizarryRLeischFLiCMaechlerMRossiniAJSawitzkiGSmithCSmythGTierneyLYangJYZhangJ2004Bioconductor: open software development for computational biology and bioinformaticsGenome Biology5R8010.1186/gb-2004-5-10-r80HendrixCWChenBAGudderaVHoesleyCJustmanJNakabiitoCSalataRSoto-TorresLPattersonKMinnisAMGandhamSGomezKRichardsonBABumpusNN2013MTN-001: randomized pharmacokinetic cross-over study comparing tenofovir vaginal gel and oral tablets in vaginal tissue and other compartmentsPLOS ONE8e5501310.1371/journal.pone.0055013HindsonBJNessKDMasquelierDABelgraderPHerediaNJMakarewiczAJBrightIJLuceroMYHiddessenALLeglerTCKitanoTKHodelMRPetersenJFWyattPWSteenblockERShahPHBousseLJTroupCBMellenJCWittmannDKErndtNGCauleyTHKoehlerRTSoAPDubeSRoseKAMontesclarosLWangSStumboDPHodgesSPRomineSMilanovichFPWhiteHEReganJFKarlin-NeumannGAHindsonCMSaxonovSColstonBW2011High-throughput droplet digital PCR system for absolute quantitation of DNA copy numberAnalytical Chemistry838604861010.1021/ac202028gHladikFLentzGAkridgeREPetersonGKelleyHMcElroyAMcElrathMJ1999Dendritic cell-T-cell interactions support coreceptor-independent human immunodeficiency virus type 1 transmission in the human genital tractJournal of Virology7358335842HladikFSakchalathornPBallweberLLentzGFialkowMEschenbachDMcElrathMJ2007Initial events in establishing vaginal entry and infection by human immunodeficiency virus type-1Immunity2625727010.1016/j.immuni.2007.01.007HosokawaMTakeharaAMatsudaKEguchiHOhigashiHIshikawaOShinomuraYImaiKNakamuraYNakagawaH2007Oncogenic role of KIAA0101 interacting with proliferating cell nuclear antigen in pancreatic cancerCancer Research6725687610.1158/0008-5472.CAN-06-4356HoustekJPickováAVojtiskováAMrácekTPecinaPJesinaP2006Mitochondrial diseases and genetic defects of ATP synthaseBiochimica Et Biophysica Acta17571400140510.1016/j.bbabio.2006.04.006Huang daWShermanBTLempickiRA2009Systematic and integrative analysis of large gene lists using DAVID bioinformatics resourcesNature Protocols4445710.1038/nprot.2008.211JainMZhangLPattersonEEKebebewE2011KIAA0101 is overexpressed, and promotes growth and invasion in adrenal cancerPLOS ONE6e2686610.1371/journal.pone.0026866KonkelJEChenW2011Balancing acts: the role of TGF-beta in the mucosal immune systemTrends in Molecular Medicine1766867610.1016/j.molmed.2011.07.002LeeCGRenJCheongISBanKHOoiLLYong TanSKanANuchprayoonIJinRLeeKHChotiMLeeLA2003Expression of the FAT10 gene is highly upregulated in hepatocellular carcinoma and other gastrointestinal and gynecological cancersOncogene222592260310.1038/sj.onc.1206337LewisWDayBJCopelandWC2003Mitochondrial toxicity of NRTI antiviral drugs: an integrated cellular perspectiveNature Reviews Drug Discovery281282210.1038/nrd1201LinSMDuPHuberWKibbeWA2008Model-based variance-stabilizing transformation for Illumina microarray dataNucleic Acids Research36e1110.1093/nar/gkm1075LiuXOryVChapmanSYuanHAlbaneseCKallakuryBTimofeevaOANealonCDakicASimicVHaddadBRRhimJSDritschiloARiegelAMcBrideASchlegelR2012ROCK inhibitor and feeder cells induce the conditional reprogramming of epithelial cellsThe American Journal of Pathology18059960710.1016/j.ajpath.2011.10.036LongADMangalamHJChanBYTolleriLHatfieldGWBaldiP2001Improved statistical inference from DNA microarray data using analysis of variance and a Bayesian statistical framework. Analysis of global gene expression in Escherichia coli K12The Journal of Biological Chemistry276199371994410.1074/jbc.M010192200LouissaintNACaoYJSkipperPLLibermanRGTannenbaumSRNimmagaddaSAndersonJREvertsSBakshiRFuchsEJHendrixCW2013Single dose Pharmacokinetics of oral tenofovir in plasma, Peripheral Blood Mononuclear cells, Colonic and vaginal tissueAIDS Research and Human Retroviruses291443145010.1089/AID.2013.0044LynnDJWinsorGLChanCRichardNLairdMRBarskyAGardyJLRocheFMChanTHShahNLoRNaseerMQueJYauMAcabMTulpanDWhitesideMDChikatamarlaAMahBMunznerTHokampKHancockREBrinkmanFS2008InnateDB: facilitating systems-level analyses of the mammalian innate immune responseMolecular Systems Biology421810.1038/msb.2008.55MarrazzoJMRamjeeGRichardsonBAGomezKMgodiNNairGPalaneeTNakabiitoCvan der StratenANoguchiLHendrixCWDaiJYGaneshSMkhizeBTaljaardMParikhUMPiperJMâsseBGrossmanCRooneyJSchwartzJLWattsHMarzinkeMAHillierSLMcGowanIMChirenjeZMVOICE Study Team2015Tenofovir-based preexposure prophylaxis for HIV infection among African womenThe New England Journal of Medicine37250951810.1056/NEJMoa1402269MartinMRehaniKJopeRSMichalekSM2005Toll-like receptor-mediated cytokine production is differentially regulated by glycogen synthase kinase 3Nature Immunology677778410.1038/ni1221McGowanIHoesleyCCranstonRDAndrewPJanockoLDaiJYCarballo-DieguezAAyudhyaRKPiperJHladikFMayerK2013A phase 1 randomized, double blind, placebo controlled rectal safety and acceptability study of tenofovir 1% gel (MTN-007)PLOS ONE8e6014710.1371/journal.pone.0060147MisakaTMiyashitaTKuboY2002Primary structure of a dynamin-related mouse mitochondrial GTPase and its distribution in brain, subcellular localization, and effect on mitochondrial morphologyThe Journal of Biological Chemistry277158341584210.1074/jbc.M109260200Modica-NapolitanoJSKulawiecMSinghKK2007Mitochondria and human cancerCurrent Molecular Medicine712113110.2174/156652407779940495NaranbhaiVAbdool KarimSSAltfeldMSamsunderNDurgiahRSibekoSAbdool KarimQCarrWHCAPRISA004 TRAPS team2012Innate immune activation enhances HIV acquisition in women, diminishing the effectiveness of tenofovir microbicide gelThe Journal of Infectious Diseases206993100110.1093/infdis/jis465NeuhoferW2010Role of NFAT5 in inflammatory disorders associated with osmotic stressCurrent Genomics1158459010.2174/138920210793360961Ni ChonghaileTSarosiekKAVoTTRyanJATammareddiAMoore VdelGDengJAndersonKCRichardsonPTaiYTMitsiadesCSMatulonisUADrapkinRStoneRDeangeloDJMcConkeyDJSallanSESilvermanLHirschMSCarrascoDRLetaiA2011Pretreatment mitochondrial priming correlates with clinical response to cytotoxic chemotherapyScience3341129113310.1126/science.1206727PerazellaMA2010Tenofovir-induced kidney disease: an acquired renal tubular mitochondriopathyKidney International781060106310.1038/ki.2010.344PinheiroLBColemanVAHindsonCMHerrmannJHindsonBJBhatSEmslieKR2012Evaluation of a droplet digital polymerase chain reaction format for DNA copy number quantificationAnalytical Chemistry841003101110.1021/ac202578xRenJKanALeongSHOoiLLJeangKTChongSSKonOLLeeCG2006FAT10 plays a role in the regulation of chromosomal stabilityThe Journal of Biological Chemistry281114131142110.1074/jbc.M507218200RenJWangYGaoYMehtaSBLeeCG2011FAT10 mediates the effect of TNF-alpha in inducing chromosomal instabilityJournal of Cell Science1243665367510.1242/jcs.087403RheinwaldJGGreenH1975Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cellsCell633134310.1016/S0092-8674(75)80001-8SaeedAIBhagabatiNKBraistedJCLiangWSharovVHoweEALiJThiagarajanMWhiteJAQuackenbushJ2006TM4 microarray software suiteMethods in Enzymology41113419310.1016/S0076-6879(06)11009-5ShidaraYYamagataKKanamoriTNakanoKKwongJQManfrediGOdaHOhtaS2005Positive contribution of pathogenic mutations in the mitochondrial genome to the promotion of cancer by prevention from apoptosisCancer Research651655166310.1158/0008-5472.CAN-04-2012SimpsonFLammerts van BuerenKButterfieldNBowlesJAdolpheCSimmsLAYoungJWalshMDLeggettBFowlesLFWickingC2006The PCNA-associated factor KIAA0101/p15(PAF) binds the potential tumor suppressor product p33ING1bExperimental Cell Research312738510.1016/j.yexcr.2005.09.020SmythGKGentlemanRCareyVDudoitSIrizarryRHuberW2005Limma: linear models for microarray dataBioinformatics and computational biology Solutions using R and bioconductorNew YorkSpringer397420SurhCDSprentJ2012TGF-beta puts the brakes on homeostatic proliferationNature Immunology1362863010.1038/ni.2345TienDSchnaareRLKangFCohlGMcCormickTJMoenchTRDoncelGWatsonKBuckheitRWLewisMGSchwartzJDouvilleKRomanoJW2005In vitro and in vivo characterization of a potential universal placebo designed for use in vaginal microbicide clinical trialsAIDS Research and Human Retroviruses2184585310.1089/aid.2005.21.845WangGChenHWOktayYZhangJAllenELSmithGMFanKCHongJSFrenchSWMcCafferyJMLightowlersRNMorseHCIIIKoehlerCMTeitellMA2010PNPASE regulates RNA import into mitochondriaCell14245646710.1016/j.cell.2010.06.035WoodgettJROhashiPS2005GSK3: an in-Toll-erant protein kinase?Nature Immunology675175210.1038/ni0805-751YuPHuangBShenMLauCChanEMichelJXiongYPayanDGLuoY2001p15(PAF), a novel PCNA associated factor with increased expression in tumor tissuesOncogene2048448910.1038/sj.onc.120411310.7554/eLife.04525.018Decision letterSgaierSemaReviewing editorBill & Melinda Gates Foundation, India
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Mucosal effects of tenofovir 1% gel” for consideration at eLife. Your article has been favorably evaluated by Prabhat Jha (Senior editor), Sema Sgaier (Reviewing editor), and 3 reviewers, one of whom, Stephen Becker, has agreed to share his identity.
The Reviewing editor and the reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
The Reviewing editor and the reviewers agree that this work is very important, timely, comprehensive, and should be published. However, before making a final decision, we would like the minor comments below to be addressed.
1) It appears that the data reported here are from the use of the reduced glycerin rectal formulation of tenofovir gel. Since there is testing on vaginal mucosa, there may have also been use of the standard formulation. Please ensure that it is clear to the reader which formulation has been used and include in the Discussion whether the authors feel that the findings are pertinent to the formulation examined in efficacy studies.
2) Methods section:
a) Paragraph 12 is a bit confusing with regard to understanding which specimen type (vaginal keratinocytes or rectal mucosa) is being tested. It appears to be rectal tissue because of the reference to time point 0, I and VII, however the description is in a paragraph that also discusses testing of the vaginal tissue.
b) Rectal study:
Please clarify when the rectal sponge samples were collected (at the same time as the biopsies?).
Were the samples blinded to the researchers?
The 1-day and 7-day treatment samples not only differed in treatment duration, but also in timing of sample collection post treatment. The 1-day samples were collected 30-60 min after tenofovir application and would have shown only acute effects, whereas the 7-day samples were collected 24 hours after the last dose and would have shown chronic effects. To more accurately compare differences in gene expression between one-dose vs 7-daily doses, the same post treatment time point(s) should be used for sampling. The significance of the difference in sample timing at the two treatment time points should be clarified in the text.
c) Vaginal cell experiments:
Please explain the use of cholera toxin and hydrocortisone in the media, and the effects they may have on inflammatory endpoints?
If possible, provide more details concerning the vaginal tissue donors (age, hormonal use etc.), and quality control measures used for cell cultures (number of passages, lack of contamination with fibroblast or other cell types etc).
Please explain the significance of the Alamar blue data shown in the supplement.
3) Discussion section:
In the fourth paragraph it states that while KIAA0101 and UBD appear to be affected by tenofovir gel, there is no actual clinical evidence for carcinogenicity. Does this refer to lack of evidence from the data presented here? If so, could it be argued that the exposure/follow up here is sufficient to make such a statement? If based on prior data, such as animal carcinogenesis studies performed as part of preclinical development of tenofovir, this is should be clarified and the studies should be cited.
10.7554/eLife.04525.019Author response
1) It appears that the data reported here are from the use of the reduced glycerin rectal formulation of tenofovir gel. Since there is testing on vaginal mucosa, there may have also been use of the standard formulation. Please ensure that it is clear to the reader which formulation has been used and include in the Discussion whether the authors feel that the findings are pertinent to the formulation examined in efficacy studies.
The vaginal mucosa was not tested; rather, we isolated primary vaginal epithelial cells from four healthy women and exposed them to tenofovir in vitro. These tenofovir exposures were not done in gel form but as tenofovir dilutions in media. We have revised the Abstract to make this clear.
2) Methods section:
a) Paragraph 12 is a bit confusing with regard to understanding which specimen type (vaginal keratinocytes or rectal mucosa) is being tested. It appears to be rectal tissue because of the reference to time point 0, I and VII, however the description is in a paragraph that also discusses testing of the vaginal tissue.
We think the confusion arose from the different nature of in vivo (MTN-007) and in vitro (epithelial cells) data. In vitro, we can treat the same cells with tenofovir or leave them untreated. Therefore, direct pairwise comparisons between treated and untreated cells are possible. In vivo, each subject only received one treatment, and statistical analysis of the in vivo data has to be done in a longitudinal fashion as pairwise longitudinal comparisons between before and after treatment in each subject. In contrast to the more straight forward analysis of the in vitro data, the in vivo data can therefore be analyzed in two different ways, as described in the Methods.
In the Methods section for statistical microarray analysis, we have now clarified that only the in vivo MTN-007 data were analyzed in two ways, whereas the in vitro data were analyzed in one way.
b) Rectal study:
Please clarify when the rectal sponge samples were collected (at the same time as the biopsies?).
The rectal sponges were at the same visits, immediately before the biopsies were taken. We added this information to the Methods section “Unbiased Mass Spectrometry of Rectal Sponge Eluates”.
Were the samples blinded to the researchers?
Yes. Unblinding of the microarray results was in fact the most surprising moment of the study. We knew (since we did have time point information) that one arm showed a particularly large number of gene expression changes. Based on our initial hypotheses we were convinced that this had to be the nonoxynol-9 (N-9) arm. When it turned out to be the tenofovir arm, we at first thought that a mistake had occurred during unblinding. However, further cross-checking confirmed that 1% tenofovir led to many more changes than 2% N-9. This was borne out by the subsequent in-depth pathway analysis, which revealed mitochondrial changes that are typical for NRTIs but not for N-9, and subsequently the good concordance between the in vivo and the in vitro tenofovir data.
The 1-day and 7-day treatment samples not only differed in treatment duration, but also in timing of sample collection post treatment. The 1-day samples were collected 30-60 min after tenofovir application and would have shown only acute effects, whereas the 7-day samples were collected 24 hours after the last dose and would have shown chronic effects. To more accurately compare differences in gene expression between one-dose vs 7-daily doses, the same post treatment time point(s) should be used for sampling. The significance of the difference in sample timing at the two treatment time points should be clarified in the text.
We have now emphasized this difference better in the text of the Methods section “Mucosal Biopsy Procedures”.
c) Vaginal cell experiments:
Please explain the use of cholera toxin and hydrocortisone in the media, and the effects they may have on inflammatory endpoints?
This media formulation is standard for culturing primary epithelial cells since publication of a classic paper in Cell by Rheinwald and Green (Cell 6:331-344; 1975), and many follow-up papers by Jim Rheinwald (Harvard) and others. However, none explains mechanistically why these two particular ingredients are needed. It is true that cholera toxin could be considered as potentially pro-inflammatory and hydrocortisone as potentially anti-inflammatory. However, they were present in all cultures, and therefore their effect should not contribute to differences between tenofovir-treated and -untreated cultures. The two factors could perhaps lead to some changes longitudinally over time, but this is not tested in the in vitro experiments. In vitro, we compare tenofovir-treated and -untreated cells independently at each time point, but we do not compare between different time points.
If possible, provide more details concerning the vaginal tissue donors (age, hormonal use etc), and quality control measures used for cell cultures (number of passages, lack of contamination with fibroblast or other cell types etc.).
We harvest discarded vaginal tissues under a waiver of consent, which prohibits us from obtaining any clinical or identifying information from the tissue donors. All we know is that the vaginal surgeries are performed for benign conditions. The strong concordance of the raw gene expression data between the three vaginal cell lines tested by microarrays (Figure 3–source data 1) indicates that unknown conditions in the tissue donors, or unknown differences of cell cultures in vitro, did not lead to identifiable outlier results in the in vitro experiments.
Please explain the significance of the Alamar blue data shown in the supplement.
The AlamarBlue data are part of Figure 4 (Figure 4E), not the supplement, and indicate that tenofovir enhanced vaginal cell proliferation in vitro.
3) Discussion section:
In the fourth paragraph it states that while KIAA0101 and UBD appear to be affected by tenofovir gel, there is no actual clinical evidence for carcinogenicity. Does this refer to lack of evidence from the data presented here? If so, could it be argued that the exposure/follow up here is sufficient to make such a statement? If based on prior data, such as animal carcinogenesis studies performed as part of preclinical development of tenofovir, this is should be clarified and the studies should be cited.
Lack of carcinogenicity cannot be inferred from the short-term data in our study. No peer-reviewed studies exist that demonstrate carcinogenicity for tenofovir, nor do studies exist that reliably rule out carcinogenicity.
diff --git a/elife04729.xml b/elife04729.xml
new file mode 100644
index 0000000..a8fd7a9
--- /dev/null
+++ b/elife04729.xml
@@ -0,0 +1 @@
+elifeeLifeeLifeeLife2050-084XeLife Sciences Publications, Ltd0472910.7554/eLife.04729Research articleGenomics and evolutionary biologyThe genetic architecture of gene expression levels in wild baboonsTungJenny1*†abcZhouXiang12†dAlbertsSusan C34StephensMatthew12GiladYoav1*Department of Human Genetics, University of Chicago, Chicago, United StatesDepartment of Statistics, University of Chicago, Chicago, United StatesInstitute of Primate Research, National Museums of Kenya, Nairobi, KenyaDepartment of Biology, Duke University, Durham, United StatesDermitzakisEmmanouil TReviewing editorUniversity of Geneva Medical School, SwitzerlandFor correspondence: jt5@duke.edu (JT);gilad@uchicago.edu (YG)
These authors contributed equally to this work
Department of Evolutionary Anthropology, Duke University, Durham, United States
Duke Population Research Institute, Duke University, Durham, United States
Institute of Primate Research, National Museums of Kenya, Nairobi, Kenya
Department of Biostatistics, University of Michigan, Ann Arbor, United States
Primate evolution has been argued to result, in part, from changes in how genes are regulated. However, we still know little about gene regulation in natural primate populations. We conducted an RNA sequencing (RNA-seq)-based study of baboons from an intensively studied wild population. We performed complementary expression quantitative trait locus (eQTL) mapping and allele-specific expression analyses, discovering substantial evidence for, and surprising power to detect, genetic effects on gene expression levels in the baboons. eQTL were most likely to be identified for lineage-specific, rapidly evolving genes; interestingly, genes with eQTL significantly overlapped between baboons and a comparable human eQTL data set. Our results suggest that genes vary in their tolerance of genetic perturbation, and that this property may be conserved across species. Further, they establish the feasibility of eQTL mapping using RNA-seq data alone, and represent an important step towards understanding the genetic architecture of gene expression in primates.
DOI:http://dx.doi.org/10.7554/eLife.04729.001
10.7554/eLife.04729.002eLife digest
Our genes contain the instructions needed to make all aspects of the body. These instructions can be changed by altering the sequence of the DNA that makes up the genes, which can account for many of the different characteristics found in humans and other animals.
However, our characteristics can also be altered by changing how often the genes issue their instructions, which is known as gene expression. For example, it is thought that changes in the expression of some genes in primates may account for the expansion of brain sizes over evolutionary time, particularly in the ancestors of modern humans. Most studies into gene expression in primates have compared different species or focused on humans. It is less clear how many, and what type of, genes vary in expression between individuals of the same species in other natural populations.
Here, Tung, Zhou et al. used a technique called RNA-sequencing to study gene expression in a population of wild baboons that have been studied for over four decades by the Amboseli Baboon Research Project in Kenya. This involved collecting blood samples from 63 individually recognized adult baboons. After RNA-sequencing, Tung, Zhou et al. were able to identify specific sections of the baboon genome where the DNA sequence an individual baboon carried could predict how highly individual genes were expressed. These sections are known as ‘expression quantitative trait loci’ (or eQTLs for short).
Tung, Zhou et al. found that there was a lot of genetically controlled variation in gene expression across the 63 baboons. Most of the eQTLs were found to be in genes that are rapidly evolving or are relatively new. There were fewer eQTLs in genes that are shared across a wide variety of species, possibly because keeping the expression of these genes stable is important for processes that are essential for life.
Many of the eQTLs found in the baboons were in genes where eQTLs are also found in humans. This suggests that the set of genes where genetic variation affects gene expression in the baboons may also be a similar set in humans.
Tung, Zhou et al. also examined how the age, sex, and social integration of the baboons affected the variation in gene expression observed in the population. They found that for most genes, these factors had only small effects on gene expression levels. However, for some genes, these factors could affect the level of expression throughout the life of the individual.
These findings demonstrate that it is feasible to study gene expression patterns in wild primates. The next challenge is to investigate how environmental and genetic factors combine to influence gene expression, and the evolutionary impact of these effects for animals as a whole.
DOI:http://dx.doi.org/10.7554/eLife.04729.002
Author keywordsbaboonallele-specific expressionRNA-seqexpression quantitative trait locusResearch organismyellow baboon (Papio cynocephalus)http://dx.doi.org/10.13039/100000049National Institute on AgingAG034513AlbertsSusan Chttp://dx.doi.org/10.13039/100000001National Science Foundation (NSF)IOS 0919200AlbertsSusan Chttp://dx.doi.org/10.13039/100000057National Institute of General Medical Sciences (NIGMS)GM077959GiladYoavhttp://dx.doi.org/10.13039/100000051National Human Genome Research Institute (NHGRI)HG006123GiladYoavhttp://dx.doi.org/10.13039/100007234University of ChicagoTungJennyhttp://dx.doi.org/10.13039/100000049National Institute on AgingAG031719AlbertsSusan Chttp://dx.doi.org/10.13039/100000001National Science Foundation (NSF)DEB 0846286AlbertsSusan CThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementRNA sequencing of individuals within a wild baboon population reveals extensive power to detect functional regulatory variation, and suggests that the set of genes affected by such variation may be conserved across species.Introduction
Gene regulatory variation has been shown to make fundamental contributions to phenotypic variation in every species examined to date. This relationship has been demonstrated most clearly at the level of gene expression, which captures the integrated output of a large suite of other regulatory mechanisms. Variation in gene expression levels has been linked to fitness-related morphological, physiological, and behavioral variation in both lab settings and natural populations (e.g., Abzhanov et al., 2004; Hammock and Young, 2005; Tishkoff et al., 2006; Chan et al., 2010; reviewed in Wray, 2007), and is a robust biomarker of disease in humans (e.g., Golub et al., 1999; Borovecki et al., 2005). In addition, patterns of gene expression are often associated with signatures of natural selection (Rifkin et al., 2003; Denver et al., 2005; Gilad et al., 2006a; Blekhman et al., 2008), suggesting their functional importance even when their phenotypic significance remains unknown.
In primates, the majority of research on the evolution of gene expression has concentrated on cross species comparisons, particularly using humans, chimpanzees, and rhesus macaques (Enard et al., 2002; Cáceres et al., 2003; Khaitovich et al., 2004; Gilad et al., 2005, 2006b; Haygood et al., 2007; Blekhman et al., 2008; Babbitt et al., 2010; Barreiro et al., 2010; Blekhman et al., 2010; Brawand et al., 2011; Perry et al., 2012). These studies—motivated by a long-standing argument about the importance of gene regulation in primate evolution (King and Wilson, 1975)—have been important for identifying patterns of constraint on gene expression phenotypes over long evolutionary time scales, and for suggesting candidate loci that might contribute to phenotypic uniqueness in humans or other species. For example, gene expression patterns associated with neurological development appear to have experienced an accelerated rate of change in primates relative to other mammals, with axonogenesis-related and cell adhesion-related genes accelerated specifically in the human lineage (Brawand et al., 2011). Similarly, differentially expressed genes in human liver are enriched for metabolic function (Blekhman et al., 2008), suggesting a potential molecular basis for arguments implicating dietary shifts in the emergence of modern humans (Kaplan et al., 2000; Ungar and Teaford, 2002; Wrangham, 2009).
Adaptively relevant changes in gene expression levels across species implicate selection on gene expression phenotypes within species, and particularly within populations, the basic unit of evolutionary change. However, in contrast to cross species comparisons, we still know little about the genetic architecture of gene expression levels in natural nonhuman primate populations. No estimates of the heritability of gene expression traits are available, even for populations that have been intensively studied for many decades. We also do not know whether segregating genetic variation that affects gene expression is common or rare, how the effect sizes of such variants are distributed, or whether they carry a signature indicative of natural selection. If gene regulatory variation has indeed been key to primate evolution, as classic arguments suggest (King and Wilson, 1975), then large gaps therefore remain in our understanding of this process.
Three primary reasons combine to account for the absence of such data. First, until relatively recently, the only feasible approach for measuring genome-wide gene expression levels on a population scale was microarray technology. This constraint limited the diversity of systems that could be assessed because cost-effective, commercially available arrays have only been developed for a handful of taxa. Second, genomic resources, especially detailed catalogs of known genetic variants (e.g., 1000 Genomes Project Consortium et al., 2010, International HapMap Consortium, 2005), are also limited to a small set of species. The lack of such resources creates major barriers to genome-scale studies of the genetics of gene expression in other organisms, which rely on complementary gene expression and genotype data. Finally, for many taxa, samples suitable for gene expression profiling can be challenging to collect. In nonhuman primates, for example, RNA samples are rarely available even for the most intensively studied natural populations.
Recently, sequencing-based methods for measuring gene expression levels (e.g., RNA-seq) have eliminated the need for species-specific arrays. Comparative genomic studies using RNA-seq have thus vastly expanded the set of taxa for which genome-wide expression data are available (including primates: Brawand et al., 2011; Perry et al., 2012). Importantly, because fragments of expressed genes are resequenced many times in RNA-seq studies, data on genetic variation are also generated in the process. Although these data can be affected by technical biases, several studies have demonstrated the generally high reliability of genotypes inferred from RNA-seq reads (Perry et al., 2012; Piskol et al., 2013). Such data can provide important insight into genetic diversity in species for which little other information exists (Perry et al., 2012). Additionally, they provide the two ingredients necessary for mapping gene expression traits to genotype, at moderate cost and without the requirement for previously ascertained genetic variants.
Here, we evaluate the potential for such work in an intensively studied wild primate population, the baboons (Papio cynocephalus) of the Amboseli basin in Kenya. 43 years of prior research on this population have established it as an important model for human social behavior, health, and aging (Alberts and Altmann, 2012), and have facilitated the development of protocols for collecting samples appropriate for gene expression analysis (Tung et al., 2009; Babbitt et al., 2012; Runcie et al., 2013). We generated RNA-seq data for 63 individually recognized members of the Amboseli study population. We used these data to explore the frequency, impact, and potential selective relevance of variants associated with variation in gene expression levels, using complementary expression quantitative trait locus (eQTL) mapping and allele-specific expression (ASE) approaches. We found evidence for abundant functional regulatory variation in the Amboseli baboons, and a surprising amount of power to detect these variants even with a modest sample size. We also found that functional variants are depleted in highly conserved genes, consistent with constraint on gene expression patterns. However, among genes with eQTL, we did not find strong support for a relationship between effect size and minor allele frequency. Such a relationship would be consistent with pervasive negative selection on gene expression phenotypes (i.e., selection against variants that produce large perturbations in gene expression levels) and has been suggested by work in humans (Battle et al., 2014). Finally, we used our data set to provide the first estimates of the heritability of gene expression levels in wild primates, including the relative contributions of cis-acting and trans-acting genetic variation.
ResultsFunctional regulatory variation is common in the Amboseli baboons
We obtained blood samples from 63 individually recognized adult baboons in the Amboseli population (Figure 1—figure supplement 1). From these samples, we produced a total of 1.89 billion RNA-seq reads (mean of 30.0 ± 4.5 s.d. million reads per individual, with 8.6 ± 1.8 s.d. million reads uniquely mapped to exons: Supplementary file 1A). On average, 67.2% of reads mapped to the most recent release of the baboon genome (Panu2.0), 69.2% of which could be assigned to a unique location. We used the set of uniquely mapped reads to estimate gene-wise gene expression levels for NCBI-annotated baboon RefSeq genes. After subsequent read processing and normalization steps (‘Materials and methods’, Figure 1—figure supplement 1 and Figure 1—figure supplement 2), we considered variation in gene expression levels for 10,409 genes expressed in whole blood (i.e., all genes for which we could test for cis-acting genetic effects on gene expression).
We also used the RNA-seq reads to identify segregating genetic variants in the Amboseli population. We considered only high confidence sites that were variable within the Amboseli population (‘Materials and methods’; Figure 1—figure supplement 3). As expected (Piskol et al., 2013), these sites were highly enriched in annotated gene bodies (Figure 1; Figure 1—figure supplement 4). Based on parallel analyses applied to human RNA-seq data, we estimated approximately 97% of these sites to be true positives, and a median correlation between true genotypes and inferred genotypes of 98.7% (‘Materials and methods’; Figure 1—figure supplements 5–6). To identify putative expression quantitative trait loci (eQTL), we focused on variants that passed quality control filters, within 200 kb of the gene of interest. Such variants represent likely cis-acting eQTL, which are more readily identifiable in small sample sizes than trans-eQTL. To identify cases of allele-specific expression, which provides independent but complementary evidence for functional cis-regulatory variation, we focused on genes for which multiple heterozygotes were identified for variants in the exonic regions of expressed genes. We also required a minimum total read depth at exonic heterozygous sites of 300 reads (which should provide high power to detect modest ASE: Fontanillas et al., 2010), resulting in a total set of 2280 genes tested for ASE.10.7554/eLife.04729.003
Baboon eQTLs are enriched in and near genes.
The locations of all SNPs tested in the eQTL analysis are shown in gold relative to the 5′ most gene transcription start site (TSS) and the 3′ most gene transcription end site (TES) for all 10,409 genes. SNPs detected as eQTL are overplotted in blue, and are enriched, relative to all SNPs tested, near transcription start sites, transcription end sites, and within gene bodies. Gray shaded rectangle denotes the region bounded by the TSS and TES, with gene lengths divided into 20 bins for visibility (because the gene body is thus artificially enlarged, SNP density within genes cannot be directly compared with SNP density outside of genes). Note that SNPs that fall outside of one focal gene may fall within the boundaries of other genes. Inset: distribution of all SNPs tested relative to the location of genes, highlighting the concentration of SNPs in genes (the peak at the center of the plot). See Figure 1—figure supplements 1–14 for additional details on workflow, variant calling validation, location of all analyzed SNPs relative to genes, agreement between eQTL and ASE detection, and effects of local structure.
DOI:http://dx.doi.org/10.7554/eLife.04729.003
10.7554/eLife.04729.004
Detailed workflow for gene expression level estimation.
DOI:http://dx.doi.org/10.7554/eLife.04729.004
10.7554/eLife.04729.005
Elimination of GC bias via quantile normalization.
Each plot shows gene GC content (x-axis) vs the log of the ratio of the individual's RPKM for that gene to mean RPKM across all individuals. Data for three individuals are shown in pairs (A and B, C and D, E and F) for prior to (left) and after (right) quantile normalization.
DOI:http://dx.doi.org/10.7554/eLife.04729.005
10.7554/eLife.04729.006
Detailed workflow for SNP genotyping.
DOI:http://dx.doi.org/10.7554/eLife.04729.006
10.7554/eLife.04729.007
Location of analyzed SNPs relative to genes.
The locations of all SNPs tested in the eQTL analysis are shown in gold relative to the 5′ most gene transcription start site (TSS) and the 3′ most gene transcription end site (TES) for all 10,409 genes. The location of all SNPs tested in association with eQTL genes is overplotted in blue. Gray shaded rectangle denotes the region bounded by the TSS and TES, with gene lengths divided into 20 bins for visibility.
DOI:http://dx.doi.org/10.7554/eLife.04729.007
10.7554/eLife.04729.008
Accuracy of genotype calls for SNPs independently typed in HapMap3.
(A) Distribution of correlations between SNPs called using RNA-seq data and SNPs called independently by HapMap3 (n = 9919 variants). (B) Estimated homozygosity levels for n = 69 YRI individuals at the same set of sites; outliers (denoted with red stars) reflect those individuals with the lowest correlation between RNA-seq-based genotypes and HapMap3 genotypes. The four starred outliers in (B) include the three lowest accuracy individuals in the boxplots in (A).
DOI:http://dx.doi.org/10.7554/eLife.04729.008
10.7554/eLife.04729.009
PCA projection of YRI samples using the RNA-seq-based pipeline vs independently typed SNPs.
PCA projection of genotype data from the RNA-seq-based pipeline and the HapMap3 data place individual samples very close together. (A) and (B) show the same data, but (B) zooms in on the central cluster for better visibility.
DOI:http://dx.doi.org/10.7554/eLife.04729.009
10.7554/eLife.04729.010
Agreement between eQTL and ASE approaches for identifying functional variants.
(A) Venn diagram depicting the overlap between genes with significant eQTL and ASE, among genes tested in both cases (note that the number of genes with eQTL is smaller in this figure than in the overall data set because we consider only the set of genes that were testable for both eQTL and ASE, n = 2280 instead of n = 10,409). Genes with significant eQTL are more likely to have significantly detectable ASE and vice-versa (n = 2280; p < 10−25). (B) eQTL SNPs in exonic regions that could also be tested for ASE reveal correlated effect sizes (n = 123; p < 10−20). (C) Similarly, ASE SNPs exhibit effect sizes that are correlated with evidence for eQTL at the same sites (n = 510; p < 10−45).
DOI:http://dx.doi.org/10.7554/eLife.04729.010
10.7554/eLife.04729.011
Power to detect ASE vs eQTL.
(A) Detection of ASE is favored for genes with higher expression levels (p = 3.99 × 10−209), (B) whereas detection of eQTL is favored for genes with greater cis-regulatory SNP density (p = 1.05 × 10−73).
DOI:http://dx.doi.org/10.7554/eLife.04729.011
10.7554/eLife.04729.012
Characteristics of YRI eQTL identified in the RNA-seq vs conventional pipelines.
Boxplot differences between eQTL identified in the YRI data set using chip-based genotype data vs RNA-seq-based genotype data for (A) gene expression levels in RPKM (Wilcoxon test p = 6.53 × 10−9); (B) conservation levels measured by average phyloP per gene (p = 0.707); (C) conservation levels measured using Homologene conservation scores (p = 0.600); and (D) magnitude of the eQTL effect size (p = 0.137).
DOI:http://dx.doi.org/10.7554/eLife.04729.012
10.7554/eLife.04729.013
Differences in the magnitude of ASE vs distance between sites.
(A) Difference in the magnitude of ASE estimated for pairs of tested sites (i.e., absolute difference of the absolute values of z-scores), by distance between sites. (B) Difference in the magnitude of ASE estimated for pairs of tested sites for genes with significant ASE only, where one site in the pair is the site with the best ASE support for the gene. In both plots, distance categories reflect the range from the previous category to the labeled max value.
DOI:http://dx.doi.org/10.7554/eLife.04729.013
10.7554/eLife.04729.014
Location of eQTL SNPs relative to genes with and without controlling for local structure.
The locations of all eQTL SNPs (n = 1787) identified in the main eQTL analysis are shown in gold relative to the 5′ most gene transcription start site (TSS) and the 3′ most gene transcription end site (TES). eQTL SNPs detected in a parallel analysis controlling for local structure (n = 1583) are overplotted in blue. Gray shaded rectangle denotes the region bounded by the TSS and TES, with gene lengths divided into 20 bins for visibility. Note that SNPs that fall outside of one focal gene may fall within the boundaries of other genes. Inset: Quantile–quantile plot of eQTL locations in models that do and do not control for local structure (Kolmogorov-Smirnov test, p = 0.577).
DOI:http://dx.doi.org/10.7554/eLife.04729.014
10.7554/eLife.04729.015
Number of eQTL identified by PCs removed from the gene expression data set.
DOI:http://dx.doi.org/10.7554/eLife.04729.015
10.7554/eLife.04729.016
Coverage by genotype call.
Mean coverage by genotype class for (A) all SNPs tested in the baboon eQTL analysis (n = 64,432), and (B) SNPs identified as eQTL (n = 1693). QQ plot of mean coverage in homozygotes for the reference allele vs homozygotes for the alternate allele for (C) all SNPs and (D) SNPs identified as eQTL. The magnitude of increased coverage in reference allele homozygotes indicates the degree of systematic reference allele mapping bias (dashed line shows the expectation for no mapping bias). Reference allele homozygotes tend to have higher coverage, on average, than alternate allele homozygotes (K-S test: p < 2.2 × 10−16 for all SNPs; p = 3.9 × 10−5 for eQTL SNPs), suggesting some degree of mapping bias; however the effect is actually smaller for eQTL SNPs than for all SNPs (K-S D = 0.167 for all SNPs; K-S D = 0.084 for eQTL SNPs).
DOI:http://dx.doi.org/10.7554/eLife.04729.016
10.7554/eLife.04729.017
Detection of ASE is not dependent on number of heterozygotes, conditional on total read depth.
SNPs within the set tested for ASE (n = 8145) were divided into deciles based on total read depth. The evidence for a relationship (−log10 of the p-value from a Wilcoxon test) between number of heterozygous individuals at each site and detection of significant ASE is shown on the y-axis for each decile. Dashed line shows a nominal significance threshold of p = 0.01. Blue numbers above each point show the number of sites that fall within the decile; purple numbers below each point show the maximum total read depth for that decile (minimum total read depth is the maximum depth for the previous decile, or 300 for the lowest decile).
DOI:http://dx.doi.org/10.7554/eLife.04729.017
Both analyses converged to reveal extensive segregating genetic variation affecting gene expression levels in the Amboseli population. At a 10% false discovery rate, we identified eQTL for 1787 (17.2%) of the genes we analyzed, and evidence for ASE for 510 (23.4%) of tested genes. Consistent with reports in humans (e.g., Veyrieras et al., 2008; Pickrell et al., 2010a), eQTL were strongly enriched near gene transcription start sites and in gene bodies (Figure 1; controlling for the background distribution of sites tested, which were also enriched in and around genes). Within gene bodies, eQTL were particularly likely to be detected near transcription end sites; this potentially reflects enrichment in 3′ untranslated regions, which are poorly annotated in baboon. Also as expected, genes with eQTL were more likely to exhibit significant ASE and vice-versa (hypergeometric test: p < 10−25; Figure 1—figure supplement 7). The magnitude and direction of ASE and eQTL were significantly correlated when an eQTL SNP could also be assessed for ASE (n = 123 genes; r = 0.719, p < 10−20, Figure 1—figure supplement 7), and when ASE SNPs were assessed as eQTL (n = 510 genes; r = 0.575, p < 10−45, Figure 1—figure supplement 7). Detection of ASE was most strongly favored for highly expressed genes (i.e., higher RPKM: Wilcoxon test: p < 10−208; Figure 1—figure supplement 8), whereas detection of eQTL was most strongly favored for genes with high local SNP density (p < 10−72; Figure 1—figure supplement 8).
Increased power to detect eQTL in baboons relative to humans
The number and effect sizes of the eQTL we detected indicate that our power to detect eQTL in the Amboseli population was surprisingly high, especially given that our genotyping data set was limited only to those sites represented in RNA-seq data (i.e., primarily within transcribed regions of moderately to highly expressed genes). Further, while thousands of cis-eQTL have been mapped in single human populations, doing so has generally required sample sizes several fold larger than ours (Lappalainen et al., 2013; Battle et al., 2014).
To provide a more informative estimate of the difference in power to detect eQTL in baboons relative to humans, we applied the same mapping, data processing, variant calling, and eQTL modeling pipeline to a similarly sized RNA-seq data set on 69 Yoruba (YRI) HapMap samples, in which samples were sequenced to a similar depth (Pickrell et al., 2010a). Using our approach for estimating and modeling the gene expression data, but obtaining the genotype data from an independent array platform, we could identify 700 genes with significant eQTL in the YRI data set at a 10% FDR. Approximately half (51%) could be recovered if we only focused on SNPs in transcribed regions. This number (n = 357) therefore reflects the likely theoretical limit of detection for performing eQTL mapping in which SNPs are called based on RNA-seq data. Indeed, when eQTL mapping for the YRI was conducted using genotype data obtained from RNA-seq reads (i.e., the same pipeline used for the baboons), we identified 290 genes with eQTL (41.4% of those identified using independently collected genotype data). eQTL identified in the RNA-seq pipeline do not differ from those identified only in the conventional pipeline in either effect size or in surrounding sequence conservation, but do tend to fall in more highly expressed genes (Wilcoxon test on RPKM values: p = 6.53 × 10−9; Figure 1—figure supplement 9), suggesting that sequencing coverage considerations reduce the number of identifiable eQTL below the theoretical maximum. The RNA-seq-based pipeline therefore reduces the number of genes with detectable eQTL by 50–60%, suggesting that if genotyping array data had been available for the baboons, we might have identified eQTL for ∼3500–4000 genes, comparable to results from human data sets with more than 350 samples (Lappalainen et al., 2013). To better understand the reasons behind this difference, we investigated three possible explanations.
Shifts in the minor allele frequency spectrum
We observed that the minor allele frequency (MAF) spectrum of variants called in the baboon data set included proportionally more intermediate frequency variants and proportionally fewer low frequency variants than in the human data set (Figure 2A, inset). To investigate the degree to which this shift conferred greater power to detect eQTL in the baboons, we simulated eQTL for 10% of the genes in the study by randomly choosing a SNP near each of these genes.10.7554/eLife.04729.018
Power to detect eQTL in the Amboseli baboons compared to the HapMap YRI population.
(A) Simulated eQTL data sets demonstrate that the baboon data set has greater power to detect eQTL (at a 10% FDR threshold) when eQTL are simulated based on effect size (solid lines and triangles) but not when eQTL are simulated based on proportion of variance in gene expression levels explained (PVE: dashed lines and circles). This result likely stems from differences in the minor allele frequency (MAF) spectrum between baboons and YRI (inset), which favors eQTL mapping in the baboons; simulations based on effect size are sensitive to MAF, but simulations based on PVE are not. (B) Masking the simulated eQTL SNP demonstrates that the baboon data set has greater power to detect eQTL due to both increased cis-regulatory SNP density and more extended LD (inset). Subsampling the SNP density in the baboon data set to the level of the YRI data set reduces the difference in power but does not remove it completely. In B, all results are shown for PVE-based simulations to exclude the effects of the MAF. See Figure 4—figure supplement 1 for power simulations for masked SNPs based on effect size.
DOI:http://dx.doi.org/10.7554/eLife.04729.018
10.7554/eLife.04729.019
Relationship between power to detect eQTL and simulated effect size, when the true eQTL is masked.
Purple line shows the baboon data; pink line shows the baboon data with SNP density subsampled to match the YRI; orange line shows the YRI data. Masking the simulated eQTL SNP demonstrates that the baboon data set has greater power to detect eQTL due to both increased cis-regulatory SNP density and more extended LD. Subsampling the SNP density in baboon to the level of the YRI data set reduces the difference in power but does not remove it completely.
DOI:http://dx.doi.org/10.7554/eLife.04729.019
We did so in two ways. First, we simulated the effect size of the eQTL, with possible effect sizes ranging from 0.25 to 2.5, in intervals of 0.25 (effect sizes are relative to a standard normal distribution). The power to detect an eQTL of a given effect size is contingent on the relative representation of different genotype classes in a population, and hence MAF (larger MAFs produce a more balanced set of alternative genotypes, and thus more power). Second, we simulated the proportion of variance in gene expression levels (PVE) explained by the eQTL, with possible PVE values ranging from 5% to 50%, in intervals of 5%. In this case, power to detect an eQTL does not depend on MAF because simulating the PVE directly integrates across the combined impact of effect size and MAF (a simulated high PVE eQTL with low MAF implies a large effect size variant). Thus, the impact of the MAF spectrum on the power to detect eQTL is reflected in the differences in power between the baboon data set and the YRI data set in the effect size-based vs the PVE-based simulations. In all cases, we calculated power as the proportion of genes with simulated eQTL recovered at a 10% FDR.
In PVE-based simulations, power to detect simulated eQTL was greater in the YRI data set (Figure 2A, dashed orange line vs dashed purple line), although this advantage disappeared when the YRI data set was subsampled to the same size as the baboon data set (Figure 2A, dashed gray line). However, the baboon data set provided more power to detect eQTL than the YRI data set (whether subsampled or not) when simulations were based on effect size, where power scales with MAF (Figure 2A, solid lines). Based on these differences, we estimate that the power to identify an eQTL of effect size equal to the mean estimated beta in baboons (0.96), is increased in the Amboseli baboons by approximately 1.34-fold (Figure 2A, solid purple line vs solid orange line) as a function of differences in the MAF spectrum alone.
Differences in genetic diversity and linkage disequilibrium
Because our RNA-seq-based approach does not identify variants outside of transcribed regions, causal SNPs were probably often not typed. To quantify the power to detect eQTL under this scenario, we again simulated eQTL among genes in the baboon and YRI data sets, but masked the causal sites. Doing so revealed much greater power to identify eQTL in baboons than in humans, across all values of simulated PVE or effect size (Figure 2B; Figure 2—figure supplement 1). One possible explanation for this observation stems from increased genetic diversity in the baboons compared to the YRI. Indeed, in baboons we tested an average of 45.4 (±57.0 s.d.) genetic variants for each gene, whereas applying the same pipeline in YRI yielded an average of 20.3 (±21.4 s.d.) testable variants per gene. An alternative explanation relates to patterns of LD, which we estimate to decay somewhat more slowly in the baboons (Figure 2B, inset). Higher SNP density in baboons increases the likelihood that, when a causal SNP is not typed, a nearby SNP will be available that tags it. Longer range LD suggests that a given SNP could also tag distant causal variants more effectively.
To assess the contributions of SNP density and LD, we refined our simulations by first thinning the SNP density in the baboons to match SNP density in the YRI, and again masking the simulated causal eQTL. As expected, reducing genetic diversity in the baboons reduced the power to detect genes with a true eQTL (Figure 2B, purple dashed line vs pink dashed line). However, it did not completely account for the difference between the human population and the baboon population, suggesting that LD patterns probably contribute to higher eQTL mapping power in baboons as well as SNP density. Specifically, for an eQTL that explains 28% of the variance in gene expression levels (the mean PVE detected in baboons for genes with significant eQTL), we estimate that SNP density and LD effects increase power by 1.21-fold (Figure 2B, purple dashed line vs pink dashed line) and 1.43-fold (Figure 2B, pink dashed line vs orange dashed line), respectively, when causal SNPs are not typed.
Together, our simulations suggest that the MAF spectrum, genetic diversity, and LD patterns increase the number of genes with detectable eQTL in baboons vs the YRI by 2.35-fold overall (1.34× from the MAF, 1.21× from SNP density effects, and 1.43× from LD effects). Further, considering that the effect size estimates in baboons tended to be larger than in the YRI (mean of 0.96 in baboons vs mean of 0.80 in YRI), the actual fold increase estimated from simulations is approximately 6-fold (Figure 4—figure supplement 1: ratio of purple vs orange lines at these effect sizes). This estimate is remarkably consistent with empirical results from our comparison of the real baboon and YRI data, in which we identified 6.16-fold the number of eQTL in the baboons. One possibility is that this difference arises from a history of known admixture in Amboseli between the dominant yellow baboon population and immigrant anubis baboons (Papio anubis: Alberts and Altmann, 2001; Tung et al., 2008). Thus, it might reflect the difference between an admixed population and an unadmixed population rather than a difference between species. However, this explanation seems unlikely because evidence for ASE does not extend further from tested genes in baboons compared to YRI (Figure 1—figure supplement 10), and because adding controls for local (chromosome-specific) structure when testing for eQTL still results in a large excess of eQTL detected in the baboon data set (∼7× higher than in YRI: ‘Materials and methods’ and Figure 1—figure supplement 11)
Mixed evidence for natural selection on gene expression levels
Interestingly, we found that genes harboring eQTL in baboons were also more likely to have detectable eQTL in the YRI (hypergeometric test, p = 2.39 × 10−7). Given the sample size limitations of the data sets we considered, this overlap suggests that large effect eQTL tend to be nonrandomly concentrated in specific gene orthologues. This pattern could arise if the regulation of some genes has been selectively constrained over long periods of evolutionary time, whereas others have been more permissible to genetic perturbation. Indeed, we found that the mean per-gene phyloP score calculated based on a 46-way primate comparison was significantly reduced (reflecting less conservation) for genes with detectable eQTL in both species, and greatest for genes in which eQTL were not detected in either case (p < 10−53; Figure 3A). We obtained similar results using phyloP scores based on a 100-way vertebrate comparison (p < 10−21; Figure 3—figure supplement 1).10.7554/eLife.04729.020
Mixed evidence for negative selection on variants affecting gene expression level.
(A) Genes that harbor detectable eQTL in baboons, the YRI, or both are more likely to be conserved across long stretches of evolutionary time, based on mean phyloP scores in a 46-way primate genome comparison (n = 7268; p < 10−53). (B) These genes are also more likely to be lineage-specific, based on Homologene annotations (n = 7065; p = 1.78 × 10−8). (C) Although we detect a strong negative correlation between eQTL effect size and eQTL minor allele frequency, in support of pervasive selection against alleles with large effects on gene expression levels, this correlation also appears when simulating constant eQTL effect sizes, suggesting winner's curse effects. See Figure 3—figure supplement 1 for phyloP results based on a 100-way vertebrate genome comparison.
DOI:http://dx.doi.org/10.7554/eLife.04729.020
10.7554/eLife.04729.021
Correlation between eQTL detection and mean phyloP scores based on 100-way vertebrate comparison.
Genes with eQTL in both data set or one data set are less conserved across vertebrates than genes for which no eQTL were detected (n = 7,268, p < 10−19).
DOI:http://dx.doi.org/10.7554/eLife.04729.021
eQTL were more likely to be identified for genes with higher genetic diversity (Figure 1—figure supplement 8), which may account for the relationship between phyloP score and eQTL across species: highly conserved genes are less likely to contain many variable sites. More conserved genes also tend to have slightly lower average minor allele frequencies (p = 0.002), which might reduce the power to detect eQTL (although the effect size is small: r2 = 0.001). However, genes with eQTL in both species were also less likely to have orthologues in deeply diverged species, based on conservation in Homologene (β = −0.036, p = 1.78 × 10−8; Figure 3B). Genetic diversity within the baboons is very weakly correlated with Homologene conservation (r2 = 0.004) and uncorrelated with average minor allele frequency (p = 0.38). Thus, sequence-level conservation scores and depth of homology across species combine to suggest that eQTL—or at least those with relatively large effect sizes—are least likely to be detected for strongly conserved loci, and most likely to be detected for lineage-specific, rapidly evolving genes. Consistent with this idea, genes involved in basic cellular metabolic processes were under-enriched among the set of genes with eQTL in both species, and enriched among the set of genes for which no eQTL were detected in either species (Supplementary file 1B–C). The set of genes with eQTL in either or both species, on the other hand, were enriched for loci involved in antigen processing, catalytic activity, and interaction with the extracellular environment (e.g., receptors, membrane-associated proteins).
Widespread selective constraint on gene expression levels has been suggested in previous eQTL analyses in humans, with evidence supplied by a strong negative correlation between minor allele frequency and eQTL effect size (Battle et al., 2014). This pattern could arise if selection acts against large genetic perturbations, such that variants of large effect would be present only at low frequencies. Consistent with this idea, plotting eQTL effect size vs MAF in the baboons results in a very strong, highly significant negative correlation (r = −0.723, p < 10−280; Figure 3C), with no large effect eQTL detected at higher MAFs. However, such a relationship could also be a consequence of the so-called winner's curse (in which sampling variance leads to upwardly biased effect size estimates: Zöllner and Pritchard, 2007) because the degree of bias in effect size estimation is itself negatively correlated with MAF. Indeed, when we simulated sets of eQTL with constant small effect sizes (β = 0.75, close to the mean effect size detected for SNPs with MAF ≥0.4), we found that the relationship between estimated effect size and MAF among detected eQTL almost perfectly recapitulated the observed negative correlation. Hence, the correlation between estimated eQTL effect size and MAF in the baboons does not provide strong support for widespread negative selection on gene expression phenotypes within species. We note, however, that our sample size of individuals is much smaller than that used for a similar analysis in humans (Battle et al., 2014: n = 922 individuals), and larger sample sizes should attenuate winner's curse effects.
Genetic and environmental contributions to gene expression variation in wild baboons
Finally, we took advantage of our data set to generate the first estimates of genetic, demographic (age and sex), and environmental contributions to gene expression variation in wild nonhuman primates (Supplementary file 1D). While our limited sample size leads to high variance around estimates for any individual gene, the median estimates across genes should be unbiased (Zhou et al., 2013), so we concentrated on these overarching patterns. We focused specifically on three social environmental variables of known importance in this population, all of which have been extensively investigated as models for human social environments. These were: (i) early life social status, which predicts growth and maturation rates (Altmann and Alberts, 2005; Charpentier et al., 2008); (ii) maternal social connectedness to other females, which predicts both adult lifespan and the survival of a female's infants (Silk et al., 2003, 2009, 2010; Archie et al., 2014); and (iii) maternal social connectedness to males, based on recent evidence that heterosexual relationships have strong effects on survival as well (Archie et al., 2014).
Overall, we found that genetic effects on gene expression levels tended to be far more pervasive than demographic and environmental effects. Specifically, the median additive genetic PVE was 28.4%, similar to, or slightly greater than, estimates from human populations (Monks et al., 2004; McRae et al., 2007; Emilsson et al., 2008; Price et al., 2011; Wright et al., 2014). We applied a Bayesian sparse linear mixed model (BSLMM: Zhou et al., 2013) to further partition this additive genetic PVE into two components: a component attributable to cis-SNPs (here, all SNPs within 200 kb of a gene) and a component attributable to trans-SNPs (all other sites in the genome). Again similar to humans (Price et al., 2011; Wright et al., 2014), we found that more of the additive genetic PVE is explained by the trans component (median PVE = 23.8%) than the cis component (median PVE = 2.9%) (Figure 4). Unsurprisingly, we estimated a larger cis-acting component for genes in which functional cis-regulatory variation was detected in our previous analysis (median PVE = 10.2% among eQTL genes and median PVE = 5.0% among ASE genes).10.7554/eLife.04729.022
Genetic contributions to variance in gene expression levels in wild baboons.
Proportion of variance in gene expression levels estimated for all genes, genes without detectable eQTL, and genes with detectable eQTL. Additive genetic effects on gene expression variation, especially cis-acting effects, are larger for eQTL genes than for other genes. See Figure 4—figure supplements 1–3 for related results on percent variance explained by genetic, environmental, and demographic variables and results using an alternative set of SNPs for estimating ptrans.
DOI:http://dx.doi.org/10.7554/eLife.04729.022
10.7554/eLife.04729.023
PVE explained by demographic and early environmental variables.
QQ plots of PVE explained by a variable of interest vs PVE explained by that variable with permuted data, for (A) age; and (B) maternal social connectedness to males (SCI-M). Bottom panels show the difference between evidence for significant PVE by sex for (C) genes on autosomes vs (D) genes on the X chromosome (bottom right).
DOI:http://dx.doi.org/10.7554/eLife.04729.023
10.7554/eLife.04729.024
Distribution of PVE explained by additive genetic variance, age, sex, and maternal social connectedness to males across all genes.
DOI:http://dx.doi.org/10.7554/eLife.04729.024
10.7554/eLife.04729.025
Genetic contributions to variance in gene expression levels, with p<sub>trans</sub> based on SNPs on other chromosomes only.
DOI:http://dx.doi.org/10.7554/eLife.04729.025
In contrast to the substantial genetic effects we detected, the median PVE explained by age and sex were 1.89% and 0.82%, respectively (Figure 4—figure supplements 1–2). The distribution of PVE explained by age was significantly greater than expected by chance (Kolmogorov–Smirnov test on binned PVEs, in comparison to permuted data: p < 10−11), whereas that explained by sex was not (p = 0.100); large sex effects tended to be constrained to a small set of genes on the X chromosome (Figure 4—figure supplement 1). Of the early environmental variables we investigated, only maternal social connectedness to males explained more variance in gene expression levels than expected by chance (p = 4.19 × 10−3), with a median PVE of 1.9%. Notably, while social connectedness to males (i.e., heterosexual bonds) and social connectedness to females (i.e., same-sex bonds) are both known predictors of longevity in the Amboseli baboons, previous analyses suggest that their effects are largely independent (Archie et al., 2014). Our result extends this observation to the early life effects of maternal social connectedness on variance in gene expression levels.
Taken together, our data suggest that while almost all genes are influenced by genetic variation, the effects of demographic and environmental parameters are generally modest for any single aspect of the environment. However, in at least some cases, we find evidence that early environmental effects on gene expression levels appear to persist across the life course, as has previously been demonstrated in laboratory settings and in response to severe early adversity in humans (e.g., Weaver et al., 2004; Miller et al., 2009; Cole et al., 2012).
Discussion
Much of what we know about genetic contributions to variation in gene expression levels in primates (and vertebrates more generally) come from the extensive body of research on humans. However, increasing evidence indicates that humans are demographically unusual: compared to other primates, humans exhibit low levels of neutral genetic diversity and a low long-term effective population size (Chen and Li, 2001; Hernandez et al., 2007; Perry et al., 2012). Further, humans are distinguished from other primates by recent explosive population growth (Keinan and Clark, 2012; Tennessen et al., 2012). While late Pleistocene population expansion has been suggested for some nonhuman primates, including chimpanzees and Chinese-origin rhesus macaques (Hernandez et al., 2007; Wegmann and Excoffier, 2010), none have undergone the extreme levels of population increase that characterized humans. Indeed, evidence from microsatellite data suggests that the long-term effective population size of baboons actually may have contracted during this period (Storz et al., 2002).
These differences are not simply of historical interest, but also important for understanding the genetic architecture of traits measured in the present day. Differences in demographic history not only affect overall levels of genetic variation and the minor allele frequency spectrum, but also the mean effect size of sites that contribute to phenotypic variation (Lohmueller, 2014). Interestingly, demographic history does not impact overall trait heritability (Lohmueller, 2014; Simons et al., 2014), perhaps explaining why we estimated mean additive genetic PVEs for gene expression levels in baboons that are similar to those estimated for humans. However, demographic history can influence the power to detect individual genetic contributions to phenotypic variation. Large-scale population expansion of the type that occurred in human history appears to reduce power to identify genotype-phenotype correlations for fitness-related traits (Lohmueller, 2014). This observation may account, in part, for our ability to identify many more functional regulatory variants in the baboons than we expected based on previous studies in humans.
However, while our analysis extends previous observations that large effect eQTL are nonrandomly distributed, we found mixed evidence for widespread negative selection on gene expression levels. Specifically, within the baboons alone, we found that the negative relationship between eQTL effect size and minor allele frequency was explicable based on winner's curse effects alone. Thus, increased power to identify functional regulatory variants in the baboons is probably not due to pervasive associations between gene expression levels and fitness. In contrast, stronger evidence for selection on gene expression patterns stems from our cross species comparisons. In particular, we observed that genes with eQTL in baboons significantly overlapped with genes with eQTL in humans, and that these genes as a class also tended to be less constrained at the sequence-level (consistent with observations for analyses of cis-eQTL in humans alone: Popadin et al., 2014). This result suggests that genes vary in their tolerance of functional regulatory genetic variation, and, intriguingly, that gene-specific robustness to genetic perturbation may be a conserved property across species.
Because no comparable data are yet available for other large mammal populations, including for other baboons, it is unclear whether our results are typical or instead a consequence of the Amboseli population's own unique history. In particular, the population has experienced recent admixture between yellow baboons, the dominant taxon, and closely related anubis baboons (P. anubis) (Alberts and Altmann, 2001; Tung et al., 2008). Admixture, which appears to be relatively common in natural populations (Mallet, 2005), can have important consequences for genetic diversity and LD patterns. While it appears to have had a modest impact on the relative ability to map gene expression phenotypes in baboons vs the YRI data set, comparison to a non-admixed baboon population could help resolve this question further. More generally, our results encourage further investigation of the relationship between demography and trait genetic architecture in other populations, as has been suggested for humans (Lohmueller, 2014) but could also be profitably extended to nonhuman model systems. Such comparisons would provide an empirical basis for testing predicted relationships between demographic history and the power to identify genotype-phenotype associations. From an applied perspective, they could also help identify animal models that favor more highly powered association mapping studies, a strategy that has already been heavily exploited in domestic dogs (Karlsson et al., 2007; Karlsson and Lindblad-Toh, 2008) and suggested for rhesus macaques (Hernandez et al., 2007). While the same sites will probably rarely be associated with the same traits across species, this strategy could help identify molecular mechanisms that are conserved across humans and animal models (e.g., Lamason et al., 2005). Comparisons that use matched sample types will be particularly informative: our study compared eQTL detection in whole blood (from the baboons) with eQTL from lymphoblastoid cell lines (YRI), which exhibit highly correlated, but not identical, patterns of overall gene expression (Spearman's rho = 0.645, p < 1 × 10−16)—these differences could affect rates of eQTL detection as well.
Finally, our data—the first profile of genome-wide gene expression levels in a wild primate population—serve as a useful proof of principle of the ability to concurrently generate genome-wide gene expression phenotype and genotype data, and to relate them to each other using eQTL and ASE approaches. Intensively studied natural primate populations—some of which have been studied continuously for 30 or more years—have emerged as important phenotypic models for human behavior, health, and aging. The approach we used here provides a way to leverage these models for complementary genetic studies as well, especially if eQTL prove to be strongly enriched for sites associated with other traits, as in humans (Nicolae et al., 2010). Although preliminary, our results highlight the increasing feasibility of integrating functional genomic data with phenotypic data on known individuals in the wild. For example, our data set revealed a number of genes in which variation in gene expression levels could be mapped to an identifiable eQTL, validated using an ASE approach, and also linked to early life environmental variation. Such cases suggest the potential for future investigations of the molecular basis of persistent environmental effects, including whether genetic and environmental effects act additively or interact.
Materials and methodsStudy subjects and blood sample collection
Study subjects were 63 individually recognized adult members (26 females and 37 males) of the Amboseli baboon population. All study subjects were recognized on sight by observers based on unique physical characteristics. To obtain blood samples for RNA-seq analysis, each baboon was anesthetized with a Telazol-loaded dart using a handheld blowpipe. Study subjects were darted opportunistically between 2009 and 2011, avoiding females with dependent infants and pregnant females beyond the first trimester of pregnancy (female reproductive status is closely monitored in this population, and conception dates can be estimated with a high degree of accuracy). Following anesthetization, animals were quickly transferred to a processing site distant from the rest of the group. Blood samples for RNA-seq analysis were collected by drawing 2.5 ml of whole blood into PaxGene Vacutainer tubes (Qiagen, Valencia, CA), which contain a lysis buffer that stabilizes RNA for downstream use. Following sample collection, study subjects were allowed to regain consciousness in a covered holding cage until fully recovered from the effects of the anesthetic. They were then released within view of their social group; all subjects promptly rejoined their respective groups upon release, without incident.
Blood samples were stored at approximately 20°C overnight at the field site. Samples were then shipped to Nairobi the next day for storage at −20°C until transport to the United States and subsequent RNA extraction.
Gene expression profiling using RNA-seq
For each RNA sample (one per individual), we constructed an RNA-seq library suitable for measuring whole genome gene expression using Dynal bead poly-A mRNA purification and a standard Illumina RNA-seq prep protocol. Each library was randomly assigned to one lane of an Illumina Genome Analyzer II instrument and sequenced to a mean depth of 30 million 76-base pair reads (±4.5 million reads s.d., Supplementary file 1A). The resulting reads were mapped to the baboon genome (Panu2.0) using the efficient short-read aligner bwa 0.5.9 (Li and Durbin, 2009), with a seed length of 25 bases, a maximum edit distance of two mismatches in the seed, a read trimming quality score threshold of 20, and the default maximum edit distance (4% after trimming). To recover reads that spanned putative exon–exon junctions, and therefore could not be mapped directly to the genome, we used the program jfinder on reads that did not initially map (Pickrell et al., 2010b). Finally, we filtered the resulting mapped reads data for low quality reads (quality score <10) and for reads that did not map to a unique position in the genome. To assign reads to genes, we used the RefSeq exon annotations for Panu 2.0 (ref_Panu_2.0_top_level.gff3, downloaded September 6, 2012). We considered the total read counts for each gene and individual as the sum of the number of reads for that individual that overlapped the union of all exon base pairs assigned to a given gene. In downstream analyses, we considered only highly expressed genes that had non-zero counts in more than 10% individuals, and that had mean read counts greater than or equal to 10 (excluding the gene for beta-globin).
We then performed quantile normalization across samples followed by quantile normalization for each gene individually, resulting in estimates of gene expression levels for each gene that were distributed following a standard normal distribution. This procedure effectively removed GC bias in gene expression level estimates (Figure 1—figure supplement 2). For eQTL mapping, ASE analysis, and PVE estimation for sex and age we used all 63 individuals. For PVE estimation for maternal rank and social connectedness, missing data meant that we conducted our analysis on n = 52 and n = 47 individuals, respectively.
Variant identification and genotype calls
To identify genetic variants in the baboon data set, we used the Genome Analysis Toolkit (v. 1.2.6; McKenna et al., 2010; DePristo et al., 2011). Because no validated reference set of known genetic variants are available for baboon, we performed an iterative bootstrapping procedure for base quality score recalibration. Specifically, we performed an initial round of base quality score recalibration and identified a set of variants using GATK's UnifiedGenotyper and VariantFiltration walker. From this call set, we constructed a set of high confidence variants with quality score ≥100 that passed all filters for variant confidence (variants failed if QD < 2.0), mapping quality (variants failed if MQ < 35.0), strand bias (variants failed if FS > 60.0), haplotype score (variants failed if HaplotypeScore >13.0), mapping quality (variants failed if MQRankSum < −12.5) and read position bias (variants failed if ReadPosRankSum < −8.0). We used this high confidence set as the set of ‘known sites’ in a second round of base quality score recalibration, repeating this procedure until the number of variants identified in consecutive rounds of recalibration stabilized. In the final call set, we removed all sites (i) that were monomorphic in the Amboseli samples; (ii) for which genotype data were missing for more than 12 individuals (19%) in the data set; (iii) that deviated from Hardy–Weinberg equilibrium; and (iv) that failed the above quality control filters. We further filtered the data set to contain only sites with a minimum quality score of 100 that were located within 200 kb of a gene of interest, and that were sequenced at a mean coverage ≥5× across all samples. We validated our quality control and filtering steps by performing the same procedure on an RNA-seq data set from the HapMap Yoruba population (see below). These steps resulted in a set of 64,432 single nucleotide polymorphisms carried forward into downstream analysis (30,938 for the YRI). For eQTL mapping analysis, missing genotypes in this final set were imputed using BEAGLE (Browning and Browning, 2009).
To estimate genome-wide LD, we followed the approach of Eberle et al. (2006), which uses allele frequency-matched SNPs to calculate pair-wise LD. Specifically, we selected SNPs with MAFs greater than 10% and divided them into four subgroups (MAF between 10%–20%; MAF between 20%–30%; MAF between 30%–40%; and MAF between 40%–50%). We then calculated pair-wise r2 for all SNP pairs within 100 kb in each subgroup using VCFtools (Danecek et al., 2011) and combined values from all four subgroups.
Estimating accuracy of SNP genotypes using human RNA-seq data
To assess the accuracy of the RNA-seq-based genotyping calls we performed in the baboons, we investigated a similarly sized data set of RNA-seq reads from a human population (Pickrell et al., 2010a). Because this data set focused on samples from the HapMap consortium (n = 69 members of the Yoruba population from Ibadan, Nigeria), we were able to compare genotypes called using the RNA-seq pipeline to independently collected genotype data from HapMap Phase 3 (r27) (International HapMap Consortium, 2010). To do so, we focused on 9919 variants that were genotyped in both data sets. We then calculated the correlation between genotypes called in the RNA-seq-based pipeline and genotypes from HapMap, for each individual (Figure 1—figure supplement 5A). We also found that low accuracy was correlated with the level of apparent homozygosity in the genotype data (Figure 1—figure supplement 5B). In the baboon data, we had no individuals with unusually low homozygosity, but six individuals with unusually high homozygosity (>80% of genotype calls). These outliers were missing a median of 10.6% of data in the unimputed genotype data set, whereas all other individuals were missing a median of 0.6% data. However, removing these six individuals from our analysis resulted in very similar results as using the full data set: 87.6% of eQTL genes (n = 1566) identified when using all individuals were also identified with this subset.
Importantly, the available data from humans also support accurate variant discovery. Of the 30,938 sites that we identified from the RNA-seq data and that passed all of our filters, only 3.1% (967) did not have an assigned rsID in dbSNP release 138. These sites were likely enriched for false positives, as the transition/transversion ratio for this set was 1.42, vs 2.80 for the set of 30,938 sites as a whole.
eQTL mapping
To identify cis-acting eQTLs in the baboon data set, we used the linear mixed model approach implemented in the program GEMMA (Zhou and Stephens, 2012). This model provides a computationally efficient method for eQTL mapping while explicitly accounting for genetic non-independence within the sample; in our case, some individuals in the data set are related (although overall relatedness was low: the median kinship coefficient across all pairs was 0.015; mean = 0.024 ± 0.033 s.d.).
For each gene, we considered all variants within 200 kb of the gene as candidate eQTLs. For each variant, we fitted the following linear mixed model:y=μ+xβ+u+ε,u∼MVN(0,σu2K),ε∼MVN(0,σe2I),and tested the null hypothesis H0: β = 0 vs the alternative H1: β ≠ 0. Here, y is the n by 1 vector of gene expression levels for the n individuals in the sample. Gene expression values were first corrected for hidden factors that could act as sources of global structure (e.g., batch effects or ancestry- or environment-related trans effects) by regressing out the first 10 principal components of the gene expression data. Consistent with previous results (e.g., Pickrell et al., 2010a), this procedure greatly improves our ability to detect eQTL (Figure 1—figure supplement 12). In the model, μ is the intercept; x is the n by 1 vector of genotypes for the variant of interest; and β is the variant's effect size. The n by 1 vector of u is a random effects term to control for individual relatedness and other sources of population structure, where the n by n matrix K = XXT/p provides estimates of pairwise relatedness derived from the complete 63 × 64,432 genotype data set X. Residual errors are represented by ε, an n by 1 vector, and MVN denotes the multivariate normal distribution.
We took the variant with the best evidence (i.e., lowest p-value) for association with gene expression levels for each gene, and then calculated corrected gene-wise q-values (with a 10% false discovery rate threshold) via comparison to the same values obtained from permuted data (similar to Barreiro et al., 2012; Pickrell et al., 2010a).
Possible confounds associated with eQTL mapping using RNA-seq data
We evaluated the potential for eQTL mapping based on RNA-seq data to introduce four possible confounds.
First, for genes with large effect cis-eQTLs, reads from heterozygotes at eQTL-linked sites might be biased towards the allele associated with higher gene expression levels. If so, heterozygotes might be mistakenly genotyped as homozygotes for the high expressing allele, resulting in an underrepresentation of heterozygous genotypes relative to neutral expectations. To control for this possibility, we eliminated sites that violated Hardy-Weinberg expectations (n = 2386) from our analyses. We note, however, that this scenario would not introduce false positives. Instead, it would lead to more conservative detection of additive eQTL effects, with the direction of an estimated eQTL effect still consistent with the true effect.
Second, SNP calling might be biased towards the reference allele. If so, more reads would be required to support a genotype call of homozygote alternate than a genotype call of homozygote reference. This bias would result in higher apparent expression levels for alternate allele homozygotes and lower expression levels for reference allele homozygotes, which could create false positive eQTLs. However, we observe no evidence for this scenario in our data set. For all tested SNPs (n = 64,432) and for eQTL SNPs only (n = 1693), alternate allele homozygotes tend to have slightly lower coverage than reference allele homozygotes, and heterozygotes tend to have the highest coverage (because more reads are required to support inference of heterozygosity) (Figure 1—figure supplement 13). Thus, coverage and genotype do not covary additively, and this potential confound is unlikely to produce false positive eQTLs.
Third, read mapping might be biased towards the reference allele, such that reads carrying the alternate allele are less likely to map because they contain more mismatches to the reference genome. This possibility is consistent with our observation that alternate allele homozygotes tend to have slightly less coverage than reference allele homozygotes (Figure 1—figure supplement 13). While this difference in coverage is significant (Kolmogorov-Smirnov test: p < 2.2 × 10−16 for all SNPs; p = 3.9 × 10−5 for eQTL SNPs), the magnitude of the effect itself is modest (Figure 1—figure supplement 13), probably because we allowed reads to map with up to three mismatches: Wittkopp and colleagues have shown that reference allele mapping bias is largely obviated by allowing reads to map with more mismatches (Stevenson et al., 2013). Further, systematic calling of false positive eQTLs due to biased read mapping would predict a bias towards negative effect sizes (i.e., eQTL effects suggesting that the alternate allele is associated with lower expression levels). Our data are not consistent with such a pattern: 47% of eQTL betas are negative, whereas 53% are positive. Reference allele mapping biases are, however, more likely to affect ASE analysis, producing a pattern of greater expression in the reference allele. Indeed, we do observe a bias towards negative betas in the ASE analysis (67.2% of n = 510 genes), although the overall magnitude and direction of ASE data agree well with eQTL evidence.
Fourth, lower mean coverage in homozygotes of either type relative to heterozygotes could induce false positive eQTLs in which the major allele was associated with lower gene expression levels. To test this possibility, we recoded eQTL effects to reflect the effect of the major allele instead of the effect of the alternate allele (i.e., a genotype of 0 = homozygous minor and a genotype of 2 = homozygous major). We observed a modest excess of eQTL for which the major allele was associated with lower gene expression levels (56%, binomial test p = 1.15 × 10−7). This bias did not differ depending on whether the major allele was the reference allele or the alternate allele (Fisher's Exact Test, p = 0.28), supporting minimal read mapping biases in our data. Instead, it appears to be primarily driven by SNPs with low minor allele frequencies (proportion of negative betas for the lowest quartile of MAFs = 62.8%, p = 7.49 × 10−8; highest quartile of MAFs = 48.6%, p = 0.602). At these sites, eQTL inference relies primarily on two genotype classes (the major allele homozygotes and heterozygotes) rather than three genotype classes. Because heterozygotes tend to have slightly higher coverage than homozygotes of both classes, spurious relationships between genotype and gene expression levels are much less likely to be observed when both types of homozygotes are well represented (i.e., MAFs are larger).
Along with the high genotype accuracy rates estimated from the Yoruba data, our analyses thus indicate that the set of eQTL we identified are largely robust to RNA-seq-specific confounds. The eQTL identified in YRI in the conventional pipeline vs the RNA-seq pipeline offer a further source of comparison. We find that eQTL identified through the RNA-seq pipeline tend to be associated with more highly expressed genes (providing greater power to call genotypes: Wilcoxon test p = 6.53−9), but otherwise do not differ in sequence conservation (phyloP scores: p = 0.707; Homologene scores: p = 0.603) or in estimated effect size (p = 0.137) (Figure 1—figure supplement 9). Further, effect size magnitude is highly correlated across pipelines when eQTL are discovered in both pipelines (r = 0.874, p < 10−57). When eQTL were discovered only in the RNA-seq pipeline (n = 104), they tended to be high on the ranked list of eQTL evidence in the conventional pipeline as well (median rank of 1395, where the top 700 were significant and 10,615 genes were tested), suggesting that many of them did not pass the threshold for eQTL detection in that analysis. Thus, the most salient source of error stems from low MAF sites, which are also the cases most vulnerable to sampling error and winner's curse effects more generally (Figure 3)—a problem that is not confined to RNA-seq-based eQTL mapping. Taken together, these analyses argue that, as a general rule, eQTL associated with lower MAF SNPs should be treated with increased caution.
ASE detection
To identify ASE, we focused on SNPs within gene exons with Phred-scaled quality scores greater than 10. We further required that these sites have more than five reads in more than two individuals and more than 300 total reads across all heterozygous individuals. This threshold is based on the observation that the power to detect ASE is dependent on sequencing read coverage at heterozygous sites (Fontanillas et al., 2010). Indeed, in our data set, power to detect ASE appeared to scale primarily with total read coverage rather than number of heterozygous individuals. Sites with more reads tended to have more heterozygotes (r = 0.266, p < 10−100); however, when sites were partitioned by total read depth (in deciles), sites with significant ASE were not more likely to harbor more heterozygotes in any decile (Wilcoxon test comparing number of heterozygotes in significant sites vs background; Figure 1—figure supplement 14).
After these filtering steps, we retained 8154 SNPs associated with 2280 genes for ASE analysis. For ASE analysis, we did not take into account possible recombination between exonic SNPs and the (unknown) cis-regulatory variants whose effects they capture, as we did not have detailed data on recombination rates across the baboon genome. However, recombination between exonic SNPs and the true causal regulatory SNPs would decrease our power to detect ASE.
For each variant, we considered a beta-binomial distribution (following Pickrell et al., 2010a) to model the number of reads from the (+) haplotype (denoted as xi+) or the number of reads from the (−) haplotype (denoted as xi−), conditional on the number of total reads (denoted as yi = xi+ + xi−), for each individual i, orxi+|yi∼binomial(yi,θ),θ∼beta(α,β).
We tested the null hypothesis H0: α = β vs the alternative H1: α ≠ β using a likelihood ratio test. For both the null model and the alternative model, beta distribution parameters (α and β) were estimated via a maximum likelihood approach, using the R function optim. Again, we took the variant with the lowest p-value for each gene, and then calculated corrected gene-wise q-values (using a 10% false discovery rate threshold) via comparison to the same values obtained from an empirical null distribution. To construct the empirical null distribution, we performed the same analysis after substituting the xi+ value for each variant of interest, for each heterozygous individual, with a randomly selected xi+ value from a heterozygous site elsewhere in the genome (contingent on that site having the same number of total reads, yi).
Power simulations
To assess the relative power of eQTL mapping in baboons vs the YRI data set, we randomly selected 10% of the genes in each data set to harbor eQTL. For each of these simulated eQTL genes, we then randomly chose a SNP among all the cis-SNPs tested (i.e., all variable sites that passed quality control filters and fell within 200 kb of a gene of interest) and assigned it as a causal eQTL. The impact of the eQTL was simulated using either effect size, in which we simulated a constant effect size between 0.25 and 2.5 (in intervals of 0.25) or PVE, in which we chose an effect size that explained a specific proportion of variance in gene expression levels (from 5% to 50%, in intervals of 5%). We then simulated gene expression levels by adding the effect of the simulated cis-eQTL SNP to residual errors drawn from a standard normal distribution. To calculate the FDR, we also simulated a set of genes with no eQTL. For each combination of effect sizes and population (baboon or YRI), and for each simulation scenario (e.g., with the causal SNP masked or unmasked, with SNP density thinned in the baboons, or using PVE vs a constant effect size), we performed 10 replicates. For each replicate, we calculated the power to detect eQTL as the proportion of simulated eQTL genes recovered at a 10% empirical FDR.
Testing the contribution of admixture to eQTL detection
To investigate whether admixture might drive our power to detect eQTL in the baboon data set, we performed three analyses.
First, we asked whether evidence for ASE remained similar across longer distances (i.e., between sites separated by more base pairs) in the baboons vs in the YRI. Such a pattern might be expected if long-distance, admixture-driven LD explained our other observations. However, the pattern of ASE similarity (the magnitude of the difference between ASE estimates) by distance between sites was highly congruent between the YRI and baboon data sets (Figure 1—figure supplement 10).
Second, we investigated whether adding a control for local structure (i.e., population structure in cis to a gene of interest, and based only on variants located on the same chromosome) asymmetrically reduced evidence for eQTL in the baboon data set relative to the YRI data set. To do so, we regressed out the top two PCs for variants on the same chromosome as the gene of interest from the gene expression data prior to fitting mixed effects models. We found that this approach modestly reduced the number of eQTL discoveries in the baboon data set (n = 1583 from n = 1787, an 11.4% difference). However, this number was still 5.4× larger than the number of eQTL detectable in the YRI, and when we applied the same local structure control to the YRI data, a comparable drop in the number of discoverable eQTL also occurred (n = 216 from n = 290, resulting in a ∼7× fold increase in eQTL in baboons vs YRI).
Third, we compared the spatial distribution of eQTL in baboon between the models with and without local structure controls. We reasoned that if admixture drove most of the signal in the data set, controlling for local structure should shift the location of discovered eQTL closer to the gene of interest, where the strongest cis effects are generally identified. However, the locations of eQTL were very similar under both models (Kolmogorov-Smirnov test, p = 0.577; Figure 1—figure supplement 11).
Evidence for patterns consistent with natural selection on gene expression levels
We investigated the relationship between conservation level and the presence of detectable eQTL in the Amboseli baboons or the YRI using phyloP conservation scores (Pollard et al., 2010) and Homologene conservation of orthology across species. For the former, we extracted the per-site phyloP score from the 46-way primate comparison or 100-way vertebrate comparison on the UCSC Genome Browser for each base contained within the annotated exons (including untranslated regions) used for mapping RNA-seq reads in the YRI. We then calculated the average phyloP score across all exons associated with a given gene. We obtained Homologene scores from the CANDID database (Hutz et al., 2008). In both cases, we used linear models to test for a relationship between conservation level and three categories of genes: those with no detectable eQTL in either the baboons or YRI; those with a detectable eQTL in one of the two species; and those with a detectable eQTL in both species.
To investigate whether the correlation between minor allele frequency and eQTL effect size could be a result of winner's curse effects, we extracted the results from our simulations in which the causal variant was masked and the true effect size was fixed at a small value (beta = 0.75). We then calculated the correlation between the estimated effect size (β) from these simulations against minor allele frequency, for detected eQTL only.
Estimation of genetic contributions to gene expression
We used the Bayesian sparse linear mixed model (BSLMM) approach implemented in the GEMMA software package (Zhou and Stephens, 2012) to estimate the genetic contribution to gene expression variation. Specifically, for each gene, we fit the following model:y=μ+xcisβcis+xtransβtrans+ε,βcis,i∼πN(0,σa2)+(1−π)δ0,βtrans,i∼N(0,σb2),where y is the n by 1 vector of gene expression levels for n individuals; μ is the intercept; xcis is an n by pcis matrix of genotypes for pciscis-SNPs and βcis are the corresponding effect sizes; xtrans is an n by ptrans matrix of genotypes for ptranstrans-SNPs and βtrans are the corresponding effect sizes; and ε is an n by 1 vector of i.i.d. residual errors. We used different priors for cis-acting effects and trans-acting effects to capture different properties for the two components. Specifically, the spike-slab prior on the cis effects βcis captures our prior belief that only a small proportion of local SNPs has cis effects and these effects are relatively large. The normal prior on the trans effects captures our prior knowledge that trans-acting SNPs tend to be relatively difficult to find and have relatively small effects. In addition, because pcis is small and ptrans approximately equals p, the number of total SNPs, we used p instead of ptrans to facilitate computation (i.e., ptrans was based on all genotyped sites used in our analyses, n = 64,432). Results are qualitatively similar if ptrans is calculated based on sites that must act in trans (i.e., sites located on a different chromosome than the chromosome containing the gene of interest: Figure 4—figure supplement 3). We used Markov chain Monte Carlo (MCMC) to fit the model with 1000 burn-in and 10,000 sampling steps. We obtained posterior samples of βcis and βtrans to calculate the PVE attributed by each of the two components, as well as the total additive genetic PVE contributed by both components.
To calculate PVE values for demographic and environmental predictors, we again used the linear mixed model approach implemented in GEMMA to control for additive genetic effects. Sex was known from direct observation of the study subjects. Ages were known to within a few days' error for 52 of the 63 individuals in the data set; six animals had birth dates estimated to be accurate within 1 year, four animals had birth dates estimated to be accurate within 2 years, and one had a birth date estimated to be less accurate than 2 years. Early social status was measured using the proportional dominance rank of the individual's mother, at the time of that individual's conception. Dominance ranks are assigned monthly using ad libitum observations of dyadic agonistic (aggressive or competitive) encounters within social groups (Hausfater, 1974; Alberts et al., 2003). Maternal social connectedness values were defined as the social connectedness of the individual's mother, in the year of that female's life during which the focal individual was born. Social connectedness is calculated on a yearly basis as the frequency with which a female was involved in affiliative interactions, relative to the median for all females in the population at the same time and controlling for observer effort (see Runcie et al., 2013; Archie et al., 2014). Social connectedness is measured for females, but can focus on either female–female relationships (SCI-F) or a female's relationship with adult males (SCI-M), which have independent effects on longevity in this population (Archie et al., 2014). For SCI-F, affiliative interactions included both grooming interactions and close spatial proximity to other females. For SCI-M, only grooming interactions were used.
For each gene, we fit the following model:y=μ+xβ+u+ε,u∼MVN(0,σu2K),ε∼MVN(0,σe2I),where x is the n by 1 vector of values for the demographic or environmental predictor of interest and β is its coefficient. The n by 1 vector of u is a random effects term with K = XXT/p controlling for additive genetic effects. We calculated the PVE estimate as var(xβ)/var(y), where var denotes the sample variance.
Acknowledgements
We thank the Kenya Wildlife Services, Institute of Primate Research, National Museums of Kenya, National Council for Science and Technology, members of the Amboseli-Longido pastoralist communities, Tortilis Camp, and Ker & Downey Safaris for their assistance in Kenya. We also thank J Altmann for general support and access to the Amboseli data set and samples; RS Mututua, S Sayialel, JK Warutere, M Akinyi, T Wango, and V Oudu for invaluable assistance with sample collection; K Michelini for assistance with RNA-seq data generation; S Mukherjee for advice on statistical analysis; and LB Barreiro, S Montgomery, the associate editor, and one anonymous reviewer for comments on an earlier draft of the manuscript. Finally, we thank the Baylor College of Medicine Human Genome Sequencing Center for access to the current version of the baboon genome assembly (Panu 2.0).
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
JT, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
XZ, Conception and design, Analysis and interpretation of data, Drafting or revising the article
YG, Conception and design, Analysis and interpretation of data, Drafting or revising the article
SCA, Acquisition of data, Drafting or revising the article, Contributed unpublished essential data or reagents
MS, Analysis and interpretation of data, Drafting or revising the article
Ethics
Animal experimentation: Samples and data used in this study were collected from wild baboons living in the Amboseli ecosystem of southern Kenya. All behavioral, environmental, and demographic data are gathered as part of noninvasive observational monitoring of known individuals within the study population. This research is conducted under the authority of the Kenya Wildlife Service (KWS), the Kenyan governmental body that oversees wildlife (permit number NCST/5/002/R/777 to SCA and NCST/RCD/12B/012/57 to JT). As the animals are members of a wild population, KWS requires that we do not interfere with injuries to study subjects inflicted by predators, conspecifics, or through other naturally occurring events (e.g., falling out of trees). To collect blood samples, we perform temporary immobilizations using the anesthetic Telazol, delivered via a handheld blowgun. Permission to perform this procedure was granted through KWS, and was performed under the supervision of a KWS-approved Kenyan veterinarian, who monitored anesthetized animals for hypothermia, hyperthermia, and trauma (no such events occurred during our sample collection efforts). Observational and sample collection protocols were approved though IACUC committees at Duke University (current protocol is A028-12-02 to SCA and JT) and the University of Chicago (ACUP 72080 to YG).
Additional files10.7554/eLife.04729.026
Supplementary tables. (A) Read mapping summary. (B) Gene Ontology analysis for genes with no eQTL in baboon or YRI. (C) Gene Ontology analysis for genes with eQTL in either or both baboon and YRI. (D) Demographic and environmental data.
DOI:http://dx.doi.org/10.7554/eLife.04729.026
Major datasets
The following data set was generated:
TungJ, ZhouX, AlbertsSC, StephensM, GiladY, 2015, The Genetic Architecture of Gene Expression Levels in Wild Baboons, GSE63788; Publically available at the NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/).
The following previously published data set was used:
PickrellJK, MarioniJC, PaiAA, DegnerJF, EngelhardtBE, NkadoriE, VeyrierasJB, StephensM, GiladY, PritchardJK, 2010, Understanding mechanisms underlying human gene expression variation with RNA sequencing, GSE19480; Publically available at the NCBI Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/).
References1000 Genomes Project ConsortiumAbecasisGRAltshulerDAutonABrooksLDDurbinRMGibbsRAHurlesMEMcVeanGA2010A map of human genome variation from population-scale sequencingNature4671061107310.1038/nature09534AbzhanovAProtasMGrantBRGrantPRTabinCJ2004Bmp4 and morphological variation of beaks in Darwin's finchesScience3051462146510.1126/science.1098095AlbertsSCAltmannJ2001Immigration and hybridization patterns of yellow and anubis baboons in and around Amboseli, KenyaAmerican Journal of Primatology5313915410.1002/ajp.1AlbertsSCAltmannJ2012The amboseli baboon research project: 40 years of continuity and changeLong-term field studies of primatesSpringerAltmannJAlbertsSC2005Growth rates in a wild primate population: ecological influences and maternal effectsBehavioral Ecology and Sociobiology5749050110.1007/s00265-004-0870-xAlbertsSCWattsHEAltmannJ2003Queuing and queue-jumping: long-term patterns of reproductive skew in male savannah baboons, Papio cynocephalusAnimal Behaviour6582184010.1006/anbe.2003.2106ArchieEATungJClarkMAltmannJAlbertsSC2014Social affiliation matters: both same-sex and opposite-sex relationships predict survival in wild female baboonsProceedings Biological sciences/The Royal Society2812014126110.1098/rspb.2014.1261BabbittCCFedrigoOPfefferleADBoyleAPHorvathJEFureyTSWrayGA2010Both noncoding and protein-coding RNAs contribute to gene expression evolution in the primate brainGenome Biology and Evolution2677910.1093/gbe/evq002BabbittCCTungJWrayGAAlbertsSC2012Changes in gene expression associated with reproductive maturation in wild female baboonsGenome Biology and Evolution410210910.1093/gbe/evr134BarreiroLBMarioniJCBlekhmanRStephensMGiladY2010Functional comparison of innate immune signaling pathways in primatesPLOS Genetics6e100124910.1371/journal.pgen.1001249BarreiroLBTailleuxLPaiAAGicquelBMarioniJCGiladY2012Deciphering the genetic architecture of variation in the immune response to Mycobacterium tuberculosis infectionProceedings of the National Academy of Sciences of USA1091204120910.1073/pnas.1115761109BattleAMostafaviSZhuXPotashJBWeissmanMMMcCormickCHaudenschildCDBeckmanKBShiJMeiRUrbanAEMontgomerySBLevinsonDFKollerD2014Characterizing the genetic basis of transcriptome diversity through RNA-sequencing of 922 individualsGenome Research24142410.1101/gr.155192.113BlekhmanRMarioniJCZumboPStephensMGiladY2010Sex-specific and lineage-specific alternative splicing in primatesGenome Research2018018910.1101/gr.099226.109BlekhmanROshlackAChabotAESmythGKGiladY2008Gene regulation in primates evolves under tissue-specific selection pressuresPLOS Genetics4e100027110.1371/journal.pgen.1000271BoroveckiFLovrecicLZhouJJeongHThenFRosasHDHerschSMHogarthPBouzouBJensenRV2005Genome-wide expression profiling of human blood reveals biomarkers for Huntington's diseaseProceedings of the National Academy of Sciences of USA102110231102810.1073/pnas.0504921102BrawandDSoumillonMNecsuleaAJulienPCsárdiGHarriganPWeierMLiechtiAAximu-PetriAKircherMAlbertFWZellerUKhaitovichPGrütznerFBergmannSNielsenRPääboSKaessmannH2011The evolution of gene expression levels in mammalian organsNature47834334810.1038/nature10532BrowningBLBrowningSR2009A unified approach to genotype imputation and haplotype-phase inference for large data sets of trios and unrelated individualsThe American Journal of Human Genetics8421022310.1016/j.ajhg.2009.01.005CáceresMLachuerJZapalaMARedmondJCKudoLGeschwindDHLockhartDJPreussTMBarlowC2003Elevated gene expression levels distinguish human from non-human primate brainsProceedings of the National Academy of Sciences of USA100130301303510.1073/pnas.2135499100ChanYFMarksMEJonesFCVillarrealGShapiroMDBradySDSouthwickAMAbsherDMGrimwoodJSchmutzJMyersRMPetrovDJónssonBSchluterDBellMAKingsleyDM2010Adaptive evolution of pelvic reduction in sticklebacks by recurrent deletion of a Pitx1 enhancerScience32730230510.1126/science.1182213CharpentierMJTungJAltmannJAlbertsSC2008Age at maturity in wild baboons: genetic, environmental and demographic influencesMolecular Ecology172026204010.1111/j.1365-294X.2008.03724.xChenFCLiWH2001Genomic divergences between humans and other hominoids and the effective population size of the common ancestor of humans and chimpanzeesThe American Journal of Human Genetics6844445610.1086/318206ColeSWContiGArevaloJMRuggieroAMHeckmanJJSuomiSJ2012Transcriptional modulation of the developing immune system by early life social adversityProceedings of the National Academy of Sciences of USA109205782058310.1073/pnas.1218253109DanecekPAutonAAbecasisGAlbersCABanksEDePristoMAHandsakerRELunterGMarthGTSherrySTMcVeanGDurbinR1000 Genomes Project Analysis Group2011The variant call format and VCFtoolsBioinformatics272156215810.1093/bioinformatics/btr330DenverDRMorrisKStreelmanJTKimSKLynchMThomasWK2005The transcriptional consequences of mutation and natural selection in Caenorhabditis elegansNature Genetics3754454810.1038/ng1554DePristoMABanksEPoplinRGarimellaKVMaguireJRHartlCPhilippakisAAdel AngelGRivasMAHannaMMcKennaAFennellTJKernytskyAMSivachenkoAYCibulskisKGabrielSBAltshulerDDalyMJ2011A framework for variation discovery and genotyping using next-generation DNA sequencing dataNature Genetics4349149810.1038/ng.806EberleMARiederMJKruglyakLNickersonDA2006Allele frequency matching between SNPs reveals an excess of linkage disequilibrium in genic regions of the human genomePLOS Genetics2e14210.1371/journal.pgen.0020142EmilssonVThorleifssonGZhangBLeonardsonASZinkFZhuJCarlsonSHelgasonAWaltersGBGunnarsdottirSMouyMSteinthorsdottirVEiriksdottirGHBjornsdottirGReynisdottirIGudbjartssonDHelgadottirAJonasdottirAJonasdottirAStyrkarsdottirUGretarsdottirSMagnussonKPStefanssonHFossdalRKristjanssonKGislasonHGStefanssonTLeifssonBGThorsteinsdottirULambJRGulcherJRReitmanMLKongASchadtEEStefanssonK2008Genetics of gene expression and its effect on diseaseNature45242342810.1038/nature06758EnardWKhaitovichPKloseJZöllnerSHeissigFGiavaliscoPNieselt-StruweKMuchmoreEVarkiARavidRDoxiadisGMBontropREPääboS2002Intra-and interspecific variation in primate gene expression patternsScience29634034310.1126/science.1068996FontanillasPLandryCRWittkoppPJRussCGruberJDNusbaumCHartlDL2010Key considerations for measuring allelic expression on a genomic scale using high‐throughput sequencingMolecular Ecology1921222710.1111/j.1365-294X.2010.04472.xGiladYOshlackARifkinSA2006aNatural selection on gene expressionTRENDS in Genetics2245646110.1016/j.tig.2006.06.002GiladYOshlackASmythGKSpeedTPWhiteKP2006bExpression profiling in primates reveals a rapid evolution of human transcription factorsNature44024224510.1038/nature04559GiladYRifkinSABertonePGersteinMWhiteKP2005Multi-species microarrays reveal the effect of sequence divergence on gene expression profilesGenome Research1567468010.1101/gr.3335705GolubTRSlonimDKTamayoPHuardCGaasenbeekMMesirovJPCollerHLohMLDowningJRCaligiuriMABloomfieldCDLanderES1999Molecular classification of cancer: class discovery and class prediction by gene expression monitoringScience28653153710.1126/science.286.5439.531HammockEAYoungLJ2005Microsatellite instability generates diversity in brain and sociobehavioral traitsScience3081630163410.1126/science.1111427HausfaterG1974Dominance and reproduction in Baboons (Papio cynocephalus)Contributions to Primatology71150HaygoodRFedrigoOHansonBYokoyamaKDWrayGA2007Promoter regions of many neural-and nutrition-related genes have experienced positive selection during human evolutionNature Genetics391140114410.1038/ng2104HernandezRDHubiszMJWheelerDASmithDGFergusonBRogersJNazarethLIndapABourquinTMcPhersonJMuznyDGibbsRNielsenRBustamanteCD2007Demographic histories and patterns of linkage disequilibrium in Chinese and Indian rhesus macaquesScience31624024310.1126/science.1140462HutzJEKrajaATMcLeodHLProvinceMA2008CANDID: a flexible method for prioritizing candidate genes for complex human traitsGenetic Epidemiology3277979010.1002/gepi.20346International Hapmap Consortium2005A haplotype map of the human genomeNature4371299132010.1038/nature04226KaplanHHillKLancasterJHurtadoAM2000A theory of human life history evolution: diet, intelligence, and longevityEvolutionary Anthropology Issues News and Reviews915618510.1002/1520-6505(2000)9:43.0.CO;2-7KarlssonEKBaranowskaIWadeCMHillbertzNHZodyMCAndersonNBiagiTMPattersonNPielbergGRKulbokasEJIIIComstockKEKellerETMesirovJPvon EulerHKämpeOHedhammarALanderESAnderssonGAnderssonLLindblad-TohK2007Efficient mapping of mendelian traits in dogs through genome-wide associationNature Genetics391321132810.1038/ng.2007.10KarlssonEKLindblad-TohK2008Leader of the pack: gene mapping in dogs and other model organismsNature Reviews Genetics971372510.1038/nrg2382KeinanAClarkAG2012Recent explosive human population growth has resulted in an excess of rare genetic variantsScience33674074310.1126/science.1217283KhaitovichPWeissGLachmannMHellmannIEnardWMuetzelBWirknerUAnsorgeWPääboS2004A neutral model of transcriptome evolutionPLOS Biology2E13210.1371/journal.pbio.0020132KingMCWilsonAC1975Evolution at two levels in humans and chimpanzeesScience18810711610.1126/science.1090005LamasonRLMohideenMAMestJRWongACNortonHLArosMCJurynecMJMaoXHumphrevilleVRHumbertJESinhaSMooreJLJagadeeswaranPZhaoWNingGMakalowskaIMcKeiguePMO'donnellDKittlesRParraEJManginiNJGrunwaldDJShriverMDCanfieldVAChengKC2005SLC24A5, a putative cation exchanger, affects pigmentation in zebrafish and humansScience3101782178610.1126/science.1116238LappalainenTSammethMFriedländerMR't HoenPAMonlongJRivasMAGonzàlez-PortaMKurbatovaNGriebelTFerreiraPGBarannMWielandTGregerLvan ItersonMAlmlöfJRibecaPPulyakhinaIEsserDGigerTTikhonovASultanMBertierGMacArthurDGLekMLizanoEBuermansHPPadioleauISchwarzmayrTKarlbergOOngenHKilpinenHBeltranSGutMKahlemKAmstislavskiyVStegleOPirinenMMontgomerySBDonnellyPMcCarthyMIFlicekPStromTMGeuvadis ConsortiumLehrachHSchreiberSSudbrakRCarracedoAAntonarakisSEHäslerRSyvänenACvan OmmenGJBrazmaAMeitingerTRosenstielPGuigóRGutIGEstivillXDermitzakisET2013Transcriptome and genome sequencing uncovers functional variation in humansNature50150651110.1038/nature12531LiHDurbinR2009Fast and accurate short read alignment with Burrows–Wheeler transformBioinformatics251754176010.1093/bioinformatics/btp324LohmuellerKE2014The impact of population demography and selection on the genetic architecture of complex traitsPLOS Genetics10e100437910.1371/journal.pgen.1004379MalletJ2005Hybridization as an invasion of the genomeTrends in Ecology & Evolution2022923710.1016/j.tree.2005.02.010McKennaAHannaMBanksESivachenkoACibulskisKKernytskyAGarimellaKAltshulerDGabrielSDalyMDePristoMA2010The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing dataGenome Research201297130310.1101/gr.107524.110McRaeAFMatigianNAVadlamudiLMulleyJCMowryBMartinNGBerkovicSFHaywardNKVisscherPM2007Replicated effects of sex and genotype on gene expression in human lymphoblastoid cell linesHuman Molecular Genetics1636437310.1093/hmg/ddl456MillerGEChenEFokAKWalkerHLimANichollsEFColeSKoborMS2009Low early-life social class leaves a biological residue manifested by decreased glucocorticoid and increased proinflammatory signalingProceedings of the National Academy of Sciences106147161472110.1073/pnas.0902971106MonksSALeonardsonAZhuHCundiffPPietrusiakPEdwardsSPhillipsJWSachsASchadtEE2004Genetic inheritance of gene expression in human cell linesThe American Journal of Human Genetics751094110510.1086/426461NicolaeDLGamazonEZhangWDuanSDolanMECoxNJ2010Trait-associated SNPs are more likely to be eQTLs: annotation to enhance discovery from GWASPLOS Genetics6e100088810.1371/journal.pgen.1000888PerryGHMelstedPMarioniJCWangYBainerRPickrellJKMicheliniKZehrSYoderADStephensMPritchardJKGiladY2012Comparative RNA sequencing reveals substantial genetic variation in endangered primatesGenome Research2260261010.1101/gr.130468.111PickrellJKMarioniJCPaiAADegnerJFEngelhardtBENkadoriEVeyrierasJBStephensMGiladYPritchardJK2010aUnderstanding mechanisms underlying human gene expression variation with RNA sequencingNature46476877210.1038/nature08872PickrellJKPaiAAGiladYPritchardJK2010bNoisy splicing drives mRNA isoform diversity in human cellsPLOS Genetics6e100123610.1371/journal.pgen.1001236PiskolRRamaswamiGLiJB2013Reliable identification of genomic variants from RNA-seq dataThe American Journal of Human Genetics9364165110.1016/j.ajhg.2013.08.008PollardKSHubiszMJRosenbloomKRSiepelA2010Detection of nonneutral substitution rates on mammalian phylogeniesGenome Research2011012110.1101/gr.097857.109PopadinKYGutierrez-ArcelusMLappalainenTBuilASteinbergJNikolaevSILukowskiSWBazykinGASeplyarskiyVBIoannidisPZdobnovEMDermitzakisETAntonarakisSE2014Gene age predicts the Strength of Purifying selection acting on gene expression variation in humansThe American Journal of Human Genetics9566067410.1016/j.ajhg.2014.11.003PriceALHelgasonAThorleifssonGMccarrollSAKongAStefanssonK2011Single-tissue and cross-tissue heritability of gene expression via identity-by-descent in related or unrelated individualsPLOS Genetics7e100131710.1371/journal.pgen.1001317RifkinSAKimJWhiteKP2003Evolution of gene expression in the Drosophila melanogaster subgroupNature Genetics3313814410.1038/ng1086RuncieDEWiedmannRTArchieEAAltmannJWrayGAAlbertsSCTungJ2013Social environment influences the relationship between genotype and gene expression in wild baboonsPhilosophical Transactions of the Royal Society of London Series B, Biological Sciences3682012034510.1098/rstb.2012.0345SilkJBAlbertsSCAltmannJ2003Social bonds of female baboons enhance infant survivalScience3021231123410.1126/science.1088580SilkJBBeehnerJCBergmanTJCrockfordCEnghALMoscoviceLRWittigRMSeyfarthRMCheneyDL2009The benefits of social capital: close social bonds among female baboons enhance offspring survivalProceedings Biological Sciences/The Royal Society2763099310410.1098/rspb.2009.0681SilkJBBeehnerJCBergmanTJCrockfordCEnghALMoscoviceLRWittigRMSeyfarthRMCheneyDL2010Strong and consistent social bonds enhance the longevity of female baboonsCurrent Biology201359136110.1016/j.cub.2010.05.067SimonsYBTurchinMCPritchardJKSellaG2014The deleterious mutation load is insensitive to recent population historyNature Genetics4622022410.1038/ng.2896StevensonKRCoolonJDWittkoppPJ2013Sources of bias in measures of allele-specific expression derived from RNA-seq data aligned to a single reference genomeBMC Genomics1453610.1186/1471-2164-14-536StorzJFBeaumontMAAlbertsSC2002Genetic evidence for long-term population decline in a savannah-dwelling primate: inferences from a hierarchical bayesian modelMolecular Biology and Evolution191981199010.1093/oxfordjournals.molbev.a004022TennessenJABighamAWO'connorTDFuWKennyEEGravelSMcgeeSDoRLiuXJunGKangHMJordanDLealSMGabrielSRiederMJAbecasisGAltshulerDNickersonDABoerwinkleESunyaevSBustamanteCDBamshadMJAkeyJMBroadGOSeattleGONHLBI Exome Sequencing Project2012Evolution and functional impact of rare coding variation from deep sequencing of human exomesScience337646910.1126/science.1219240TishkoffSAReedFARanciaroAVoightBFBabbittCCSilvermanJSPowellKMortensenHMHirboJBOsmanMIbrahimMOmarSALemaGNyamboTBGhoriJBumpsteadSPritchardJKWrayGADeloukasP2006Convergent adaptation of human lactase persistence in Africa and EuropeNature Genetics39314010.1038/ng1946TungJCharpentierMJEGarfieldDAAltmannJAlbertsSC2008Genetic evidence reveals temporal change in hybridization patterns in a wild baboon populationMolecular Ecology171998201110.1111/j.1365-294X.2008.03723.xTungJPrimusABouleyAJSeversonTFAlbertsSCWrayGA2009Evolution of a malaria resistance gene in wild primatesNature46038839110.1038/nature08149UngarPSTeafordMF2002Human diet: its origin and evolutionGreenwood Publishing GroupVeyrierasJBKudaravalliSKimSYDermitzakisETGiladYStephensMPritchardJK2008High-resolution mapping of expression-QTLs yields insight into human gene regulationPLOS Genetics4e100021410.1371/journal.pgen.1000214WeaverICCervoniNChampagneFAD'alessioACSharmaSSecklJRDymovSSzyfMMeaneyMJ2004Epigenetic programming by maternal behaviorNature Neuroscience784785410.1038/nn1276WegmannDExcoffierL2010Bayesian inference of the demographic history of chimpanzeesMolecular Biology and Evolution271425143510.1093/molbev/msq028WranghamR2009Catching fire: how cooking made us humanBasic BooksWrayGA2007The evolutionary significance of cis-regulatory mutationsNature Reviews Genetics820621610.1038/nrg2063WrightFASullivanPFBrooksAIZouFSunWXiaKMadarVJansenRChungWZhouYHAbdellaouiABatistaSButlerCChenGChenTHD'AmbrosioDGallinsPHaMJHottengaJJHuangSKattenbergMKocharJMiddeldorpCMQuAShabalinATischfieldJToddLTzengJYvan GrootheestGVinkJMWangQWangWWangWWillemsenGSmitJHde GeusEJYinZPenninxBWBoomsmaDI2014Heritability and genomics of gene expression in peripheral bloodNature Genetics4643043710.1038/ng.2951ZhouXCarbonettoPStephensM2013Polygenic modeling with Bayesian sparse linear mixed modelsPLOS Genetics9e100326410.1371/journal.pgen.1003264ZhouXStephensM2012Genome-wide efficient mixed-model analysis for association studiesNature Genetics4482182410.1038/ng.2310ZöllnerSPritchardJK2007Overcoming the winner's curse: estimating penetrance parameters from case-control dataThe American Journal of Human Genetics8060561510.1086/51282110.7554/eLife.04729.027Decision letterDermitzakisEmmanouil TReviewing editorUniversity of Geneva Medical School, Switzerland
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “The Genetic Architecture of Gene Expression Levels in Wild Baboons” for consideration at eLife. Your article has been favorably evaluated by Aviv Regev (Senior editor), a Reviewing editor, and two reviewers, one of whom, Stephen Montgomery, has agreed reveal his identity.
The Reviewing editor and the reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
Please find below a summary of the comments of the reviewers that we request that you address in your revised manuscript:
Figure 1 and 3 should not be main figures. The pipeline should be moved to the supplement. Also the ASE and eQTL effect sizes is a general technical issue and does not relate specifically to the study of the genetics of gene expression in baboons. The text around these sections is long and generally distracting.
There is a comment regarding the lack of an association between effect size and minor allele frequency compared to Battle et al., 2014. However, Battle et al. had 922 individuals compared to 66 individuals here (among several technical differences). We are not sure that these studies can be easily compared to provide a definitive statement regarding negative selection in baboons here. A caveat regarding study differences might suffice.
The authors bring up the possibility of admixture in the baboon population. This is slightly concerning as it could increase the number of hets and has the potential to increase the length of haplotypes (both observations of the study). It further could explain many of the observations of the study. Three possible suggestions to address this are to test whether: (i) one sees an excess of trans-associations on the same chromosome compared to across chromosomes for cis-eQTL in baboons versus YRI, (ii) apply a surrogate variable or hidden factor correction to eQTL analysis on a single chromosome or (iii), our reviewers’ favorite, test if ASE is correlated over a longer distance (more independent genes) than in humans—in this case, since there is no biological basis for why the locations of causal variants should be further away from the genes they regulate between baboons and humans, the decay of the correlation of ASE measured in proximal genes should be similar. We realize the authors have a model to control for individual relatedness and population structure, but this is derived from the entire genetic data set and does not address local patterns of admixture.
The gene expression data was quantile normalized. Why was a hidden factor correction not applied? Typically, these types of corrections dramatically improve eQTL discovery. Our concern is if there is some structure to the data that is both present and correlated in genotype and gene expression space, the number discoveries will be artificially inflated.
For ASE analyses: (i) the authors assume no recombination, this is not stated, (ii) how is beta in theta∼beta (alpha, beta) estimated, and (iii) detection of ASE correlates with expression level (Figure 3D), this is not a surprise, but given their model we are concerned whether this estimate is more extreme, because ASE has different variances in effect size when it is estimated from a few individuals for very highly expressed genes (2 het individuals with 150 reads = 300 total) compared to lots of estimates from intermediately expressed genes (10 het individuals with 30 reads = 300 total). For robustness, the authors should show whether detection of ASE in their study is independent of the number of input individuals once a testable site has been selected using their criterion.
The authors should discuss the possibility that the negative correlation between conservation and probability of eQTL in a gene in baboons at least may be driven by the technical issue that only coding SNPs were tested and therefore conserved genes will tend to have low MAF and therefore low power.
The authors indicate a large component of expression variability is in trans. Is trans defined as on other chromosomes? In particular, the authors should clarify what goes into the ptrans matrix.
10.7554/eLife.04729.028Author response
Figure 1 and 3should not be main figures. The pipeline should be moved to the supplement. Also the ASE and eQTL effect sizes is a general technical issue and does not relate specifically to the study of the genetics of gene expression in baboons. The text around these sections is long and generally distracting.
We have removed the original Figure 1 from the manuscript, as it provided an overview version of the more detailed pipelines provided in the original Figure 1–figure supplements 1-6. These supplements are now attached to the original Figure 2 (now Figure 1), along with other figures that were previously included as Figure 1’s supplements. The plots in former Figure 3 have also been changed to Figure 1 supplements and removed from the main text. In addition, we have streamlined our discussion of the power to detect ASE versus eQTL in the main text, largely removing this entire section.
There is a comment regarding the lack of an association between effect size and minor allele frequency compared toBattle et al., 2014. However, Battle et al. had 922 individuals compared to 66 individuals here (among several technical differences). We are not sure that these studies can be easily compared to provide a definitive statement regarding negative selection in baboons here. A caveat regarding study differences might suffice.
We agree and have added such a caveat to this section of the Results.
The authors bring up the possibility of admixture in the baboon population. This is slightly concerning as it could increase the number of hets and has the potential to increase the length of haplotypes (both observations of the study). It further could explain many of the observations of the study. Three possible suggestions to address this are to test whether: (i) one sees an excess of trans-associations on the same chromosome compared to across chromosomes for cis-eQTL in baboons versus YRI, (ii) apply a surrogate variable or hidden factor correction to eQTL analysis on a single chromosome or (iii), our reviewers’ favorite, test if ASE is correlated over a longer distance (more independent genes) than in humans—in this case, since there is no biological basis for why the locations of causal variants should be further away from the genes they regulate between baboons and humans, the decay of the correlation of ASE measured in proximal genes should be similar. We realize the authors have a model to control for individual relatedness and population structure, but this is derived from the entire genetic data set and does not address local patterns of admixture.
Thanks for these helpful suggestions. We agree that the evolutionary history of this population could have influenced our findings. In the revised manuscript, we investigate this possibility in substantially greater detail (eleventh paragraph in the Results section and under the heading “Testing the contribution of admixture to eQTL detection”, in a new Materials and methods section). Overall, we found little evidence that admixture drove our results. These conclusions are based on three sets of findings:
1) Distance between sites tested for ASE predicts the magnitude of the difference between ASE estimates, as expected; however, this relationship does not differ between baboons and YRI (new Figure 1–figure supplement 10). This analysis, motivated by the reviewers’ third suggestion, suggests that ASE is not correlated over a longer distance in baboons than in humans (note that we could not directly test how correlations between ASE estimates decay with distance because our ASE estimates are site-specific, not individual-specific).
2) Controlling for local structure in addition to global ancestry (suggestion 2) modestly reduces the number of eQTL discoveries in the baboon data set, but not more so than the same procedure in YRI. When controlling for local structure using the top two principal components for variants on the same chromosome, the number of detectable eQTL drops from 1787 to 1583. However, this number is still 5.4-fold larger than the number detected in the YRI. A drop also occurs when running a parallel model in YRI (from 290 to 216), so that the number of eQTL detected in baboons is ∼7-fold larger when controls for local structure are used in both. Thus, more extensive local structure does not appear to explain the increased power in baboons.
3) In support of this idea, the spatial distribution of baboon eQTLs relative to genes is almost identical between the models with and without local structure controls (new Figure 1–figure supplement 11). If admixture drove most of the signal in the data set, we would expect to observe greater enrichment of eQTL within or near genes when adding controls for local structure; we do not.
The gene expression data was quantile normalized. Why was a hidden factor correction not applied? Typically, these types of corrections dramatically improve eQTL discovery. Our concern is if there is some structure to the data that is both present and correlated in genotype and gene expression space, the number discoveries will be artificially inflated.
This comment reflects our mistake in writing the original version of our Methods. We indeed corrected for hidden factors by regressing out the first 10 principal components of the overall gene expression data. As the reviewers note, this process greatly improved our ability to detect eQTL and minimized the possibility that the eQTL we detected reflect global structure in gene expression space (we did the same for the YRI data, so our methods remained comparable). We explain our procedure in the revised Methods (tenth paragraph) and have also added a figure supplement (new Figure 1–figure supplement 12) showing the relationship between the number of PCs we removed and eQTL detection.
For ASE analyses: (i) the authors assume no recombination, this is not stated, (ii) how is beta in theta∼beta (alpha, beta) estimated, and (iii) detection of ASE correlates with expression level (Figure 3D), this is not a surprise, but given their model we are concerned whether this estimate is more extreme, because ASE has different variances in effect size when it is estimated from a few individuals for very highly expressed genes (2 het individuals with 150 reads = 300 total) compared to lots of estimates from intermediately expressed genes (10 het individuals with 30 reads = 300 total). For robustness, the authors should show whether detection of ASE in their study is independent of the number of input individuals once a testable site has been selected using their criterion.
With regards to (i), we did not make any explicit assumptions about recombination in the ASE analysis. Recombination between the exonic SNPs we used to test for ASE and true causal regulatory SNPs would decrease our power to detect ASE; we make this point clear in the revision (in the subsection “ASE detection” of the Materials and methods). For (ii), we used a maximum likelihood approach to estimate both the alpha and beta parameters in the beta component of the beta-binomial model; we have clarified this point in the same subsection of the Materials and methods of the revised manuscript. To address point (iii), we have now tested whether ASE is more likely to be detected for sites with more heterozygous individuals, conditional on total read depth (overall, number of hets and total read depth are correlated: r = 0.266, p < 10-100). We find no evidence for such an effect across the ten deciles of total read depth values we tested (please see the aforementioned subsection and Figure 1–figure supplement 14).
The authors should discuss the possibility that the negative correlation between conservation and probability of eQTL in a gene in baboons at least may be driven by the technical issue that only coding SNPs were tested and therefore conserved genes will tend to have low MAF and therefore low power.
Thanks for pointing out this possible interpretation. We have now calculated the correlation between levels of conservation and average minor allele frequency for each gene, for SNPs used in our analysis. The correlation between a gene’s average phyloP score (from the 46-way primate comparison) and the average MAF of all SNPs tested in association with that gene is nominally significant (p = 0.002) but explains a very small fraction of the variance (r = -0.037). Thus, while more conserved genes do tend to have lower average MAFs (among SNPs tested), this relationship is weak, probably because we filter out sites with very low MAFs and because many sites occur in non-protein coding regions of the transcript or in the transcribed regions of other genes. Further, we observe no significant correlation between a gene’s Homologene score and average minor allele frequency among SNPs tested (p = 0.38). These analyses suggest that the relationship between conservation and eQTL discovery is probably not driven by a relationship between conservation and MAF (at least within the data set we ultimately analyzed). We have discussed this point in the revised manuscript (under the heading “Mixed evidence for nature selectin of gene expression levels”, in the Results section). We also cite a recent paper by Popadin et al. (third paragraph of the Discussion) that also suggests that fewer cis-eQTL are found in older genes, similar to our findings for the Homologene analysis.
The authors indicate a large component of expression variability is in trans. Is trans defined as on other chromosomes? In particular, the authors should clarify what goes into the ptransmatrix.
Because the number of sites that are included in ptrans is almost equal to p (the total number of sites in the genome), we calculated the ptrans matrix based on all SNPs typed. However, in the revision we also have run ptrans based on all other chromosomes except the chromosome containing the focal gene. These results are almost identical to the results obtained when using all SNPs; we include them as a new supplemental figure (please see the subsection “Estimation of genetic contributions to gene expression”, in the Materials and methods, and new Figure 4–figure supplement 3).
Previously, we identified ISRIB as a potent inhibitor of the integrated stress response (ISR) and showed that ISRIB makes cells resistant to the effects of eIF2α phosphorylation and enhances long-term memory in rodents (Sidrauski et al., 2013). Here, we show by genome-wide in vivo ribosome profiling that translation of a restricted subset of mRNAs is induced upon ISR activation. ISRIB substantially reversed the translational effects elicited by phosphorylation of eIF2α and induced no major changes in translation or mRNA levels in unstressed cells. eIF2α phosphorylation-induced stress granule (SG) formation was blocked by ISRIB. Strikingly, ISRIB addition to stressed cells with pre-formed SGs induced their rapid disassembly, liberating mRNAs into the actively translating pool. Restoration of mRNA translation and modulation of SG dynamics may be an effective treatment of neurodegenerative diseases characterized by eIF2α phosphorylation, SG formation, and cognitive loss.
DOI:http://dx.doi.org/10.7554/eLife.05033.001
Author keywordsintegrated stress responseeIF2unfolded protein responseISRIBprotein synthesisribosome profilingResearch organismHumanMousehttp://dx.doi.org/10.13039/100000011Howard Hughes Medical Institute (HHMI)SidrauskiCarmelaWalterPeterhttp://dx.doi.org/10.13039/100005665Kinship FoundationSearle Scholars Program 11-SSP-229IngoliaNicholas Thttp://dx.doi.org/10.13039/100000002National Institutes of Health (NIH)T32GM007231McGeachyAnna Mhttp://dx.doi.org/10.13039/100005843Carnegie Institution of WashingtonIngoliaNicholas TThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementBuilding on previous work which showed that the small molecule ISRIB potently blocks the integrated stress response (Sidrauski et al., 2013), we report on ISRIB's remarkable specificity and fast action in vivo, underscoring its proposed direct effect on translation.Introduction
Diverse cellular conditions activate an integrated stress response (ISR) that rapidly reduces overall protein synthesis while sustaining or enhancing translation of specific transcripts whose products support adaptive stress responses. The ISR is mediated by diverse stress-sensing kinases that converge on a common target, serine 51 in eukaryotic translation initiation factor alpha (eIF2α) eliciting both global and gene-specific translational effects (Harding et al., 2003; Wek et al., 2006). Mammalian genomes encode four eIF2α kinases that drive this response: PKR-like endoplasmic reticulum (ER) kinase (PERK) is activated by the accumulation of unfolded polypeptides in the lumen of the ER, general control non-derepressible 2 (GCN2) kinase by amino acid starvation and UV light, protein kinase RNA-activated (PKR) by viral infection, and heme-regulated eIF2α kinase (HRI) by heme deficiency and redox stress. The eIF2α kinase PERK is also part of the unfolded protein response (UPR). This intracellular stress signaling network is comprised of three ER-localized transmembrane sensors, IRE1, ATF6, and PERK, which initiate unique signaling cascades upon sensing an increase in unfolded proteins in the ER lumen (Walter and Ron, 2011; Pavitt and Ron, 2012).
The common mediator of the ISR, eIF2α, is a subunit of an essential translation initiation factor conserved throughout eukaryotes and archaea. The heterotrimeric eIF2 complex (composed of subunits α, β and γ) brings initiator methionyl tRNA (Met-tRNAi) to translation initiation complexes and mediates start codon recognition. It binds GTP along with Met-tRNAi to form a ternary complex (eIF2-GTP-Met-tRNAi) that assembles, along with the 40S ribosomal subunit and several other initiation factors, into the 43S pre-initiation complex (PIC). The 43S PIC is recruited to the 5′ methylguanine cap of an mRNA and scans the 5′UTR for an AUG initiation codon (Hinnebusch and Lorsch, 2012). Start site codon recognition triggers GTP hydrolysis and phosphate release, which is followed by release of eIF2 from the 40S subunit, allowing binding of the 60S ribosomal subunit to join. After these events, the elongation phase of protein synthesis ensues. To engage in a new round of initiation, the newly released eIF2 complex has to be re-loaded with GTP, a reaction catalyzed by its dedicated guanine nucleotide exchange factor (GEF), the heteropentameric eukaryotic initiation factor 2B (eIF2B). Phosphorylation of eIF2α does not directly affect its function in the PIC, but rather inhibits eIF2B, thereby depleting ternary complex and reducing translation initiation (Krishnamoorthy et al., 2001). eIF2B complex is limiting in cells, present in lower abundance than eIF2; a small amount of phospho-eIF2α therefore acts as a competitive inhibitor with dramatic effects on eIF2B activity. When eIF2B is inhibited and ternary complex is unavailable, the rate of translation initiation decreases.
Unimpaired elongation in the face of reduced initiation allows translating ribosomes to run off of their mRNAs, generating naked mRNAs that can then bind to RNA-binding proteins (RBPs) and form messenger ribonucleoproteins, which can further assemble into stress granules (SGs). These cytoplasmic, non-membrane bounded organelles contain translationally stalled and silent mRNAs, 40S ribosomal subunits and their associated pre-initiation factors and RBPs; these RBPs facilitate the nucleation and reversible aggregation of SGs through reversible, low-affinity protein–protein interactions mediated by their low complexity domains (Buchan and Parker, 2009; Kedersha and Anderson, 2009; Kato et al., 2012).
Paradoxically, under conditions of reduced ternary complex formation and protein synthesis, a group of mRNAs is translationally up-regulated. These mRNAs contain short upstream open reading frames (uORFs) in their 5′ UTRs, which are required for their ISR-responsive translational control (Hinnebusch, 2005; Jackson et al., 2010). These target transcripts include mammalian ATF4 (a cAMP response element binding transcription factor) and CHOP (a pro-apoptotic transcription factor) (Harding et al., 2000; Vattem and Wek, 2004; Palam et al., 2011). ATF4 regulates the transcription of many genes involved in metabolism and nutrient uptake and thus is a major regulator of the transcriptional changes that ensue upon eIF2α phosphorylation and ISR induction (Harding et al., 2003). Although activation of this cellular program can initially mitigate the stress and confer cytoprotection, persistent and severe stress and its associated reduction in protein synthesis and CHOP activation lead to apoptosis (Tabas and Ron, 2011; Lu et al., 2014).
In animals, the ISR has been implicated in diverse processes ranging from the regulation of insulin production to learning and memory. These effects were studied first using genetics by generating knock-out mice lacking individual eIF2α kinases as well as a knock-in of the non-phosphorylatable allele eIF2αS51A (Eif2s1S51A). Homozygous loss of eIF2α phosphorylation leads to perinatal death but heterozygous eIF2α+/S51A animals, which have reduced levels of eIF2α phosphorylation, grow into healthy adults showing phenotypes that demonstrate the importance of translation initiation in establishment of long-term memories (Scheuner et al., 2001). Behavioral tests demonstrated that PKR−/−, GCN2−/− and eIF2α+/S51A animals display enhanced memory consolidation in learning paradigms of light training (Costa-Mattioli et al., 2005, 2007; Zhu et al., 2011). Pharmacological modulation of eIF2α phosphorylation represented an important advance, allowing easier discrimination between developmental and acute effects of ISR reduction and circumventing the lethal phenotype of homozygous eIF2αS51A/S51A. Recent work identified small molecules that modulate the ISR pathway at distinct steps: (1) kinase inhibitors that target PERK or PKR (Jammi et al., 2003; Atkins et al., 2013); (2) an activator of HRI (Chen et al., 2011); (3) salubrinal, an inhibitor of eIF2α phosphatases (Boyce et al., 2005); and (4) ISRIB (Sidrauski et al., 2013). By a yet unknown mechanism, ISRIB blunts the effects of eIF2α phosphorylation in cells and thus represents the first bona fide ISR inhibitor acting downstream of all eIF2α kinases.
Here, we show that ISRIB reverses comprehensively and specifically the effects of eIF2α phosphorylation. By profiling the genome-wide translational program downstream of the ISR, we present the application of ribosome profiling to the ISR in mammalian cells, which allowed us to identify and quantify the translational changes that take place upon its induction and ISRIB treatment. Moreover, live cell imaging revealed that ISRIB addition can trigger a remarkably fast dissolution of phospho-eIF2α-dependent SGs in stressed cells, restoring translation.
ResultsRibosome profiling of ER stress in mammalian cells
We used ribosome profiling to characterize translational changes induced by ER stress. Deep sequencing of ribosome-protected mRNA fragments provides global, quantitative measurements of translation and reveals the precise location of ribosomes on each mRNA (Ingolia et al., 2009, 2011). We triggered the UPR in HEK293T cells by treating them with tunicamycin (Tm), a toxin that blocks N-linked glycosylation of ER-resident proteins. We chose to analyze an early time point (1 hr) in order to focus on translational changes preceding the extensive transcriptional induction that takes place upon activation of the three branches of the UPR (for a time course of UPR induction, see Figure 3—figure supplement 1 in 10.7554/eLife.00498). After 1 hr of Tm or mock treatment, we added cycloheximide (CHX) to arrest translating ribosomes, lysed the cells, and digested the extract with nuclease to degrade mRNAs not protected by ribosomes. In parallel, we isolated total mRNAs to monitor any changes in mRNA levels. Ribosome profiling data revealed a discrete subset of mRNAs that were translationally up- or down-regulated more than twofold after UPR induction (Figure 1A, above or below box) as seen by changes in abundance of ribosome-protected fragments (RPF) (‘Ribo-Seq’, y-axis) without corresponding changes in mRNA levels (‘mRNA-Seq’, x-axis). Data points representing statistically significant changes in expression between Tm-treated and untreated (‘UT’) samples are highlighted in black.10.7554/eLife.05033.002
Translational regulation upon ER stress in mammalian cells.
(A) Translational and mRNA changes in HEK293T cells upon ER stress. HEK293T cells were treated with or without 1 μg/ml of Tm for 1 hr. The y-axis represents fold changes in ribosome-protected fragments (Ribo-Seq) between Tm-treated and control samples. The x-axis represents fold changes in mRNA levels (mRNA-Seq) between Tm-treated and control samples. Data points reflecting significant changes (FDR-corrected p-value < 0.1) between Tm treated and untreated (‘UT’) samples are shown in black and non-significant changes are shown in light grey. Note that genes with significant changes (black circles) are numerous in Tm-treated cells and thus the cloud of genes with no significant changes (grey circles) is mostly hidden in the background. Genes with substantially enhanced RFPs and uORFs that are known to be phospho-eIF2α-dependently regulated are labeled pink. ISR-translational targets that contain previously identified uORFs are labeled in green. Triangles denote genes that fall beyond the axis range. The genes inside the grey box are those that change less than twofold in RPF or mRNA reads. Figure 1—source data 2A contains a list of all genes that change more than twofold in RPFs during Tm induction (FDR-corrected p-value < 0.1, corresponding to black circles above and below the box). (B) Translational and mRNA changes in cells co-treated with Tm and ISRIB. HEK293T cells were treated with or without 1 μg/ml of Tm and 200 nM ISRIB for 1 hr. The y-axis represents fold changes in ribosome-protected fragments (Ribo-Seq) between Tm + ISRIB-treated and control samples. The x-axis represents fold changes in mRNA levels (mRNA-Seq) between Tm + ISRIB-treated and control samples. Genes that significantly change when ISRIB co-administration modulates the effects of Tm treatment are shown in black (FDR-corrected p-value < 0.1). Figure 1—source data 2C contains a list of all genes that change more than twofold in RPFs during Tm and ISRIB treatment (FDR-corrected p-value < 0.1). The identity of the ISR-translational targets that contain previously identified uORFs (labeled in green) was not included in this panel as they all collapsed to the center of the plot. (C) Translational and mRNA changes in ISRIB-treated cells. HEK293T cells were treated with or without 200 nM ISRIB for 1 hr. The y-axis represents fold changes in ribosome-protected fragments (Ribo-Seq) between ISRIB-treated and control samples. The x-axis represents fold changes in mRNA levels (mRNA-Seq) between ISRIB-treated and control samples. Data points reflecting significant changes (FDR-corrected p-value < 0.1) between ISRIB-treated and untreated (‘UT’) samples are shown in black and non-significant changes are shown in light grey. Figure 1—source data 2D contains a list of all genes that change more than twofold in RPFs during ISRIB treatment (FDR-corrected p-value < 0.1, corresponding to black circles outside of the box). ATF4 and SLC35A4 (labeled in this panel) showed reduced translational efficiency upon addition of ISRIB. Two biological replicates were analyzed per condition. Number of reads aligned to the genome and ORFs for all samples are found in Figure 1—source data 2E. Correlation plots for the replicates for each condition are found in Figure 1—figure supplement 3. mRNA abundance for all ORFs mapped are found in Figure 1—figure supplement 4. Read counts for all conditions and each individual transcript are found in Figure 1—source data 1. The Ribo-seq and mRNA-seq data have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO series accession number GSE65778.
DOI:http://dx.doi.org/10.7554/eLife.05033.002
10.7554/eLife.05033.003
Read counts for all conditions and each individual transcript.
DOI:http://dx.doi.org/10.7554/eLife.05033.003
10.7554/eLife.05033.004
Source data for <xref ref-type="fig" rid="fig1">Figure 1</xref>.
(A) Significant translational changes in Tm-treated cells. List of genes that change more than twofold in translational efficiency after 1 hr of Tm treatment. mRNA (Tm/UT) represents log2-fold changes in mRNA levels between Tm and untreated samples, and Ribo-Seq (Tm/UT) represents log2-fold changes in RPFs between Tm- and un-treated samples (UT). Only genes with a log2-fold change >|1| and a FDR-corrected p-value <0.1 were included in the table. Known ISR-translational targets are highlighted in pink and targets containing previously identified uORFs are highlighted in green. (B) ISR-translational targets containing uORFs. List of genes that change more than twofold in translational efficiency after 1 hr of Tm treatment and contain previously identified uORFs. Ribo-Seq (Tm/UT) represents log2-fold changes in RPFs between Tm- and un-treated samples (UT). (C) Significant translational changes in Tm and ISRIB-treated cells. List of genes that change more than twofold in translational efficiency after 1 hr of Tm and ISRIB treatment. mRNA [(Tm + ISRIB)/UT] represents log2-fold changes in mRNA levels between Tm and ISRIB-treated and untreated samples and Ribo-Seq [(Tm + ISRIB)/UT)] represents log2-fold changes in RPFs between Tm and ISRIB-treated and untreated samples. Only genes with a log2-fold change >|1| and a FDR-corrected p-value < 0.1 were included in the table. (D) Significant translational changes in ISRIB-treated cells. List of genes that change more than twofold in translational efficiency after 1 hr of ISRIB treatment. mRNA (ISRIB/UT) represents log2-fold changes in mRNA levels between ISRIB-treated and untreated samples and[SPACE]Ribo-Seq (ISRIB/UT) represents log2-fold changes in RPFs between ISRIB-treated and untreated samples. Only genes with a log2-fold change >|1| and a FDR-corrected p-value < 0.1 were included in the table. (E) Alignment of sequencing libraries. Total number of reads and the percentage of reads for both mRNA-seq and ribo-seq libraries that align to the genome and ORF for all conditions (UT, Tm, Tm + ISRIB, ISRIB) and for each replicate.
DOI:http://dx.doi.org/10.7554/eLife.05033.004
10.7554/eLife.05033.005
Ribosome and mRNA densities in the 5’UTR of ATF4 and SLC35A4.
mRNA reads (y-axis) are represented along the sequence of each gene (x-axis) in the upper panel. Ribosome footprint (ribo) reads (y-axis) are represented along the sequence of each gene (x-axis) in the lower panel. The known and predicted uORFs are indicated along the sequence in green. The ORF is indicated along the sequence in blue.
DOI:http://dx.doi.org/10.7554/eLife.05033.005
10.7554/eLife.05033.006
Translational regulation of mTOR targets upon ER-stress.
(A) Cells were treated with or without 1 μg/ml of Tm for 1 hr. The y-axis represents fold changes in ribosome-protected fragments (Ribo-Seq) between Tm-treated and control samples. The x-axis represents fold changes in mRNA levels (mRNA-Seq) between Tm-treated and control samples. Only mTOR translational targets are plotted (colored light green). Significant changes in mTOR genes (FDR-corrected p-value < 0.1) between Tm-treated and untreated (UT) are highlighted in black. (B) Cells were treated with or without 1 μg/ml of Tm and 200 nM ISRIB for 1 hr. The y-axis represents fold changes in ribosome-protected fragments (Ribo-Seq) between Tm + ISRIB-treated and control samples. The x-axis represents fold changes in mRNA levels (mRNA-Seq) between Tm + ISRIB-treated and control samples. Only mTOR translational targets are plotted (colored light green). Genes that significantly change when ISRIB co-administration modulates the effects of Tm treatment are shown in black (FDR-corrected p-value < 0.1). (C) Cells were treated with or without 200 nM ISRIB for 1 hr. The y-axis represents fold changes in ribosome-protected fragments (Ribo-Seq) between ISRIB-treated and control samples. The x-axis represents fold changes in mRNA levels (mRNA-Seq) between ISRIB-treated and control samples. Only mTOR translational targets are plotted (colored light green). Significant changes in mTOR genes (FDR-corrected p-value < 0.1) between ISRIB-treated and untreated (UT) are highlighted in black.
DOI:http://dx.doi.org/10.7554/eLife.05033.006
10.7554/eLife.05033.007
Correlation plots for duplicate ribosome profiling experiments.
The detected ribosome (ribo) or mRNA density is plotted for each gene in each experimental condition (untreated [UT], tunicamycin [Tm], tunicamycin + ISRIB [Tm + ISRIB] and ISRIB). Correlation coefficients (r2) between replicates (A and B) in each condition are indicated in the lower right for each panel.
DOI:http://dx.doi.org/10.7554/eLife.05033.007
10.7554/eLife.05033.008
Mean mRNA abundance of all genes mapped.
The x-axis represents log2 (mean normalized count) for each mRNA mapped and the y-axis represents log2 fold changes in mRNA abundance in the different experimental conditions: Tm (panel A), Tm + ISRIB (panel B), and ISRIB (panel C). Previously known phospho-eIF2α-dependent ISR translational targets are highlighted in pink.
DOI:http://dx.doi.org/10.7554/eLife.05033.008
Consistent with the well-established presence of regulatory uORFs in their 5′-UTRs, this genome-wide analysis identified four previously extensively studied mRNAs that displayed significant translational upregulation: ATF4, ATF5, CHOP, and GADD34, (Figure 1A, colored pink). The mRNAs encoding the closely paralogous transcription factors ATF4 and ATF5 are known translational targets of the ISR (Lu et al., 2004; Vattem and Wek, 2004; Zhou et al., 2008). They contain two uORFs (the second one overlapping with the coding sequence [CDS]) that govern their enhanced translational efficiency. The mRNAs encoding the pro-apoptotic transcription factor CHOP and the regulatory subunit of the eIF2α phosphatase GADD34 were also significantly upregulated at the translational level. Although both CHOP and GADD34 are also known transcriptional targets of ATF4, we did not detect significant induction of their mRNAs at this early time point (Figure 1A, lack of displacement along x-axis) indicating that at the time point chosen our analysis exclusively reports on translational effects. CHOP and GADD34 mRNAs also contain uORFs that allow for translational regulation upon eIF2α phosphorylation (Lee et al., 2009; Palam et al., 2011).
We identified a total of 78 mRNAs whose translation changed significantly and substantially (more than twofold) upon ER stress in HEK293T cells (listed in Figure 1—source data 2A). GO term analysis revealed the involvement of these genes in diverse functions and several encode for proteins with entirely unknown functions. Besides the four known ISR translational targets described above, six mRNAs in the list contain previously mapped uORFs as validated by ribosome profiling in the presence of a translation initiation inhibitor to mark initiation sites (Figure 1A, colored green and Figure 1—source data 2B) (Lee et al., 2012). Whereas 5% of the non-significantly changed genes in the Tm sample contain previously identified AUG-initiated uORFs, 14% of genes in the list of ISR-translational targets contain uORFs, indicating a significant enrichment (p < 0.003, chi-squared test with Yates correction).
A seventh and novel uORF-containing translational target of the ISR encodes SLC35A4, a putative nucleotide-sugar transporter (Song, 2013). It was recently shown that the longest uORF of SLC35A4 is indeed translated because peptides corresponding to the encoded polypeptide were found in a whole proteome mass spectrometry study (Kim et al., 2014). Analysis of RPFs in the uORFs of the SLC35A4 and ATF4 mRNAs revealed significant ribosome density, further confirming that these regulatory uORFs are normally translated (Figure 1—figure supplement 1). Due to the reduced mRNA expression levels of ATF5, CHOP, and GADD34 in the absence of stress or at early time-points of UPR activation, we did not analyze the RPFs or mRNA reads at specific locations along these genes, as the read numbers were low.
Interestingly, there was a slight reduction in translation of mRNAs encoding ribosomal proteins and translation elongation factors (Figure 1—figure supplement 2, panel A). The translation of this functionally related class of ∼100 abundant mRNAs, which have a 5′ terminal oligopyrimidine (5′ TOP) motif, is controlled by the activity of the mTOR kinase (Meyuhas, 2000; Tang et al., 2001; Hsieh et al., 2012; Thoreen et al., 2012). The concerted changes that we observed in their translation upon UPR activation suggest that ER stress and eIF2α phosphorylation affects 5′ TOP translation in HEK293T cells.
ISRIB substantially reduced the translational effects elicited by stress and eIF2α phosphorylation
To study the translational effects of the small molecule ISRIB at a genome-wide level, we analyzed changes in RPFs and mRNA levels after addition of the drug to both ER-stressed and unstressed cells. As seen in Figure 1B, ISRIB comprehensively blocked the translational changes that take place upon ER-stress. A large number of genes with a significant change in expression upon stress collapsed to the center of the plot with ISRIB and Tm co-treatment (Figure 1B, highlighted in black). Importantly, ISRIB abolished the induction of the known phospho-eIF2α-dependent translational targets (Figure 1B, colored pink) and the seven ISR-translational targets with previously identified uORFs (Figure 1B, colored green). The mRNAs that remained translationally induced in the presence of ISRIB are listed in Figure 1—source data 2C. In addition, ISRIB reversed the reduction in translation of mTOR target mRNAs upon ER stress (Figure 1—figure supplement 2, panel B).
Importantly, ISRIB treatment alone did not have general effects on translation in non-stressed cells, as revealed by the lack of substantial changes in RPFs in most cellular mRNAs, nor did it cause any significant changes in mRNA levels (Figure 1C) and mTOR target expression (Figure 1—figure supplement 2, panel C). In the absence of ER stress, ISRIB-treated cells behaved like untreated cells with the exception of a reduction in the basal level of translation of ATF4 and SLC35A4 mRNAs and a few additional mRNAs (Figure 1C and Figure 1—source data 2C). Taken together, these data strongly support the notion that ISRIB does not have global effects on translation, transcription, or mRNA stability in non-stressed cells and underscores its remarkable ability to counteract selectively the translational changes elicited by eIF2α phosphorylation in stressed cells.
ISRIB prevents formation of stress granules exclusively triggered by eIF2α phosphorylation
Phosphorylation of eIF2α and reduction of ternary complex formation are tightly linked to the formation of stress granules (SGs) (Kedersha et al., 2002). ISRIB renders cells insensitive to the effects of eIF2α phosphorylation, thus leading to the prediction that it prevents SG formation as well. We tested this hypothesis by inducing SG formation using thapsigargin (Tg), a potent ER stressor that inhibits the ER calcium pump and was recently shown by ribosome profiling to yield analogous translational effects to tunicamycin (Reid et al., 2014). Microscopic detection of SGs required a stronger induction of ER stress than commonly achieved with Tm, making Tg the preferred inducer. We monitored SGs by performing immunofluorescence on eIF3a, a translation initiation factor that is recruited into SGs. As expected, we found that ISRIB significantly reduced their assembly upon co-treatment with Tg (Figure 2A,B). In addition, ISRIB prevented SG formation induced by arsenite (Ars), another widely used inducer of eIF2α phosphorylation via activation of HRI. As expected, both treatments induced eIF2α phosphorylation but only Tg induced the ER-resident kinase, PERK, as seen by its shift in mobility that is due to its extensive auto-phosphorylation (Figure 2C). Both the PERK mobility shift and eIF2α phosphorylation elicited by Tg treatment were blocked by a PERK inhibitor (GSK707800; Axten et al., 2012) but not by ISRIB. Like ISRIB, and as expected by the block in eIF2α phosphorylation, the GSK PERK inhibitor prevented SG induction upon Tg addition (Figure 2A).10.7554/eLife.05033.009
ISRIB blocks stress granule formation induced by eIF2α phosphorylation.
(A) Immunofluorescence analysis (eIF3a) of U2OS cells treated with 200 nM Tg for 1 hr, 250 μM Ars for 30 min, or 100 nM Pat A for 30 min in the presence or absence of 200 nM ISRIB or 1 μM GSK797800 PERK inhibitor. A secondary Alexa Dye 488 anti-rabbit antibody was used to visualize eIF3a and DAPI was used to visualize nuclei. Representative images of at least two biological replicates are shown. (B) Quantitation of the percentage of cells containing stress granules in the different conditions described in A. Images were collected from at least two independent experiments and the number of cells with SGs or no SGs counted. The total number of cells counted for each condition was (sum of all replicates): Control (N = 81), ISRIB (N = 94), Tg (N = 122), Tg + ISRIB (N = 71), Ars (N = 85), Ars + ISRIB (N = 84), Pat A (N = 47) and Pat A + ISRIB (N = 64). No cells had SGs in Tg + PERK inh (N = 71). p-values are derived from a Student's t-test, *p < 0.05. (C) Immunoblot analysis of PERK, phospho eIF2α, and total eIF2α in U2OS cells treated as in A. Hippuristanol (Hipp) was used at 300 nM for 30 min. The right blot was overexposed to confirm the absence of induction of eIF2α phosphorylation upon Pat A and Hipp treatment. A representative blot of three independent experiments is shown. The asterisk (*) represents a background band or degradation product.
DOI:http://dx.doi.org/10.7554/eLife.05033.009
SG formation can also be induced in the absence of eIF2α phosphorylation by inhibiting the eIF4A helicase, which is part of the cap-binding eIF4F complex (Mazroui et al., 2006). Pateamine A (Pat A) binds to and inhibits this enzyme and blocks scanning of the PIC and translation initiation (Dang et al., 2006). In agreement, Pat A-induced SG formation but it did not cause eIF2α phosphorylation (Figure 2A,C). In contrast to phospho-eIF2α-induced SGs, Pat-A-induced SGs were not reduced by ISRIB (Figure 2A,B). Thus, ISRIB blocks phospho-eIF2α-dependent SG induction selectively.
ISRIB triggers rapid disassembly of stress granules and restores translation
To visualize SG formation in living cells and to assess the effects of ISRIB on pre-formed SGs, we took advantage of a stable cell line expressing G3BP fused to GFP (Kedersha et al., 2008). In contrast to cell lines that overexpress SG-associated RNA binding proteins like G3BP, in this single clone-derived cell line, low expression of the fusion protein preserves stress-dependent regulation of SG assembly. We confirmed that in this cell line ISRIB significantly reduces SG formation driven by stresses that cause eIF2α phosphorylation (Tg and Ars) but not by phospho-eIF2α-independent induction through eIF4A inhibition (Pat A and hippuristanol [Hipp]) (Figure 3A,B) (Cencic et al., 2012). To match the strength of the stresses used in these experiments and minimize the toxic effects of these agents, we used the shortest incubation time and the lowest concentration of each stressor that resulted in SG formation in the majority of cells. ISRIB has an EC50 of 5 nM as previously measured using an uORFs-ATF4-driven luciferase reporter in HEK293T cells (Sidrauski et al., 2013). In close agreement with the high potency measured in the reporter assay, ISRIB significantly reduced SG formation even at concentrations as low as 2 nM in U2OS cells and as expected, an inactive analog, ISRIBinact, did not reduce their formation (Figure 3—figure supplement 1).10.7554/eLife.05033.010
ISRIB addition rapidly dissolves pre-formed stress granules in live cells restoring translation.
(A) Live cell imaging of stress granules in U2OS cells stably expressing G3BP-GFP (SG marker). Cells were treated with 200 nM Tg for 40 min, 250 μM Ars for 30 min, 100 nM Pat A for 30 min, or 300 nM Hipp in the presence or absence of 200 nM ISRIB. Cells were imaged using an epifluorescence microscope. Representative images of at least two biological replicates are shown. (B) Quantitation of the percentage of cells containing stress granules in the different conditions described in A. Images were collected from at least two independent experiments and the number of cells with SGs or no SGs counted. The number of cells analyzed for each condition was (sum of replicates): Control (N = 98), ISRIB (N = 81), Tg (N = 101), Tg + ISRIB (N = 84), Ars (N = 80), Ars + ISRIB (N = 55), Pat A (N = 58), Pat A + ISRIB (N = 50), Hipp (N = 41) and Hipp + ISRIB (N = 52). p-values are derived from a Student's t-test, *p < 0.05. (C) Stress granules were pre-formed with Tg for 40 min (as in Figure 3A) and then CHX (50 μg/ml) or ISRIB (200 nM) was added to the well, incubated for 10 min and images were collected. Representative images of at least two biological replicates are shown. (D) ISRIB quickly restores mRNA translation upon disassembly of stress granules. Cells were treated as in C with 200 nM Tg for 40 min and then DMSO, CHX (50 μg/ml), or ISRIB (200 nM) was added at the same time as [35S]-methionine. Cells were lysed after 15 min, protein was run in an SDS-PAGE gel and radioactivity was measured in each lane (N = 2, mean ± SD). (E) ISRIB quickly dissolves stress granules but does not affect P-bodies. Live cell imaging of U2OS cells stably expressing G3BP-GFP (SG marker) and Dcp1-RFP (P-body marker). Cells were treated with 200 nM Tg for 45 min followed by addition of 200 nM ISRIB at t = 0 min to the well and then imaged using spinning disk confocal microscopy. Images were collected every 30 s. The red arrows point to two representative P-bodies. Representative images of at least three biological replicates are shown.
DOI:http://dx.doi.org/10.7554/eLife.05033.010
10.7554/eLife.05033.011
ISRIB dose response and inactive analog in stress granule assay.
(A) Live cell imaging of SGs in U2OS cells stably expressing G3BP-GFP. Cells were treated with 200 nM Tg and different doses of ISRIB (as indicated) or 2 μM of an inactive analog of ISRIB (ISRIBinact). Representative images of at least two biological replicates are shown. (B) Quantitation of the percentage of cells containing stress granules in the different conditions. The number of cells analyzed for each condition were: ISRIBinact (N = 37), 0.5 nM ISRIB (N = 81), 2 nM ISRIB (N = 91), 10 nM ISRIB (N = 43).
DOI:http://dx.doi.org/10.7554/eLife.05033.011
10.7554/eLife.05033.012
Representative SDS-PAGE gel of [<sup>35</sup>S]-methionine pulse as described in <xref ref-type="fig" rid="fig3">Figure 3D</xref>.
Top panel is an autoradiogram and bottom panel is total protein of the same gel as shown by Coomassie staining.
DOI:http://dx.doi.org/10.7554/eLife.05033.012
Treatment of cells with CHX disassembles SGs in the presence of ongoing stress (Kedersha et al., 2000; Mollet et al., 2008). This observation as well as other pharmacological and microscopy data revealed that SGs are highly dynamic structures with mRNAs quickly shuttling in and out. When these mRNAs leave SGs, translation is reinitiated; CHX then immobilizes elongating ribosomes and prevents mRNA re-entry into SGs. Because polyribosome disassembly is blocked by CHX yet required for SG assembly, CHX treatment dissolves pre-formed SGs. As seen in Figure 3C, a 10-min treatment with CHX following Tg induction of SGs (40 min) was sufficient to observe disassembly. Like CHX, ISRIB addition disassembled SGs within 10 min, even in the prolonged presence of the stressor Tg (Figure 3C). Whereas ISRIB restored translation of mRNAs that are liberated from SGs, as seen by the quick recovery in [35S]-methionine incorporation, CHX further reduced protein synthesis (Figure 3D and Figure 3—figure supplement 2). These experiments demonstrate that ISRIB triggers disassembly of pre-formed SGs by loading dissociating mRNAs with actively translating ribosomes.
We next looked at the kinetics of SG disassembly upon ISRIB addition. Strikingly, after only 5 min of ISRIB treatment, Tg-induced SGs were no longer observed in cells (Figure 3E and Video 1). We also investigated the impact of ISRIB on P-bodies, a molecularly distinct class of RNA aggregates that serve as centers of mRNA decay (Kedersha and Anderson, 2009). The mRNA decay factor Dcp1 serves as a marker for these structures, and we visualized them in living cells using the fusion protein Dcp1-RFP. We saw that P-bodies were constitutively present in a percentage of the cells and were not affected by ISRIB treatment or by the stressors used to induce SGs over the time-course experiments explored here (Figure 3E red arrows, Video 2 and data not shown).10.7554/eLife.05033.013
ISRIB triggers stress granule disassembly.
U2OS cells stably expressing G3BP-GFP (SG marker) and Dcp1-RFP (P-body marker) were treated with 200 nM Tg for 40 min and then 200 nM ISRIB was added at t = 0 min to the well and imaged using an epifluorescence microscope. Images of G3BP-GFP (SGs) were collected every 30 s.
DOI:http://dx.doi.org/10.7554/eLife.05033.013
10.7554/eLife.05033.014
ISRIB does not trigger disassembly of P-bodies.
Images of Dcp1-RFP (P-bodies) corresponding to the same field of cells as in Video 1 were collected every 30 s.
DOI:http://dx.doi.org/10.7554/eLife.05033.014
Discussion
ISRIB is the first reported antagonist of the ISR that blocks signaling downstream of all eIF2α kinases. It was shown to have good pharmacokinetic properties and brain penetration, making it a useful tool to study the systemic effects of acute inhibition of the pathway. We showed that ISRIB administration enhances long-term memory in rodents (Sidrauski et al., 2013). More recently, we showed by electrical recordings in brain slices that by preventing AMPAR down-regulation in the post-synaptic neuron, ISRIB blocks mGluR-mediated long-term depression (LTD), an effect that is dependent on eIF2α phosphorylation (Di Prisco et al., 2014). Comprehensive analyses of the cellular effects and kinetics of action of ISRIB are critical for interpretation of its in vivo effects and assessment of its therapeutic potential.
Our translational and transcriptional profiling confirmed that ISRIB treatment of ER-stressed cells substantially and comprehensively blocks the translational effects of eIF2α phosphorylation. ISRIB blocked SG formation that was triggered by eIF2α phosphorylation but did not abolish their assembly upon eIF4A inhibition; eIF4A inhibitors do not cause eIF2α phosphorylation and can induce SGs in eIF2αS51A/S51A cells (Mazroui et al., 2006; Mokas et al., 2009). These data further support the notion that ISRIB solely inhibits cellular events that are a consequence of eIF2α phosphorylation. In agreement with these observations, we previously showed by polyribosome sedimentation analysis that ISRIB does not reverse bulk translational down-regulation triggered by inhibition of CAP-dependent initiation (Sidrauski et al., 2013). Moreover, ISRIB treatment alone did not induce overall changes in translation or mRNA levels. Taken together these data demonstrate that ISRIB is a pharmacological agent that acutely and specifically blocks the ISR and is thus an invaluable tool for in vivo studies.
Translational regulation upon ISR induction
The method of ribosome profiling can monitor in vivo translation comprehensively and with nucleotide resolution (Ingolia et al., 2009). We used this method to monitor translation of all cellular mRNAs upon ISR activation. We found that a limited set of mRNAs is preferentially translated in a substantial manner upon a reduction in ternary complex assembly. Although previous large-scale analyses have revealed that almost 45% of all 5′ UTRs have at least one upstream uORF (Calvo et al., 2009; Ingolia et al., 2011), our data revealed that only a few of these mRNAs contain uORFs with regulatory properties that significantly enhance translation of their downstream coding sequences upon eIF2α phosphorylation. The canonical ISR translational targets, ATF4, ATF5, CHOP, and GADD34 mRNAs were significantly induced upon 1 hr treatment with the ER-stressor tunicamycin. The stress-induced, uORF-mediated regulation of GCN4 translation in yeast established the paradigm for this mode of regulation (Dever et al., 1995; Grant et al., 1995). As in mammalian cells, GCN2 is activated in amino acid-starved yeast by the accumulation of uncharged tRNAs, catalyzing eIF2α phosphorylation. The transcript encoding GCN4, a bZIP transcription factor with homology to mammalian ATF4, has four uORFs that modulate translation of its coding sequence upon stress. GCN4 induction is thought to occur via a re-scanning mechanism that allows 40S ribosomal subunits to remain mRNA-bound after completing the translation of short reading frames and subsequently reinitiate in the downstream coding sequence after reloading with ternary complex (Hinnebusch, 2005). The select mRNAs that are translationally upregulated in mammalian cells have uORFs that vary in number, length, and distance from the coding sequences. As was observed for GCN4, the uORF2 of ATF4 mRNA showed ribosome density, supporting the notion that it is translated under normal growth conditions (Ingolia et al., 2009). Whether the same mechanism of rescanning is utilized by all these mRNAs is not known but, like ATF4, their regulation depends on their uORFs (Vattem and Wek, 2004; Zhou et al., 2008; Lee et al., 2009; Palam et al., 2011).
SLC35A4 is a novel translational target of the ISR. Ribosome profiling of HEK293T cells upon arsenite treatment, a potent inducer of eIF2α phosphorylation, also revealed the increased synthesis of SLC35A4 (Andreev et al., 2015). It belongs to a large family of nucleotide sugar transporters (NSTs) that are highly conserved transmembrane antiporters localized to the ER or Golgi apparatus (Song, 2013). The role of SLC35A4 in cells is unknown but it may function as the elusive ER-localized UDP-glucose transporter. This hypothesis is particularly attractive in the context of our data because unfolded ER proteins, which trigger the ISR, are continuously de- and re-glucosylated on their N-glycans using UDP-glucose as the glucose donor. Proteins with monoglucosylated N-glycans bind calnexin or calreticulin which promote protein folding. Translational induction of SLC35A4 may thus quickly enhance UDP-glucose transport into the ER lumen upon the accumulation of unfolded proteins in order to promote this pro-folding pathway.
Ribosome profiling upon activation of the UPR uncovered additional mRNAs induced upon eIF2α phosphorylation (our data, Reid et al., 2014; Andreev et al., 2015). Several of these mRNAs encode for proteins with entirely unknown functions and the remaining targets are involved in a wide range of cellular processes. Whether these ISR-induced translational targets are similarly regulated by the presence of uORFs in their 5′ UTRs remains to be determined with the construction of synthetic translational reporters. ISRIB blocked their differential translation, suggesting that these changes were due to phospho-eIF2α. There may be additional transcripts that are synthesized later during ISR activation, downstream of the early transcription factor targets such as ATF4, as well as tissue-specific mRNAs that are controlled by phospho-eIF2α. For example, OPHN1 is a neuron-specific mRNA containing uORFs that is translationally upregulated after mGluR engagement and eIF2α phosphorylation and induces LTD. By blocking the effects of phospho-eIF2α in cells, ISRIB also blocks mGluR-dependent LTD (Di Prisco et al., 2014). Ribosome profiling of glutamatergic neurons upon ISR induction may reveal additional transcripts whose translational control contributes to the molecular events underlying memory.
The ribosome profiling data presented here revealed that eIF2α phosphorylation modestly, yet significantly, decreased translation of a large number of ribosomal proteins and elongation factors. Although the decrease in translation of ribosomal proteins and elongation factors upon eIF2α phosphorylation is small in magnitude, its effects on bulk protein synthesis in the cell are significant as these represent a large number of highly expressed proteins. Translation of these mRNAs was previously shown to be under control of mTOR kinase, which regulates mRNA cap-binding factor eIF4E via phosphorylation of inhibitory eIF4E-binding proteins, thereby adjusting protein synthesis in cells in response to the cell's energy and nutrient status (Ma and Blenis, 2009). In this way, mTOR preferentially regulates translation of a group of mRNAs characterized by 5′ TOP motifs (Hsieh et al., 2012; Thoreen et al., 2012). Upon mTOR inhibition, translation of 5′ TOP mRNAs is reduced and the expression of factors required for protein synthesis is diminished.
The observed effect that eIF2α phosphorylation preferentially decreased translation of 5′ TOP mRNAs could, in principle, be due to inhibition of mTOR in response to ER stress. However, ISRIB reversed the translational changes, indicating that they are likely to be downstream consequences of eIF2α phosphorylation. Thus, if these translational changes do reflect altered mTOR activity, then the change in mTOR signaling must result from reduced translation mediated by eIF2α phosphorylation. Alternatively, eIF2α phosphorylation may lead to silencing of these mRNAs by recruiting them into SGs. The RNA binding proteins TIA-1 and TIAR, which are prominently SG-associated, were previously shown to bind to TOP mRNAs, leading to their translational downregulation upon amino acid starvation. This effect required both mTOR inhibition and GCN2 activation, the latter resulting in eIF2 phosphorylation (Damgaard and Lykke-Andersen, 2011). SGs also have been shown to recruit signaling molecules including upstream negative regulators of mTORC1 (raptor and DYRK3) and mTORC1 itself, and thus, SG formation may reduce their presence in the cytosol and impede translation of 5′ TOP mRNAs (Thedieck et al., 2013; Wippich et al., 2013).
Stress granule dynamics and ISRIB
The dynamic nature of SGs allowed us to monitor the action of ISRIB upon its addition to live cells in real time. Strikingly, addition of ISRIB to stressed cells with pre-formed SGs lead to their quick dissolution (less than 5 min), liberating mRNAs back into the translational pool. A pulse of [35S]-methionine confirmed the fast recovery in protein synthesis even in the presence of stress. Although the molecular target of ISRIB remains unknown, its quick action suggests a direct effect on translation initiation. Phospho-eIF2α resistance has been observed both in yeast and in mammalian cells. In yeast, mutations in eIF2B (the GEF for eIF2) and eIF5 (the 48S PIC-associated GTPase-activating protein for eIF2) have been reported to make cells insensitive to this phosphorylation event (Vazquez de Aldana and Hinnebusch, 1994; Pavitt et al., 1997, 1998). In mammalian cells, TLR4 engagement in macrophages leads to increased eIF2B activity by removal of an inhibitory phosphorylation and insensitivity to ISR activation (Woo et al., 2012). Thus, ISRIB may directly or indirectly enhance the activity of eIF2B, eIF5, or other initiation factors, thus quickly reversing the cellular effects of phosphorylated eIF2α.
SGs contain a large number of RBPs that harbor low complexity sequence domains that nucleate through transient, low affinity interactions (Kato et al., 2012). These RBPs usually contain several RNA-binding domains and can associate with more than one mRNA; this multi-valency further favors the coalescence of RNA-protein granules. A conspicuous feature of some degenerative diseases is the cytoplasmic or nuclear aggregation of RBPs, driven in some cases by pathogenic mutations. TDP-43 and FUS mutations are found in amyetropic lateral sclerosis (ALS) and fronto temporal lobar degeneration (FTLD) (Li et al., 2013), and mutations in hnRNPA1 and hnRPNPA2/B1 have also been found in ALS (Kim et al., 2013). Recent reports have also described the presence of RNA and RBPs in aggregates that form in prion disease, taupathies, and Alzheimer's (Vanderweyde et al., 2012; Ash et al., 2014). The impact of these cytosolic aggregates on SG dynamics is not known, though they may hamper the ability of SGs to properly dissolve, thereby contributing to sustained translational attenuation and neurodegeneration. By quickly disassembling SGs even in the presence of stress, ISRIB may provide a useful therapeutic intervention in these diseases by antagonizing the cellular effects of pathogenic RNA-protein assemblies.
Materials and methodsChemicals
Tunicamycin was obtained from Calbiochem EMB Bioscience. Thapsigargin, cycloheximide and sodium arsenite were obtained from Sigma–Aldrich. Hippuristanol and pateamine A were a kind gift from Jerry Pelletier. GSK797800 (PERK inhibitor) was obtained from TRC Inc. ISRIB (Sidrauski et al., 2013) and an inactive analog (754125) (Di Prisco et al., 2014) were synthesized in-house.
Cell culture
HEK293T, U2OS, and U2OS GFP-G3BP/Dcp1-RFP cells were maintained at 37°C, 5% CO2 in DMEM media supplemented with 10% FBS, L-glutamine and antibiotics (penicillin and streptomycin). U2OS cells stably expressing G3BP-GFP/Dcp1-RFP cells were a kind gift from Nancy Kedersha (Kedersha et al., 2008).
Isolation of ribosome footprints and RNA
HEK293T cells were treated with or without 1 μg/ml of tunicamycin, tunicamycin and ISRIB (200 nM), or ISRIB for 1 hr. Cycloheximide (CHX) (100 μg/ml) was added for 2 min, cells were washed with ice cold PBS (with 100 μg/ml of CHX) and lysed in 20 mM Tris pH = 7.4 (RT), 200 mM NaCl, 15 mM MgCl, 1 mM DTT, 8% glycerol, 100 μg/ml CHX, 1% Triton and protease inhibitors (Roche complete EDTA-free). A syringe (25G5/8) was used to triturate cells, the lysate was clarified at 12,000 rpm for 10 min and half of the lysate was used for RNA extraction (Trizol, Invitrogen, Carlsbad, CA) and the other half was digested with RNase I (Ambion). The amount of RNase I and time of incubation was optimized for each sample based on the collapse of polyribosomes to the monosome peak as analyzed by analytical polyribosome gradients. The reaction was quenched with SUPERaseIn (Ambion, Life Technologies) and the digested lysate was then loaded on an 800 μl sucrose cushion (1.7 g of sucrose was dissolved in 3.9 ml of lysis buffer without Triton) and centrifuged in a TLA100.2 rotor at 70,000 rpm for 4 hr. The pellet was resuspended in 10 mM Tris pH = 7 (RT), and RNA was extracted (phenol/chloroform).
Generation of sequencing libraries and data analysis
Sequencing libraries were generated as described in Ingolia et al., 2012. For data analysis, we used DESeq as described by Anders and Huber (2010). P-adj values (p-values) were calculated using the R command ‘p.adjust’ for multiple comparisons and the BH method (Benjamini and Hochberg, 1995) to correct for false discovery rate. The data in this publication have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO series accession number GSE65778.
Immunofluorescence
U2OS cells were seeded on 4-well chamber slides (Lab-Tek) 18 hr prior to processing for immunofluorescence. Cells (80% confluent) were fixed with ice-cold methanol. The cells were then rinsed with PBS (Sigma) and blocked for 1 hr at room temperature in 0.5% BSA in PBS. The cells were then incubated overnight at 4°C with an anti-eIF3A rabbit antibody (#3411; Cell Signaling Technology) at a 1:1000 dilution in blocking buffer. The next morning the slides were washed three times (5 min each time) with PBS and then incubated for 1 hr at room temperature in a 1:1000 dilution (in 0.5% BSA in PBS) of secondary anti-rabbit antibody labeled with Alexa Dye 488 (Molecular Probes). The slides were washed three additional times with PBS. The slides were then mounted with antifade reagent with DAPI (Life Technologies P-36931). Lastly, the slides were imaged using a Zeiss Axiovert 200M epifluorescence microscope.
Live cell microscopy
U2OS G3BP-GFP/Dcp1-RFP cells were plated in 8-well Lab-Tek chamber slides and switched to imaging media (lacking phenol red) upon addition of different stress inducers. Cells were either imaged using a Zeiss Axiovert 200M epifluorescence microscope or in a heated chamber using a spinning confocal epifluorescence microscope (Eclipse Ti-Nikon) and an Andor iXon3 camera.
Protein analysis
Cells were washed with PBS and lysed in SDS-PAGE loading buffer (1% SDS, 62.5 mM Tris–HCl pH 6.8, 10% glycerol). Lysates were sonicated and loaded on Any-kD SDS-PAGE gels (BioRad). Proteins were transferred onto nitrocellulose and probed with primary antibodies diluted in Tris-buffered saline supplemented with 0.1% Tween 20 and 5% BSA. The following antibodies were used: PERK (D11A8) (1:1000), eIF2α (#9722; Cell Signaling technology) (1:1000), phospho-eIF2α (Ser51) (44728G; Invitrogen). An HRP-conjugated secondary antibody (Amersham) was employed to detect immune-reactive bands using enhanced chemiluminescence (SuperSignal, Thermo Scientific).
[<sup>35</sup>S]-methionine incorporation
U2OS GFP-G3BP/mRFP-DCP1a cells were seeded on 12-well plates, allowed to recover overnight and treated with 100 nM Tg for 40 min. ISRIB (200 nM) or CHX (50 µg/ml) was added at the same time as 50 µCi of [35S]-methionine (Perkin Elmer) and incubated for 15 min. Cells were lysed by addition of SDS-PAGE loading buffer. Lysates were sonicated and equal amounts were loaded on SDS-PAGE gels (BioRad). The gel was dried and radioactive methionine incorporation was detected by exposure to a phosphor-screen and visualized with a Typhoon 9400 Variable Mode Imager (GE Healthcare).
Acknowledgements
We thank Margaret Elvekrog, Voytek Okreglak, Shelley Starck, and Jirka Peschek for editing the manuscript and the members of the Walter lab for helpful discussions.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
CS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
NTI, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
AMMG, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
PW, Analysis and interpretation of data, Drafting or revising the article
ReferencesAndersSHuberW2010Differential expression analysis for sequence count dataGenome Biology11R10610.1186/gb-2010-11-10-r106AndreevDEO'ConnorPBFaheyCKennyEMTereninIMDimitrievSECormicanPMorrisDWShatskyINBaranovPV2015Translation of 5' leaders is pervasive in genes resistant to eIF2 repressioneLife4e0397110.7554/eLife.03971AshPEVanderweydeTEYoumansKLApiccoDJWolozinB2014Pathological stress granules in Alzheimer's diseaseBrain Research1584525810.1016/j.brainres.2014.05.052AtkinsCLiuQMinthornEZhangSFigueroaDJMossKGStanleyTBSandersBGoetzAGaulNChoudhryAEAlsaidHJuckerBMAxtenJMKumarR2013Characterization of a novel PERK kinase inhibitor with anti-tumor and anti-angiogenic activityCancer Research731993200210.1158/0008-5472.CAN-12-3109AxtenJMMedinaJRFengYShuARomerilSPGrantSWLiWHHeerdingDAMinthornEMenckenTAtkinsCLiuQRabindranSKumarRHongXGoetzAStanleyTTaylorJDSigethySDTomberlinGHHassellAMKahlerKMShewchukLMGampeRT2012Discovery of 7-methyl-5-(1-{[3-(trifluoromethyl)phenyl]acetyl}-2,3-dihydro-1 H-indol-5-yl)-7 H-pyrrolo[2,3- d]pyrimidin-4-amine (GSK2606414), a potent and selective first-in-class Inhibitor of protein kinase R (PKR)-like endoplasmic reticulum kinase (PERK)Journal of Medicinal Chemistry557193720710.1021/jm300713sBenjaminiYHochbergY1995Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple TestingJournal of the Royal Statistical SocietySeries B57289300BoyceMBryantKFJousseCLongKHardingHPScheunerDKaufmanRJMaDCoenDMRonDYuanJ2005A selective inhibitor of eIF2alpha dephosphorylation protects cells from ER stressScience30793593910.1126/science.1101902BuchanJRParkerR2009Eukaryotic stress granules: the ins and outs of translationMolecular Cell3693294110.1016/j.molcel.2009.11.020CalvoSEPagliariniDJMoothaVK2009Upstream open reading frames cause widespread reduction of protein expression and are polymorphic among humansProceedings of the National Academy of Sciences of USA1067507751210.1073/pnas.0810916106CencicRGalicia-VázquezGPelletierJ2012Inhibitors of translation targeting eukaryotic translation initiation factor 4AMethods in Enzymology51143746110.1016/B978-0-12-396546-2.00020-6ChenTOzelDQiaoYHarbinskiFChenLDenoyelleSHeXZverevaNSupkoJGChorevMHalperinJAAktasBH2011Chemical genetics identify eIF2α kinase heme-regulated inhibitor as an anticancer targetNature Chemical Biology760861410.1038/nchembio.613Costa-MattioliMGobertDHardingHHerdyBAzziMBrunoMBidinostiMBen MamouCMarcinkiewiczEYoshidaMImatakaHCuelloACSeidahNSossinWLacailleJCRonDNaderKSonenbergN2005Translational control of hippocampal synaptic plasticity and memory by the eIF2alpha kinase GCN2Nature4361166117310.1038/nature03897Costa-MattioliMGobertDSternEGamacheKColinaRCuelloCSossinWKaufmanRPelletierJRosenblumKKrnjevićKLacailleJCNaderKSonenbergN2007eIF2alpha phosphorylation bidirectionally regulates the switch from short- to long-term synaptic plasticity and memoryCell12919520610.1016/j.cell.2007.01.050DamgaardCKLykke-AndersenJ2011Translational coregulation of 5'TOP mRNAs by TIA-1 and TIARGenes & Development252057206810.1101/gad.17355911DangYKedershaNLowWKRomoDGorospeMKaufmanRAndersonPLiuJO2006Eukaryotic initiation factor 2alpha-independent pathway of stress granule induction by the natural product pateamine AThe Journal of Biological Chemistry281328703287810.1074/jbc.M606149200DeverTEYangWAströmSByströmASHinnebuschAG1995Modulation of tRNA(iMet), eIF-2, and eIF-2B expression shows that GCN4 translation is inversely coupled to the level of eIF-2.GTP.Met-tRNA(iMet) ternary complexesMolecular and Cellular Biology1563516363Di PriscoGVHuangWBuffingtonSAHsuCCBonnenPEPlaczekANSidrauskiCKrnjevićKKaufmanRJWalterPCosta-MattioliM2014Translational control of mGluR-dependent long-term depression and object-place learning by eIF2αNature Neuroscience171073108210.1038/nn.3754GrantCMMillerPFHinnebuschAG1995Sequences 5' of the first upstream open reading frame in GCN4 mRNA are required for efficient translational reinitiationNucleic Acids Research233980398810.1093/nar/23.19.3980HardingHPNovoaIZhangYZengHWekRSchapiraMRonD2000Regulated translation initiation controls stress-induced gene expression in mammalian cellsMolecular Cell61099110810.1016/S1097-2765(00)00108-8HardingHPZhangYZengHNovoaILuPDCalfonMSadriNYunCPopkoBPaulesRStojdlDFBellJCHettmannTLeidenJMRonD2003An integrated stress response regulates amino acid metabolism and resistance to oxidative stressMolecular Cell1161963310.1016/S1097-2765(03)00105-9HinnebuschAG2005Translational regulation of GCN4 and the general amino acid control of yeastAnnual Review of Microbiology5940745010.1146/annurev.micro.59.031805.133833HinnebuschAGLorschJR2012The mechanism of eukaryotic translation initiation: new insights and challengesCold Spring Harbor Perspectives in Biology410.1101/cshperspect.a011544HsiehACLiuYEdlindMPIngoliaNTJanesMRSherAShiEYStumpfCRChristensenCBonhamMJWangSRenPMartinMJessenKFeldmanMEWeissmanJSShokatKMRommelCRuggeroD2012The translational landscape of mTOR signalling steers cancer initiation and metastasisNature485556110.1038/nature10912HurtaudCGellyCBouillaudFLévi-MeyrueisC2006Translation control of UCP2 synthesis by the upstream open reading frameCellular and Molecular Life Sciences631780178910.1007/s00018-006-6129-0IngoliaNTBrarGARouskinSMcGeachyAMWeissmanJS2012The ribosome profiling strategy for monitoring translation in vivo by deep sequencing of ribosome-protected mRNA fragmentsNature Protocols71534155010.1038/nprot.2012.086IngoliaNTGhaemmaghamiSNewmanJRWeissmanJS2009Genome-wide analysis in vivo of translation with nucleotide resolution using ribosome profilingScience32421822310.1126/science.1168978IngoliaNTLareauLFWeissmanJS2011Ribosome profiling of mouse embryonic stem cells reveals the complexity and dynamics of mammalian proteomesCell14778980210.1016/j.cell.2011.10.002JacksonRJHellenCUTPestovaTV2010The mechanism of eukaryotic translation initiation and principles of its regulationNature Reviews Molecular Cell Biology1111312710.1038/nrm2838JammiNVWhitbyLRBealPA2003Small molecule inhibitors of the RNA-dependent protein kinaseBiochemical and Biophysical Research Communications308505710.1016/S0006-291X(03)01318-4KatoMHanTWXieSShiKDuXWuLCMirzaeiHGoldsmithEJLonggoodJPeiJGrishinNVFrantzDESchneiderJWChenSLiLSawayaMREisenbergDTyckoRMcKnightSL2012Cell-free formation of RNA granules: low complexity sequence domains form dynamic fibers within hydrogelsCell14975376710.1016/j.cell.2012.04.017KedershaNChoMRLiWYaconoPWChenSGilksNGolanDEAndersonP2000Dynamic shuttling of TIA-1 accompanies the recruitment of mRNA to mammalian stress granulesThe Journal of Cell Biology1511257126810.1083/jcb.151.6.1257KedershaNAndersonP2009Chapter 4-Regulation of translation by stress granules and processing bodiesElsevier incKedershaNChenSGilksNLiWMillerIJStahlJAndersonP2002Evidence that ternary complex (eIF2-GTP-tRNA(i)(Met))-deficient preinitiation complexes are core constituents of mammalian stress granulesMolecular Biology of the Cell1319521010.1091/mbc.01-05-0221KedershaNTisdaleSHickmanTAndersonP2008Real-time and quantitative imaging of mammalian stress granules and processing bodiesMethods in Enzymology44852155210.1016/S0076-6879(08)02626-8KimHJKimNCWangYDScarboroughEAMooreJDiazZMacLeaKSFreibaumBLiSMolliexAKanagarajAPCarterRBoylanKBWojtasAMRademakersRPinkusJLGreenbergSATrojanowskiJQTraynorBJSmithBNToppSGkaziASMillerJShawCEKottlorsMKirschnerJPestronkALiYRFordAFGitlerADBenatarMKingODKimonisVERossEDWeihlCCShorterJTaylorJP2013Mutations in prion-like domains in hnRNPA2B1 and hnRNPA1 cause multisystem proteinopathy and ALSNature49546747310.1038/nature11922KimMSPintoSMGetnetDNirujogiRSMandaSSChaerkadyRMadugunduAKKelkarDSIsserlinRJainSJainSThomasJKMuthusamyBLeal-RojasPKumarPSahasrabuddheNABalakrishnanLAdvaniJGeorgeBRenuseSSelvanLDPatilAHNanjappaVRadhakrishnanAPrasadSSubbannayyaTRajuRKumarMSreenivasamurthySKMarimuthuASatheGJChavanSDattaKKSubbannayyaYSahuAYelamanchiSDJayaramSRajagopalanPSharmaJMurthyKRSyedNGoelRKhanAAAhmadSDeyGMudgalKChatterjeeAHuangTCZhongJWuXShawPGFreedDZahariMSMukherjeeKKShankarSMahadevanALamHMitchellCJShankarSKSatishchandraPSchroederJTSirdeshmukhRMaitraALeachSDDrakeCGHalushkaMKPrasadTSHrubanRHKerrCLBaderGDIacobuzio-DonahueCAGowdaHPandeyA2014A draft map of the human proteomeNature50957558110.1038/nature13302KrishnamoorthyTPavittGDZhangFDeverTEHinnebuschAG2001Tight binding of the phosphorylated alpha subunit of initiation factor 2 (eIF2alpha) to the regulatory subunits of guanine nucleotide exchange factor eIF2B is required for inhibition of translation initiationMolecular and Cellular Biology215018503010.1128/MCB.21.15.5018-5030.2001LeeSLiuBLeeSHuangSXShenBQianSB2012Global mapping of translation initiation sites in mammalian cells at single-nucleotide resolutionProceedings of the National Academy of Sciences of USA109E2424E243210.1073/pnas.1207846109LeeYYCevallosRCJanE2009An upstream open reading frame regulates translation of GADD34 during cellular stresses that induce eIF2alpha phosphorylationThe Journal of Biological Chemistry2846661667310.1074/jbc.M806735200LiYRKingODShorterJGitlerAD2013Stress granules as crucibles of ALS pathogenesisThe Journal of Cell Biology20136137210.1083/jcb.201302044LuMLawrenceDAMarstersSAcosta-AlvearDKimmigPMendezASPatonAWPatonJCWalterPAshkenaziA2014Cell death. Opposing unfolded-protein-response signals converge on death receptor 5 to control apoptosisScience3459810110.1126/science.1254312LuPDHardingHPRonD2004Translation reinitiation at alternative open reading frames regulates gene expression in an integrated stress responseThe Journal of Cell Biology167273310.1083/jcb.200408003MaXMBlenisJ2009Molecular mechanisms of mTOR-mediated translational controlNature Reviews Molecular Cell Biology1030731810.1038/nrm2672MandalSMandalAJohanssonHEOrjaloAVParkMH2013Depletion of cellular polyamines, spermidine and spermine, causes a total arrest in translation and growth in mammalian cellsProceedings of the National Academy of Sciences of USA1102169217410.1073/pnas.1219002110MazrouiRSukariehRBordeleauMEKaufmanRJNorthcotePTanakaJGallouziIPelletierJ2006Inhibition of ribosome recruitment induces stress granule formation independently of eukaryotic initiation factor 2alpha phosphorylationMolecular Biology of the Cell174212421910.1091/mbc.E06-04-0318MeyuhasO2000Synthesis of the translational apparatus is regulated at the translational levelEuropean Journal of Biochemistry2676321633010.1046/j.1432-1327.2000.01719.xMokasSMillsJRGarreauCFournierMJRobertFAryaPKaufmanRJPelletierJMazrouiR2009Uncoupling stress granule assembly and translation initiation inhibitionMolecular Biology of the Cell202673268310.1091/mbc.E08-10-1061MolletSCougotNWilczynskaADautryFKressMBertrandEWeilD2008Translationally repressed mRNA transiently cycles through stress granules during stressMolecular Biology of the Cell194469447910.1091/mbc.E08-05-0499PalamLRBairdTDWekRC2011Phosphorylation of eIF2 facilitates ribosomal bypass of an inhibitory upstream ORF to enhance CHOP translationJournal of Biological Chemistry286109391094910.1074/jbc.M110.216093PavittGDRamaiahKVKimballSRHinnebuschAG1998eIF2 independently binds two distinct eIF2B subcomplexes that catalyze and regulate guanine-nucleotide exchangeGenes & Development1251452610.1101/gad.12.4.514PavittGDYangWHinnebuschAG1997Homologous segments in three subunits of the guanine nucleotide exchange factor eIF2B mediate translational regulation by phosphorylation of eIF2Molecular and Cellular Biology1712981313PavittGDRonD2012New insights into translational regulation in the endoplasmic reticulum unfolded protein responseCold Spring Harbor Perspectives in Biology410.1101/cshperspect.a012278ReidDWChenQTayASShenolikarSNicchittaCV2014The unfolded protein response triggers selective mRNA release from the endoplasmic reticulumCell1581362137410.1016/j.cell.2014.08.012ScheunerDSongBMcEwenELiuCLaybuttRGillespiePSaundersTBonner-WeirSKaufmanRJ2001Translational control is required for the unfolded protein response and in vivo glucose homeostasisMolecular Cell71165117610.1016/S1097-2765(01)00265-9SidrauskiCAcosta-AlvearDKhoutorskyAVedanthamPHearnBRLiHGamacheKGallagherCMAngKKWilsonCOkreglakVAshkenaziAHannBNaderKArkinMRRensloARSonenbergNWalterP2013Pharmacological brake-release of mRNA translation enhances cognitive memoryeLife2e0049810.7554/eLife.00498SongZ2013Roles of the nucleotide sugar transporters (SLC35 family) in health and diseaseMolecular Aspects of Medicine3459060010.1016/j.mam.2012.12.004TabasIRonD2011Integrating the mechanisms of apoptosis induced by endoplasmic reticulum stressNature Cell Biology1318419010.1038/ncb0311-184TangHHornsteinEStolovichMLevyGLivingstoneMTempletonDAvruchJMeyuhasO2001Amino acid-induced translation of TOP mRNAs is fully dependent on phosphatidylinositol 3-kinase-mediated signaling, is partially inhibited by rapamycin, and is independent of S6K1 and rpS6 phosphorylationMolecular and Cellular Biology218671868310.1128/MCB.21.24.8671-8683.2001ThedieckKHolzwarthBPrentzellMTBoehlkeCKläsenerKRufSSonntagAGMaerzLGrellscheidS-NKremmerENitschkeRKuehnEWJonkerJWGroenAKRethMHallMNBaumeisterR2013Inhibition of mTORC1 by astrin and stress granules prevents apoptosis in cancer cellsCell15485987410.1016/j.cell.2013.07.031ThoreenCCChantranupongLKeysHRWangTGrayNSSabatiniDM2012A unifying model for mTORC1-mediated regulation of mRNA translationNature48510911310.1038/nature11083VanderweydeTYuHVarnumMLiu-YesucevitzLCitroAIkezuTDuffKWolozinB2012Contrasting pathology of the stress granule proteins TIA-1 and G3BP in tauopathiesJournal of Neuroscience328270828310.1523/JNEUROSCI.1592-12.2012VattemKMWekRC2004Reinitiation involving upstream ORFs regulates ATF4 mRNA translation in mammalian cellsProceedings of the National Academy of Sciences of USA101112691127410.1073/pnas.0400541101Vazquez de AldanaCRHinnebuschAG1994Mutations in the GCD7 subunit of yeast guanine nucleotide exchange factor eIF-2B overcome the inhibitory effects of phosphorylated eIF-2 on translation initiationMolecular and Cellular Biology1432083222WalterPRonD2011The unfolded protein response: from stress pathway to homeostatic regulationScience3341081108610.1126/science.1209038WekRCJiangHYAnthonyTG2006Coping with stress: eIF2 kinases and translational controlBiochemical Society Transactions34711WippichFBodenmillerBTrajkovskaMGWankaSAebersoldRPelkmansL2013Dual specificity kinase DYRK3 Couples stress granule condensation/dissolution to mTORC1 signalingCell15279180510.1016/j.cell.2013.01.033WooCWKutzlerLKimballSRTabasI2012Toll-like receptor activation suppresses ER stress factor CHOP and translation inhibition through activation of eIF2BNature Cell Biology1419220010.1038/ncb2408YuLKellyUEbrightJNMalekGSaloupisPRickmanDWMcKayBSArshavskyVYBowes RickmanC2007Oxidative stress-induced expression and modulation of Phosphatase of Regenerating Liver-1 (PRL-1) in mammalian retinaBiochimica Et Biophysica Acta17731473148210.1016/j.bbamcr.2007.06.005ZhouDPalamLRJiangLNarasimhanJStaschkeKAWekRC2008Phosphorylation of eIF2 directs ATF5 translational control in response to diverse stress conditionsThe Journal of Biological Chemistry2837064707310.1074/jbc.M708530200ZhuPJHuangWKalikulovDYooJWPlaczekANStoicaLZhouHBellJCFriedlanderMJKrnjevićKNoebelsJLCosta-MattioliM2011Suppression of PKR promotes network excitability and enhanced Cognition by Interferon-γ-mediated disinhibitionCell1471384139610.1016/j.cell.2011.11.02910.7554/eLife.05033.015Decision letterRonDavidReviewing editorUniversity of Cambridge, United Kingdom
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “The Small Molecule ISRIB Reverses eIF2α-phosphorylation-dependent Effects on Translation and Stress Granule Formation” for consideration at eLife. Your article has been evaluated by Randy Schekman (Senior editor) and three reviewers, one of whom is a member of our Board of Reviewing Editors.
The Reviewing editor and the other reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
All three reviewers recognized the significance of your attempt to further characterize the consequences of ISRIB application to stressed cells and the potential for such a study to serve as an important advance. The methodology you chose to evaluate the impact on mRNA translation by ribosome profiling and on stress granule formation, by imaging, was likewise deemed appropriate to the task. However, the individual reviews and the consultative process that followed uncovered important problems concerning the interpretability of the data and the validity of the conclusions drawn. These problems are rather pervasive and it will not be possible for the Reviewing editor to decide alone if a revised version would be suited for publication. Thus, if you decide to revise your paper in accordance with the stipulations below, please bear in mind that it will need to be reviewed again by all three reviewers.
1) Detailed information concerning the Ribo-Seq experiments is missing: How many million mapped reads and how many replicates were done for each sample? The worry that data set may be inadequate is compounded by an impression that the read density for those elements of data that are presented in detail (Figure 1–figure supplement 1, for example) is rather low.
2) Significance figures for RNA transcript abundance are not reported.
3) The full power of single base resolution of the RPF analysis is not harnessed to de-convolute the effects of stress and ISRIB on the translation of uORFs in ATF4 and ATF5. It should be possible to assign reads to either the uORF or the main ORF even when they are overlapping as the reading frame is different.
4) The central conclusions of the paper, that ISRIB eliminated the effects of stress on translational regulation and that ISRIB is equivalent to mutations that abolish the ISR are not supported by the data. Numerous mRNA are differentially engaged by ribosomes in stressed & ISRIB treated sample(s) (Figure 2C) compared to the stressed PERK-/- cells (the benchmark used here for an ISR-inhibited system, Figure 2A). These experiments may be further confounded by a methodological issue in that the cells shown in Figure 2C were apparently treated with tunicamycin for 1 hour, whereas those in Figure 2A for 30 minutes. These issue needs to be dealt with in some detail.
5) The effects of ISRIB on stress granule formation in the images shown seem convincing. But these need to be quantified and analyzed with appropriate statistical tools.
The reviewers’ individual comments are noted below.
Reviewer #1
This manuscript provides new experiments building on those published in a 2013 eLife article concerning the mechanism of action of an ISR inhibitor compounds called ISRIB.
The authors use complimentary techniques to those used in their original study to provide additional support to their model that ISRIB acts to nullify the downstream signalling events following induction of eIF2 phosphorylation. Two experimental strategies are employed here: 1) Ribosome profiling (Ribo-Seq) of cells to examine global translational control responses to ISR and its inhibition and 2) cell imaging via immunofluorescence and also live-cell GFP to examine the appearance of stress granules and P bodies-markers of cellular stress. Both of these analyses appear to confirm and extend the findings reported in the original study, but do not yet identify the mechanism by which the inhibitor ISRIB nullifies the ISR.
My major concerns are with the robustness of the data reported. In its present version this is not possible to fully assess.
1) For RNA and Ribo-Seq experiments, there is no information given concerning the depth of sequencing. How many million mapped reads and how many replicates were done for each sample?
2) The significant changes in ribosome occupancy are reported in supplementary tables, but changes in RNA transcript abundance are not reported. This data should also be added to the manuscript.
3) For the Ribo-Seq fold changes reported, how are reads mapping to overlapping ORFs handled? Eg Figure 1–figure supplement 1, shows reads covering the uORFs (green) and main ORFs (blue) that overlap.
4) Is the read density data in Figure 1–figure supplement 1 reporting a single representative sample, total summed reads from replicates (if done) or mean reads from multiple replicates? The read density is not very high for any of the genes shown, except for ATF4.
5) Can the authors confirm and comment on the presence of RPFs between the annotated uORF and main ORF for CHOP mRNA?
6) One perhaps unexpected finding was an apparent increase in uORF RPFs following stress. The authors have chosen not to elaborate on this point. The usual translational control models depicted often show an all or nothing extreme response, but it is not surprising at all that there are still ribosomes on uORFs that may get skipped more frequently under stress conditions. Perhaps an important measure of control is the proportion of ribosomes that skip uORF2 in ATF4 and 5. This could be quantified as the Ribo-Seq read data for the overlapping ORF regions should be in separate reading frames. It should be possible to assign reads to either the uORF or the main ORF. As a measure of stress mediated translational control, the authors could quantify the number and proportion of RPF within the ORFs key for translational control. For ATF4 and ATF5 ORF/uORF2 reads would give an alternative readout of the relative translational control.
7) Blots showing eIF2 phosphorylation relative to total eIF2 should be in each type of cell used (plus minus) treatment for the time points used in the Ribo Seq study. This could be added to the ATF4 blot which uses different time points.
8) For the immunofluorescence experiments and for the GFP experiments in Figures 3 and 4 and Figure 4–figure supplement 1, each experiment requires quantification of a larger number of cells than are shown in each representative image. For example in 100 cells in each of three separate experiments what proportion of cells (plus minus) error contain stress granules or P bodies as appropriate. Such an analysis would enable statistical treatment of the significance of the observations made.
9) What is the effect of the ISR and ISRIB on the other arms of the UPR at the times used here. Are sampling times too early to observe XBP1 splicing and changes in ribosome binding? If they are this should be stated, if they are not it would helpful to show XBP1 reads.
Reviewer #2
In this manuscript, the authors study the role of ISRIB, a small molecule targeting the integrated stress response (ISR) in two experimental systems: Ribosome profiling and stress-granules formation. The study of the ISR by ribosome profiling essentially confirms earlier studies. The effect of ISRIB and stress granules is surprising and novel, but the underlying mechanisms remain unknown. While the study is potentially interesting, there are some important issues to address.
Major issues:
Throughout the manuscript, there is a disconnection between the data and the conclusions. I will only highlight the major ones here:
1) One of the central conclusions of the paper is that ISRIB eliminated the effects of eIF2α phosphorylation. The authors wrote: “The translational output of ISRIB-treated cells to ER stress was remarkably similar to that of cells with a genetically ablated ISR”. This contradicts the data. In fact, there are major differences between the two data sets (Figure 2A and 2C). I recommend providing a complete list of genes that change after ISRIB treatment and to study these changes, as it may shed light on ISRIB's function and/or target.
Supporting the idea that the changes caused by ISRIB (Figure 2) are actually tractable, ISRIB caused more changes (Figure 2) than Tm (Figure 1A). The secret of ISRIB's function may be hidden in this dataset. This needs to be documented and exploited.
A related issue: The authors conclude that “ISRIB does not have general off-targets effects on translation, transcription or mRNA stability.”
Again, this contradicts the data. Beside, “off-target” is a not suitable here, since we don't know what is the target of ISRIB.
2) Figure 3B: What are the vertical lines on PERK blots? Between first and second lane in the top blot and between the 9th and 10th lane in the bottom blot.
3) Figure 3A: It would be good to see images with better resolution to appreciate the localization of eIF3α, which looks very interesting. I still see some stress-granules on Ars+ISRIB in some cells but in most cells, the signal is too strong to be analyzed. It would be good to see images of better resolution and some quantitative assessment of the effects.
4) The disappearance of stress-granules is the impressive finding of the manuscript but how does this all happen? It is difficult to follow what might be going on because of the lack of consistency in the conditions used and because adequate controls are not presented. From this paper and the previous one, it would appear that ISRIB acts upstream of ATF4 translation and is dependent upon eIF2α-P (but without affecting the latter)? This needs to be explored further to get some understanding of what is happening.
Figure 4C: It would be good to present gels to show the effects, as the authors have the images already. It would be interesting to appreciate the qualitative changes in translation. Does this match, what is seen by ribosome profiling?
5) The Discussion is very broad and not connected to the current dataset. The first sentence of the Discussion is incorrect, as there are other antagonists of the ISR. I suggest two options for a revised discussion: Either to focus the discussion on the current dataset or to provide additional data inline with the discussion (mTOR signaling and the ISR, memory and stress granules, memory and ribosome profiling data).
6) Figure 1B: I didn't understand this panel. It needs to be clarified.
Reviewer #3
This study provides an important extension validating the compound ISRIB as a highly specific inhibitor of the canonical eIF2α-phosphorylation-dependent Integrated Stress Response. Given that original report was published in eLife, this present paper is most suitable as a linked research advance.
The most important finding is that the effects of ISRIB on mRNA translation, revealed by the unbiased tool of ribosome foot-print profiling and mRNA sequencing, are limited to effacement of the translational induction of a small set (five in total) of mRNAs that are translationally upregulated by the ISR. In this regard, ISRIB discretely mimics the effects of mutations that block eIF2α phosphorylation, either by interfering with the action of the upstream kinase (PERK-KO) or by precluding substrate phosphorylation in cis (eIF2αS51A).
Further support for a comprehensive defect in the ISR, introduced by ISRIB, is provided by evidence that stress granules (these are mysterious collections of mRNA binding proteins and translation factors that assemble in cells experiencing high levels of phosphorylated eIF2α) are rapidly disassembled by ISRIB.
In passing, this paper makes an important contribution to the study of the ISR: By confirming that its known positively regulated targets, cobbled together from biased searches (ATF4, ATF5, CHOP, GADD34), comprise a nearly complete list; the single newcomer being SLC35A. And by drawing attention to the fact that their translational induction by the ISR does not entail the loss of footprints on the short repressive uORFs (as predicted by the regulated translation re-initiation model).
An important caveat to these complementary comments is that this reviewer lacks the expertise to judge the technical validity of the RNA seq and ribosome footprinting and defers to other reviewers' expertise in this regard.
The lack of any measureable effect of ISRIB on baseline translation is surprising as there is evidence for basal activity of the ISR: For example PERK_KO, eIF2a_S51A and ATF4_KO cells all share a strong baseline requirement for amino acid supplementation (likely indicating that the ISR contributes to baseline ATF4-mediated gene expression). The authors may wish to comment on this point.
10.7554/eLife.05033.016Author response
All three reviewers recognized the significance of your attempt to further characterize the consequences of ISRIB application to stressed cells and the potential for such a study to serve as an important advance. The methodology you chose to evaluate the impact on mRNA translation by ribosome profiling and on stress granule formation, by imaging, was likewise deemed appropriate to the task. However, the individual reviews and the consultative process that followed uncovered important problems concerning the interpretability of the data and the validity of the conclusions drawn. These problems are rather pervasive and it will not be possible for the Reviewing editor to decide alone if a revised version would be suited for publication. Thus, if you decide to revise your paper in accordance with the stipulations below, please bear in mind that it will need to be reviewed again by all three reviewers.
By now delivering a more focused message that directly relates to the parent eLife paper, we feel that these changes substantially improve the clarity and quality of this contribution and render it more appropriate as a Research advance.
We have made the following major changes in the manuscript:
1) To align the paper better with the previous publication, we excluded the Ribo-Seq and mRNA-seq data generated in mouse embryonic fibroblasts. Due to the intrinsic differences in kinetics and degree of UPR induction between MEFs and HEK293T cells, we now focus only on the ribosome profiling data generated in the latter. As reported in our previous eLife manuscript (10.7554/eLife.00498), in HEK293T cells, we observe a complete block both in ATF4 translational induction and bulk translational down regulation upon co-treatment with ISRIB, making it our cell line of choice to study the genome-wide effects of ISRIB. As requested by the reviewers, we include biological replicates for both Ribo-Seq and paired mRNA seq-data in all conditions reported in this cell type. Correlation coefficients between replicates and mRNA abundance for all genes are provided in new supplementary figures.
2) We included uORF ribosome occupancy for only the two highest expressed ISR-translational targets for which we obtained sufficient coverage in reads for both mRNA and Ribo-Seq data. We removed the uORF ribosome occupancy upon ER-stress plots as new insights pertaining to the mechanism of translation and rescanning of the uORFs in translationally stress-induced genes will require substantial further experimentation and would extend beyond the scope of this work. The primary goal of this Research Advance is to further characterize the biological effects of the small molecule ISRIB.
3) We have quantitated all stress granule data.
As requested, we have reworded the title to read “The Small Molecule ISRIB Reverses the Effects of eIF2α Phosphorylation on Translation and Stress Granule Assembly”.
1) Detailed information concerning the Ribo-Seq experiments is missing: How many million mapped reads and how many replicates were done for each sample? The worry that data set may be inadequate is compounded by an impression that the read density for those elements of data that are presented in detail (Figure 1–figure supplement 1, for example) is rather low.
The Ribo-Seq and mRNA-seq data in HEK293T cells was performed in duplicate (biological replicates) for each condition. Figure 1–figure supplement 7 contains the number of reads, number of reads mapped and the number of reads mapped to ORFs for each replicate sample. We will prepare a GEO submission for all datasets.
We have now included only ATF4 and SLC35A4, the two highest expressed ISR-targets, in our ribosome footprinting and mRNA read density plots along the genes (now Figure 1–figure supplement 3), for which we have sufficient coverage.
2) Significance figures for RNA transcript abundance are not reported.
We have included mean normalized counts for all transcripts in all conditions tested in Figure 1–figure supplement 9.
3) The full power of single base resolution of the RPF analysis is not harnessed to de-convolute the effects of stress and ISRIB on the translation of uORFs in ATF4 and ATF5. It should be possible to assign reads to either the uORF or the main ORF even when they are overlapping as the reading frame is different.
Due to the fact that all Ribo-Seq data in this manuscript was generated using cycloheximide to freeze translating ribososomes, and this may lead to an artifactual increase in 5’UTR ribosome occupancy, we have not quantitatively analyzed changes in uORF occupancy. Carefully de-convoluting uORF occupancy upon stress and ISRIB treatment will require further experimentation (including the use of a translation initiation inhibitor treatment to generate libraries). This analysis extends beyond the scope of the current Research Advance.
4) The central conclusions of the paper, that ISRIB eliminated the effects of stress on translational regulation and that ISRIB is equivalent to mutations that abolish the ISR are not supported by the data. Numerous mRNA are differentially engaged by ribosomes in stressed & ISRIB treated sample(s) (Figure 2C) compared to the stressed PERK-/-cells (the benchmark used here for an ISR-inhibited system,Figure 2A). These experiments may be further confounded by a methodological issue in that the cells shown inFigure 2Cwere apparently treated with tunicamycin for 1 hour, whereas those inFigure 2Afor 30 minutes. These issue needs to be dealt with in some detail.
We have excluded the MEF data, and, because of differences observed between cell lines, we no longer benchmark PERK-/- or eIF2α as an ISR-inhibited system. Importantly, we show that ISRIB comprehensively blocks translational upregulation of the vast majority of the mRNAs in HEK293T cells, including the well-established ISR translational targets ATF4, ATF5, CHOP and GADD34. Both the previously submitted manuscript as well as the current one contain tables that list for each condition the translationally upregulated mRNAs.
5) The effects of ISRIB on stress granule formation in the images shown seem convincing. But these need to be quantified and analyzed with appropriate statistical tools.
We have quantified all stress granule data and added the corresponding quantitation panels to Figures 2 and 3 with their corresponding statistics in each figure legend.
The reviewers’ individual comments are noted below.
Reviewer #1
[…] My major concerns are with the robustness of the data reported. In its present version this is not possible to fully assess.
1) For RNA and Ribo-Seq experiments, there is no information given concerning the depth of sequencing. How many million mapped reads and how many replicates were done for each sample?
We have included biological replicates for all Ribo-Seq and mRNA-seq data for all conditions. Figure 1–figure supplement 7 contains the reads mapped information and Figure 1–figure supplement 8 shows the correlation coefficient between replicates.
2) The significant changes in ribosome occupancy are reported in supplementary tables, but changes in RNA transcript abundance are not reported. This data should also be added to the manuscript.
RNA transcript abundance is shown in Figure 1–figure supplement 9.
3) For the Ribo-Seq fold changes reported, how are reads mapping to overlapping ORFs handled? EgFigure 1–figure supplement 1, shows reads covering the uORFs (green) and main ORFs (blue) that overlap.
We removed Ribo-Seq-fold changes of uORFs from the manuscript.
4) Is the read density data inFigure 1–figure supplement 1reporting a single representative sample, total summed reads from replicates (if done) or mean reads from multiple replicates? The read density is not very high for any of the genes shown, except for ATF4.
The read density represents the total reads of two replicates. We have only included ATF4 and SLC35A4, which have sufficient overall higher read density to draw strong conclusions.
5) Can the authors confirm and comment on the presence of RPFs between the annotated uORF and main ORF for CHOP mRNA?
The data have been removed.
6) One perhaps unexpected finding was an apparent increase in uORF RPFs following stress. The authors have chosen not to elaborate on this point. The usual translational control models depicted often show an all or nothing extreme response, but it is not surprising at all that there are still ribosomes on uORFs that may get skipped more frequently under stress conditions. Perhaps an important measure of control is the proportion of ribosomes that skip uORF2 in ATF4 and 5. This could be quantified as the Ribo-Seq read data for the overlapping ORF regions should be in separate reading frames. It should be possible to assign reads to either the uORF or the main ORF. As a measure of stress mediated translational control, the authors could quantify the number and proportion of RPF within the ORFs key for translational control. For ATF4 and ATF5 ORF/uORF2 reads would give an alternative readout of the relative translational control.
The data are no longer presented in the manuscript. The analysis of remaining uORF occupancy in the face of downstream ORF translation will require extensive additional studies.
7) Blots showing eIF2 phosphorylation relative to total eIF2 should be in each type of cell used (plus minus) treatment for the time points used in the Ribo-Seq study. This could be added to the ATF4 blot which uses different time points.
The ATF4 blot has been removed as it related to the MEFs data. We have reported time courses of eIF2α phosphorylation and ATF4 production in HEK293T cells in Figure 3–figure supplement 1 of our original eLife paper (10.7554/eLife.00498). We do not detect phospho-eIF2α phosphorylation or ATF4 production after only 1 hour of Tm treatment by Western blot analysis, which is the condition used for ribo and mRNA-seq analysis. Ribo-Seq analysis is a more sensitive method than Western blotting to detect the early translational changes upon ER-stress.
8) For the immunofluorescence experiments and for the GFP experiments inFigures 3and 4 and Figure 4–figure supplement 1, each experiment requires quantification of a larger number of cells than are shown in each representative image. For example in 100 cells in each of three separate experiments what proportion of cells (plus minus) error contain stress granules or P bodies as appropriate. Such an analysis would enable statistical treatment of the significance of the observations made.
The quantitation was added to the stress granule data.
9) What is the effect of the ISR and ISRIB on the other arms of the UPR at the times used here. Are sampling times too early to observe XBP1 splicing and changes in ribosome binding? If they are this should be stated, if they are not it would helpful to show XBP1 reads.
At the time points analyzed, it is too early to detect changes in XBP1 mRNA splicing. In Figure 3–figure supplement 1 of our original eLife paper (10.7554/eLife.00498), we looked at XBP1s induction upon Tm treatment in HEK293T cells. XBP1s production lags ATF4 translational upregulation and is not detected at 1 hour of Tm treatment. We also reported in our previous manuscript that ISRIB treatment or blocking the PERK branch of the UPR does not alter activation of IRE1 and XBP1 splicing but it prolongs activation of this branch as measured both by IRE1-GFP foci formation and XBP1 splicing (Figure 5C and Figure 5–figure supplement 2).
Reviewer #2
[…] While the study is potentially interesting, there are some important issues to address.
Major issues:
Throughout the manuscript, there is a disconnection between the data and the conclusions. I will only highlight the major ones here:
1) One of the central conclusions of the paper is that ISRIB eliminated the effects of eIF2α phosphorylation. The authors wrote: “The translational output of ISRIB-treated cells to ER stress was remarkably similar to that of cells with a genetically ablated ISR”. This contradicts the data. In fact, there are major differences between the two data sets (Figure 2A and 2C). I recommend providing a complete list of genes that change after ISRIB treatment and to study these changes, as it may shed light on ISRIB's function and/or target.
We have removed the MEF data and thus we do no longer compare ISR-ablated cells with ISRIB-treated cells. In both the original submitted eLife Research Advance and in the revised manuscript, we have provided a list of genes that are significantly and substantially translationally upregulated in all conditions reported (Tm, Tm + ISRIB and ISRIB alone). We have carefully analyzed the genes that remain translationally upregulated in the presence of Tm + ISRIB. Unfortunately they are all hypothetical proteins with unknown functions and thus do not shed light on ISRIB function. Some of these uncharacterized genes are also down regulated in the presence of ISRIB alone.
Supporting the idea that the changes caused by ISRIB (Figure 2) are actually tractable, ISRIB caused more changes (Figure 2) than Tm (Figure 1A). The secret of ISRIB's function may be hidden in this dataset. This needs to be documented and exploited.
As stated above, ISRIB changes have been documented. Compared to Tm treatment, ISRIB induced changes are small.
A related issue: The authors conclude that “ISRIB does not have general off-targets effects on translation, transcription or mRNA stability.” Again, this contradicts the data. Beside, “off-target” is a not suitable here, since we don't know what is the target of ISRIB.
We have removed off-target from the text. The ribosome and mRNA genome-wide profiling data demonstrates that ISRIB comprehensively blocks the translational effects driven by UPR activation and eIF2α phosphorylation and does not lead to overall and spurious changes in mRNA levels or translation.
2)Figure 3B: What are the vertical lines on PERK blots? Between first and second lane in the top blot and between the 9th and 10th lane in the bottom blot.
All lanes were contiguous in the blots. The film was rescanned and the lines are no longer present.
3)Figure 3A: It would be good to see images with better resolution to appreciate the localization of eIF3α, which looks very interesting. I still see some stress-granules on Ars+ISRIB in some cells but in most cells, the signal is too strong to be analyzed. It would be good to see images of better resolution and some quantitative assessment of the effects.
Quantitative assessment was included for all stress granule data. Immunofluorescence data of eIF3α with all stressors used in this paper has been previously published.
4) The disappearance of stress-granules is the impressive finding of the manuscript but how does this all happen? It is difficult to follow what might be going on because of the lack of consistency in the conditions used and because adequate controls are not presented. From this paper and the previous one, it would appear that ISRIB acts upstream of ATF4 translation and is dependent upon eIF2α-P (but without affecting the latter)? This needs to be explored further to get some understanding of what is happening.
The goal of using different stressors was to distinguish between phospho-eIF2α-dependent (thapsigargin and arsenite) and independent (panteamine A and hippuristanol) stress granule formation. As expected by the ability of ISRIB to make cells resistant to the effects of eIF2α phosphorylation, the small molecule was able to block only phospho-eIF2α-dependent stress granule formation. ATF4 is downstream of eIF2α phosphorylation and thus it is also blocked by ISRIB. The mechanism of action of the drug will await identification of its molecular target.
Figure 4C: It would be good to present gels to show the effects, as the authors have the images already. It would be interesting to appreciate the qualitative changes in translation. Does this match, what is seen by ribosome profiling?
We have included the gel in Figure 3–figure supplement 2. The live cell imaging and 35S-methionine incorporation and recovery experiments were performed in U2OS cells expressing G3BP-GFP using the potent ER-stressor, thapsigargin. It is not the same cell type or the same ER-stressor (tunicamycin) used in the ribosome footprinting analysis.
5) The Discussion is very broad and not connected to the current dataset. The first sentence of the Discussion is incorrect, as there are other antagonists of the ISR. I suggest two options for a revised discussion: Either to focus the discussion on the current dataset or to provide additional data inline with the discussion (mTOR signaling and the ISR, memory and stress granules, memory and ribosome profiling data).
We have substantially rewritten the Discussion. To the best of our knowledge however, ISRIB is indeed the first and to date only known ISR inhibitor. Although there are inhibitors of specific eIF2α kinases (PERK and PKR inhibitors), ISRIB is the only molecule that can block signaling downstream of all eIF2α kinases by making cells resistant to eIF2α phosphorylation. An ISR agonist exists, salubrinal, which prolongs eIF2α phosphorylation. An HRI activator also exists which can also act as an agonist of the ISR in cells that express HRI.
6)Figure 1B: I didn't understand this panel. It needs to be clarified.
This figure was removed.
Reviewer #3
[…] The lack of any measureable effect of ISRIB on baseline translation is surprising as there is evidence for basal activity of the ISR: For example PERK_KO, eIF2a_S51A and ATF4_KO cells all share a strong baseline requirement for amino acid supplementation (likely indicating that the ISR contributes to baseline ATF4-mediated gene expression). The authors may wish to comment on this point.
HEK293T cells were only treated with ISRIB for one hour and thus the basal levels of ISR activation may not be as evident as in cells under prolonged ablation of PERK (PERK-/-), ATF4 (ATF4-/-) or non-phosphorylatable eIF2α (eIF2α S51A/S51A).
diff --git a/elife05216.xml b/elife05216.xml
new file mode 100644
index 0000000..b3a3d1e
--- /dev/null
+++ b/elife05216.xml
@@ -0,0 +1 @@
+elifeeLifeeLifeeLife2050-084XeLife Sciences Publications, Ltd0521610.7554/eLife.05216Short reportDevelopmental biology and stem cellsNeuroscienceEye morphogenesis driven by epithelial flow into the optic cup facilitated by modulation of bone morphogenetic proteinHeermannStephan12*†SchützLucas1LemkeSteffen1KrieglsteinKerstin2WittbrodtJoachim1*Centre for Organismal Studies Heidelberg, Ruprecht Karls Universität, Heidelberg, GermanyDepartment of Molecular Embryology, Institute of Anatomy and Cell Biology, University Freiburg, Freiburg, GermanyWhitfieldTanya TReviewing editorUniversity of Sheffield, United KingdomFor correspondence: stephan.heermann@cos.uni-heidelberg.de (SH);jochen.wittbrodt@cos.uni-heidelberg.de (JW)
Anatomie und Zellbiologie, Heidelberg University, Heidelberg
The hemispheric, bi-layered optic cup forms from an oval optic vesicle during early vertebrate eye development through major morphological transformations. The overall basal surface, facing the developing lens, is increasing, while, at the same time, the space basally occupied by individual cells is decreasing. This cannot be explained by the classical view of eye development. Using zebrafish (Danio rerio) as a model, we show that the lens-averted epithelium functions as a reservoir that contributes to the growing neuroretina through epithelial flow around the distal rims of the optic cup. We propose that this flow couples morphogenesis and retinal determination. Our 4D data indicate that future stem cells flow from their origin in the lens-averted domain of the optic vesicle to their destination in the ciliary marginal zone. BMP-mediated inhibition of the flow results in ectopic neuroretina in the RPE domain. Ultimately the ventral fissure fails to close resulting in coloboma.
DOI:http://dx.doi.org/10.7554/eLife.05216.001
10.7554/eLife.05216.002eLife digest
The eye is our most important organ for sensing and recognizing our environment. In humans and other vertebrates, the eye forms from an outgrowth of the brain as the embryo develops. This outgrowth is called the optic vesicle and it is rapidly transformed into a cup-shaped structure known as the optic cup. Defects in this process prevent the optic cup from closing completely, which leads to a severe condition called Coloboma—one of the most frequent causes of blindness in children.
The optic cup has two distinct layers: the inside layer—known as the neuroretina—contains light sensitive cells and is surrounded by the other layer called the pigmented epithelium. It is thought that the neural retina is made from cells from the side of the optic vesicle that faces the lens, and the pigmented epithelium is formed by cells from the other side of the vesicle. This is a plausible explanation and is well accepted, but it cannot explain how the neuroretina can become five times larger as the cup forms.
Heermann et al. addressed this problem by using four-dimensional in vivo microscopy to follow individual cells as the optic cup forms in living zebrafish embryos. The experiments show that the neuroretina is made of cells from both sides of the optic vesicle. Cells from the back of the optic vesicle (furthest away from the lens) join the rest of the cells by moving around the outside rim of the cup.
Further experiments found that a signaling molecule called BMP—which is crucial to the normal development of the eye—controls the flow of cells around the developing optic cup. This factor needs to be carefully controlled during the development of the eye; when BMP activity was artificially increased, the flow of cells stopped, resulting in neuroretinal tissue developing in the wrong place (in the outer layer of the optic cup).
The experiments also reveal that the stem cells in the retina—which divide to produce new cells throughout the life of the zebrafish—originate from two distinct areas in the optic vesicle.
Heermann et al.'s findings challenge the textbook model of eye development by revealing that cells from both sides of the optic vesicle contribute to the neuroretina and that retinal stem cells originate from a specific place in the developing eye. A future challenge will be to understand how the movement of the cells into the neuroretina is coordinated to make a perfectly shaped eye.
DOI:http://dx.doi.org/10.7554/eLife.05216.002
Author keywordsoptic vesicleoptic cupBMP antagonistneuroretinal flowoptic fissurecolobomaResearch organismzebrafishhttp://dx.doi.org/10.13039/501100001659Deutsche ForschungsgemeinschaftDFG/ Molecular mechanisms regulating optic fissure closureKrieglsteinKerstinWittbrodtJoachimhttp://dx.doi.org/10.13039/501100000781European Research Council (ERC)Advanced Grant (AdG), LS3, ERC-2011-ADGWittbrodtJoachimThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementThe lens-averted domains of the optic vesicle are reservoirs of neuroretinal cells that flow into the developing optic cup in a process that is critically influenced by BMP signaling.Main Text
The bi-layered optic vesicles of vertebrates are formed through a bilateral evagination of the late prosencephalon. In teleosts, this process is driven by a migration of single cells that undergo a subsequent intercalation into the epithelium of the expanding optic vesicle (Rembold et al., 2006, England et al., 2006, Sinn and Wittbrodt, 2013, Ivanovitch et al., 2013). The oval optic vesicle develops into a hemispheric bi-layered optic cup through a process that involves major morphological transformations. A long-held view of this process proposes that the lens-averted epithelium of the optic vesicle differentiates into the retinal pigmented epithelium (RPE), while the epithelium facing the lens gives rise to the neuroretina, which subsequently bends around the developing lens (Chow and Lang, 2001; Fuhrmann, 2010; Walls, 1942). This neuroepithelial bending is driven by a basal constriction of lens-facing retinal progenitor cells (RPC) (Martinez-Morales et al., 2009) (Bogdanović et al., 2012), which ultimately reduces the space occupied by an individual RPC at the basal surface. However, we observed that this is accompanied by a 4.7-fold increase in the overall basal optic cup surface area (Figure 1A–C). To identify the cellular origin of this massive increase, we performed in vivo time-lapse microscopy in zebrafish at the corresponding stages, starting at 16.5 hpf (Figure 1D–L, Video 1), in a transgenic line expressing a membrane-coupled GFP in retinal stem and progenitor cells (Rx2::GFPcaax).10.7554/eLife.05216.003
Neuroretinal surface increases during optic cup formation by epithelial flow.
(A) Scheme showing the orientation of the pictures presented in B–L. (B) Basal neuroretinal surface increases from early to late optic cup stage (dashed yellow lines). (C) Basal neuroretinal surface was measured in 3D (superimposed orange lines), although RPCs undergo basal constriction during optic cup formation, the surface increases 4.7 fold from early to late optic cup stage, (D–L) transition from optic vesicle to optic cup over time, shown at a ventral (D–F), a central (G–I), and a dorsal (J–L) level. The membrane localized GFP is driven by an rx2 promoter (rx2::GFPcaax), which is active in RPCs. The optic vesicle is bi-layered (D, G, J) with a prospective lens-facing (arrows in D and E) and a prospective lens-averted (arrowheads in D, G, J) epithelium, connected to the forebrain by the optic stalk (os in D), at a ventral level both are connected at the distal site (circle in D), at a central level both are connected distally and proximally (circles in G), notably the morphology of the lens-averted epithelium at a dorsal level is different from central and ventral levels (arrowhead in J). Over time at ventral and central levels (D–F and G–I, respectively), the lens-averted epithelium is being integrated into the forming optic cup (arrowheads in D, E, G, H, I and arrow in H). A patch of cells in the lens-averted domain gives rise to the RPE (asterisks in H and arrowhead in J, K), le: developing lens, os: optic stalk, B and D–L were derived from 4D imaging data starting at 16.5 hpf (D, G, J), one focal plane is presented as Video 1, scalebar in B and C = 50 µm.
DOI:http://dx.doi.org/10.7554/eLife.05216.003
10.7554/eLife.05216.004
(related to <xref ref-type="fig" rid="fig1">Figure 1</xref>) (control) Optic vesicle to optic cup transition visualized with rx2::GFPcaax (orientation as in <xref ref-type="fig" rid="fig1">Figure 1</xref>) (imaging starts at 16.5 hpf, framerate 1/15 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.004
Strikingly, and in contrast to the former model (Chow and Lang, 2001; Fuhrmann, 2010; Walls, 1942), our analysis shows that almost the entire bi-layered optic vesicle gives rise to the neural retina (Figure 1D–I), with the marked exception of a small lens-averted patch (see below). The majority of the lens-averted epithelium (Figure 1D,G, between arrowheads) serves as a neuro-epithelial reservoir, which eventually is fully integrated into the lens-facing neuro-epithelium (Video 1). This occurs through a sheet-like flow of lens-averted cells into the forming optic cup (Figure 1E,H). This epithelial flow is independent of cell proliferation (Figure 2—figure supplement 1, Video 2) as demonstrated by aphidicolin treatment. The process is highly reminiscent of gastrulation movements and explains the marked increase of the lens-facing basal neuroretinal surface area. Notably, a small domain of the lens-averted epithelium exhibits a different morphology and behavior. As optic cup formation proceeds, this region flattens, enlarges, exhibits the morphological characteristics of RPE, and eventually ceases expressing RX2, a marker for retinal stem and early progenitor cells (Figure 1H, asterisks, Video 3, in between arrows).10.7554/eLife.05216.005
(related to <xref ref-type="fig" rid="fig1">Figure 1</xref> and <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>) Aphidicolin treated embryo. (imaging started at 17 hpf, framerate 1/10 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.005
10.7554/eLife.05216.006
(related to <xref ref-type="fig" rid="fig1">Figure 1</xref>) (control) Optic vesicle to optic cup transition visualized by H2BGFP RNA into rx2::GFPcaax (orientation as in <xref ref-type="fig" rid="fig2">Figure 2</xref>), data derived from same imaging data as <xref ref-type="other" rid="video4">Video 4</xref>, 3D rendered. Arrows mark the border between future RPE and Neuroretina (imaging starts at 16.5 hpf, framerate 1/10 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.006
Our data highlight that almost the entire optic vesicle contributes to the formation of the neural retina. This new perspective on optic cup formation raises the question of how the elongated oval optic vesicle is transformed into the hemispheric optic cup. We addressed this by 4D imaging of optic cup formation using a nuclear label (H2BGFP) (Figure 2A). We found, concomitant with lens formation, a prominent epithelial flow around the temporal perimeter of the forming optic cup. An involution of cells from the domain of the retinal pigmented epithelium (RPE) into the domain of the neuroretina had been proposed (Li et al., 2000). Such reorganization of the lens-averted and the lens-facing epithelia, affecting the temporal optic cup, has been subsequently described (Picker et al., 2009) and confirmed (Kwan et al., 2012). It was proposed that such ‘rim movements’ could occur around most of the optic vesicle circumference (Kwan et al., 2012).10.7554/eLife.05216.007
Neuroepithelial flow drives morphological changes from optic vesicle to optic cup: the role of the optic fissure and the impact on the forming stem cell domain.
(A) Dorsal view on optic cup development over time visualized by mosaic nuclear GFP (H2BGFP) (data are derived from 4D imaging data started at 16.5 hpf, one optical section is provided as Video 4), while the lens-facing neuroepithelium is starting to engulf the developing lens (asterisk), the lens-averted epithelium is largely integrated into the lens-facing epithelium by flowing around the distal nasal and temporal rims (arrows). A white dotted line marks the border between lens-facing and lens-averted epithelium. (B) Scheme showing the key findings of A, the lens (asterisk) facing epithelium is marked with red bars. The lens-averted epithelium, which over time is integrated into the lens-facing epithelium is marked with green dots (except the cells at the edges are additionally marked with a yellow core). In between the last cells, which are integrated into the optic cup, the RPE will form in the lens averted domain. (C) shows the percentage of movements with a considerable share in dorso-ventral direction for the dorsal, central, and ventral area of the developing eye. In the ventral area of the eye, there is significantly more movement in the dorso-ventral axis, than in the central or dorsal area. (D) scheme demonstrating the optic vesicle to optic cup transition (lateral view). Notably, the morphological change from the elongated oval optic vesicle to the hemispheric optic cup is driven mainly by the ventral regions (arrows mark the orientation of epithelial flow) (C and D).
DOI:http://dx.doi.org/10.7554/eLife.05216.007
10.7554/eLife.05216.008
Epithelial flow is independent of cell division.
(A) Retinal cell division was inhibited by application of aphidicolin, a well-established DNA polymerase inhibitor. Aphidicolin efficiently inhibited cell proliferation shown by drastically reduced pHH3 positive nuclei (upper panel, average of 6 pHH3 positive nuclei) compared to the control (lower panel, average of 21 pHH3 positive nuclei) (the optic cup is encircled with a dotted white line, 21.5 hpf). (B) We addressed the epithelial flow of aphidicolin-treated wild-type embryos, injected with H2BGFP RNA at the one cell stage; please see Figure 2A as control. The embryo was preincubated with aphidicolin 5 hr prior to the start of imaging (17 hpf, see also Video 2). The application of aphidicolin did not affect the epithelial flow. As a low level side effect of aphidicolin we observed cell death, in line with previous reports, importantly also not affecting the epithelial flow.
DOI:http://dx.doi.org/10.7554/eLife.05216.008
Our data confirm a flow around the temporal perimeter and additionally demonstrate epithelial flow around the nasal perimeter into the forming optic cup. We uncover that the direction of the epithelial flow primarily establishes two distinct neuroretinal domains (nasal and temporal) separated by the static dorsal and ventral poles of the forming eye (Figure 2D, Figure 3A). We use these poles as dorsal and ventral reference points throughout the manuscript. Importantly, the prominent rotation of the eye cup only occurs after the epithelial flow has ceased (24–36 hpf, Schmitt and Dowling, 1994).10.7554/eLife.05216.009
Development of the CMZ and quantification of the flow towards this domain.
(A) scheme of optic cup development (lateral view over time) including the results of nuclear tracking from the presumptive CMZ back in time to the lens-averted epithelium, remarkably two distinct domains became apparent within the lens-averted epithelium as the source for the presumptive CMZ. (B) Establishment of the presumptive CMZ domain (dorsal view), nuclear tracking of cells (maximum projection) from the lens-averted domain (encircled in upper picture) eventually residing in the forming CMZ (additionally encircled in lower picture), scalebar = 50 µm. (C) Scheme showing the optic cup from the lateral side. For quantification four domains were selected, nasal–dorsal, nasal–ventral, temporal–dorsal, and temporal–ventral. Note that the dorsal distal domain is only assembled secondarily and the ventral pole shows the optic fissure. (D) Based on differential effective distance, effective speed, and directionality, the migration distance was divided in two phases in the nasal and temporal domain, respectively.
DOI:http://dx.doi.org/10.7554/eLife.05216.009
The prospective RPE remains in the lens-averted domain and expands in conjunction with the bi-furcated flow of the neuroretina from the lens-averted into the lens-facing domain (Figure 2A,B, Video 3). To further address the transformation of the elongated, oval optic vesicle into the hemispheric optic cup, we quantified cellular movements along the dorso-ventral axis. We found that the most prominent movements leading to the extension in the dorsal ventral axis occurred in the ventral domain (Figure 2C). A key step in the formation of the ventral neuroretina is the formation of the optic fissure at the ventral pole of the optic vesicle. Lens-averted epithelium flows through this fissure into the forming optic cup to constitute the ventral neuroepithelium (Figure 2D). Taken together, we present a model of optic cup formation, driven by gastrulation-like epithelial flow from the lens-averted into the lens-facing epithelium of the forming optic cup. The epithelium flows in two domains around the temporal and nasal rim, respectively and through the optic fissure of the forming optic cup. Overall, this has far-reaching implications for different aspects of eye development. One is the establishment of the retinal stem cell niche in the ciliary marginal zone (CMZ) (Centanin et al., 2011), the distal rim of the optic cup/retina.
To address whether the CMZ domain originates from a mixed population of progenitor cells that have been ‘set aside’, or from a predefined coherent domain, we analyzed the transition from optic vesicle to optic cup in 3D over time (4D) (Video 4). By tracking individual cells, we identified the origin of the distal retinal domain, the future CMZ, as two distinct domains (nasal and temporal) within the lens-averted epithelium at the optic vesicle stage (Figure 3A,B, Video 5). Based on tracking information, we noticed distinct phases during the flow from the lens-averted domain towards the CMZ (Figure 3D). Although cells show high motility in an early phase (Figure 3D, 1), the directed flow is established only in a later phase (Figure 3D, 2), in which cells ultimately flow to the rim of the forming optic cup (Figure 3D).10.7554/eLife.05216.010
(related to <xref ref-type="fig" rid="fig2">Figure 2</xref>) (control) Optic vesicle to optic cup transition visualized by H2BGFP RNA into rx2::GFPcaax (orientation as in <xref ref-type="fig" rid="fig2">Figure 2</xref>) (imaging starts at 16.5 hpf, framerate 1/10 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.010
10.7554/eLife.05216.011
(related to <xref ref-type="fig" rid="fig2">Figure 2</xref>) (control) Optic vesicle to optic cup transition visualized by H2BGFP RNA into rx2::GFPcaax (orientation as in <xref ref-type="fig" rid="fig2">Figure 2</xref>), data as in <xref ref-type="other" rid="video4">video 4</xref> with tracked cells (maximum projection) to the presumptive CMZ.
DOI:http://dx.doi.org/10.7554/eLife.05216.011
As indicated above, the dorsal pole of the optic vesicle remains static (Figure 2D). Thus, the presumptive dorsal CMZ domain either originates from the lens-facing neuroretina or, alternatively, is established secondarily at a later time point, like the ventral CMZ in the region of the optic fissure. The identification of lens-averted domains as the source of the future nasal and temporal CMZ is consistent with the hypothesis of a distinct origin of retinal stem cells. Our data support a scenario in which the entire optic vesicle is initially composed of stem cells that at the lens-facing side respond to a signal to take a progenitor fate.
We propose a tight coupling of morphogenesis with cell determination by inductive signals derived from the surface ectoderm to explain the successive spreading of retinal differentiation from the center to the periphery (Sinn and Wittbrodt, 2013). Accordingly, lens-averted stem cells might retain their stem cell fate because they are exposed to that signal at the latest point in time. An alternative hypothesis is that stemness might require an active process at the interface to the RPE; it is also possible that both scenarios are involved. Both scenarios are consistent with the expression pattern of rx2, which is initially found in the entire optic vesicle and subsequently is confined to the CMZ. Strikingly, rx2-positive cells of the CMZ represent multipotent retinal stem cells (Reinhardt, Centanin et al., submitted).
We demonstrated that cell motility and thus tissue fluidity are a prerequisite for neuroretinal flow. These characteristics are likely maintained through signaling, raising the question of which system might be involved. A likely candidate might be BMP, which has been linked to mobility in other tissues during development. In heart jogging, for example, BMP has an ‘antimotogenic’ effect (Veerkamp et al., 2013). BMP signaling is important for various aspects of vertebrate eye development such as the enhancement of RPE and the inhibition of optic cup/neuroretina development (Fuhrmann et al., 2000; Hyer et al., 2003; Müller et al., 2007; Steinfeld et al., 2013), the formation of the dorso-ventral axis (Behesti et al., 2006; Holly et al., 2014; Koshiba-Takeuchi et al., 2000; Sasagawa et al., 2002), and the induction of the optic fissure (Morcillo et al., 2006). Specific regions of the eye also seem to depend on the modulation of BMP signaling by the expression of a BMP antagonist (Sakuta et al., 2001, French et al., 2009).
We analyzed BMP signaling activity by assays based on the phosphorylation of the Smads 1/5/8 and the activation of a BMP signaling reporter (Laux et al., 2011). BMP signaling is mainly elevated in the temporal domain and to a lesser degree in the nasal domain of the optic vesicle (16.5hpf, Figure 4A,B,D,E). At 21.5 hpf BMP signaling is confined to the dorsal pole of the optic cup (Figure 4C,F). The transcriptional BMP sensor is activated with a delay and shows a more confined area of activity (compare Figure 4A–C to Figure 4D–F).10.7554/eLife.05216.012
Analyses of BMP signaling and expression of BMP antagonists during development at 16.5 hpf, 19 hpf, and 21 hpf embryos are presented in a lateral view nasal left.
(A–C) pSmad 1/5/8 immunohistochemistry (red) and DAPI nuclear staining. Activated BMP signaling can be appreciated mainly in the temporal domain of the optic vesicle (arrows) (A–B) and in the dorsal domain of the optic cup (arrows) (C). At 16.5 hpf, a small domain of activated BMP signaling is visible in the nasal optic vesicle (arrows) (A). (D–F) Immunohistochemically enhanced BRE::GFP (green) and DAPI nuclear staining. Activated BMP signaling can be appreciated in the temporal late optic vesicle (arrows) (E) and the dorsal optic cup (arrows) (F). Hardly any activity can be detected in the optic vesicle at 16.5 hpf. Note the delay of activity in comparison to pSmad 1/5/8. (G–I) Whole mount in situ hybridizations with a fsta probe (Fast Red) and DAPI nuclear staining. In the optic vesicle as well as in the optic cup two domains (nasal and temporal) of fsta expression can be seen (arrows). (J–L) Whole mount in situ hybridizations with a bambia probe (Fast Red) and DAPI nuclear staining. Bambi expression can be seen in the temporal domain of the optic vesicle (arrows) (J–K) and in the dorsal domain of the optic cup (arrows) (L).
DOI:http://dx.doi.org/10.7554/eLife.05216.012
To address the means by which BMP activity is restricted, we analyzed the activity of prominent BMP antagonists follistatin a (fsta) (Thompson et al., 2005), and bambi (bambia) (French et al., 2009). Fsta was expressed in two domains, a nasal and a temporal domain (Figure 4G–I and Figure 5A), whereas bambi was only expressed in the temporal domain of the optic vesicle (Figure 4J,K) and the dorsal domain of the optic cup (Figure 4L). The regions of fsta expression correspond to the domains showing neuroretinal flow during optic cup formation.10.7554/eLife.05216.013
BMP antagonism drives neuroepithelial flow during optic cup formation.
(A) whole mount in situ hybridization for fsta (NBT/BCIP) (17.5 hpf). (B) GFP expressed in the optic vesicle (arrows) of an rx2::GFPcaax zebrafish embryo (16.5 hpf), (C–D) GFP driven by the BRE and transmission/brightfield image for orientation. Strong GFP expression can be observed in the eye when BMP is driven under rx2 (arrows in D), whereas only mild GFP can be observed in controls (arrows in C). (E) Scheme showing the orientation of the pictures presented in F, (F) optic cup development over time of an rx2::BMP4 embryo. Cells are visualized by nuclear GFP (H2BGFP). A dotted line is indicating the border between lens-averted and lens-facing epithelium. Remarkably, the pan-ocular driven BMP resulted in persisting lens-averted domains. The data presented in F are derived from 4D imaging data (start at 16.6 hpf) one optical section is also presented as Video 6.
DOI:http://dx.doi.org/10.7554/eLife.05216.013
To address the importance of localized BMP signaling in wild-type embryos, we expressed BMP4 in the entire eye using an Rx2 proximal cis regulatory element (Figure 5B), which overrides the localized BMP antagonist in the optic vesicle and optic cup.
In BMP reporter fish (Laux et al., 2011), we addressed BMP signaling activity under control and experimental conditions. At the optic cup stage, moderate BMP signaling activity was observed in the dorsal retina of control fish (Figures 4F and 5C). The pan-ocular expression of BMP4 resulted in a strong response of the reporter, indicating pan-ocular BMP4 signaling (Figure 5D).
Strikingly resembling the BMP dependent ‘antimotogenic’ effect (Veerkamp et al., 2013), pan-ocular BMP expression arrested epithelial flow during optic cup formation. Time-lapse in vivo microscopy revealed that cells in the lens-averted part of the future neuroretina remained in the prospective RPE domain and did not contribute to the optic cup (Video 6 and 7). This ultimately resulted in an apparently ectopic domain of neuroretina that arose from a morphogenetic failure, rather than from a trans-differentiation of RPE (Figure 6D, Fig. 6—figure supplement 1,2,3, Videos 8,9,10). The severity of the phenotype correlated well with levels of fsta expression in the optic vesicle and was most prominent in the temporal domain of the optic vesicle. These findings highlight the importance of the modulation of BMP signaling for epithelial fluidity during the transformation from optic vesicle to optic cup. We propose that the repression of BMP signaling is crucial to mobilize the lens-averted retinal epithelium, causing it to flow and eventually constitute the neural retina to a large extent.10.7554/eLife.05216.014
(related to <xref ref-type="fig" rid="fig4">Figure 4</xref>) (rx2::BMP4) Optic vesicle to optic cup transition visualized by H2BGFP RNA (orientation as in <xref ref-type="fig" rid="fig3">Figure 3</xref>) (imaging starts at 16.5 hpf, framerate 1/15 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.014
10.7554/eLife.05216.015
(related to <xref ref-type="fig" rid="fig4">Figure 4</xref>) (rx2::BMP4) Optic vesicle to optic cup transition visualized by rx2::GFPcaax (orientation as in <xref ref-type="fig" rid="fig3">Figure 3</xref>) (imaging starts at 19 hpf, framerate 1/15 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.015
10.7554/eLife.05216.016
Impaired eye gastrulation results in coloboma.
(A–B) Membrane-localized GFP (rx2::GFPcaax) in a developing eye during optic fissure closure (A = early, B = late) (lateral view, derived from imaging data, (A) start at 24 hpf, (B) after 34 hr imaging at 22°C). Rx2 is expressed in retinal stem cells/RPCs (A) and after NR differentiation is additionally expressed in photoreceptors and Müller Glia (B) while its expression is maintained in retinal stem cells of the CMZ (Reinhardt and Centanin et al., submitted). The optic fissure margins are still undifferentiated (arrows in B), (C) developing eye of rx2::BMP4 fish (lateral view), membrane-localized GFP (rx2::GFPcaax, anti-GFP immunointensified), DAPI nuclear labeling and anti-laminin immunostaining, the optic fissure is visible, noteworthy the temporal retina is mis-shaped and folded into the RPE domain (best visible in DAPI, arrowheads), and located on a basal membrane (arrowheads in anti-laminin), especially the temporal optic fissure margin (arrowheads in GFPcaax) is located in the folded part of the temporal retina and not facing the optic fissure (arrows in GFPcaax) (24 hpf) (D–E) impaired optic fissure closure in rx2::BMP4 embryos over time at a proximal (E) and a distal (D) level. (Data obtained from 4D imaging of rx2 ::BMP4/ rx2::GFPcaax started at 21.5 hr. Data are also presented as Video 10.) Importantly, next to the affected temporal optic cup also the nasal optic cup is mis-folded (arrowheads in D). Remarkably, however, the nasal optic fissure margin extents into the optic fissure (dashed arrow in E) but the temporal optic fissure margin does not, likely being the result of the intense mis-bending of the temporal optic cup. This results in a remaining optic fissure (asterisk in E). (F–G) Brightfield images of variable phenotype intensities observed in rx2::BMP4 hatchlings.
DOI:http://dx.doi.org/10.7554/eLife.05216.016
10.7554/eLife.05216.017
Postembryonic eye development of rx2::BMP4 hatchlings.
Although the temporal optic cup is largely malformed and folded it can be seen clearly, that vsx1 as well as vsx2 transgenes (intensified by wholemount immunohistochemistry) are expressed in the folded epithelium (arrows). This indicates at least a partial correct differentiation into neuroretinal tissue.
DOI:http://dx.doi.org/10.7554/eLife.05216.017
10.7554/eLife.05216.018
Lateral view on optic cup development over time, rx2::GFPcaax (control) is compared to rx2::BMP4 at proximal and distal levels.
While in controls the lens-averted domain (yellow dotted line) is integrated into the developing optic cup it persists in rx2::BMP4 embryos. Note the increasing optic fissure (arrows) in rx2::BMP4 embryos. These data were obtained by 4D imaging (start at 20 hpf) and are also presented as Video 8 and Video 9.
DOI:http://dx.doi.org/10.7554/eLife.05216.018
10.7554/eLife.05216.019
Dorsal view on optic cup development of an rx2::BMP4 embryo over time at ventral vs central/ dorsal levels.
The yellow dotted line indicates the border between the lens-facing and the lens-averted domain. Remarkably, the lens-averted domain is not integrated into the optic cup (compare to Figures 1 and 2). Notably, an altered morphology of the ventral optic vesicle can be observed showing the domain which is not going to be integrated (arrows). These data are derived from imaging data (start at 19 hpf) which is presented in Video 7.
DOI:http://dx.doi.org/10.7554/eLife.05216.019
10.7554/eLife.05216.020
Ventral retinal identity remains in rx2::BMP4 embryos.
Whole mount in situ hybridization with a vax2 probe (NBT/BCIP) of control (left) and rx2::BMP4 embryos (right) in a lateral view (upper pictures) and a ventral view (lower pictures) (28 hpf). Note that the ventral retinal marker remains expressed in the forming ventral optic cup even if BMP4 is expressed panocularly (rx2::BMP4). Also note that the vax2 domain in control is broader than in the rx2::BMP4 embryos (arrows) in which it is more prominent in the optic stalk region (arrowhead) (the dotted line indicates the optic fissure).
DOI:http://dx.doi.org/10.7554/eLife.05216.020
10.7554/eLife.05216.021
Control to <xref ref-type="other" rid="video9">video 9</xref>, optic cup development recorded with rx2::GFPcaax (lateral view) (imaging starts at 20 hpf, framerate 1/10 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.021
10.7554/eLife.05216.022
(rx2::BMP4) Optic cup development recorded with rx2::GFPcaax (lateral view) (imaging starts at 20 hpf, framerate 1/10 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.022
10.7554/eLife.05216.023
Proximal domain of an rx2::BMP4 embryo showing an impaired optic fissure closure (orientation as in <xref ref-type="fig" rid="fig5">Figure 5E</xref>) (imaging starts at 21.5 hpf, framerate 1/10 min).
DOI:http://dx.doi.org/10.7554/eLife.05216.023
We further investigated the implications of impaired epithelial flow for subsequent steps of eye development (e.g., fate of the optic fissure). After initiation of neuroretinal differentiation in control embryos, the undifferentiated domains are restricted to the un-fused optic fissure margins and the forming CMZ. Both can be visualized by the expression of Rx2 (Figure 6A,B). The impairment of neuroretinal flow, however, resulted in a mis-organization of the optic fissure. Here, the undifferentiated Rx2-expressing domain was found at the ultimate tip of the lens-averted neuroretinal domain, which failed to flow into the optic cup and persisted in the prospective RPE (Figure 6C). As a result, the temporal optic fissure margin, in particular, failed to extend into the optic fissure (Figure 6D–E). This also holds true, but to a lesser extent, to the nasal optic fissure margin (Figure 6D). As a result, the two fissure margins cannot converge resulting in a persisting optic fissure, a coloboma. Macroscopically, the pan-ocular expression of BMP4 results in phenotypes including a ‘Plattauge’ (flat-eye) (Figure 6G), in which the ventral part of the eye is strongly affected and a milder phenotype (Figure 6F), in which the ventral retina develops, but with a persisting optic fissure.
It was previously shown that exposing the developing eye to high levels of ectopically applied BMP can cause dorsalization, concomitant with a loss of ventral cell identities (Behesti et al., 2006). This is likely the cause for coloboma (Behesti et al., 2006; Koshiba-Takeuchi et al., 2000, Sasagawa et al., 2002). Our data based on stable BMP4 expression (rx2::BMP4) in the entire optic vesicle, however, conclusively show that early BMP4 exposure arrests neuroepithelial flow, resulting in a morphologically affected ventral retina. The ventral expression of vax2 in optic cups of rx2::BMP4 embryos indicates the maintenance of ventral retinal fates and argues against early transdifferentiation/dorsalization induced by BMP (Figure 6—figure supplement 4). Remarkably, the remaining lens-averted domain of those embryos, which was ectopically localized and was not integrated into the optic cup, eventually differentiated into neuroretina (Figure 6—figure supplement 1), as indicated by the expression of vsx1 (Kimura et al., 2008; Shi et al., 2011; Vitorino et al., 2009) and vsx2 (formerly Chx10) (Vitorino et al., 2009). Notably, a localization of neuroretina within the RPE domain might be mistaken for an RPE to neuroretina trans-differentiation, as proposed for other phenotypes (Araki et al., 2002; Azuma et al., 2005; Sakaguchi et al., 1997, Bankhead et al., 2015).
Even in amniotes, the histological analyses of consecutive stages of optic cup development are best interpreted as epithelial flow that also enlarges the retinal surface. This can even be appreciated during in vitro optic cup formation using mammalian embryonic stem cells (Eiraku et al., 2011).
Taken together, our data clearly show that during optic vesicle to optic cup transformation, the lens-averted part of the optic vesicle is largely integrated into the lens-facing optic cup by flowing around the distal rim of the optic cup including the forming optic fissure. Our data have far-reaching implications on the generation of the retinal stem cell niche of teleosts, as the last cells flowing into the optic cup will eventually constitute the CMZ. We identify a part of the lens-averted epithelium as the primary source of the RPE. The arrest of neuroepithelial flow by the ‘antimotogenic’ effect of BMP (Veerkamp et al., 2013) results in coloboma and thus highlights the importance of the flow through the fissure for the establishment of the ventral optic cup.
It is unlikely that the bending of the neuroretina provides the motor for the epithelial flow; in the opo mutant no ectopic neuroretina can be found, indicating that the flow persists, even in the absence of optic cup bending (Bogdanović et al., 2012). Consequently, forces established outside the neuroretina are likely to drive the flow. One tissue potentially involved is the mono-layered-forming RPE. We speculate that this tissue contributes to the flow by changing its shape from a columnar to a flat epithelium, massively enlarging its surface (Figure 1J–K, Video 3). This remains an interesting point, in particular given that epithelial flow is maintained even if cell proliferation is inhibited in both neuroretina and RPE.
Materials and methodsTransgenic zebrafish and Injections
BMP4 was cloned via directional Gateway from zebrafish cDNA into a pEntr D-TOPO (Invitrogen, Germany) vector with the following primers: forw: 5′ CACCGTCTAGGGATCCCTTGTTCTTTTTGCAGCCGCCACCATGATTCCTGGTAATCGAATGCTG 3′, rev: 5′ TTAGCGGCA GCCACACCCCTCGACCAC 3′.
The expression construct was assembled via a Gateway reaction using Tol2 destination vector containing a cmlc: GFP (Kwan et al., 2007), a 5′Entry vector containing an rx2 promoter (Martinez-Morales et al., 2009), the vector containing the BMP4 and a 3′Entry vector containing a pA sequence (Kwan et al., 2007). The construct was co-injected with mRNA encoding Tol2 transposase into the cytoplasm of zebrafish eggs at the one cell stage. Stable lines were preselected based on GFP expression in the heart (cmlc2::GFP), raised and validated in F1 and subsequent generations. Lines were maintained as closed stocks and crossed to other lines as indicated in the manuscript.
The rx2::GFPcaax construct was assembled with the 5′ and 3′ components described above and GFPcaax in the pEntr D-topo vector via Gateway (Invitrogen) and co-injected with mRNA encoding Tol2 transposase into the cytoplasm of zebrafish eggs at the one cell stage. Stable lines were preselected based on GFP expression in the heart (cmlc2::GFP), raised, and validated in F1 and subsequent generations. Lines were maintained as closed stocks and crossed to other lines as indicated in the manuscript.
The BRE::GFP zebrafish line (Laux et al., 2011) was kindly provided by Beth Roman. The Vsx1::GFP zebrafish line (Kimura et al., 2008; Shi et al., 2011; Vitorino et al., 2009) was kindly provided by Lucia Poggi. The Vsx2::RFP zebrafish line (Vitorino et al., 2009) was kindly provided by the lab of William Harris.
Where indicated RNA for H2BGFP (nuclear localized GFP) (37 ng/µl) was injected into 1–8 cell staged zebrafish embryos enabling 4D imaging of mosaically nuclear labeled zebrafish.
Drug treatment with aphidicolin
Zebrafish embryos were treated with aphidicolin (10 µg/ml, Serva, Germany) in order to inhibit cell proliferation. 12 embryos were treated with aphidicolin. 4D imaging was performed on one with an aphidicolin pretreatment of 5 hr. The efficacy of the treatment was addressed by analyzing nuclei in mitosis (positive for the expression of phospho-histone H3. At 21.5 hpf pHH3 positive nuclei were counted in central sections of four control (untreated embryos from the same batch) (average: 21) and experimental (average: 6) retinae, respectively.
Quantification of optic cup surface
Optic cup surfaces were measured with the help of FIJI (ImageJ NIH software). The mean of the length of the measured lines (Figure 1C) of two adjacent optical sections was multiplied by the optical section interval.
Microscopy
Confocal data of whole mount immunohistochemical stainings a Leica (Germany) SPE microscope was used. Samples were mounted in glass bottom dishes (MaTek, Ashland, MA). Olympus (Germany) stereomircoscope was used for recording brightfield images of rx2::BMP4 hatchlings and the overview of the expression of rx2::GFPcaax. For whole mount in situ data acquisition, a Zeiss (Germany) microscope was used. Time-lapse imaging was performed with a Leica SP5 setup which was upgraded to a multi photon microscope (Mai Tai laser, Spectra Physics, Germany). It was recorded in single photon modus and multi photon modus. For time-lapse imaging, embryos were embedded in 1% low melting agarose and covered with zebrafish medium, including tricaine for anesthesia. Left and right eyes were used and oriented to fit the standard dorsal view or view from the side.
Whole mount in situ hybridization
Whole mount in situ hybridization was performed with probes for fsta bambia and vax2. The probes were selfmade. Sequences were amplified by PCR from zebrafish cDNA and subcloned into pGEMTeasy vector (Promega, Germany). In vitro transcription was performed with Sp6/T7 Polymerase. Hybridization was largely performed according to Quiring et al. (2004). The Probe bas visualized with NBT/BCIP (Roche, Switzerland) or Fast Red (Roche) as indicated.
Whole mount immunohistochemistry
Immunohistochemistry was performed according to a standard whole mount immunohistochemistry protocol. Briefly, embryos/hatchlings were fixed, washed, bleached (KOH/H2O2 in PTW), and blocked (BSA [1%], DMSO [1%], Triton X-100 [0.1%], NGS [4%], PBS [1×]). In case of anti-pSmad 1/5/8 immunohistochemistry embryos were additionally treated with proteinase K (10 µg/ml, 16.5 hpf: 5 min, 19 hpf and 21.5 hpf: 6 min). Samples were incubated in primary antibody solution (anti-laminin, 1:50, Abcam, Germany) (anti-GFP 1:200, life technologies, Germany) (anti-dsRED, Clontech, Germany) (anti-pHH3, 1:100, Milipore, Germany) (anti-pSmad1/5/8, 1:25, Cell Signaling, Germany) in blocking solution. Samples were washed and incubated in secondary antibody solution (anti-rabbit Dylight, 1:300, anti-chicken Alexa 488, 1:300, Jackson, UK) with DAPI (stock: 2 µg/ml, 1:500) added. Consecutively, samples were washed and mounted for microscopy.
Quantification of dorso-ventral movement
The amount of movement in the dorso-ventral axis was quantified using a supervoxel based Optical Flow algorithm (Amat et al., 2013). The pixel wise output was visualized by applying a spherical coordinate system to the eye using a custom made ImageJ plugin (Source code 1: file plugin). The color coding is based on the sign of the polar angle theta and the sign of the azimuth angle phi, as well as on their respective combinations. The quantification was performed by counting the labeled pixels in an ImageJ macro (Source code 1: file macro).
Cell tracking
Cells were tracked manually using MtrackJ (Meijering et al., 2012) in Fiji (ImageJ) (Schindelin et al., 2012) back in 4D stacks to their original location or until lost. Only tracks with a significant length were used for the visualizations. Centered on the track cells are represented as spheres. Partially results are presented in a side view where the dorso-ventral axis originally represented as the z-axis has now become the y-axis. A factor of 10.5703 is introduced in order to adjust the data of the former z-axis to the other two axes. The color coding is done by choosing colors from an 8 bit lookup table and applying them from the dorsal to the ventral side based on the end of the track. Partially tracking results are presented as tailed spheres. The spheres are based on the tracking data using an average over the last three timepoints. The image is stretched in the z-axis using a factor of 10.5703, to adjust the scale to the x and y axes. Tails are created using a lookup table with 16 different shades per color. The respective shade is defined by the distance and difference in time between the recent position and the position on the tail.
Acknowledgements
We want to thank Karin Schumacher for generously providing space for finishing the work on this manuscript, Lea Schertel for excellent technical assistance, members of the Wittbrodt lab for material (P Haas and A Schmidt for the rx2: GFP construct and zebrafish line, B Höckendorf and M Stemmer for H2BGFP mRNA, B Höckendorf for FIJI related input) and constructive stimulating feedback, B Roman for the BRE: GFP zebrafish, William Harris for the vsx2: RFP zebrafish line and Lucia Poggi for making the vsx1: GFP zebrafish line available. We also want to thank Russ Hodge for valuable input on editing the manuscript. LS is recipient of a fellowship from the Hartmut-Hoffman-Berling International Graduate School (HBIGS). This work was supported by the Deutsche Forschungsgemeinschaft (JW, KK, SL) and the ERC (JW).
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
SH, Final approval of the version to be published, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
LS, Final approval of the version to be published, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
SL, Final approval of the version to be published, Analysis and interpretation of data, Drafting or revising the article
KK, Final approval of the version to be published, Conception and design, Drafting or revising the article
JW, Final approval of the version to be published, Conception and design, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents
Additionl files10.7554/eLife.05216.024
Source Code .zip contains: Plugin: imageJ plugin for visualization of dorso-ventral movements (please see ‘Materials and methods’ section: quantification of dorso-ventral movement). Macro: imageJ macro for counting of labeled pixels (please see ‘Materials and methods’ section: quantification of dorso-ventral movement).
DOI:http://dx.doi.org/10.7554/eLife.05216.024
ReferencesAmatFMyersEWKellerPJ2013Fast and robust optical flow for time-lapse microscopy using super-voxelsBioinformatics2937338010.1093/bioinformatics/bts706ArakiMTakanoTUemonsaTNakaneYTsudzukiMKanekoT2002Epithelia-mesenchyme interaction plays an essential role in transdifferentiation of retinal pigment epithelium of silver mutant quail: localization of {FGF} and related molecules and aberrant migration pattern of neural crest cells during eye rudiment forDevelopmental Biology24435837110.1006/dbio.2002.0591AzumaNTadokoroKAsakaAYamadaMYamaguchiYHandaHMatsushimaSWatanabeTKidaYOguraTToriiMShimamuraKNakafukuM2005Transdifferentiation of the retinal pigment epithelia to the neural retina by transfer of the Pax6 transcriptional factorHuman Molecular Genetics141059106810.1093/hmg/ddi098BankheadEJColasantoMPDyorichKMJamrichMMurtaughLCFuhrmannS2015Multiple requirements of the focal dermal hypoplasia gene porcupine during ocular morphogenesisThe American Journal of Pathology18519721310.1016/j.ajpath.2014.09.002BehestiHHoltJKSowdenJC2006The level of BMP4 signaling is critical for the regulation of distinct T-box gene expression domains and growth along the dorso-ventral axis of the optic cupBMC Developmental Biology66210.1186/1471-213X-6-62BogdanovićODelfino-MachínMNicolás-PérezMGavilánMPGago-RodriguesIFernández-MiñánALilloCRíosRMWittbrodtJMartínez-MoralesJR2012Numb/Numbl-Opo antagonism controls retinal epithelium morphogenesis by regulating integrin endocytosisDevelopmental Cell2378279510.1016/j.devcel.2012.09.004CentaninLHoeckendorfBWittbrodtJ2011Fate restriction and multipotency in retinal stem cellsCell Stem Cell955356210.1016/j.stem.2011.11.004ChowRLLangRA2001Early eye development in vertebratesAnnual Review of Cell and Developmental Biology1725529610.1146/annurev.cellbio.17.1.255EirakuMTakataNIshibashiHKawadaMSakakuraEOkudaSSekiguchiKAdachiTSasaiY2011Self-organizing optic-cup morphogenesis in three-dimensional cultureNature472515610.1038/nature09941EnglandSJBlanchardGBMahadevanLAdamsRJ2006A dynamic fate map of the forebrain shows how vertebrate eyes form and explains two causes of cyclopiaDevelopment1334613461710.1242/dev.02678FrenchCREricksonTFrenchDVPilgrimDBWaskiewiczAJ2009Gdf6a is required for the initiation of dorsal-ventral retinal patterning and lens developmentDevelopmental Biology333374710.1016/j.ydbio.2009.06.018FuhrmannSLevineEMRehTA2000Extraocular mesenchyme patterns the optic vesicle during early eye development in the embryonic chickDevelopment12745994609FuhrmannS2010Eye morphogenesis and patterning of the optic vesicleCurrent Topics in Developmental Biology93618410.1016/B978-0-12-385044-7.00003-5HollyVLWidenSAFamulskiJKWaskiewiczAJ2014Sfrp1a and Sfrp5 function as positive regulators of Wnt and {BMP} signaling during early retinal developmentDevelopmental Biology38819220410.1016/j.ydbio.2014.01.012HyerJKuhlmanJAfifEMikawaT2003Optic cup morphogenesis requires pre-lens ectoderm but not lens differentiationDevelopmental Biology25935136310.1016/S0012-1606(03)00205-7IvanovitchKCavodeassiFWilsonSW2013Precocious acquisition of neuroepithelial character in the eye field underlies the onset of eye morphogenesisDevelopmental Cell2729330510.1016/j.devcel.2013.09.023KimuraYSatouCHigashijimaSI2008V2a and V2b neurons are generated by the final divisions of pair-producing progenitors in the zebrafish spinal cordDevelopment1353001300510.1242/dev.024802Koshiba-TakeuchiKTakeuchiJKMatsumotoKMomoseTUnoKHoepkerVOguraKTakahashiNNakamuraHYasudaKOguraT2000Tbx5 and the retinotectum projectionScience28713413710.1126/science.287.5450.134KwanKMFujimotoEGrabherCMangumBDHardyMECampbellDSParantJMYostHJKankiJPChienCB2007The Tol2kit: a multisite gateway-based construction kit for Tol2 transposon transgenesis constructsDevelopmental Dynamics2363088309910.1002/dvdy.21343KwanKMOtsunaHKidokoroHCarneyKRSaijohYChienCB2012A complex choreography of cell movements shapes the vertebrate eyeDevelopment13935937210.1242/dev.071407LauxDWFebboJARomanBL2011Dynamic analysis of BMP-responsive smad activity in live zebrafish embryosDevelopmental Dynamics24068269410.1002/dvdy.22558LiZJosephNMEasterSS2000The morphogenesis of the zebrafish eye, including a fate map of the optic vesicleDevelopmental Dynamics21817518810.1002/(SICI)1097-0177(200005)218:1<175::AID-DVDY15>3.0.CO;2-KMartinez-MoralesJRRemboldMGregerKSimpsonJCBrownKEQuiringRPepperkokRMartin-BermudoMDHimmelbauerHWittbrodtJ2009ojoplano-mediated basal constriction is essential for optic cup morphogenesisDevelopment1362165217510.1242/dev.033563MeijeringEDzyubachykOSmalI2012Methods for cell and particle trackingMethods Enzymol50418320010.1016/B978-0-12-391857-4.00009-4MorcilloJMartínez-MoralesJRTrousseFFerminYSowdenJCBovolentaP2006Proper patterning of the optic fissure requires the sequential activity of BMP7 and SHHDevelopment1333179319010.1242/dev.02493MüllerFRohrerHVogel-HöpkerA2007Bone morphogenetic proteins specify the retinal pigment epithelium in the chick embryoDevelopment1343483349310.1242/dev.02884PickerACavodeassiFMachateABernauerSHansSAbeGKawakamiKWilsonSWBrandM2009Dynamic coupling of pattern formation and morphogenesis in the developing vertebrate retinaPLOS Biology7e100021410.1371/journal.pbio.1000214QuiringRWittbrodtBHenrichTRamialisonMBurgtorfCLehrachHWittbrodtJ2004Large-scale expression screening by automated whole-mount in situ hybridizationMechanisms of Development12197197610.1016/j.mod.2004.03.031RemboldMLoosliFAdamsRJWittbrodtJ2006Individual cell migration serves as the driving force for optic vesicle evaginationScience3131130113410.1126/science.1127144SakaguchiDSJanickLMRehTA1997Basic fibroblast growth factor (FGF-2) induced transdifferentiation of retinal pigment epithelium: generation of retinal neurons and gliaDevelopmental Dynamics20938739810.1002/(SICI)1097-0177(199708)209:4<387::AID-AJA6>3.0.CO;2-ESakutaHSuzukiRTakahashiHKatoAShintaniTIemuraSiYamamotoTSUenoNNodaM2001Ventroptin: a BMP-4 antagonist expressed in a double-gradient pattern in the retinaScience29311111510.1126/science.1058379SasagawaSTakabatakeTTakabatakeYMuramatsuTTakeshimaK2002Axes establishment during eye morphogenesis in Xenopus by coordinate and antagonistic actions of BMP4, Shh, and RAGenesis33869610.1002/gene.10095SchindelinJArganda-CarrerasIFriseEKaynigVLongairMPietzschTPreibischSRuedenCSaalfeldSSchmidBTinevezJYWhiteDJHartensteinVEliceiriKTomancakPCardonaA2012Fiji: an open-source platform for biological-image analysisNature Methods967668210.1038/nmeth.2019SchmittEADowlingJE1994Early eye morphogenesis in the zebrafish, Brachydanio rerioJournal of Comparative Neurology34453254210.1002/cne.903440404ShiZTrenholmSZhuMBuddinghSStarENAwatramaniGBChowRL2011Vsx1 regulates terminal differentiation of type 7 ON bipolar cellsThe Journal of Neuroscience31131181312710.1523/JNEUROSCI.2331-11.2011SinnRWittbrodtJ2013An eye on eye developmentMechanisms of Development13034735810.1016/j.mod.2013.05.001SteinfeldJSteinfeldICoronatoNHampelMLLayerPGArakiMVogel-HöpkerA2013RPE specification in the chick is mediated by surface ectoderm-derived BMP and Wnt signallingDevelopment1404959496910.1242/dev.096990ThompsonTBLerchTFCookRWWoodruffTKJardetzkyTS2005The structure of the follistatin:activin complex reveals antagonism of both type I and type II receptor bindingDevelopmental Cell953554310.1016/j.devcel.2005.09.008VeerkampJRudolphFCseresnyesZPrillerFOttenCRenzMSchaeferLAbdelilah-SeyfriedS2013Unilateral dampening of Bmp activity by nodal generates cardiac left-right asymmetryDevelopmental Cell2466066710.1016/j.devcel.2013.01.026VitorinoMJusufPRMaurusDKimuraYHigashijimaSHarrisWA2009Vsx2 in the zebrafish retina: restricted lineages through derepressionNeural Development41410.1186/1749-8104-4-14WallsGL1942The vertebrate eye and its adaptive radiationNew York LondonHafner Publishing Company 196310.7554/eLife.05216.025Decision letterWhitfieldTanya TReviewing editorUniversity of Sheffield, United Kingdom
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Epithelial flow into the optic cup facilitated by suppression of BMP drives eye morphogenesis” for consideration at eLife. Your article has been favorably evaluated by Diethard Tautz (Senior editor) and three reviewers, one of whom, Tanya Whitfield, is a member of our Board of Reviewing Editors.
The Reviewing editor and the other reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
While all three reviewers find the study interesting, there are some substantive reservations over the manuscript in its current form. A major criticism is the lack of citation and acknowledgement of previous literature. A number of papers have been suggested by the reviewers, all of which should be considered, cited and discussed carefully. In addition, the conclusions and interpretation of the findings should be adjusted to take full account of these earlier studies.
In addition, some further experimental work should be done to strengthen the manuscript. A fuller description, at multiple embryonic stages, of the BRE:GFP expression domains in the eye, and correlation with expression of both P–Smads and BMP inhibitors (fsta, bambi), is required. A later stage in situ showing vax2 expression following BMP overexpression should also be shown. If additional data (such as the origin of the stem cell compartment, or tracing the patch of lens-averted epithelium that forms the RPE) can be extracted from the existing movies (or generated with new ones), this information would be extremely valuable and would enhance the significance and novelty of the findings.
A number of other suggestions have been suggested by the reviewers to improve the manuscript, including adding more detail to the Experimental Procedures section and clarifying the data provided in the Table. It will be essential to provide accurate embryo staging and labeling throughout.
The full reviews are appended below.
Reviewer #1:
This is an interesting report documenting the cell movements that contribute to formation of the zebrafish neuroretina. The authors demonstrate that much of the retina derives from lens-averted epithelium, which moves round and into the lens-adjacent layer in directed movements of the cell sheet akin to gastrulation. These movements continue in the absence of cell proliferation, and account for the apparently paradoxical increase in surface area of the neuroretina, despite the decrease in apical surface area of individual cells in this tissue. It is proposed that a small patch of the lens-averted epithelium gives rise to the retinal pigmented epithelium (RPE).
Some of the imaging is spectacular and supports the conclusions of the paper well. However, there are also a number of deficits that should be addressed. These are listed below.
General points:
The authors place great importance on the retinal stem cell niche, but have no markers to show this population. Can they be visualised in some way?
They also place importance on the role of the developing RPE in driving the cell movements, but the cell movements of this layer are not described well at all, and the small patch of cells that is proposed to become the RPE is not identified or followed in any of the figures or movies. It will be important to show this to confirm the assertion that only a small patch of the lens-averted epithelium develops into the RPE.
Numbers and quantitation: more detail is needed throughout, e.g. for the phenotypes of 'variable expressivity' (eleventh paragraph) resulting from pan-ocular expression of BMP4.
Figures:
Scale bars are needed on the figures, especially on Figure 1B and C, where the surface area is estimated from linear dimensions drawn onto sections. The scale is needed to ensure the data shown in the figure panels match up with the values listed in the Table.
Stages/hours post fertilization should also be listed in the legend or directly on the panels for all figures.
Figure 1–figure supplement 1 should show a control (H2BGFP without aphidicolin). If this is provided by Figure 2A, there should be a reference to this figure in the legend to Figure 1–figure supplement 1, so that a comparison can be made. Are the small bright spots the dying cells? How many embryos were treated, and with what concentration of aphidicolin? This information does not appear to be provided anywhere. How was the block in cell proliferation confirmed? Was this by counting nuclei, or lack of mitotic figures?
Figure 4–figure supplement 1A, panel 2: This picture is not of publication quality, it is completely out of focus, with no morphological detail visible.
Figure 4–figure supplement 2: It would be helpful to show here a stage series, showing when the BMP reporter normally begins to be expressed in the eye, and how levels of expression compare with other areas of BMP activity in the embryo. Ideally, an additional confirmation of BMP activity, e.g. by P-Smad levels, should be shown; alternatively, more information about this transgenic line or a citation is needed in the Experimental Procedures.
Figure 5–figure supplement 1 merge: The box drawn in the first panel does not appear to correspond to the area shown in the enlargements.
Figure 5–figure supplement 4 needs repeating and improving. A control is needed to show expression levels in non-transgenic embryos. A higher power picture of the staining in the eye would be useful; the outline and arrows currently overlie the staining, making it difficult to see.
Experimental Procedures: More detail is needed for description of the transgenic lines. Are these already published? In which case, a citation is needed. If not published, more description of the constructs used is needed. A ZFIN reference should be given for each line, if available.
Immunohistochemistry: Not enough detail is given here for others to be able to replicate the experiments. What do the acronyms stand for, what dilutions of primary and secondary antibodies were used, and what concentrations of DAPI, DMSO, Triton, etc.?
Table:
The Table is difficult to understand. What are the numbers? Do they refer to measurements from a single embryo, or are they mean values from multiple embryos? (In which case, SEM or SD should be given). If from a single embryo, it is difficult to draw any firm conclusions from the data. In the text, phase 2 is described as a fast, smooth flow, but in all cases the effective speed shown during phase 2 in the table appears to be slower than the values shown for phase 1. The total distance moved in phase 2 needs to correlate with the distances shown on the figures, which is why scale bars on the figures are needed. It would be helpful to have a full stop instead of a comma for the decimal point, as 10,000 could be read as 'ten thousand'.
Movies:
The movies are very helpful, but more information is needed in the legends that describe these. What are the times over which they were taken, what was the frame rate, and what stage are the embryos? As for the figures, this information is needed to correlate with the values shown in the table, and should be given in the movie legends. Movie 8 is of lower quality and resolution than the others, and should be improved.
Reviewer #2:
This manuscript analyses the cell dynamics accompanying optic cup formation in the zebrafish and shows that cells in the “lens-averted” region of the optic vesicle are incorporated into the “lens-facing” region. It further shows that the inhibition of BMP is required for this process to occur efficiently. The study has interesting implications to understand the origin of optic cup domains. In addition, it provides a new interpretation for phenotypes that up to now had been interpreted as transdifferentiation of RPE into retina, and highlight the power of live imaging to provide an accurate interpretation of phenotypic outcomes. I consider the work a valuable contribution to our current knowledge on eye morphogenesis and patterning. However, I have important concerns with the way the significance of previous works analyzing this same process has been minimised by the authors, and with the interpretation of some of their observations.
The finding of optic vesicle cells ingressing into the future retinal layer from the prospective RPE layer is not new. It has been extensively described by Picker et al., PLoS Biology, 2009 and Kwan et al., Development, 2012. I disagree with the way the authors seem to remove significance from those papers, and I consider they should be appropriately cited. In particular, Picker et al. clearly and extensively showed that prior to optic cup formation, the optic vesicle is mostly formed by future retinal cells, and that the lens-averted layer of the primordium constitutes the origin of the temporal retinal domain.
This does not mean that I do not consider the current work valuable. The authors do provide a detailed analysis of the consequences this morphogenetic process has to understand the origin of different optic cup domains (in particular the CMZ and the RPE), and dissect part of the molecular mechanisms controlling it.
The authors state that follistatin is expressed in the temporal domain, yet their in-situ (Figure 4) clearly shows expression in the prospective dorsal portion of the optic vesicle. At early stages, the optic vesicle in zebrafish has not acquired its final disposition and temporal is located along the ventral portion of the optic vesicle. Follistatin expression is stronger at the distal-most region of the optic vesicle, both dorsal and ventral, which corresponds to prospective dorsal (see Veien et al., Development, 2008). Indeed, this would be more consistent with previous reports showing a requirement of differential BMP signalling along the dorso-ventral, and not the naso-temporal, axis.
In the same line with my first comment, I find the authors should be more rigorous with their citation of previous work. In the first sentence of their manuscript and based on a previous report from the authors from 2006, they state that optic vesicle evagination occurs by single cell migration. However, since then there have been additional papers analyzing dynamically this process and showing results that support alternative explanations for the active behaviours of eye field cells. These papers (England et al., Development, 2006 and Ivanovitch et al., Dev Cell, 2013) should be acknowledged, and the first statement softened.
Reviewer #3:
Given the involvement of BMP signaling in ocular patterning, it is exciting to find that overexpression of BMP can affect specific cell movements. The manuscript contains some interesting observations, but the authors have not adequately integrated their results with the existing literature. I have concerns about the conclusions and interpretations.
1) An anterior rotational movement characterized by Picker et al., 2009, is largely overlooked, but has significant implications for these analyses: what the authors refer to as the optic vesicle temporal domain (where fsta is expressed) gives rise to the dorsal domain (180° away from the optic fissure). Yet the authors describe the dorsal region as having notably less flow. Is this the prospective nasal region? Clarification, and placing these results in the context of previously characterized movements would be useful.
2) The authors show BRE reporter expression in the dorsal domain at optic cup stage. At least one other BMP inhibitor, bambi, is already known to be expressed in the prospective dorsal domain (first temporal, then dorsal) throughout optic cup development (French et al., 2009). The coincidence of BMP inhibitor expression and BRE reporter expression suggests a potential negative feedback loop, though BMP signaling is not completely abolished in the dorsal domain, based on phospho-Smad staining (French et al., 2009) and BRE reporter expression. Therefore, fsta expression (and expression of other BMP inhibitors) may not mark a zone of repressed BMP signaling, as the authors have proposed.
3) The authors argue that BMP overexpression driven by rx2 does not lead to fate changes in the eye, by demonstrating unchanged expression of vax2. However, the embryo appears to be ∼18 somite stage (this is a guess given the lack of staging)? This reviewer was under the impression that the rx2 promoter does not begin to induce transcription until ∼14 somite stage, therefore, it is unclear if vax2 expression would be altered in such a short time. A later in situ (prim–5) would be a better control, especially given strong induction of the BRE and later phenotypes. Any clarification on promoter activity and timing would be welcome.
4) The authors argue that there is no BMP activity in the early optic vesicle by using the BRE reporter line. Is it clear that this line reports all BMP activity? There is evidence that BMP signaling is active in the early optic vesicle: French et al., 2009, demonstrate strong phospho-Smad staining in the optic vesicle temporal region (which gives rise to the dorsal domain) at 6 and 10 somite stages.
5) Cell movement from lens-averted to lens-facing domain has previously been analyzed via live imaging (Picker et al., 2009), cell tracking and volume measurements (Kwan et al., 2012). A novel finding would be when such flow ceases, thereby defining the definitive stem cell compartment; this was never demonstrated previously. The authors' movies are likely to contain that information, as the last embryo in Figure 2A appears older than prim–5, based on lens morphology.
6) Embryo ages (somite stages or hours post fertilization) are lacking throughout the manuscript, making it difficult to place these movements in context of other known events.
10.7554/eLife.05216.026Author response
While all three reviewers find the study interesting, there are some substantive reservations over the manuscript in its current form. A major criticism is the lack of citation and acknowledgement of previous literature. A number of papers have been suggested by the reviewers, all of which should be considered, cited and discussed carefully. In addition, the conclusions and interpretation of the findings should be adjusted to take full account of these earlier studies.
In addition, some further experimental work should be done to strengthen the manuscript. A fuller description, at multiple embryonic stages, of the BRE:GFP expression domains in the eye, and correlation with expression of both P–Smads and BMP inhibitors (fsta, bambi), is required. A later stage in situ showing vax2 expression following BMP overexpression should also be shown. If additional data (such as the origin of the stem cell compartment, or tracing the patch of lens-averted epithelium that forms the RPE) can be extracted from the existing movies (or generated with new ones), this information would be extremely valuable and would enhance the significance and novelty of the findings.
A number of other suggestions have been suggested by the reviewers to improve the manuscript, including adding more detail to the Experimental Procedures section and clarifying the data provided in the Table. It will be essential to provide accurate embryo staging and labeling throughout.
The full reviews are appended below.
Reviewer #1:
This is an interesting report documenting the cell movements that contribute to formation of the zebrafish neuroretina. The authors demonstrate that much of the retina derives from lens-averted epithelium, which moves round and into the lens-adjacent layer in directed movements of the cell sheet akin to gastrulation. These movements continue in the absence of cell proliferation, and account for the apparently paradoxical increase in surface area of the neuroretina, despite the decrease in apical surface area of individual cells in this tissue. It is proposed that a small patch of the lens-averted epithelium gives rise to the retinal pigmented epithelium (RPE).
Some of the imaging is spectacular and supports the conclusions of the paper well. However, there are also a number of deficits that should be addressed. These are listed below.
General points:
The authors place great importance on the retinal stem cell niche, but have no markers to show this population. Can they be visualised in some way?
This is a very interesting aspect that we have elaborated further in the revised manuscript. The retina specific homeobox transcriptions factor Rx2 in fact represents a marker for retinal stem cells at the relevant stages of the manuscript (see attached manuscript by Reinhardt, Centanin et al., submitted elsewhere). It is initially expressed in all cells of the evaginated optic vesicle (likely stem cells) and as development proceeds is limited to the CMZ. At later stages (once the retinal cell types enter terminal differentiation after the completion of optic cup formation) its expression is additionally found in Müller Glia cells and photoreceptors. The stem cells of the neuroretina and the Retinal Pigmented Epithelium (RPE) are located within the CMZ. Using an inducible CRE based lineage analysis we show in the accompanying manuscript that Rx2 cells of the CMZ represent stem cells. An individual Rx2 positive cell is multipotent and gives rise to all retinal cell types (Reinhardt, Centanin et al., submitted). This is now explicitly stated in the revised version of the manuscript where we refer to rx2 as retinal stem cell marker. The submitted manuscript of Reinhardt, Centanin et al. has been attached.
They also place importance on the role of the developing RPE in driving the cell movements, but the cell movements of this layer are not described well at all, and the small patch of cells that is proposed to become the RPE is not identified or followed in any of the figures or movies. It will be important to show this to confirm the assertion that only a small patch of the lens-averted epithelium develops into the RPE.
This is an important point, and we thank the referee for this comment. In the revised version of the manuscript we now put emphasis on the developing RPE. We use two features to identify early RPE: The absence of Rx2 expression as well as the specific flat morphology of the RPE cells (and nuclei). Using those criteria we follow the edge of the RPE, best visualized in Movie 3. We show the developing RPE in between the arrows, which mark the border of RPE. In addition, Figure 1J-L is highlighting the change in cell shape the RPE precursors undergo, concomitant with their loss of Rx2 expression (between asterisks in Figure 1H).
We have taken care to clearly state these important points in the revised manuscript.
Numbers and quantitation: more detail is needed throughout, e.g. for the phenotypes of 'variable expressivity' (eleventh paragraph) resulting from pan-ocular expression of BMP4.
The referee is right. Our statements were misleading and the variability was due to genetic variability in the analyzed lines. We had initially mentioned the variable expressivity to indicate a graded activity of BMP4, which however was apparently not fully explained.
In order to avoid that confusion we now only focus on one phenotypic class and have removed the misleading term from the revised manuscript.
Figures:
Scale bars are needed on the figures, especially on Figure 1B and C, where the surface area is estimated from linear dimensions drawn onto sections. The scale is needed to ensure the data shown in the figure panels match up with the values listed in the Table.
Scalebars have been added to the figures, which were used for quantitation.
Stages/hours post fertilisation should also be listed in the legend or directly on the panels for all figures.
Information about the developmental stages has been added to the figure legends.
Figure 1–figure supplement 1 should show a control (H2BGFP without aphidicolin). If this is provided by Figure 2A, there should be a reference to this figure in the legend to Figure 1–figure supplement 1, so that a comparison can be made.
The new figure (Figure 2–figure supplement 1) presenting this data is connected to Figure 2, which is showing the control data. As requested we have put a reference to the corresponding figure legend.
Are the small bright spots the dying cells? How many embryos were treated, and with what concentration of aphidicolin? This information does not appear to be provided anywhere. How was the block in cell proliferation confirmed? Was this by counting nuclei, or lack of mitotic figures?
The referee is right, we also believe that the small bright spots represent condensed nuclei of apoptotic cells. Notably this is not affecting the described neuroretinal flow. We added the necessary information regarding the aphidicolin treatment in the revised manuscript.
Briefly, we pretreated for 5 hours before imaging with a concentration of 10µg/ml (Serva, cat:13696) 12 embryos. One of those was subjected to in vivo time-lapse analysis.
The efficacy of the treatment was addressed by analyzing nuclei in mitosis (positive for the expression of phospho-histone H3. At 21.5 hpf pHH3 positive nuclei were counted in central sections of four control (average: 21) and experimental (average: 6) retinae respectively. We have now included the anti-pHH3 immunohistochemical stainings to the revised manuscript (Figure 2–figure supplement 1) showing the aphidicolin mediated inhibition of cell proliferation.
Figure 4–figure supplement 1A, panel 2: This picture is not of publication quality, it is completely out of focus, with no morphological detail visible.
Figure 4–figure supplement 2: It would be helpful to show here a stage series, showing when the BMP reporter normally begins to be expressed in the eye, and how levels of expression compare with other areas of BMP activity in the embryo. Ideally, an additional confirmation of BMP activity, e.g. by P-Smad levels, should be shown; alternatively, more information about this transgenic line or a citation is needed in the Experimental Procedures.
The referee is right and we have restructured that entire part also in response to the general request and the comments of the other referees. Figure 4–figure supplement 1A has been removed. Instead we show a new Figure 4 which represents the dynamics of fsta and bambi expression along with indicators for BMP activity (anti-pSmad 1/5/8 immunohistochemical stainings and BRE::GFP expression) at relevant stages for optic cup formation. The BRE:GFP zebrafish have been provided by Beth Roman. The original paper (Laux et al., 2011) has (and had) been cited. The combination of BMP signaling modulators fsta and bambi with indicators for BMP activity at the key stages of the vesicle to cup transition now provides all necessary evidence at a glance. The text and figure legends have been revised accordingly.
Figure 5–figure supplement 1 merge: The box drawn in the first panel does not appear to correspond to the area shown in the enlargements.
We have taken care that the box moved perfectly corresponds to the close up, thanks for pointing this out.
Figure 5–figure supplement 4 needs repeating and improving. A control is needed to show expression levels in non-transgenic embryos. A higher power picture of the staining in the eye would be useful; the outline and arrows currently overlie the staining, making it difficult to see.
We followed the referee’s advice and repeated the entire series of in situs of vax2 on wild type and rx2::BMP4 embryos. Incorporating the suggestions of the other referees we now show embryos at a later stage (28 hpf) to address the potential impact of ectopic BMP4 on ventral patterning of the retina (Figure 6–figure supplement 4). We show the eyes of control and rx2::BMP4 embryos from different angles (lateral and ventral). The expression of vax2 in the ventral part of the retina persists even under conditions of BMP expression in the entire optic vesicle. This indicates that the observed morphological alterations observed under those conditions are not due to a miss-patterning of the ventral retina but can rather be attributed to the antimotogenic effect of BMP4. We have taken care to place arrows and outline outside of the staining.
Experimental Procedures: More detail is needed for description of the transgenic lines. Are these already published? In which case, a citation is needed. If not published, more description of the constructs used is needed. A ZFIN reference should be given for each line, if available.
This is an important point. All published transgenic lines were properly referenced in the submitted version and are as well in the revised version. All of the components of the lines generated in the context of this manuscript are referenced in detail in the experimental procedures section of the revised manuscript. The details are found in the revised experimental procedures.
Immunohistochemistry: Not enough detail is given here for others to be able to replicate the experiments. What do the acronyms stand for, what dilutions of primary and secondary antibodies were used, and what concentrations of DAPI, DMSO, Triton, etc.?
We have taken care that the information now presented allows full reproduction of the experiments. The relevant information is provided in the experimental procedures section of the revised manuscript.
Table:
The Table is difficult to understand. What are the numbers? Do they refer to measurements from a single embryo, or are they mean values from multiple embryos? (In which case, SEM or SD should be given). If from a single embryo, it is difficult to draw any firm conclusions from the data. In the text, phase 2 is described as a fast, smooth flow, but in all cases the effective speed shown during phase 2 in the table appears to be slower than the values shown for phase 1. The total distance moved in phase 2 needs to correlate with the distances shown on the figures, which is why scale bars on the figures are needed. It would be helpful to have a full stop instead of a comma for the decimal point, as 10,000 could be read as 'ten thousand'.
The referee is right, the table is difficult to understand. It reflects the attempt to provide quantitation of the movies to allow a more intuitive perception of the situation. Apparently that attempt failed. Since the table does not provide any information that is not also provided by the movies and it appears more confusing than enlightening, we have decided to remove it from the manuscript to enhance the clarity of our arguments. We thank the referee for pointing out this problem.
Movies:
The movies are very helpful, but more information is needed in the legends that describe these. What are the times over which they were taken, what was the frame rate, and what stage are the embryos? As for the figures, this information is needed to correlate with the values shown in the table, and should be given in the movie legends. Movie 8 is of lower quality and resolution than the others, and should be improved.
We now provide the information requested for each of the movie files in the revised version of the manuscript. Movie 8 (Movie 10 in the revised manuscript) represents imaging deep inside the optic cup. It has been taken under the same imaging settings as all the other movies. Due to scattering of the overlying tissue the quality could only be partially improved. This improved movie is now provided as revised Movie 10.
Reviewer #2:
This manuscript analyses the cell dynamics accompanying optic cup formation in the zebrafish and shows that cells in the “lens-averted” region of the optic vesicle are incorporated into the “lens-facing” region. It further shows that the inhibition of BMP is required for this process to occur efficiently. The study has interesting implications to understand the origin of optic cup domains. In addition, it provides a new interpretation for phenotypes that up to now had been interpreted as transdifferentiation of RPE into retina, and highlight the power of live imaging to provide an accurate interpretation of phenotypic outcomes. I consider the work a valuable contribution to our current knowledge on eye morphogenesis and patterning. However, I have important concerns with the way the significance of previous works analyzing this same process has been minimised by the authors, and with the interpretation of some of their observations.
The finding of optic vesicle cells ingressing into the future retinal layer from the prospective RPE layer is not new. It has been extensively described by Picker et al., PLoS Biology, 2009 and Kwan et al., Development, 2012. I disagree with the way the authors seem to remove significance from those papers, and I consider they should be appropriately cited. In particular, Picker et al. clearly and extensively showed that prior to optic cup formation, the optic vesicle is mostly formed by future retinal cells, and that the lens-averted layer of the primordium constitutes the origin of the temporal retinal domain.
We fully agree with the referee that Picker et al., as well as Kwan et al., described flow over the distal rim of the developing optic cup and are sorry for not adequately pointing that out in the initial version. It was never our intention not to acknowledge these findings or to remove significance from them.
In the revised version of the manuscript we substantially extended this paragraph to give full credit to previous research. We added the reference to a paper published before Picker et al. and Kwan et al. speculating about such a flow (Li et al., 2000). Importantly the flow in the temporal domain was then shown by Picker et al. and confirmed by Kwan et al. The nasal flow was not demonstrated and the connection of flow to the origin of the optic fissure had not been addressed before. The aspect of the dorsal pole, showing no flow, had also not been noted before.
This does not mean that I do not consider the current work valuable. The authors do provide a detailed analysis of the consequences this morphogenetic process has to understand the origin of different optic cup domains (in particular the CMZ and the RPE), and dissect part of the molecular mechanisms controlling it.
We thank the referee for this supportive statement and hope that our changes have clarified the issue.
The authors state that follistatin is expressed in the temporal domain, yet their in-situ (Figure 4) clearly shows expression in the prospective dorsal portion of the optic vesicle. At early stages, the optic vesicle in zebrafish has not acquired its final disposition and temporal is located along the ventral portion of the optic vesicle. Follistatin expression is stronger at the distal-most region of the optic vesicle, both dorsal and ventral, which corresponds to prospective dorsal (see Veien et al., Development, 2008). Indeed, this would be more consistent with previous reports showing a requirement of differential BMP signalling along the dorso-ventral, and not the naso-temporal, axis.
We agree with the referee that this is an important issue. Since the orientation of the vesicle and the localization of BMP activity with respect to the expression of BMP modulators was also brought up by the other referees we now provide a new Figure 4 for further clarification. We added more data showing the expression of fsta, bambi, as well as phosopho SMAD and the active BMP sensor at different developmental stages (please see revised Figure 4).
The point of eye rotation, which has been described in zebrafish does not pose a problem for the relative annotation of expression patterns during eye cup formation.
In a manuscript of Schmitt and Dowling (1994) two important phases of “eye rotation” have been described, a slight early rotation (10-12SS) and a more intense rotation at a later phase (24-36 hpf). Our analysis clearly shows that the neuroretinal flow described in our manuscript starts after the first, slight rotation and has already ceased before onset of the intense rotation of the optic cup (24 to 36 hpf, Schmitt and Dowling, 1994). We describe the formation of the optic fissure by retinal flow in between the two phases of eye rotations. This is in line with the timing of development described by Schmitt and Dowling (1994).
In the paper by Veien et al., the analyses started at an earlier developmental stage than our analyses likely addressing the first phase of “eye rotation”. As described above, the neuroretinal flow described in our manuscript occurs in between the rotational stages. Thus the relative positions remain stable in the phase of eye formation we are discussing here. Therefore all the patterns described are not apparently affected by the known phases of eye rotation. Our analysis furthermore indicates that expression patterns are dynamic with respect to transcriptional control and are therefore of limited value for linking fates and position (compare the expression of fsta and bambi in Figure 4).
With respect to the impact of the modulation of BMP signaling in the nasal and temporal domain the new Figure 4 now provides compelling expression data that fully support the hypothesis presented in the manuscript (please see the revised Figure 4).
In the same line with my first comment, I find the authors should be more rigorous with their citation of previous work. In the first sentence of their manuscript and based on a previous report from the authors from 2006, they state that optic vesicle evagination occurs by single cell migration. However, since then there have been additional papers analyzing dynamically this process and showing results that support alternative explanations for the active behaviours of eye field cells. These papers (England et al., Development, 2006 and Ivanovitch et al., Dev Cell, 2013) should be acknowledged, and the first statement softened.
It was not our intention not to acknowledge previous work. Although this aspect is only part of our Introduction, we extended this paragraph and cited the mentioned papers. We have tried to present a synthesis of the not so different perspectives promoted by the respective authors.
Reviewer #3:
Given the involvement of BMP signaling in ocular patterning, it is exciting to find that overexpression of BMP can affect specific cell movements. The manuscript contains some interesting observations, but the authors have not adequately integrated their results with the existing literature. I have concerns about the conclusions and interpretations.
1) An anterior rotational movement characterized by Picker et al., 2009, is largely overlooked, but has significant implications for these analyses: what the authors refer to as the optic vesicle temporal domain (where fsta is expressed) gives rise to the dorsal domain (180° away from the optic fissure). Yet the authors describe the dorsal region as having notably less flow. Is this the prospective nasal region? Clarification, and placing these results in the context of previously characterized movements would be useful.
We thank the referee for this comment. We realized that this issue needs further clarification. In the paper of Schmitt and Dowling (1994) two important phases of “eye rotation” have been described, one slight rotation (10-12SS) and a more intense rotation at a later phase (24-36 hpf).
Our analysis clearly show that the neuroretinal flow, which we describe, starts after the first, slight rotation and has already ceased before onset of the intense rotation of the optic cup (24 to 36 hpf, Schmitt and Dowling, 1994). Further, based on our 4D data we describe the formation of the fissure by the retinal flow, in between the two phases of eye rotations. This is in line with the timing of development described by Schmitt and Dowling (1994). However, our 4D data clearly show a different developmental process behind the formation of the optic fissure.
Importantly Picker et al. started their analyses at 10SS, therefore including at least the first phase of “eye rotation”. It is also important to consider the axis of the developing eye in relation to the axis of the body. Especially the heavy rotation of the optic cup aligns the axis of the eye and the axis of the body at this late developmental stage. With respect to the onset of neuroretinal flow, the dorsal and ventral poles relate to the optic vesicle and later define also the axis of the eye. However, as described above, the axis of the eye is only later aligned to the axis of the zebrafish body.
Based on our 4D analysis of the eye it is apparent that gene expression patterns are of limited value as cell/tissue fate markers (please compare the expression patterns of bambia to fsta in the revised Figure 4).
For addressing the neuroretinal flow by the expression of the BMP antagonists, their expression pattern at the optic vesicle stage are of importance. Here especially the expression pattern of fsta can explain the modulation of BMP signaling in the nasal and temporal domain facilitating the neuroretinal flow (please see the revised Figure 4).
2) The authors show BRE reporter expression in the dorsal domain at optic cup stage. At least one other BMP inhibitor, bambi, is already known to be expressed in the prospective dorsal domain (first temporal, then dorsal) throughout optic cup development (French et al., 2009). The coincidence of BMP inhibitor expression and BRE reporter expression suggests a potential negative feedback loop, though BMP signaling is not completely abolished in the dorsal domain, based on phospho-Smad staining (French et al., 2009) and BRE reporter expression. Therefore, fsta expression (and expression of other BMP inhibitors) may not mark a zone of repressed BMP signaling, as the authors have proposed.
We are thankful for this comment. We included the BMP antagonist bambi into our analyses and noted the strikingly overlapping patterns of BMP signaling activity and bambi expression (revised Figure 4). The referee is right in stating that bambi is not totally antagonizing BMP signaling but rather is modulating it or may be even modulated by it. With bambi alone it would be hard to explain the modulation of BMP signaling to facilitate the neuroretinal flow. However, fsta shows a different expression pattern (please see the revised Figure 4), able to explain the facilitation of the flow nicely.
3) The authors argue that BMP overexpression driven by rx2 does not lead to fate changes in the eye, by demonstrating unchanged expression of vax2. However, the embryo appears to be ∼18 somite stage (this is a guess given the lack of staging)? This reviewer was under the impression that the rx2 promoter does not begin to induce transcription until ∼14 somite stage, therefore, it is unclear if vax2 expression would be altered in such a short time. A later in situ (prim–5) would be a better control, especially given strong induction of the BRE and later phenotypes. Any clarification on promoter activity and timing would be welcome.
In situ hybridizations with a vax2 probe have been added (Figure 6–figure supplement 4). This figure shows the staining of control vs. rx2::BMP4 embryos at a later developmental stage (28 hpf). Although the vax2 domain is smaller in the rx2::BMP4 embryos, compared to the controls, it is not absent. Hence, the reason for the arrested flow (16.5–21.5) is unlikely an early mis-differentiation of the ventral domain.
4) The authors argue that there is no BMP activity in the early optic vesicle by using the BRE reporter line. Is it clear that this line reports all BMP activity? There is evidence that BMP signaling is active in the early optic vesicle: French et al., 2009, demonstrate strong phospho-Smad staining in the optic vesicle temporal region (which gives rise to the dorsal domain) at 6 and 10 somite stages.
We agree with the referee and included next to the analysis of BRE::GFP anti-pSmad 1/5/8 immunohistochemical stainings, both at different developmental stages (Figure 4). Notably the BMP signaling reporter shows a delay of activity, which can be explained by its nature as a transcriptional reporter. This suggests that BMP signaling still is active. This is in favour of a modulative action of the BMP antagonists rather than a total block. We thank the referees for this comment.
5) Cell movement from lens-averted to lens-facing domain has previously been analyzed via live imaging (Picker et al., 2009), cell tracking and volume measurements (Kwan et al., 2012). A novel finding would be when such flow ceases, thereby defining the definitive stem cell compartment; this was never demonstrated previously. The authors' movies are likely to contain that information, as the last embryo in Figure 2A appears older than prim–5, based on lens morphology.
We agree with the referee that Picker et al., as well as Kwan et al., described flow over the distal rim of the developing optic cup. It was not our intention not to acknowledge these findings. We extended the paragraph discussing this point in the revised version of the manuscript. In addition we added a reference to an additional manuscript speculating about such a flow (Li et al., 2000) even before Picker and Kwan.
The flow in the temporal domain was then shown by Picker et al. and confirmed by Kwan et al. The nasal flow was not demonstrated and the flow with respect to the origin of the optic fissure was not addressed before. The aspect of the dorsal pole, showing no flow, was also neither described nor addressed before.
Regarding the forming stem cell compartment (CMZ), this was part of our analyses. We describe the origin of the cells, which eventually reside in the CMZ domain, by tracking them backwards in 4D (Figure 3, Movie 5).
6) Embryo ages (somite stages or hours post fertilization) are lacking throughout the manuscript, making it difficult to place these movements in context of other known events.
The developmental stages of embryos have been added to the figure legends.
Neuropeptides released from dense-core vesicles (DCVs) modulate neuronal activity, but the molecules driving DCV secretion in mammalian neurons are largely unknown. We studied the role of calcium-activator protein for secretion (CAPS) proteins in neuronal DCV secretion at single vesicle resolution. Endogenous CAPS-1 co-localized with synaptic markers but was not enriched at every synapse. Deletion of CAPS-1 and CAPS-2 did not affect DCV biogenesis, loading, transport or docking, but DCV secretion was reduced by 70% in CAPS-1/CAPS-2 double null mutant (DKO) neurons and remaining fusion events required prolonged stimulation. CAPS deletion specifically reduced secretion of stationary DCVs. CAPS-1-EYFP expression in DKO neurons restored DCV secretion, but CAPS-1-EYFP and DCVs rarely traveled together. Synaptic localization of CAPS-1-EYFP in DKO neurons was calcium dependent and DCV fusion probability correlated with synaptic CAPS-1-EYFP expression. These data indicate that CAPS-1 promotes fusion competence of immobile (tethered) DCVs in presynaptic terminals and that CAPS-1 localization to DCVs is probably not essential for this role.
DOI:http://dx.doi.org/10.7554/eLife.05438.001
10.7554/eLife.05438.002eLife digest
Our ability to think and act is due to the remarkable capacity of the brain to process complex information. This involves nerve cells (or neurons) communicating with each other in a rapid and precise manner by releasing synaptic vesicles containing neurotransmitters across the gaps—called synapses—between neurons. In addition to this fast neurotransmitter signalling, neurons can transmit signals by releasing chemical signals called neuropeptides. Neuropeptides are major regulators of human brain function, including mood, anxiety, and social interactions.
Neuropeptides and other neuromodulators such as serotonin and dopamine are normally packaged into bubble-like compartments called dense-core vesicles. Compared to synaptic vesicles we know much less about how dense-core vesicles are trafficked and released. Dense-core vesicles are generally mobile and move around the inside of cells to release neuropeptides where and when they are needed. However, some vesicles are stationary and may even be loosely tethered to the cell membrane. Most of the sites where dense-core vesicles can fuse with the cell membrane are at synapses.
Previous work has suggested that the protein CAPS-1 is important for moving dense-core vesicles to the correct sites on the cell membrane, and for releasing neuropeptides across the synapses of worms and flies. However, detailed insights into this process in mammalian neurons are lacking.
By examining neurons from both normal mice and mice lacking the CAPS-1 protein, Farina et al. have now analyzed the role CAPS-1 plays in releasing neuropeptides. In cells lacking CAPS-1 fewer dense-core vesicles merged with the cell membrane than in cells containing the protein. However, a new technique that tracks the movement of individual vesicles revealed that only stationary dense-core vesicles had difficulties fusing; mobile vesicles continued to fuse with the cell membrane in the normal manner. Introducing CAPS-1 into cells lacking this protein corrected the fusion defect experienced by the stationary vesicles.
Farina et al. also showed that CAPS-1 was present at most—but not all—synapses, and synapses that had more CAPS-1 released more neuropeptides. This work shows that CAPS proteins strongly influence the probability of dense-core vesicle release and that neurons can tune this probability at individual synapses by controlling the expression of CAPS. Future work will be aimed at understanding how neurons can achieve this and which protein domains in CAPS are required.
DOI:http://dx.doi.org/10.7554/eLife.05438.002
Author keywordsCAPS proteinexocytosisneuronsdense-core vesicletraffickingResearch organismmousehttp://dx.doi.org/10.13039/501100001826Netherlands Organisation for Health Research and Development (ZonMw)916-66-101ToonenRuud Fhttp://dx.doi.org/10.13039/501100001826Netherlands Organisation for Health Research and Development (ZonMw)91208017ToonenRuud Fhttp://dx.doi.org/10.13039/501100001826Netherlands Organisation for Health Research and Development (ZonMw)90342095VerhageMatthijshttp://dx.doi.org/10.13039/501100001826Netherlands Organisation for Health Research and Development (ZonMw)90001001VerhageMatthijshttp://dx.doi.org/10.13039/501100000785Education, Audiovisual and Culture Executive Agency (EACEA)Erasmus Mundus Joint Doctorate grant EU 2011-1632/001-001-EMJDFarinaMargheritahttp://dx.doi.org/10.13039/501100000781European Research Council (ERC)322966VerhageMatthijshttp://dx.doi.org/10.13039/501100004965Sixth Framework ProgrammeEUSynapse 019055VerhageMatthijshttp://dx.doi.org/10.13039/501100004963Seventh Framework ProgrammeHEALTH-F2-2009-241498VerhageMatthijshttp://dx.doi.org/10.13039/501100004963Seventh Framework ProgrammeSynSys project HEALTH-F2-2009-242167VerhageMatthijsThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementStationary dense-core vesicles depend on the CAPS-1 protein to fuse with the presynaptic membrane.Introduction
Neuropeptides and neurotrophic factors are essential for brain development and synaptic plasticity (McAllister et al., 1996; Huang and Reichardt, 2001; Poo, 2001; Samson and Medcalf, 2006; van den Pol, 2012). These neuromodulators are transported in dense-core vesicles (DCVs). Dysregulation of DCV transport and fusion is associated with cognitive and post-traumatic stress disorders (Sadakata et al., 2007b; Meyer-Lindenberg et al., 2011; Sah and Geracioti, 2013). DCVs bud off at the Golgi network (Kim et al., 2006) and are transported via microtubule-based motor proteins (Hirokawa et al., 2009; Schlager and Hoogenraad, 2009). High-frequency firing facilitates DCV fusion and the resultant calcium influx triggers SNARE complex-dependent DCV secretion (Bartfai et al., 1988; Hartmann et al., 2001; de Wit et al., 2009; van de Bospoort et al., 2012). Unlike synaptic vesicles (SVs), DCVs lack a local recycling mechanism. To ensure a constant and uniform supply of DCVs at release sites, DCVs are generally very dynamic although some DCVs are stationary. Stimulation triggers arrest of moving DCVs (de Wit et al., 2006; Shakiryanova et al., 2006; Matsuda et al., 2009; Wong et al., 2012), probably promoting their local availability for secretion. DCV fusion sites can be located in the entire neuron but DCVs preferentially fuse at presynaptic terminals and release at extra-synaptic sites requires more robust stimulation (van de Bospoort et al., 2012). Recently, we have shown that Munc13 is an important regulator of DCV fusion at synapses (van de Bospoort et al., 2012). However, in contrast to SV release, a comprehensive insight in the molecular mechanisms of DCV secretion is still lacking.
Previous studies in Caenorhabditis elegans and Drosophila have implicated calcium-activator protein for secretion (CAPS) proteins in DCV secretion. Mutants of the C. elegans CAPS ortholog UNC-31 show reduced peptide release without affecting synaptic vesicle fusion (Speese et al., 2007). In C. elegans neurons, UNC-31 is required for the docking of DCVs at the plasma membrane (Zhou et al., 2007; Hammarlund et al., 2008; Lin et al., 2010). In Drosophila, deletion of dCAPS also affects DCV release but in contrast to C. elegans, leads to an increased DCV presence in terminals (Renden et al., 2001). Mammals express two CAPS genes, CAPS-1 and CAPS-2, which are complementarily expressed in brain (Speidel et al., 2003) and are essential for synaptic transmission (Jockusch et al., 2007). In adrenal chromaffin cells, CAPS-1 deletion affects catecholamine uptake in chromaffin granules (Speidel et al., 2005; Brunk et al., 2009) and deletion of CAPS-1 and CAPS-2 abolishes their fusion without affecting docking (Liu et al., 2010). CAPS-2 is important for cerebellar development and neuron survival (Sadakata et al., 2004, 2007a), and deletion of CAPS-1 in cerebellar neurons perturbs DCV trafficking (Sadakata et al., 2010, 2013). Hence, these studies suggest that CAPS proteins are involved in several aspects of DCV trafficking and release in invertebrates and in mammalian chromaffin cells, but their role in mammalian versus invertebrate systems appears to differ considerably, especially regarding synaptic transmission and cell survival.
In this study, we analyzed the distribution and function of CAPS proteins in DCV trafficking and fusion in mammalian neurons using wild type (WT), CAPS-2 null mutant and CAPS-1/2 null mutant mice. Endogenous CAPS-1 was present in puncta that partially overlapped with synaptic markers and also co-localized with DCV markers. CAPS deletion did not affect DCV biogenesis, neuropeptide loading or average DCV transport velocity. In CAPS double null mutant (DKO) neurons, DCV fusion was strongly reduced at synapses and at extra-synaptic sites. We developed a novel release assay to track single DCVs prior to fusion and found that CAPS deletion strongly affected DCV release of stationary, presumably tethered vesicles. We provide evidence that CAPS-1 localization at synapses is calcium dependent and that DCV release probability correlates with synaptic CAPS-1 expression levels.
ResultsCAPS-1 is present at synapses and overlaps with secretory vesicles markers
To understand CAPS-1 function in neuronal DCV release, we first investigated its sub-cellular localization in cultured neurons. Hippocampal neurons at 14 days in vitro (DIV 14) were stained with a novel, CAPS-1-specific antibody (Figure 1—figure supplement 1). CAPS-1 was present in the cytosol and in dendritic and axonal puncta, probably membrane domains (Figure 1A). Approximately 45% of these CAPS-1 puncta co-localized with the synaptic marker VGLUT1 in the entire neuron (Figure 1B, C, Pearson's coefficient: 0.42±0.04, n=7). CAPS-1 immunoreactivity was detectable in approximately 60% of VGLUT1 positive synapses and vice versa approximately 60% of the CAPS-1 puncta co-localized with VGLUT1 (Figure 1B, D, Manders' coefficients CAPS-1 in VGLUT1: 0.67±0.08, and VGLUT1 in CAPS-1: 0.64±0.05, n=8). CAPS-1 domains were also found at extra-synaptic sites (Figure 1B, E, white arrowheads). The CAPS-1 expression pattern differed from the sub-cellular localization of the DCV priming protein Munc13-1, which was much more restricted to the synapse (Figure 1E, F, Pearson's coefficient in the entire neuron: 0.64±0.04, n=10) with 96% of synapses containing M13-1 (Figure 1G, number of synapses containing M13-1: 96.3±0.7%, n=4, number of synapses containing CAPS-1: 58.1±10.9%, n=7). Since CAPS proteins have been initially identified as DCV resident proteins (Berwin et al., 1998), we tested the co-localization of CAPS-1 with DCVs in hippocampal neurons. Using the endogenous DCV protein chromogranin B (ChrB) we found that the majority of DCVs are located in the axon (Figure 1—figure supplement 2). Antibody incompatibility precluded co-staining of CAPS-1 antibody with antibodies against ChrB. Therefore, we used the DCV cargo neuropeptide Y (NPY) fused to Venus (Nagai et al., 2002), which showed more than 80% co-localization with this endogenous marker (Figure 1H, I, Manders' coefficients for chromogranin B in NPY-Venus puncta: 0.97±0.02, and NPY-Venus in ChrB puncta: 0.84±0.01, n=14). Approximately 35% of NPY-Venus labeled DCVs co-localized with endogenous CAPS-1 (Figure 1J–L, Pearson's coefficient: 0.45±0.05, n=21, number of NPY-labeled DCVs co-expressing CAPS-1: 34.71±3.03%, n=6). These data show that endogenous CAPS-1 is present at many but not all synapses. In addition, CAPS-1 domains are found at extra-synaptic regions and CAPS-1 partly co-localizes with DCV markers.10.7554/eLife.05438.003
CAPS-1 clusters are present at synaptic and extra-synaptic sites and partly co-localize with DCVs.
(A) Example image of a hippocampal neuron (DIV 14) stained for endogenous CAPS-1 (green), dendrite marker MAP2 (blue) and synapse marker VGLUT1 (red). Scale bar 10 µm. (B) Zoom of a neurite stained for CAPS-1 and VGLUT1. CAPS-1 rich domains not overlapping with VGLUT1 (filled arrowhead), VGLUT1 punctum not enriched for CAPS-1 (open arrowhead), VGLUT1 puncta overlapping with a CAPS-1 rich domain (stars). Scale bar 2 µm. (C) Co-localization of CAPS-1 with VGLUT1 in the entire neuron quantified by Pearson's correlation. Co-localization of VAMP2 with VGLUT1 was used as a positive control (VGLUT1-VAMP2: 0.8±0.02, n=7 neurons; VGLUT1-CAPS-1: 0.4±0.05, n=7 neurons, ***p<0.0001). (D) Mander's coefficients for the proportion of CAPS-1 immuno-reactivity in VGLUT1 positive locations: 0.67±0.08, n=8 neurons or proportion of VGLUT1 immunoreactivity in CAPS-1 positive locations: 0.64±0.05, n=8 neurons. (E) Example images of neurites from hippocampal neurons (DIV 14) stained for endogenous CAPS-1 (green, left panel) and VGLUT1 (red) or for Munc13-1 (M13-1, green, right panel) and VGLUT1 (red). CAPS-1 domains not overlapping with VGLUT1 (filled arrowheads), VGLUT1 puncta not enriched for CAPS-1 (open arrowheads). Synapses (VGLUT1 puncta) overlapping with CAPS-1 rich domains (stars). Scale bar 5 µm. (F) CAPS-1 co-localization with VGLUT1 in the entire neuron is lower compared to co-localization of Munc13-1 and VGLUT1. Pearson's correlation M13-1-VGLUT1: 0.6±0.04, n=10; CAPS-1-VGLUT1: 0.4±0.08, n=12, *p<0.05. (G) Percentage of VGLUT1 labeled synapses expressing CAPS-1 is lower than VGLUT1 labeled synapses expressing (VGLUT1/CAPS-1: 58.5±10.9%, n=7 neurons, number of synapses = 239; VGLUT1/M13-1: 96.2±0.7%, n=4 neurons, number of synapses = 350). (H) Example images of neurites from hippocampal neurons (DIV 14) infected with lentivirus encoding NPY-Venus (green) and stained for chromogranin B (ChgB, red). Scale bar 2 µm. (I) Mander's coefficients for the proportion of endogenous ChrB immuno-reactivity in NPY-Venus puncta: 0.97±0.02, n=14 neurons or proportion of NPY-Venus immunoreactivity in ChrB puncta: 0.84±0.01, n=14 neurons. (J) Example images of neurites from hippocampal neurons (DIV 14) infected with lentivirus encoding NPY-Venus and stained for CAPS-1 (red) and MAP2 (blue). Scale bar 2 µm. (K) Quantification of co-localization of CAPS-1 and NPY-Venus in the entire neuron. Pearson's coefficient: 0.45±0.05; n=21 neurons. (L) Percentage of NPY-Venus labeled DCVs co-localizing with CAPS-1: 34.71±3.03%, n=6, number of DCVs = 414.
DOI:http://dx.doi.org/10.7554/eLife.05438.003
10.7554/eLife.05438.004
Specificity of CAPS-1 antibody.
(A) CAPS-1 immunoreactivity (green) and co-localization with synaptobrevin (VAMP, red) in CAPS-2 KO neurons stained for the dendritic marker MAP2 (blue) (top panel). Bottom panel: absence of immunoreactivity in CAPS DKO neurons. Scale bar 10 μm. (B) CAPS-1 immunoreactivity (green) and co-localization with VGLUT1 (red). Bottom panel: zoom indicated by the box in top panel. Scale bar 10 μm.
DOI:http://dx.doi.org/10.7554/eLife.05438.004
10.7554/eLife.05438.005
Endogenous DCV marker (chromogranin <bold>B</bold>) distribution in isolated single neurons.
(A) Confocal image of a WT hippocampal neuron (DIV 14) labeled with antibodies for the dendritic marker MAP2 (green), endogenous DCV cargo protein ChgB (red) and the synaptic marker VGLUT1 (blue). A punctate distribution of ChgB is seen in dendritic (arrow) and axonal regions (MAP2-negative; arrowheads) and accumulations of ChgB are found in growth cones (asterisks). (B) Higher magnification of boxed area 1 (zoom 1) shows ChgB puncta in axonal regions. Zoom 2 shows co-localization of ChgB and VGLUT1, indicating the presence of DCVs in synapses.
DOI:http://dx.doi.org/10.7554/eLife.05438.005
DCV secretion is severely reduced upon CAPS deletion in hippocampal neurons
CAPS proteins are implicated in catecholamine uptake into secretory vesicles in chromaffin cells (Speidel et al., 2005). As a defect in vesicle loading could influence our analysis of release events, we first analyzed possible effects of CAPS deletion on the loading of proteins into DCVs. The fluorescence intensity distribution of single DCV release events measured with the DCV cargo Semaphorin-3A coupled to pH-sensitve GFP (SemapHluorin, see below) in WT and CAPS DKO cells was similar; indicating that SemapHluorin loading of fusing DCVs was not affected in CAPS DKO neurons (Figure 2A). To analyze SemapHluorin loading of all DCVs in the cell, we quantified the fluorescence intensity of SemapHluorin labeled DCVs upon application of NH4+ (which instantly de-quenches all intra-vesicular pHluorin, Figure 2B) and of antibody-labeled endogenous DCV cargo protein, secretogranin II, in WT and CAPS DKO neurons. NH4+ application showed that loading of SemapHluorin was comparable between WT and CAPS DKO neurons (Figure 2C, D). Also, fluorescence intensity levels of secretogranin II were unchanged between WT and CAPS DKO neurons (Figure 2E). As neuronal viability is affected in CAPS-2 null mutant cerebellum (Sadakata et al., 2004, 2007a), we analyzed neuronal morphology and DCV numbers in CAPS DKO neurons and did not find differences between WT and DKO neurons (Figure 2F–H). Hence, neuronal morphology, DCV biogenesis and protein loading in hippocampal neurons are not affected by deletion of CAPS-1 and CAPS-2.10.7554/eLife.05438.006
CAPS deletion does not influence DCV peptide loading or DCV biogenesis in hippocampal neurons.
(A) Frequency distribution of fluorescence intensity increase (ΔF) of individual DCV fusion events in WT and CAPS DKO neurons. No major differences are observed in ΔF of fusion events between WT and CAPS DKO neurons indicating similar Semaphluorin loading per DCV in WT and CAPS DKO neurons. The number of DCVs per bin is normalized to the total number of DCVs released. (B) Inverted wide-field image of a neuron expressing Semaphluorin upon NH4+ application to reveal all DCVs present in the cell. Zoom shows the effect of vesicle de-acidification upon NH4+ application (−NH4+ before application, +NH4+ during application). (C) Average number of DCV puncta per neuron quantified from the NH4+ response is similar in WT (n=16 neurons) and CAPS DKO (n=24) neurons. (D) Average intensity (in arbitrary units, AU) of single DCV puncta quantified from the NH4+ response in WT and CAPS DKO neurons is similar (WT n=16 neurons and 4435 puncta, CAPS DKO n=24 neurons and 4892 puncta). (E) Average intensity (in AU) of single DCV puncta in the field of view of confocal images is similar in non-transfected WT (n=10 neurons) and CAPS DKO (n=10) neurons stained for the endogenous DCV cargo secretogranin II and the dendritic marker MAP2. (F) Average number of DCV puncta per field of view. (G) Average dendritic length per field of view. (H) Number of DCV puncta per dendritic length.
DOI:http://dx.doi.org/10.7554/eLife.05438.006
To examine the function of CAPS proteins in DCV secretion we used two different fluorescent DCV cargo proteins in two secretion assays, hippocampal mass cultures and isolated single neuron cultures. First, DCVs were labeled with the secreted axon-guidance protein Semaphorin 3A coupled to the pH-sensitive enhanced green fluorescent protein (EGFP) variant, pHluorin (SemapHluorin, Figure 3A, B) and release was measured in CAPS-1/CAPS-2 DKO (Jockusch et al., 2007) and WT hippocampal neurons in DIV 14 mass cultures (Figure 3A). We have previously shown that SemapHluorin labels all DCVs in cultured neurons and reports single DCV fusion events as a sudden increase in fluorescence upon opening of the fusion pore (de Wit et al., 2009; van de Bospoort et al., 2012). DCV release was triggered by electrical stimulation using a protocol known to elicit robust neuropeptide release from neurons (16 bursts of 50 action potentials at 50 Hz, Hartmann et al., 2001; de Wit et al., 2009; van de Bospoort et al., 2012). Figure 3B top panel shows a typical fusion event reported by SemapHluorin: upon fusion pore opening intravesicular pH rises sharply unquenching pHluorin, which results in a strong increase in fluorescence. Fluorescence increase of two standard deviations above initial fluorescence (Figure 3B, grey dotted line) was scored as fusion event in panels C and E. After this increase, fluorescence may remain high (left trace, persistent event) or may decay (middle and right trace, transient events). Both persistent and transient events were counted as fusion events in panels C and E. Figure 3—figure supplement 1 explains the different fusion modes reported by SemapHluorin in more detail (see also de Wit et al., 2009).10.7554/eLife.05438.007
CAPS deletion reduces DCV fusion in hippocampal neurons.
(A) Schematic drawing of a neuronal mass culture used in these experiments in which one neuron expresses a fluorescently tagged morphology marker (green) and a DCV marker (red). Non-labeled neurons are indicated in light blue. (B) Top panel: two stills showing a typical example of a fusion event of a SemapHluorin labeled DCV with the corresponding schematic drawing. Fusion pore opening causes a sudden increase of fluorescence intensity corresponding to the increase in intravesicular pH. Bottom panel: example traces of SemapHluorin labeled DCV fusion events. Fluorescence increase of two standard deviations above initial fluorescence (grey dotted line) was scored as fusion event in C and E. After which, fluorescence may decrease (transient events) or remain high (persistent events). Figure 3—figure supplement 1 explains this behavior in detail. Scale bar 1 μm. (C) Average number of DCV fusion events per field of view during electrical stimulation with 16 bursts of 50 AP at 50 Hz with 0.5 s interval (WT: 23.7±4.5, n=24 neurons; CAPS DKO: 9.6±1.5, n=32 neurons, N=4, independent experiments, **p<0.01). (D) Cumulative frequency plot of DCV fusion events during stimulation (average vesicle fusion rate per cell during stimulation WT: 1.2±0.4 vesicles/s; CAPS-1/2 DKO: 0.15±0.1 vesicles/s). Blue bars represent stimulation of 16×50 AP at 50 Hz with 0.5 s interval. (E) Number of DCV fusion events per field of view during the first four bursts of the stimulation in C, (WT: 4.8±1.5, n=24; CAPS DKO: 1.4±0.3, n=32, *p<0.05). (F) Cumulative frequency plot of DCV fusion events during the first four bursts of the stimulation, showing that the initial release rate is slower in CAPS DKO neurons.
DOI:http://dx.doi.org/10.7554/eLife.05438.007
10.7554/eLife.05438.008
Different fusion events reported by SemapHluorin.
Cartoon depicting the different fusion events reported by SemapHluorin and analyzed in depth in de Wit et al., 2009. Electrical stimulation triggers calcium and SNARE protein-dependent membrane fusion. Opening of the fusion pore results in a sudden dequenching of vesicular SemapHluorin and increase in fluorescence (asterisk). This is scored as fusion event in Figure 3C,vE. Upon the sudden increase in fluorescence the signal either remains high (A: persistent event) or dims to baseline (B: transient event). Transient events represent incomplete release followed by vesicle retrieval and re-acidification (B1) or full release followed by cargo diffusion (B2). In B2 the vesicle may integrate into the plasma membrane or re-seal without SemapHluorin cargo (B2, grey box top and bottom panel, respectively). Persistent events reflect the continuous presence of cargo at the cell surface and may either represent stable deposits of SemapHluorin or vesicles with a permanently open fusion pore (A, grey box top and bottom panel, respectively). Persistent events are typical for cargo that interacts with the intra-vesicular matrix like Semaphorin. These events are very rare when using NPY as cargo. Our previous analysis of fusion events reported by SemapHluorin showed that 60% of fusion events are transient events of which half result in full release and 40% of events are persistent (de Wit et al., 2009). The typical example traces show persistent and transient fusion events. Both types of events were scored as fusion events in Figure 3C, E when fluorescence increased above two times the standard deviation of the initial fluorescence (grey dotted line).
DOI:http://dx.doi.org/10.7554/eLife.05438.008
Upon stimulation, CAPS DKO neurons showed a more than 60% reduction in the number of DCV fusion events compared to WT (Figure 3C, WT: 23.7±4.5 events/field of view, n=24; CAPS DKO: 9.6±1.5 events/field of view, n=32, N=4, p<0.01). The average vesicle fusion rate during stimulation, calculated from the cumulative release plots (Figure 3D), was 1.2±0.4 vesicles/s for WT compared to 0.15±0.1 vesicles/s for CAPS DKO neurons. Also when we zoomed in on the first four bursts of 50 action potentials, the number of fusion events was significantly lower in CAPS DKO neurons compared to WT (Figure 3E, F). Together, these findings show that deletion of both CAPS isoforms strongly reduces activity-dependent secretion of neuronal DCVs by decreasing the total number of fusing vesicles and vesicle release rate during stimulation.
CAPS-1 is the CAPS isoform responsible for loss of DCV secretion in hippocampal neurons
Mass cultures can be used to assess the average number of release events per field of view, not per neuron. Therefore, we adapted a single neuron culture protocol using micro-islands of astrocytes (Bekkers and Stevens, 1991; Wierda et al., 2007), which enables imaging of the entire axonal and dendritic arbor of a single neuron to perform quantitative single-cell DCV release measurements (Figure 4A). DCVs were labeled with NPY-pHluorin, and vesicle fusion was triggered by electrical stimulation (as in Figure 3). As CAPS-2 expression in hippocampal neurons is very low and deletion of CAPS-2 does not affect SV release (Jockusch et al., 2007), we used neurons from CAPS-2 null mutant (CAPS-2KO) littermates as controls. The number of DCV fusion events per cell reported by the sudden increase in fluorescence upon fusion pore opening was strongly decreased in CAPS DKO compared to controls (Figure 4B, C, CAPS-2KO: 49.6±16.3 events/cell, n=9; CAPS DKO: 10.6±7.9 events/cell, n=5, N=3; p<0.05). Hence, CAPS-1 deletion strongly inhibits DCV fusion in isolated hippocampal neurons.10.7554/eLife.05438.009
CAPS-1 deletion impairs DCV fusion in isolated neurons.
(A) Schematic drawing of a single isolated neuron grown on a micro island of astrocytes and expressing fluorescently tagged morphology marker (green) and a synapse marker (red) in addition to either NPY-pHluorin or NPY-mCherry (not shown). Average island diameter is 375 μm. (B) Average number of DCV fusion events per cell upon electrical stimulation of 16 bursts of 50 AP at 50 Hz using NPY-pHluorin as DCV marker (CAPS-2KO: 49.6±16.3, n=9; CAPS DKO: 10.6±7.9, n=5, number of independent experiments (N)=3, *p<0.05). Inset shows a typical example of a fusion event reported by NPY-pHluorin. Fluorescence increase of two standard deviations above initial fluorescence (grey dotted lines) was scored as fusion event in B. (C) Cumulative frequency plot of DCV fusion events in B, showing that DCV release is triggered by electrical stimulation paradigm (blue bars represent 16 bursts of 50 AP at 50 Hz, 16×50 AP at 50 Hz). (D) White arrowhead: Typical example of a fusion event with complete cargo release reported by the sudden and complete disappearance of NPY-mCherry fluorescence intensity. These fusion events were in measured in E and K. Time in seconds (s) after start of stimulation. Neurite marker is ECFP (blue). Scale bar 1 μm. (E) Average number of fusion events with complete cargo release per cell upon electrical stimulation using NPY-mCherry as DCV marker (CAPS-2KO: 14.6±3.3, n=14; CAPS DKO: 3.4±1.4, n=8, N=3, **p<0.01). (F) Cumulative frequency plot of DCV fusion events with complete cargo release in E, (blue bars represent 16 bursts of 50 AP at 50 Hz, 16×50 AP at 50 Hz). (G) Percentage of non-secreting cells is increased in CAPS DKO neurons (CAPS-2KO: 15±12.4%, n=26; CAPS DKO 53.6±12.6%, n=26, *p<0.05). Non-secreting cells were excluded from the analyses in B and E. (H) Example images showing NPY-pHluorin labeled DCVs fusing at synapsin-mCherry labeled synapses, indicated by the red dashed lines and DCV fusion events at extra-synaptic sites, indicated by the green dashed lines. Bar is 1 μm. (I) Percentage of synaptic and extra-synaptic DCV release events measured with NPY-pHluorin shows a similar distribution in CAPS DKO compared to CAPS-2KO (CAPS-2KO synaptic: 67.0±3.6, CAPS-2KO extra-synaptic: 0.32±1.4, n=20, **p<0.01; CAPS DKO synaptic: 64.7±2.3, CAPS DKO extrasynaptic: 35.6±0.9, n=28, **p<0.01). (J) CAPS-1 expression levels assessed by semi-quantitative immunofluorescence in CAPS-2KO, CAPS DKO, and CAPS DKO expressing CAPS-1 (Rescue) (CAPS-2KO: 2.8±0.7 AU, n=5; CAPS DKO: 0.1±0.0, n=4; Rescue: 2.4±0.7, n=3, ***p<0.0001). (K) Average number of fusion events leading to complete release per cell upon electrical stimulation with NPY-mCherry as DCV marker in CAPS-2KO, CAPS DKO and Rescue (CAPS-2KO: 8.5±1.7, n=25; CAPS DKO: 0.7±0.3, n=16; Rescue: 6.4±1.7, n=23, N=6). (L) Cumulative frequency plot of complete release DCV fusion events in K, (blue bars represent 16 bursts of 50 AP at 50 Hz, 16×50 AP at 50 Hz).
DOI:http://dx.doi.org/10.7554/eLife.05438.009
DCV fusion pore opening can progress to complete release of cargo or to partial release and fusion pore resealing (Alabi and Tsien, 2013; Wu et al., 2014). To investigate these release modes in CAPS-2 KO and CAPS DKO neurons, we labeled DCVs with NPY-mCherry, which reports full cargo release events as complete disappearance of fluorescent puncta (Figure 4D). The average number of such fusion events in control cells was much lower compared to events reported by NPY-pHluorin, which reports all fusion events irrespective of complete or incomplete release of cargo (compare Figure 4E and Figure 4B, NPY-mCherry CAPS-2KO: 14.6±3.2 events/cell; NPY-pHluorin CAPS-2KO: 49.6±16.3 events/cell). However, also with this reporter a more than 50% decrease in the number of fusion events per cell was observed in CAPS DKO neurons compared to CAPS-2KO (Figure 4E, F, CAPS-2KO: 14.6±3.2 events/cell, n=14; CAPS DKO: 3.4±1.3 events/cell, n=8, N=3, p<0.01). In addition to the strong decrease in DCV fusion events in CAPS DKO neurons, in almost 50% of CAPS DKO cells expressing either NPY-pHluorin or NPY-mCherry electrical stimulation failed to induce vesicle fusion (Figure 4G, these cells were excluded from analysis in Figure 4B–F).
We previously showed that DCVs preferentially fuse at synapses and that Munc13 plays a crucial role in synaptic preference of DCV release (van de Bospoort et al., 2012). To test whether CAPS deletion affected synaptic preference of DCV fusion we labeled presynaptic terminals with the live cell marker synapsin-mCherry and analyzed NPY-pHluorin fusion events (Figure 4H). Unlike Munc13 deletion, deletion of CAPS affected extra-synaptic and synaptic DCV release to the same extent (Figure 4I).
To confirm that the secretion defect in CAPS DKO neurons was due to lack of CAPS-1, CAPS-1 was re-introduced in CAPS DKO cells at 10 DIV via lenti-virus infection (‘rescue’) and DCV release was tested 4 days later. CAPS-1 expression levels in rescued neurons were comparable to endogenous levels (Figure 4J). Introduction of CAPS-1 in CAPS DKO cells rescued the secretion defect (Figure 4K) and release kinetics were similar to control cells (Figure 4L). Thus, deletion of CAPS-1 resulted in a strong reduction of DCV fusion events reported by NPY-PHluorin or NPY-mCherry and fusion at synapses and extra-synaptic sites was affected to the same extent.
CAPS-1 deletion strongly affects fusion of stationary DCVs
DCVs can be highly mobile or stationary (de Wit et al., 2006; Wong et al., 2012; Goodwin and Juo, 2013). Both stationary and mobile vesicles fused upon electrical stimulation (Figure 5A). In control neurons the majority of DCV fusion events were stationary DCVs (Figure 5B, CAPS-2KO: stationary 12.2±2.8 fusion events per cell, moving 4.0±1.0 fusion events per cell, n=10, N=3, p<0.05). However, in CAPS DKO neurons, release of stationary vesicles was strongly decreased while release of mobile vesicles was similar to CAPS-2KO (Figure 5B, CAPS DKO: stationary 2.0±0.8 fusion events per cell, moving 3.0±1.1 events per cell, n=5, N=2).10.7554/eLife.05438.010
Deletion of CAPS-1 affects fusion of stationary DCVs.
(A) Stills from DCV fusion assay. Electrical stimulation starts at second 10. Yellow arrowheads indicate stationary vesicles that fuse, red arrowhead indicates a stationary vesicle that does not fuse and white arrow shows a moving DCV that fuses. Kymograph shows the trajectories of the stationary and moving vesicle over time. (B) Fusing DCVs classified as stationary or moving showing that CAPS deletion strongly affects fusion from stationary vesicles (CAPS-2KO stationary prior to fusion: 12.2±2.8, moving: 4±1.0, n=10, *p<0.05; CAPS DKO stationary: 2.3±1.3, moving: 4.0±1.7, n=3, ns, N=3). (C) Non-fusing DCVs classified as stationary or moving showing that CAPS deletion does not affect general trafficking behavior of non-fusing vesicles (CAPS-2KO stationary: 33.8±1.2, moving: 17.6±0.7, n=5, *p<0.05; CAPS DKO stationary: 33.7±3.0, moving: 19.3±2.4, n=3, *p<0.05). (D) Average velocity of DCVs classified as moving prior to fusion in B, before stimulation (PreStim = 30 s; CAPS-2KO: 193.6±9.9 nm/s, n=169 vesicles, CAPS DKO: 180.4±9.2 nm/s, n=18) and during electrical stimulation (Stim: from second 30 to the onset of fusion; CAPS-2KO = 180.9±6.6 nm/s, n=169; CAPS DKO: 169.2±7.5, n=18). (E) Average velocity of non-fusing DCVs in C, before stimulation (PreStim = 30 s; CAPS-2KO: 190.8±13.7 nm/s, n=168, CAPS DKO 186.8±13.5 nm/s, n=91), during stimulation (Stim = 24 s; CAPS-2KO: 175.3±6.7 nm/s, n=168; CAPS DKO 201.8±11.8 nm/s, n=91) and after stimulation (PostStim = 36 s CAPS-2KO: 182.2±6.2 nm/s, n=168, CAPS DKO 196.2±8.5 nm/s, n=91). (F) Typical examples of electron micrographs of neuronal DCVs in synapses. Scale bar 50 nm. (G) Number of synapses containing one or more DCVs (CAPS-2KO: 53.9±12.1%, n=198 synapses; CAPS DKO 47.6±5.4%, n=152 synapses, N=4). (H) Percentage of docked DCVs per synapse (CAPS-2KO: 10.3±1.7%, DCVs = 285, CAPS DKO: 11.1±3.7%, DCVs = 201). (I) Average distance of DCVs to the closest plasma membrane (CAPS-2KO: 81.2±4.8; CAPS DKO: 84.9±10.4). (J) Average number of DCVs per synapse (CAPS-2KO: 1.6±0.2; CAPS DKO: 1.5±0.2). (K) Frequency distribution of number of DCVs per synapse (% synapses, normalized to the total number of DCVs per group).
DOI:http://dx.doi.org/10.7554/eLife.05438.010
10.7554/eLife.05438.011
DCV docking definition and zooms of <xref ref-type="fig" rid="fig5">Figure 5F</xref>.
Zooms of three example DCVs showing the classification of docked versus undocked vesicles. We define a docked DCV as a vesicle containing a dense core that has no detectable distance between the vesicular membrane and the plasma membrane. The minimal distance between membranes we can visualize is 0.52 nm. (A) Electron micrograph showing an example of docked DCV. (B) Electron micrograph showing an example of DCV close but not docked to the plasma membrane. (C) Electron micrograph showing an example DCV far away to the plasma membrane. Scale bars 100 nm. (D) Enlarged high-resolution electron microscopy images of Figure 5F. Scale bar 50 nm.
DOI:http://dx.doi.org/10.7554/eLife.05438.011
To test if CAPS deletion affected general DCV trafficking, we analyzed the dynamics of DCVs that did not fuse during the 90s imaging protocol. No differences between CAPS-2KO and CAPS DKO neurons were observed for these vesicles (Figure 5C, CAPS-2KO: stationary 33.8±1.2, moving: 17.6±0.7, n=5, p<0.05; CAPS DKO: stationary 33.66±3.0, moving 19.33±2.4, n=3, p<0.05). We also did not detect differences in average velocity of fusing and non-fusing DCVs prior to stimulation, during stimulation or in the period after stimulation (Figure 5D, E).
CAPS/UNC-31 deletion reduces the number of docked DCVs in C. elegans neuromuscular junctions (Hammarlund et al., 2008) while the total number of DCVs in mammalian synapses is not different between control and CAPS DKO (Jockusch et al., 2007). We examined electron micrographs of synapses from CAPS DKO and control neurons to test if DCV localization was affected by deletion of CAPS. No morphological differences between the two groups were observed (Figure 5F and Figure 5—figure supplement 1D). The number of synapses containing DCVs (Figure 5G), the percentage of docked DCVs (Figure 5H, see Figure 5—figure supplement 1A–C for our definition of docked vesicles), the distance of DCVs to the closest plasma membrane (Figure 5I), and the average number of DCVs per synapse (Figure 5J, K) were similar between CAPS DKO and control synapses. Thus, deletion of CAPS-1 does not affect transport of DCVs or DCV localization at synapses but specifically affects fusing vesicles, by reducing the number of fusion-competent stationary vesicles.
CAPS-1-EYFP fusion protein replaces endogenous CAPS-1 in DCV secretion and synaptic transmission
To investigate the intracellular dynamics of CAPS-1, we introduced CAPS-1 fused to enhanced yellow fluorescent protein (EYFP) (CAPS-1-EYFP) in CAPS DKO neurons (Figure 6A). As for endogenous CAPS-1 (Figure 1), the majority of CAPS-1-EYFP co-localized with the live-synaptic marker synapsin-ECFP (Figure 6A, bottom panel, stars). In addition, we observed, again similar to endogenous CAPS-1, synapses without detectable CAPS-1-EYFP (Figure 6A, bottom panel, open arrowhead) and CAPS-1 rich domains that did not co-localize with synapses (Figure 6A, bottom panel, filled arrowhead).10.7554/eLife.05438.012
CAPS-1-EYFP fusion protein replaces CAPS-1 in DCV secretion and synaptic transmission.
(A) Isolated single CAPS DKO neuron grown on glia micro-island expressing CAPS-1-EYFP (green) and synapsin-ECFP (blue). Top panels: maximum projection of the entire neuron, Scale bar 10 µm. Bottom panels: zooms of top panels showing synapses without detectable CAPS-1-EYFP (open arrowhead), CAPS-1 rich domain not overlapping with synapsin-ECFP (filled arrowhead) and a CAPS-1 rich synapse (star). Scale bar 5 µm. (B) Average number of DCV fusion events per cell (events/cell) upon electrical stimulation (16×50 AP at 50 Hz) using NPY-pHluorin as DCV marker in CAPS DKO neurons with (+CAPS-1-EYFP) and without (CAPS DKO) CAPS-1-EYFP (+CAPS-1-EYFP: 23.0±6.8, n=8, CAPS DKO: 0.7±0.3, n=16, ***p<0.0001, Mann–Whitney test). (C) Example traces of evoked EPSCs in CAPS-2KO (black), CAPS DKO (dark green) and CAPS-1-EYFP (light green) rescued CAPS DKO neurons. (D) Expression of CAPS-EYFP rescues EPSC amplitude in CAPS DKO neurons (CAPS-2KO: 2.4±0.3 nA, n=4; CAPS DKO: 0.48±0.15 nA, n=3; CAPS-1-EYFP: 2.0±0.4 nA, n=3, **p<0.01). (E) Example traces of spontaneous release (mEPSCs). (F) Mean mEPSC frequency. (CAPS-2KO: 7.5±1.1 Hz, n=3; CAPS DKO: 1.7±1.0 Hz, n=3; CAPS-1-EYFP: 6.9±1.3 Hz, n=2). (G) Mean mEPSC amplitude. (CAPS-2KO: 29.0±4.1 pA, n=3; CAPS DKO: 27.1±5.0 pA, n=3; CAPS-1-EYFP: 26.8±7.2 pA, n=2). (H) Changes in EPSC amplitude induced by 100-pulse train at 40 Hz during low frequency (0.1 Hz) stimulation. The interval between low- and high-frequency stimulation is 3 s (I) 100-pulses at 40 Hz induced rundown of normalized EPSC amplitude (zoom of H).
DOI:http://dx.doi.org/10.7554/eLife.05438.012
CAPS-1-EYFP efficiently rescued DCV secretion in CAPS DKO neurons (Figure 6B) and also rescued synaptic transmission. Whole cell patch-clamp recordings of single neurons on micro-dot astrocyte islands revealed no differences between control and CAPS-1-EYFP rescued CAPS DKO neurons. Evoked postsynaptic current amplitude (EPSC, Figure 6C, D) and spontaneous release characteristics (mEPSC frequency and amplitude, Figure 6E–G) were similar to controls. Also EPSC rundown during high-frequency stimulation (40 Hz) and its recovery (Figure 6H, I) were completely rescued. Hence, CAPS-1-EYFP mimics endogenous CAPS-1 and was used for further experiments to study the dynamics of this protein in living cells.
Localization of CAPS-1 at synapses is calcium dependent
We studied the dynamics of CAPS-1-EYFP upon electrical stimulation using the same stimulation paradigm used to elicit DCV release (16×50 AP at 50 Hz). We infected hippocampal neurons with lentiviral particles encoding CAPS-1-EYFP and synapsin-ECFP and imaged the two fluorophores simultaneously (Figure 7A, B). During stimulation, the fluorescence intensity of CAPS-1-EYFP and synapsin-ECFP at synapses strongly decreased while the extra-synaptic intensities of both proteins increased (Figure 7A–F). A large variability of intensity changes upon stimulation was observed between cells. In some neurons fluorescence intensity of synapsin-ECFP and CAPS-1-EYFP returned to baseline within 3 min (Figure 7C, D), while in other cells the intensities of both proteins remained below their initial fluorescence (Figure 7E, F). Fluorescence intensities of membrane-bound EYFP did not change at synapses during stimulation showing that the decrease of CAPS-1-EYFP fluorescence at synapses reported dispersion of the protein into neurites (Figure 7D, F, insets). The average response of synaptic CAPS-1-EYFP showed a similar dynamic profile as synapsin-ECFP, dispersing from synapses during and up until 3 min after calcium influx, while extra-synaptic CAPS-1-EYFP fluorescence increased during stimulation (Figure 7G, H). Thus, during stimulation a fraction of synaptic CAPS-1 redistributes from synapses into neurites similar to the synaptic vesicle protein synapsin 1.10.7554/eLife.05438.013
Localization of CAPS-1 at synapses is calcium dependent.
(A and B) Grey scale images of synapsin-ECFP (left) and CAPS-1-EYFP (right) labeled synapses showing the same region before (pre-15s) during (stim-22s) and after (post-150s) electrical stimulation (16×50 AP at 50 Hz). Synapsin-ECFP and CAPS-1-EYFP were imaged simultaneously at 0.5 Hz. (C and E) Traces of relative intensity changes (ΔF/F0) of synapsin-ECFP at synapses and extra-synaptic locations (79 synaptic and 31 extra-synaptic) showing increased extra-synaptic and decreased synaptic fluorescence upon stimulation (16×50 AP at 50 Hz). Open arrowheads in C indicate the pre- and post stimulations time points (same for D–F). (D and F) Example traces of relative intensity changes (ΔF/F0) of CAPS-1-EYFP at synapsin-ECFP labeled synapses from C. showing increased extra-synaptic and decreased synaptic fluorescence upon stimulation (16×50 AP at 50 Hz). Inset: synaptic fluorescence of membrane associated EYFP (EYFP control) as control. (G) Average relative intensity profiles of synapsin-ECFP, and CAPS-1-EYFP at synapses and CAPS-1-EYFP extra-synaptic, (395 synaptic regions, 155 extrasynaptic regions, n=5 cells). (H) Maximum relative intensity changes (max ΔF/F0) of synapsin-ECFP, and CAPS-1 at synapses and CAPS-1 at extra synaptic regions at t=160 s calculated from G (**p<0.01, n=5 cells each).
DOI:http://dx.doi.org/10.7554/eLife.05438.013
DCV release probability correlates with CAPS-1 expression at single synapses
We showed that on a cellular level, removing CAPS-1 strongly impairs DCV fusion (Figure 3 and Figure 4). To test whether at the single synapse level CAPS expression levels correlated with DCV release probability we simultaneously imaged NPY-mCherry labeled DCVs and CAPS-1-EYFP (Figure 8A). Like endogenous CAPS-1, CAPS-1-EYFP was present in puncta (Figure 8A, left top panel). The majority of these were stationary throughout the imaging experiment (Figure 8A, B). Only 5% of all CAPS-1-EYFP puncta were mobile during our imaging paradigm (Figure 8B). Furthermore, the number of mobile CAPS-1-EYFP puncta that co-trafficked with dynamic NPY-mCherry labeled DCVs was very low (seven out of 106 mobile DCVs showed co-trafficking of CAPS-1-EYFP, Figure 8C).10.7554/eLife.05438.014
Presence of CAPS-1 increases DCV release probability at single synapses.
(A) Left top and bottom panels: kymographs of CAPS-1-EYFP and NPY-mCherry imaged simultaneously (acquisition frequency 2 Hz) for 90 s and stimulated at second 30 (dashed box, 16×50 AP at 50 Hz). Note the dispersion of the majority of CAPS-1-EYFP puncta upon stimulation and the disappearance of NPY-mCherry puncta upon stimulation. Top right panel shows the merge of the two channels. The bottom right panel shows a schematic drawing of the merged CAPS-1-EYFP and NPY-mCherry channels to aid in the interpretation of the kymographs. Open arrowhead indicates DCV not co-localizing with CAPS-1-EYFP. Filled arrowhead indicates a CAPS-1-EYFP punctum co-trafficking with a mobile DCV. Scale bar 5 µm. (B) Percentage of stationary and moving CAPS-1-EYFP puncta during image acquisition as described in (A). (Stationary: 96.6±1.7, moving: 3.3±1.7, number of cells = 9; number of kymographs per cell = 3, total number of puncta = 116). (C) Percentage of mobile DCVs co-trafficking with CAPS-1-EYFP (mobile DCVs not co-trafficking with CAPS-1-EYFP (green bar): 99 or 93.4%; mobile DCVs co-trafficking with CAPS-1-EYFP: 7 or 6.3%, total number of cells = 20, moving DCVs analyzed = 106). (D) Typical examples of synapsin-ECFP labeled synapses with high expression levels of CAPS-1EYFP (star) or low expression levels of CAPS-1-EYFP (open arrowhead). (E) Percentage of DCV release events occurring at synapses enriched for CAPS-1 (black bar) or depleted for CAPS-1 (green bar), (synapses + CAPS-1: 84.0±3.3%, synapses − CAPS-1: 16.0±3.3%, total DCVs released = 84, n=5, ***p<0.0001). (F) Cumulative frequency plot of the DCV release events in C. showing that release at CAPS-1 deficient synapses is reduced and delayed. Blue box represents 16×50 AP at 50 Hz stimulation. (G) Released DCVs categorized in stationary or moving before fusion (synapses + CAPS-1: stationary DCVs = 46, moving DCVs = 23, total DCVs released = 69; synapse − CAPS-1: stationary DCVs = 6 moving DCVs = 8, total DCVs released = 14).
DOI:http://dx.doi.org/10.7554/eLife.05438.014
As CAPS-1 co-localizes with synaptic markers but is not enriched at every synapse (Figure 1A–C, Figure 8D), we tested whether CAPS-1 positive synapses (Figure 8D, star) have a higher DCV release probability than CAPS-1-EYFP deficient synapses (Figure 8D, open arrowhead). Synapses expressing CAPS-1-EFYP secreted more DCVs upon stimulation than synapses with no detectable CAPS-1-EYFP expression (Figure 8E, percentage of DCVs released at CAPS-1-EYFP positive synapses: 84.0±3.3%, −CAPS-1: 16.0±3.1%, n=83 DCVs from five cells, ***p<0.0001). Furthermore, DCV fusion in synapses with no detectable CAPS-1-EYFP expression required more prolonged stimulation than in CAPS-1-EYFP positive synapses (Figure 8F). Also, CAPS-1 enriched synapses released more DCVs from a stationary pool than from a mobile pool compared to CAPS-1 deficient synapses (Figure 8G, CAPS-1-EYFP positive synapses: stationary DCVs: 46; moving DCVs: 23; total DCVs counted: 69; CAPS-1-EYFP negative synapses: stationary DCVs: 6; moving DCVs: 8; DCVs counted: 14). Together these results show that DCVs preferentially reside and fuse at CAPS-1 rich domains and that CAPS-1-EYFP and DCVs rarely travel together in DIV 14 neurons. Furthermore, DCV release probability correlated with synaptic CAPS-1 expression levels.
Discussion
Here we report that, in mammalian neurons, CAPS-1 is an important regulator of DCV fusion. CAPS-1 is present at synapses and co-localizes with DCV markers but rarely travels together with dynamic DCVs. Using two DCV secretion assays we found that CAPS-1 deletion strongly impaired synaptic and extra-synaptic DCV release in hippocampal neurons. CAPS-1 deletion did not reduce the presence or docking of DCVs at synaptic sites, but strongly affected DCV secretion of stationary DCVs. Synaptic expression levels of CAPS-1 were modulated by neuronal activity and correlated with DCV release probability. CAPS-1 thus functions as a priming protein at synaptic and extra-synaptic sites, promoting the fusion competence of stationary DCVs.
CAPS-1 is present at mammalian synapses and overlaps with secretory vesicles markers
CAPS-1 was identified as a brain protein (Walent et al., 1992) that binds membranes associated with DCV secretion (Berwin et al., 1998) and stimulates calcium-dependent fusion of secretory vesicles in PC12 cells (Mennerick et al., 1995; Loyet et al., 1998; Grishanin et al., 2002). This suggests that CAPS-1 might be a DCV resident protein that is delivered to sites of exocytosis on DCVs. However, we observed that in mammalian neurons endogenous CAPS-1 localized in axonal and dendritic domains that often, but not always, co-localized with synaptic and DCV markers. Cell-wide CAPS-1 co-localization with the canonical DCV marker NPY yielded an average co-localization of only 35%. Furthermore, in CAPS DKO neurons rescued with CAPS-1-EYFP, DCVs and CAPS-1-EYFP only rarely traveled together. Finally, the increased dynamics of synaptic CAPS-1-EYFP with CAPS-1 redistribution upon stimulation are best explained by diffusion (see below) but not with a redistribution of vesicular CAPS-1. Thus, in mature mammalian neurons, CAPS-1 does not appear to be a general DCV-resident protein, but a dynamic, cytosolic factor translocating to synapses and extra-synaptic sites in an activity-dependent manner to promote fusion of pre-docked (tethered) DCVs.
A post-docking role for CAPS-1 in DCV fusion
Figure 3 and Figure 4 show that in the absence of CAPS-1, DCV fusion in hippocampal neurons is severely impaired, without defects in vesicle biogenesis and loading of endogenous or exogenous DCV cargo proteins. In addition, analysis of DCV dynamics in our live cell imaging experiments and intra-synaptic localization on electron micrographs of CAPS DKO neurons revealed that CAPS-1 deletion does not affect DCV trafficking along microtubules, synaptic accumulation and membrane docking of DCVs. Together, this suggests that in mature neurons most CAPS-1 molecules interact with DCVs after these vesicles arrived at the plasma membrane. This conclusion is consistent with observations in Drosophila neuromuscular junctions where deletion of dCaps results in an increase in the number of DCVs present at synapses (Renden et al., 2001) and the lack of a docking defect upon deletion of CAPS-1/-2 in adrenal chromaffin cells (Liu et al., 2010). However, chemical fixation used in our study might produce artifacts that change the precise distance between vesicles and the membrane. As a consequence, vesicles touching the membrane in fixed tissue may in fact have been at a short distance from the membrane. Recent excellent work using cryo-fixation and EM tomography to analyse docking of SVs and DCVs in central synapses of several mutant mouse strains including CAPS DKO showed that SV docking is impaired in CAPS DKO neurons. In line with our findings, DCVs were present in similar numbers in CAPS DKO synapses. However, although not statistically significant, the distribution of DCVs within 200 nm of the active zone appeared to be reduced (Imig et al., 2014). Hence, general consensus exists on a post-synapse delivery role for CAPS but future studies using cryo-fixation might unmask (subtle) docking defects upon CAPS loss.
CAPS-1 likely functions at the final stages of DCV fusion, interacting with proteins and lipids that function in DCV docking, priming and fusion (Grishanin et al., 2004; Sadakata et al., 2007b; James et al., 2009; Parsaud et al., 2013; Sah and Geracioti, 2013). CAPS-1 changes the conformation of syntaxin-1 from a state incompatible with SNARE-complex formation (‘closed-state’), to a state that allows formation of functional SNARE complexes (‘open-state’), as release in CAPS DKO chromaffin cells is rescued by expression of an ‘open’-variant of syntaxin-1 (Liu et al., 2010) and overexpression of open syntaxin can bypass the requirement for CAPS in DCV docking in C. elegans (Hammarlund et al., 2008). CAPS-1 shares this characteristic with Munc13-1 (Augustin et al., 1999; van de Bospoort et al., 2012). CAPS and Munc13 proteins may operate in the same molecular priming pathway (Richmond et al., 2001; Jockusch et al., 2007; Zhou et al., 2007) in a non-redundant manner. The fact that both proteins appear to bind syntaxin-1 via different binding modes may account for this non-redundancy (Parsaud et al., 2013). A recent paper indeed showed that CAPS-2 and Munc13 use different mechanisms to prime vesicles, whereas Munc13-dependent priming requires its MUN domain this domain in CAPS-2 is dispensable for priming. Instead CAPS-2 appears to require its PIP2 binding pleckstrin homology domain (Nguyen Truong et al., 2014). We found that both proteins play important stimulatory roles in DCV release from mammalian neurons. However, functional differences were evident: CAPS-1 deletion resulted in a larger reduction of DCV release events compared to deletion of Munc13-1 with a ±70% reduction in CAPS DKO (this study) and ±60% reduction in Munc13 DKO neurons (van de Bospoort et al., 2012). For synaptic vesicle fusion the situation is opposite: Munc13 DKO neurons clearly show larger defects in synaptic transmission than CAPS DKO neurons (Varoqueaux et al., 2002; Jockusch et al., 2007).
CAPS-1 displayed a different expression pattern than Munc13-1. Munc13-1 expression is strictly synaptic and specifically supports DCV secretion from synapses, while being dispensable for extra-synaptic secretion (van de Bospoort et al., 2012). In contrast, CAPS-1 accumulations are found at many synapses, but not all, and at extra-synaptic sites along neurites. Also, CAPS-1 deletion equally affects DCV release from synapses and from extra-synaptic sites. Finally, CAPS-1 redistributes from synapses during activity while Munc13-1 does not (Kalla et al., 2006). These differences in (re)distribution suggest that (1) CAPS-1 domains at extra-synaptic sites promote DCV release independently of Munc13-1. However, CAPS-1-dependent extra-synaptic release is less efficient than synaptic DCV release, which indicates that the concerted action of CAPS-1 and Munc13-1 is most efficient in priming DCVs for release. (2) At synapses, CAPS-1 and Munc13-1 have non-redundant roles in promoting DCV release, similar to their non-redundancy in SV release (Jockusch et al., 2007). The larger effect of CAPS deletion on DCV release may be explained by the additional reduction of extra-synaptic release events compared to Munc13-1 deletion. Finally, some extra-synaptic DCV release remained in CAPS DKO neurons indicating that extra-synaptic DCV release can occur in the absence of CAPS and Munc13 albeit with very low release probability.
Strong stimulation triggered redistribution of synaptic CAPS-1 into the axonal shaft. This redistribution was very similar to synapsin-ECFP (our work and Richmond et al., 2001; Parsaud et al., 2013) and is also reported for other synaptic proteins like Rab3a (Tsuriel et al., 2009), Syntaxin-1 (Ribrault et al., 2011) and Munc18 (Cijsouw et al., 2014) and for SVs (Cheung and Cousin, 2011). In contrast, the mobility of active zone proteins like Munc13-1 and Bassoon is not affected by acute stimulation (Kalla et al., 2006; Tsuriel et al., 2009). Hence, synapses rapidly exchange part of their components during high frequency stimulation. This allows synapses to adapt their release probability during and directly after stimulation as we showed for Munc18-1 (Cijsouw et al., 2014). This is also an attractive explanation for the re-distribution of CAPS-1 as we found that synapses with increased CAPS-1 expression levels had a higher release probability than synapses with low/no CAPS-1. CAPS-1 binds to PI 4,5-P2 (PIP2) (Loyet et al., 1998) and localizes to PIP2 clusters in the plasma membrane via its PIP2-binding pleckstrin homology (PH) domain (James et al., 2008). The PH domain is also required to prime secretory vesicles (Kabachinski et al., 2014; Nguyen Truong et al., 2014). Robust Ca2+ influx in our experiments likely activates phospholipase C, which hydrolysis PIP2 and may trigger CAPS-1 dispersion from synapses. Hence, calcium-dependent PIP2 hydrolysis may act as negative feedback mechanism reducing CAPS-1 availability at the synapse after robust stimulation. In addition to direct membrane interaction, CAPS-1 also binds syntaxin via its MUN domain. Syntaxin also disperses from synapses resulting in CAPS-1 co-dispersion. As Munc13-1 does not show activity-dependent redistribution among synapses but instead increases its membrane-bound fraction upon calcium influx, synapses appear to utilize two distinct mechanisms to control release probabilities during/after high frequency stimulation.
Materials and methodsPlasmids
SemapHluorin was generated by replacing EGFP in Sema3A-EGFP (De Wit et al., 2005) with super-ecliptic pHluorin (SpH) (de Wit et al., 2009). NPY-Venus was previously described (Nagai et al., 2002) and NPY-SpH was generated by replacing Venus with SpH (de Wit et al., 2009). Synapsin-mCherry was a kind gift of Dr A Jeromin (Allen Brain Institute, Seattle, USA) and synapsin-ECFP was obtained by replacing mCherry with ECFP. CAPS-1 (KIAA1121-Kazusa DNA) was sequence verified and cloned as CAPS-1-ires-EGFP and EYFP-CAPS-1. All constructs but SemapHluorin were subcloned into pLenti vectors that were produced as described (Naldini et al., 1996). Transduction efficiencies were tested on HEK cells.
Laboratory animals
CAPS-1/2 double knockout mice have been described before (Jockusch et al., 2007). Mouse embryos were obtained by caesarean section of pregnant females from timed mating. Animals were housed and bred according to institutional, Dutch and US governmental guidelines.
Primary neuron cultures
Dissociated hippocampal neurons were prepared from embryonic day 18 mice as described (de Wit et al., 2009). Hippocampi were dissected in Hanks buffered salts solution (HBSS, Sigma, The Netherlands) and digested with 0.25% trypsin (Invitrogen, The Netherlands) for 20 min at 37°C. Hippocampi were washed and triturated with fire-polished Pasteur pipettes, counted and plated in neurobasal medium (Invitrogen) supplemented with 2% B-27 (Invitrogen), 1.8% HEPES, 1% glutamax (Invitrogen) and 1% Pen-Strep (Invitrogen). High-density cultures (25,000 neurons/well) were seeded on pre-grown cultures of rat glia cells (37,500 cells/well) on 18 mm glass coverslips in 12-well plates. For micro-island culture originally described by (Mennerick et al., 1995), hippocampal neurons were plated at a density of 2000 neurons/well of a 12-well plate on micro-islands of rat glia as in Wierda et al., 2007. These micro-islands were generated by plating 8000/well rat glia on UV-sterilized agarose-coated etched glass coverslips stamped with a 0.1 mg/ml poly-d-lysine (Sigma) and 0.2 mg/ml rat tail collagen (BD Biosciences, The Netherlands) solution.
Infection and transfection
At 10 DIV, neuronal cultures were infected with a combination of lentiviruses encoding NPY-pHluorin, NPY-mCherry, synapsin-mCherry, CAPS-ires-EGFP, CAPS-YFP or synapsin-ECFP. Alternatively, neurons were transfected using calcium phosphate and expression plasmids for SemapHluorin and synapsin-mCherry. Neurons were imaged at DIV 14–DIV 18.
Imaging
Coverslips were placed in an imaging chamber perfused with Tyrode's solution (2 mM CaCl2, 2.5 mM KCl, 119 mM NaCl, 2 mM MgCl2, 20 mM glucose and 25 mM HEPES, pH 7.4). All live imaging experiments were performed on a custom-built tandem illumination microscope (TIM; Olympus, The Netherlands) consisting of an inverted imaging microscope (IX81; Olympus) and an upright laser-scanning microscope. The inverted microscope part was used for imaging fluorescence using an MT20 light source (Olympus), appropriate filter sets (Semrock, Rochester, NY), and a 40× oil objective (NA 1.3), or 60× (NA 1.49) for experiments in Figure 7, on an EM charge-coupled device camera (C9100-02; Hamamatsu Photonics, Japan). Xcellence RT imaging software (Olympus) was used to control the microscope and record the images.
In pHluorin experiments intracellular pH was neutralized with Tyrode's solution containing 50 mM ammonium chloride (NH4Cl), which replaced sodium chloride (NaCl) on an equimolar basis. Ammonium ion (NH4+) solution was delivered by gravity flow through a capillary placed onto the cells. To stimulate the cells electrically, parallel platinum electrodes placed close to the cell soma delivered 30 mA, 1 ms pulses controlled by a Master 8 system (AMPI, Germany) and a stimulus generator (A385RC, World Precision Instruments, Germany). The stimulus used was 16 trains of 50 action potentials at 50 Hz with 0.5 s interval. All imaging experiments were performed at room temperature (RT; 21–24°C). For DCV fusion assays imaging frequency used was 2 Hz. For SemapHluorin experiments in continental cultures fields of view were selected for presence of SemapHluorin-positive somata, which were placed in the center of the field of view. For protein dispersion experiments, CAPS-EYFP and Synapsin-ECFP were imaged at 0.5 Hz simultaneously for 3 min and stimulated with field electrodes (16×50 AP at 50 Hz). In Figure 8, NPY-mCherry and CAPS-EYFP were imaged simultaneously at 2 Hz.
Image analysis
Stacks from time-lapse recordings acquired with 0.5 s intervals were used to analyze DCV release. A 2×2 pixel region (0.4×0.4 μm) was analyzed according to the experiment as follows. Sema and NPY-pHluorin: fluorescent traces were expressed as fluorescence change (∆F) compared to initial fluorescence (F0), obtained by averaging the first four frames of the time-lapse recording. A fusion event was counted when fluorescence showed a sudden increase two standard deviations above F0. Onset of fusion was defined as the first frame with an increase of fluorescence of two standard deviations above F0. A cargo-pHl release event or punctum was scored as synaptic when the fluorescence center of such a release event/punctum was within 200 nm (±1 pixel, the approximate minimal point spread function of our system) of the Synapsin-mCherry fluorescence centroid. Extra-synaptic events were all events that did not meet this criterion. We only measured release events from neurites and excluded somatic release events. Somatic release events cannot be reliably measured using wide-field fluorescence microcopy due to the bright fluorescence from vesicles in/near the Golgi apparatus in which the intraluminal pH is not yet acidic. The total number of vesicles was automatically analyzed from the NH4+ application time lapse using SynD software (Schmitz et al., 2011). When using NPY-mCherry: only the fusion events were scored in which NPY-mCherry fluorescence completely disappeared from a 2×2 pixel punctum after bleaching correction (ImageJ Bleaching correction plug-in). DCVs were categorized as stationary or moving based on the slope of the Kymopgraph (ImageJ, MultipleKymograph), if the slope of the line over the kymograph was different from 0 at any point of the movie, the DCV was considered moving.
CAPS-1-ires-EGFP was used to rescue CAPS DKO neurons in combination with NPY-mCherry for DCV fusion assays in Figure 4.
Protein dispersion was analyzed by placing regions of interest (ROIs) at the synapses and analyzing the ΔF over F0 (average of the first four frames) of Synapsin-ECFP and CAPS-1-EYFP over time. ROIs not overlapping with Synapsin-ECFP were chosen for analyzing CAPS-1-EYFP dispersion at extra-synaptic sites (ΔF as above). These analyses were performed after bleaching correction (ΔF of the soma over time was used as bleaching and subtracted to the measurements). Membrane associated myristoylated EYFP was used as negative control. For co-localization analysis we used ImagJ software (National Institute of Health, USA, Plug-in JACoP). Pearson's coefficients were calculated to obtain cell wide correlation of fluorescent intensities and Mander's coefficients to obtain co-occurrence in VGLUT positive synapses, CAPS-1 puncta or NPY-puncta (Figure 1).
Electrophysiological recordings
Electrophysiological recordings were performed on single isolated glutamatergic hippocampal neurons between 14 and 18 DIV at RT (21–24°C). The patch-pipette was filled with a solution containing 135 mM potassium gluconate, 10 mM HEPES, 1 mM ethylene glycol tetra acetic acid (EGTA), 4.6 mM magnesium chloride (MgCl2), 4 mM sodium-Adenosine 5′-triphosphate (Na-ATP), 15 mM creatine phosphate, 50 U/ml phosphocreatine kinase, and 300 milliosmole (mOsm), pH 7.3. The standard extracellular medium consisted of 140 mM NaCl, 2.4 mM potassium chloride (KCl), 10 mM HEPES, 10 mM glucose, 4 mM calcium chloride (CaCl2), 4 mM MgCl2, and 300 mOsm, pH 7.3. Recordings were performed with an Axopatch 200A amplifier (Molecular Devices, Sunnyvale, CA). Digidata 1322A and Clampex 9.0 (Molecular Devices) were used for signal acquisition. After whole-cell mode, only cells with access resistance of <12 MΩ and leak current of <500 pA were accepted for analysis. Pipette resistance ranged from 4 to 6 MΩ. EPSCs were evoked by depolarizing the cell from −70 to +30 mV for 0.5 ms.
Fixation and immunocytochemistry
Cells were fixed in 4% formaldehyde (Electron Microscopies Sciences, Germany) in phosphate-buffered saline (PBS), pH 7.4, for 20 min at RT and washed in PBS. First cells were permeabilized for 5 min in PBS containing 0.5% Triton X-100 (Sigma–Aldrich) then incubated for 30 min with PBS (Gibco, The Netherlands) containing 2% normal goat serum and 0.1% Triton X-100. Incubations with primary and secondary antibodies were done for 1–2 hr at RT. Primary antibodies used were: polyclonal MAP2 (Abcam, United Kingdom, 1:500), monoclonal VAMP2 (SySy, Germany, 1:2000) and polyclonal Munc13 (SySy, 1:1000), polyclonal chromogranin B (SySy, 1:500), VGLUT1 (SySy, 1:5000), CAPS-1 (SySy,1:200), and polyclonal secretogranin II (kind gift from P Rosa, Institute of Neuroscience, Milan, Italy). Alexa Fluor conjugated secondary antibodies were from Invitrogen. Coverslips were mounted in Mowiol and examined on a Zeiss LSM 510 confocal laser-scanning microscope with a 40× objective (NA 1.3) or 60× (NA 1.4).
Electron microscopy
Neurons were fixed at DIV 14 for 1–2 hr at RT with 0.1 M cacodylate buffer, 0.25 mM CaCl2, 0.5 mM MgCl2 (pH 7.4) and processed as described (Wierda et al., 2007). Cells were washed three times for 5 min with 0.1 M cacodylate buffer (pH 7.4), post-fixed for 2 hr at RT with 1% osmium tetroxide/1% potassium ferro-cyanide, washed and stained with 1% uranyl acetate for 40 min in the dark. Cells were dehydrated with a series of increasing ethanol concentration steps and embedded in Epon and polymerized for 24 hr at 60°C. Cells of interest were selected by observing the flat Epon embedded cell monolayer under the light microscope, and mounted on pre-polymerized Epon blocks for thin sectioning. Ultrathin sections (∼90 nm) were cut parallel to the cell monolayer and collected on single-slot, formvar-coated copper grids, and stained in uranyl acetate and lead citrate. Synapses were selected at low magnification using a JEOL 1010 electron microscope. All analyses were performed on single ultrathin sections of randomly selected synapses. The distribution of DCVs was measured with ImageJ on digital images of synapses taken at 100,000× magnification using analysis software (Soft Imaging System, Gmbh, Germany). The observer was blinded for the genotype. For all morphological analyses we selected only synapses with intact synaptic plasma membranes with a recognizable pre and postsynaptic density. Docked DCVs had a distance of 0 nm from the vesicle membrane to the plasma membrane.
Statistics
Student's t tests for unpaired data were used, throughout the paper, unless otherwise specified. If deviations differed significantly, t tests were Welch corrected. The Mann–Whitney test was used to compare two groups when one or both groups did not pass the normality test. To test more than two groups, Kruskal–Wallis, Bonferroni corrected, analysis of variance was used. Kolmogorov–Smirnov test was used to test whether distributions were normally distributed. Data are plotted as mean with standard error of the mean; n represents number of neurons, N the number of independent experiments.
Acknowledgements
The authors would like to thank Prof. Nils Brose for CAPS-1/2 mice. They also thank R Zalm for cloning and producing viral particles, D Schut, I Saarloos and B Beuger for cell cultures, J Hoetjes and F den Oudsten for genotyping and R Dekker for electron microscopy. They are grateful to J Wortel and C van der Meer for animal breeding. Electron microscopy was performed at the VU/VUmc EM facility. This work is supported by the Netherlands Organization for Scientific Research (ZonMw-VENI 916-66-101, ZonMW-TOP 91208017 to RFT; Pionier/VICI 900-01-001 and ZonMW 903-42-095 to MV). This work is also supported by the EU (EUSynapse project 019055, EUROSPIN project HEALTH-F2-2009-241498, HEALTH-F2-2009-242167 SynSys project and ERC advanced grant 322966 to MV); MF is a recipient of an Erasmus Mundus Joint Doctorate grant (EU 2011-1632/001-001-EMJD) and CMP is supported by ERC Advanced grant 322966 to MV.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
MF, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
RB, Acquisition of data, Analysis and interpretation of data
EH, Performed electrophysiological recordings
CMP, Performed immunocytochemistry experiments
JRTW, Contributed EM data
JHB, Data processing and analysis
MV, Conception and design, Drafting or revising the article
RFT, Conception and design, Analysis and interpretation of data, Drafting or revising the article
Ethics
Animal experimentation: Animals were housed, handled and bred according to institutional, Dutch and U.S. governmental guidelines. All animals were handled according to approved VU University Animal Ethics and Welfare Committee protocols (DEC-FGA-13-03 and DEC-FGA-14-01).
ReferencesAlabiAATsienRW2013Perspectives on kiss-and-run: role in exocytosis, endocytosis, and neurotransmissionAnnual Review of Physiology7539342210.1146/annurev-physiol-020911-153305AugustinIRosenmundCSudhofTCBroseN1999Munc13-1 is essential for fusion competence of glutamatergic synaptic vesiclesNature40045746110.1038/22768BartfaiTIverfeldtKFisoneGSerfözöP1988Regulation of the release of coexisting neurotransmittersAnnual Review of Pharmacology and Toxicology2828531010.1146/annurev.pa.28.040188.001441BekkersJMStevensCF1991Excitatory and inhibitory autaptic currents in isolated hippocampal neurons maintained in cell cultureProceedings of the National Academy of Sciences of USA887834783810.1073/pnas.88.17.7834BerwinBFloorEMartinTF1998CAPS (mammalian UNC-31) protein localizes to membranes involved in dense-core vesicle exocytosisNeuron2113714510.1016/S0896-6273(00)80521-8BrunkIBlexCSpeidelDBroseNAhnert-HilgerG2009Ca2+-dependent activator proteins of secretion promote vesicular monoamine uptakeThe Journal of Biological Chemistry2841050105610.1074/jbc.M805328200CheungGCousinMA2011Quantitative analysis of synaptic vesicle pool replenishment in cultured cerebellar granule neurons using FM dyesJournal of Visualized Experiments10.3791/3143CijsouwTWeberJPBroekeJHBroekJASchutDKroonTSaarloosIVerhageMToonenRF2014Munc18-1 redistributes in nerve terminals in an activity- and PKC-dependent mannerThe Journal of Cell Biology20475977510.1083/jcb.201308026De WitJDe WinterFKloosterJVerhaagenJ2005Semaphorin 3A displays a punctate distribution on the surface of neuronal cells and interacts with proteoglycans in the extracellular matrixMolecular and Cellular Neurosciences29405510.1016/j.mcn.2004.12.009de WitJToonenRFVerhaagenJVerhageM2006Vesicular trafficking of semaphorin 3A is activity-dependent and differs between axons and dendritesTraffic71060107710.1111/j.1600-0854.2006.00442.xde WitJToonenRFVerhageM2009Matrix-dependent local retention of secretory vesicle cargo in cortical neuronsThe Journal of Neuroscience29233710.1523/JNEUROSCI.3931-08.2009GoodwinPRJuoP2013The scaffolding protein SYD-2/Liprin-alpha regulates the mobility and polarized distribution of dense-core vesicles in C. elegans motor neuronsPLOS ONE8e5476310.1371/journal.pone.0054763GrishaninRNKlenchinVALoyetKMKowalchykJAAnnKMartinTF2002Membrane association domains in Ca2+-dependent activator protein for secretion mediate plasma membrane and dense-core vesicle binding required for Ca2+-dependent exocytosisThe Journal of Biological Chemistry277220252203410.1074/jbc.M201614200GrishaninRNKowalchykJAKlenchinVAAnnKEarlesCAChapmanERGeronaRRMartinTF2004CAPS acts at a prefusion step in dense-core vesicle exocytosis as a PIP2 binding proteinNeuron4355156210.1016/j.neuron.2004.07.028HammarlundMWatanabeSSchuskeKJorgensenEM2008CAPS and syntaxin dock dense core vesicles to the plasma membrane in neuronsThe Journal of Cell Biology18048349110.1083/jcb.200708018HartmannMHeumannRLessmannV2001Synaptic secretion of BDNF after high-frequency stimulation of glutamatergic synapsesThe EMBO Journal205887589710.1093/emboj/20.21.5887HirokawaNNodaYTanakaYNiwaS2009Kinesin superfamily motor proteins and intracellular transportNature Reviews Molecular Cell Biology1068269610.1038/nrm2774HuangEJReichardtLF2001Neurotrophins: roles in neuronal development and functionAnnual Review of Neuroscience2467773610.1146/annurev.neuro.24.1.677ImigCMinSWKrinnerSArancilloMRosenmundCSüdhofTCRheeJBroseNCooperBH2014The morphological and molecular nature of synaptic vesicle priming at presynaptic active zonesNeuron8441643110.1016/j.neuron.2014.10.009JamesDJKhodthongCKowalchykJAMartinTF2008Phosphatidylinositol 4,5-bisphosphate regulates SNARE-dependent membrane fusionThe Journal of Cell Biology18235536610.1083/jcb.200801056JamesDJKowalchykJDailyNPetrieMMartinTF2009CAPS drives trans-SNARE complex formation and membrane fusion through syntaxin interactionsProceedings of the National Academy of Sciences of USA106173081731310.1073/pnas.0900755106JockuschWJSpeidelDSiglerASørensenJBVaroqueauxFRheeJSBroseN2007CAPS-1 and CAPS-2 are essential synaptic vesicle priming proteinsCell13179680810.1016/j.cell.2007.11.002KabachinskiGYamagaMKielar-GrevstadDMBruinsmaSMartinTF2014CAPS and Munc13 utilize distinct PIP2-linked mechanisms to promote vesicle exocytosisMolecular Biology of the Cell2550852110.1091/mbc.E12-11-0829KallaSSternMBasuJVaroqueauxFReimKRosenmundCZivNEBroseN2006Molecular dynamics of a presynaptic active zone protein studied in Munc13-1-enhanced yellow fluorescent protein knock-in mutant miceThe Journal of Neuroscience26130541306610.1523/JNEUROSCI.4330-06.2006KimTGondre-LewisMCArnaoutovaILohYP2006Dense-core secretory granule biogenesisPhysiology2112413310.1152/physiol.00043.2005LinXGMingMChenMRNiuWPZhangYDLiuBJiuYMYuJWXuTWuZX2010UNC-31/CAPS docks and primes dense core vesicles in C. elegans neuronsBiochemical and Biophysical Research Communications39752653110.1016/j.bbrc.2010.05.148LiuYSchirraCEdelmannLMattiURheeJHofDBrunsDBroseNRiegerHStevensDRRettigJ2010Two distinct secretory vesicle-priming steps in adrenal chromaffin cellsThe Journal of Cell Biology1901067107710.1083/jcb.201001164LoyetKMKowalchykJAChaudharyAChenJPrestwichGDMartinTF1998Specific binding of phosphatidylinositol 4,5-bisphosphate to calcium-dependent activator protein for secretion (CAPS), a potential phosphoinositide effector protein for regulated exocytosisThe Journal of Biological Chemistry2738337834310.1074/jbc.273.14.8337MatsudaNLuHFukataYNoritakeJGaoHMukherjeeSNemotoTFukataMPooMM2009Differential activity-dependent secretion of brain-derived neurotrophic factor from axon and dendriteThe Journal of Neuroscience29141851419810.1523/JNEUROSCI.1863-09.2009McAllisterAKKatzLCLoDC1996Neurotrophin regulation of cortical dendritic growth requires activityNeuron171057106410.1016/S0896-6273(00)80239-1MennerickSQueJBenzAZorumskiCF1995Passive and synaptic properties of hippocampal neurons grown in microcultures and in mass culturesJournal of Neurophysiology73320332Meyer-LindenbergADomesGKirschPHeinrichsM2011Oxytocin and vasopressin in the human brain: social neuropeptides for translational medicineNature Reviews Neuroscience1252453810.1038/nrn3044NagaiTIbataKParkESKubotaMMikoshibaKMiyawakiA2002A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applicationsNature Biotechnology20879010.1038/nbt0102-87NaldiniLBlömerUGageFHTronoDVermaIM1996Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vectorProceedings of the National Academy of Sciences of USA93113821138810.1073/pnas.93.21.11382Nguyen TruongCQNestvogelDRataiOSchirraCStevensDRBroseNRheeJRettigJ2014Secretory vesicle priming by CAPS is independent of its SNARE-binding MUN domainCell Reports990290910.1016/j.celrep.2014.09.050ParsaudLLiLJungCHParkSSawNMParkSKimMYSugitaS2013Calcium-dependent activator protein for secretion 1 (CAPS1) binds to syntaxin-1 in a distinct mode from Munc13–1The Journal of Biological Chemistry288230502306310.1074/jbc.M113.494088PooMM2001Neurotrophins as synaptic modulatorsNature Reviews Neuroscience2243210.1038/35049004RendenRBerwinBDavisWAnnKChinCTKreberRGanetzkyBMartinTFBroadieK2001Drosophila CAPS is an essential gene that regulates dense-core vesicle release and synaptic vesicle fusionNeuron3142143710.1016/S0896-6273(01)00382-8RibraultCReingruberJPetkovićMGalliTZivNEHolcmanDTrillerA2011Syntaxin1A lateral diffusion reveals transient and local SNARE interactionsThe Journal of Neuroscience31175901760210.1523/JNEUROSCI.4065-11.2011RichmondJEWeimerRMJorgensenEM2001An open form of syntaxin bypasses the requirement for UNC-13 in vesicle primingNature41233834110.1038/35085583SadakataTKakegawaWMizoguchiAWashidaMKatoh-SembaRShutohFOkamotoTNakashimaHKimuraKTanakaMSekineYItoharaSYuzakiMNagaoSFuruichiT2007aImpaired cerebellar development and function in mice lacking CAPS2, a protein involved in neurotrophin releaseThe Journal of Neuroscience272472248210.1523/JNEUROSCI.2279-06.2007SadakataTWashidaMIwayamaYShojiSSatoYOhkuraTKatoh-SembaRNakajimaMSekineYTanakaMNakamuraKIwataYTsuchiyaKJMoriNDetera-WadleighSDIchikawaHItoharaSYoshikawaTFuruichiT2007bAutistic-like phenotypes in Cadps2-knockout mice and aberrant CADPS2 splicing in autistic patientsThe Journal of Clinical Investigation11793194310.1172/JCI29031SadakataTKakegawaWShinodaYHosonoMKatoh-SembaRSekineYSatoYTanakaMIwasatoTItoharaSFuruyamaKKawaguchiYIshizakiYYuzakiMFuruichiT2013CAPS1 deficiency perturbs dense-core vesicle trafficking and Golgi structure and reduces presynaptic release probability in the mouse brainThe Journal of Neuroscience33173261733410.1523/JNEUROSCI.2777-13.2013SadakataTMizoguchiASatoYKatoh-SembaRFukudaMMikoshibaKFuruichiT2004The secretory granule-associated protein CAPS2 regulates neurotrophin release and cell survivalThe Journal of Neuroscience24435210.1523/JNEUROSCI.2528-03.2004SadakataTShinodaYSekineYSarutaCItakuraMTakahashiMFuruichiT2010Interaction of calcium-dependent activator protein for secretion 1 (CAPS1) with the class II ADP-ribosylation factor small GTPases is required for dense-core vesicle trafficking in the trans-Golgi networkThe Journal of Biological Chemistry285387103871910.1074/jbc.M110.137414SahRGeraciotiTD2013Neuropeptide Y and posttraumatic stress disorderMolecular Psychiatry1864665510.1038/mp.2012.101SamsonALMedcalfRL2006Tissue-type plasminogen activator: a multifaceted modulator of neurotransmission and synaptic plasticityNeuron5067367810.1016/j.neuron.2006.04.013SchlagerMAHoogenraadCC2009Basic mechanisms for recognition and transport of synaptic cargosMolecular Brain22510.1186/1756-6606-2-25SchmitzSKHjorthJJJoemaiRMWijntjesREijgenraamSde BruijnPGeorgiouCde JongAPvan OoyenAVerhageMCornelisseLNToonenRFVeldkampWJ2011Automated analysis of neuronal morphology, synapse number and synaptic recruitmentJournal of Neuroscience Methods19518519310.1016/j.jneumeth.2010.12.011ShakiryanovaDTullyALevitanES2006Activity-dependent synaptic capture of transiting peptidergic vesiclesNature Neuroscience989690010.1038/nn1719SpeeseSPetrieMSchuskeKAilionMAnnKIwasakiKJorgensenEMMartinTF2007UNC-31 (CAPS) is required for dense-core vesicle but not synaptic vesicle exocytosis in Caenorhabditis elegansThe Journal of Neuroscience276150616210.1523/jneurosci.1466-07.2007SpeidelDBruederleCEEnkCVoetsTVaroqueauxFReimKBechererUFornaiFRuggieriSHolighausYWeiheEBrunsDBroseNRettigJ2005CAPS1 regulates catecholamine loading of large dense-core vesiclesNeuron46758810.1016/j.neuron.2005.02.019SpeidelDVaroqueauxFEnkCNojiriMGrishaninRNMartinTFHofmannKBroseNReimK2003A family of Ca2+-dependent activator proteins for secretion: comparative analysis of structure, expression, localization, and functionThe Journal of Biological Chemistry278528025280910.1074/jbc.M304727200TsurielSFisherAWittenmayerNDresbachTGarnerCCZivNE2009Exchange and redistribution dynamics of the cytoskeleton of the active zone molecule bassoonThe Journal of Neuroscience2935135810.1523/JNEUROSCI.4777-08.2009van de BospoortRFarinaMSchmitzSKde JongAde WitHVerhageMToonenRF2012Munc13 controls the location and efficiency of dense-core vesicle release in neuronsThe Journal of Cell Biology19988389110.1083/jcb.201208024van den PolAN2012Neuropeptide transmission in brain circuitsNeuron769811510.1016/j.neuron.2012.09.014VaroqueauxFSiglerARheeJSBroseNEnkCReimKRosenmundC2002Total arrest of spontaneous and evoked synaptic transmission but normal synaptogenesis in the absence of Munc13-mediated vesicle primingProceedings of the National Academy of Sciences of USA999037904210.1073/pnas.122623799WalentJHPorterBWMartinTF1992A novel 145 kd brain cytosolic protein reconstitutes Ca(2+)-regulated secretion in permeable neuroendocrine cellsCell7076577510.1016/0092-8674(92)90310-9WierdaKDToonenRFde WitHBrussaardABVerhageM2007Interdependence of PKC-dependent and PKC-independent pathways for presynaptic plasticityNeuron5427529010.1016/j.neuron.2007.04.001WongMYZhouCShakiryanovaDLloydTEDeitcherDLLevitanES2012Neuropeptide delivery to synapses by long-range vesicle circulation and sporadic captureCell1481029103810.1016/j.cell.2011.12.036WuLGHamidEShinWChiangHC2014Exocytosis and endocytosis: modes, functions, and coupling mechanismsAnnual Review of Physiology7630133110.1146/annurev-physiol-021113-170305ZhouKMDongYMGeQZhuDZhouWLinXGLiangTWuZXXuT2007PKA activation bypasses the requirement for UNC-31 in the docking of dense core vesicles from C. elegans neuronsNeuron5665766910.1016/j.neuron.2007.09.01510.7554/eLife.05438.015Decision letterBrungerAxel TReviewing editorStanford University, United States
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “CAPS-1 promotes fusion competence of stationary dense-core vesicles in presynaptic terminals of mammalian neurons” for consideration at eLife. Your article has been generally favorably evaluated by Vivek Malhotra (Senior editor), a Reviewing editor, and 3 reviewers. However, several issues were identified that require major revisions before a final decision can be made.
The Reviewing editor and the reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
In this article, the authors convincingly demonstrate that CAPS positively regulates the exocytosis of stationary large-dense core granules in presynaptic nerve terminals. They used mouse KO, cultured neurons particularly autotaptic, electrophysiology and time-lapsed imaging to reach this conclusion. The authors examine subcellular CAPS-1 localization, DCV localization, DCV loading, DCV docking and fusion in CAPS DKO mutants. The main conclusion of this paper is that CAPS is involved in priming DCVs in neurons, but the manuscript explores CAPS function in detail at many levels and touches on many controversies. There are a large number of conclusions from these data, many at odds with existing literature. The experiments in this manuscript are well designed and executed. Although a role for CAPS in DCV priming is well-established, this manuscript represents the most complete study of CAPS function in DCV secretion in neurons.
Major comments:
1) The authors attempt to distinguish what they term kiss-and-run from full fusion by employing NPY-pHluorin and NPY-mCherry. They show time course for fluorescent decrease for an event with the latter (Figure 3D) but not the former. There are several mistaken notions in these studies. Full release of cargo (shown in Figure 3D) implies nothing about the mode of fusion other than the fusion pore has dilated sufficiently to allow full cargo release. The fusion pore may dilate and reclose and still allow full release of NPY-mCherry (or NPY-pHluorin, not shown). Differences in numbers of events detected with each NPY fusion protein is very likely due to sensitivity of detection, looking for a decrease in red compared to a 100x increase in green fluorescence. This entire section seems to be mistakenly rationalized and inadequately documented.
2) Figure 2B: What is going on in the bottom cartoon? The text says that semaphorin is secreted, but the figure indicates that there is no secretion. It also indicates that reacidification is instantaneous. Also, it is unclear why the cartoon shows a kiss and run type of event given the rapid spike-like nature of the semaphluorin fluorescence. The authors may wish to consider removing all points and discussions about kiss-and-run vs. full fusion if their available data cannot distinguish between these mechanisms.
3) There are several errors in citations. The authors cite Renden et al. 2001, as evidence that there is no docking defect in Drosophila dCAPS mutants. This study did not analyze docking directly, and rather report an increase in clustering of presynaptic DCVs. Furthermore, they cite Liu et al. 2010 (Introduction section) as a Drosophila study; in fact this paper assays fusion in chromaffin cells. As the authors suggest, no defect in DCV docking was observed in chromaffin cells by Liu et al., 2010. The authors reference Hammarlund (2008) for demonstrating that the MUN domains of CAPS and Munc13 bind syntaxin differently. The correct reference is Parsaud et al., 2013. The Hammarlund (2008) paper is in complete disagreement with this manuscript, nevertheless the authors should cite this manuscript for first demonstrating that open syntaxin bypasses the requirement for CAPs (paragraph two of the subsection “A post-docking role for CAPS-1 in DCVs Fusion” in the Discussion). The authors cite Lin et al., 2010, for demonstrating that CAPS is required for docking DCVs at C. elegans NMJs (“CAPS/UNC-31 deletion […] between control and CAPS DKO” and “Hence, with the exception of C. elegans unc-31 null mutants that do show docking defects […] CAPS-1 functions as a post-docking priming protein for DCV release”); in fact this manuscript describes a TIRF assay in cultured neurons, moreover the correct reference for the TIRF assay is Zhou, et al. 2007. The reference for DCV docking at C. elegans neuromuscular junctions in CAPS mutants is Hammarlund, et al., 2008. Finally, the authors should cite Imig et al., 2014, which demonstrates that there is no defect in docking DCVs in CAPS DKOs by electron tomography, which is a more sensitive assay for docking than TEM.
4) Cargo loading (mentioned in the subsection “DCV secretion is severely reduced upon CAPS deletion in hippocampal neurons”): The authors claim that they demonstrate that loading of cargo is not affected in DCVs in hippocampal neurons in conflict from the results of Speidel, et al. 2005). Speidel et al. were looking at catecholamine loading in chromaffin cells; these authors did not claim that there were defects in loading peptide cargo, and the current results are not incompatible with these results. The authors should make this clear. Nevertheless, it is worth stating clearly which results support or contradict existing literature.
5) Abstract: What is “synaptic retention of CAPS”? Maybe “localization of CAPS at synapses” would be better.
6) Figure 2C: Is “field of view” a legitimate denominator, given that the objective could be parked over a region devoid of axons? That should be dealt with in the Methods, “Fields were selected for presence of labeled sema-positive cells”. A more informative Y-axis would note that these were stimulated samples, “fusions/24 seconds stimulation”. The pause interval should be noted in the legend “16 bursts 50AP at 50Hz every 0.5 seconds”.
7) In CAPS DKO release occurs “only after the second burst of stimulation” and “delaying the onset of DCV release upon stimulation.” It appears from the curves to be half the rate for the CAPS DKO at all time points, even within the first burst.
8) Figure 4F: A figure with larger electron micrographs should be included to show context. Additionally, an enlargement showing what the authors score as docked and close but undocked DCVs should be included.
9) In the Discussion the authors discuss the different roles of CAPS and Munc13 for the release of DCVs at extrasynaptic versus synaptic sites. In addition there seems to be a third component mediating DCV fusion, since some release at extrasynaptic sites and secretion of mobile vesicles remains in CAPS mutants. If that is correct, it is worth stating in the Discussion.
10) “DCV secretion was reduced by 70% in CAPS-1/CAPS-2 DKO neurons and remaining fusion events required prolonged stimulation”. This was stated in the Abstract and at several points in the manuscript corresponding to Figures 2D, 2F, 3C, 3F, 3L. Any differences between control and CAPS DKO time courses of cumulative events were not at all evident to this reviewer.
11) “CAPS deletion specifically reduced secretion of stationary DCVs”. This was stated in the Abstract and at several points in the manuscript corresponding to Figure 4B. Whereas fusion events for stationary DCVs were substantially decreased in DKO, the number of events for moving DCVs is very small and it was not clear that a decrease would be observable above background.
12) “… but CAPS-1-EYFP and DCVs rarely traveled together” (in the Abstract). This is only anecdotally documented. In fact, the text noted such an occurrence that contrasts with this statement. Because the authors suggest that some CAPS-1 associates with some DCVs (see below), this becomes an important point but it is not documented.
13) “Synaptic retention of CAPS-1-EYFP in DKO neurons was calcium-dependent…” (in the Abstract). This corresponds to Figure 6, which is the most interesting observation in this study. However, it shows a stimulation-dependent decrease of CAPS-1-EYFP from synapses. So the loss of CAPS-1-EYFP from synapses, not its retention, was calcium-dependent or calcium-stimulated. While the authors comment on this observations from a physiological perspective of adaptation, there is no suggestion about underlying mechanism for CAPS-1 or other proteins in which similar calcium-dependent synaptic loss has been observed.
14) “Deletion of CAPS-1 and CAPS-2 did not affect DCV… docking…”. Docking corresponds to a putative functional state of DCVs. It has been measured by conventional fixation EM, rapid freeze EM, and TIRF microscopy for which there is frequent disagreement. The authors used the first of these techniques and note in the text that their results (lack of effect of CAPS KO on docking) differ from results in C. elegans. Importantly, the C. elegans results utilized rapid freeze EM. Because of known shrinkage and fixation artifacts in conventional EM, the authors should refrain from using the term docking but describe their results as DCVs close to the membrane by EM. It may well be the case when re-examined using other techniques, that there are effects of CAPS deletion on DCV docking as has been shown for Munc13-1.
15) In the single cell studies, the authors use the CAPS2 KO as controls to compare with CAPS DKO cells. While they cite Jockusch, et al., 2007, as the basis for this choice, the Jockusch et al. study did not view DCV exocytosis so the rationale for this is unclear. Labeling the figures as “control” provides confusion and should be labeled CAPS2 KO. The authors should compare cells in their assay from all genotypes to rationalize this choice.
[Editors' note: further revisions were requested prior to acceptance, as described below.]
Thank you for resubmitting your work entitled “CAPS-1 promotes fusion competence of stationary dense-core vesicles in presynaptic terminals of mammalian neurons” for further consideration at eLife. Your revised article has been favorably evaluated by Vivek Malhotra (Senior editor), a Reviewing editor, and three reviewers. The manuscript has been improved but there are some remaining issues that need to be addressed before acceptance, as outlined below:
Comments by reviewer 1:
1) Figure 2: In the rebuttal (point #2) the authors state that they are showing NPY-pHluorin in Figure 2, but they probably mean semaphorin-pHluorin.
2) In Figure 2B, the plasma membrane for the docked vesicle on the left is broken. What happened to the membrane between the TM domains of syntaxin? Is this meant to illustrate a hemi-fusion?
3) The cartoon shows a vesicle opening and the contents rushing from it. In the bottom row, three events are shown. One appears to be a full fusion that does not result in reacidification of the cargo. Two more appear to indicate that the vesicle reacidified without any release of cargo. These seem to have no relationship with the cartoon. There is no explanation of these two kinds of events in the figure legend or in the text (Results section).
4) In the legend it says that this sudden increase in fluorescence is counted as a fusion event in panels C and E. Does that mean only the one with a permanent increase was scored as fusion events? Or were any increases counted as fusion events? Should these be labeled as “complete release” or “incomplete release”, as described for NPY in the rebuttal? The legend and text needs to be less ambiguous, for example: “Both complete and incomplete release of Semaphorin as labeled in Figure 3B were counted as fusion events in panels C and E.”
5) Figure 3. The different assays scored in the top row versus second row in Figure 3 are not clear from looking at the figure. The authors should label the top row 'NPY-pHluorin' and the second row 'NPY-mCherry'.
6) The events in Figure 3B and C are all fusions, regardless of whether they were complete or incomplete release. Can the authors indicate what the complete and incomplete fusions look like for NPY-pHluorin. The authors indicate what NPY-mCherry complete release events look like In Figure 3D, but there is no indication what the events in the top row might look like. This is the exact same problem as in Figure 2. It is just not clear what is being scored and how these events are assembled into the panels in the figure.
7) Figure 3B is labeled 'fusion events', and Figure 3E is labeled 'events'. One assumes from looking at the figure that they are the same events. Using 'fusions / cell' and 'complete release / cell' would be clearer.
8) Figure 3G shows that 50% of the DKO cells show no release. The authors should make it clear that these cells were excluded from the analyses showing the 50% decrease in the rate of fusions in the top rows. (Otherwise the data can be explained simply by the fact that a fraction of cells failed to respond, not that their rates of fusion were affected).
Comments by reviewer 3:
9) There remains some confusion about studies in Figure 3 and their re-interpretation. This issue can be considered minor but some further clarification would be useful. In the original version of the manuscript, increased fluorescence of a pHluorin construct (Figure 3B) versus complete loss of fluorescence of an mCherry construct (Figure 3E) were interpreted to indicate fusion pore opening and full fusion, respectively. The revised manuscript more appropriately describes what these data indicate but it remains unclear whether these data support the interim conclusion that “incomplete and complete cargo release events were affected (by CAPS KO) to the same extent”. It depends on how these experiments were done. If they were done as suggested in the rebuttal letter: “we compared fusion pore opening (step-like increases in the green channel) to full cargo release (complete loss of signal in the red channel) and concluded that fusion pore openings occur more often than full cargo release” using both channels to detect single events, then the authors' conclusion would stand because they would show pHluorin brightening events that were accompanied by complete or incomplete mCherry loss. If this is the case, the figure legend should indicate this and the authors could consider showing traces pHluorin brightening that were accompanied by either complete or incomplete mCherry loss. If, however, the number of pHluorin brightening’s and mCherry complete losses were determined independently, then it could be the case that sensitivity of event detection favors pHluorin brightening events. From what is currently described in the manuscript, it is unclear.
10) In the revised manuscript, there is an expanded discussion of how CAPS functions with an emphasis on its PH domain and PIP2. This section needs editing and greater attention to the references cited (see the Nguyen Truong paper for references). What is PLC16? See Kabachinski et al. for discussion of PLCs in CAPS function.
10.7554/eLife.05438.016Author response
We thank the reviewers and the editor for their insightful comments and constructive suggestions. We have made changes to our manuscript to fully comply with the reviewers’ suggestions and added new data (Figure 1D, I; 2B; 4F; 7 B, C; and Figure 4–figure supplement 1) and better analysis and discussion of the data, as requested. Together, we feel this has considerably strengthened the conclusions of the data.
Major comments:
1) The authors attempt to distinguish what they term kiss-and-run from full fusion by employing NPY-pHluorin and NPY-mCherry. They show time course for fluorescent decrease for an event with the latter (Figure 3D) but not the former. There are several mistaken notions in these studies. Full release of cargo (shown inFigure 3D) implies nothing about the mode of fusion other than the fusion pore has dilated sufficiently to allow full cargo release. The fusion pore may dilate and reclose and still allow full release of NPY-mCherry (or NPY-pHluorin, not shown). Differences in numbers of events detected with each NPY fusion protein is very likely due to sensitivity of detection, looking for a decrease in red compared to a 100x increase in green fluorescence. This entire section seems to be mistakenly rationalized and inadequately documented.
The reviewers argue that our interpretation of the data in terms of vesicle ‘fusion modes’ is not justified and that partial release might not be detected in the red channel due to sensitivity issues. We agree on both these issues. We have now removed interpretation in terms of fusion modes and made the text more factual. It is not the aim of this study to interpret fusion modes and this should not interfere with the main message on CAPS function. Throughout the manuscript we changed the text: completely disappearing NPY-mCherry puncta (Figure 3 D-F) represent “complete cargo release” (avoiding the term ‘full fusion’), and NPY-pHluorin events (Figure 3 B-C) represent fusion pore opening (avoiding the term ‘kiss and run’) with either “complete release” or “incomplete release”, depending on loss in the red channel or not. Regarding the sensitivity issues: we compared fusion pore opening (step-like increases in the green channel) to full cargo release (complete loss of signal in the red channel) and concluded that fusion pore openings occur more often than full cargo release. We changed the text (” DCV fusion pore opening can progress to complete release […] synaptic preference of DCV release.”) to make this clearer.
2)Figure 2B: What is going on in the bottom cartoon? The text says that semaphorin is secreted, but the figure indicates that there is no secretion. It also indicates that reacidification is instantaneous. Also, it is unclear why the cartoon shows a kiss and run type of event given the rapid spike-like nature of the semaphluorin fluorescence. The authors may wish to consider removing all points and discussions about kiss-and-run vs. full fusion if their available data cannot distinguish between these mechanisms.
We changed the cartoon to better represent the data, as suggested by the reviewers. For our analysis of fusion events in Figure 2 we used the sudden increase in NPY-pHluorin fluorescence as signal for fusion pore opening. We added text in figure legend and results to better explain this. In line with comment #1 we removed our comments on vesicle fusion modes.
3) There are several errors in citations. The authors citeRenden et al. 2001, as evidence that there is no docking defect in Drosophila dCAPS mutants. This study did not analyze docking directly, and rather report an increase in clustering of presynaptic DCVs. Furthermore, they citeLiu et al. 2010(Introduction section) as a Drosophila study; in fact this paper assays fusion in chromaffin cells. As the authors suggest, no defect in DCV docking was observed in chromaffin cells byLiu et al., 2010. The authors referenceHammarlund (2008)for demonstrating that the MUN domains of CAPS and Munc13 bind syntaxin differently. The correct reference isParsaud et al. 2013. TheHammarlund (2008)paper is in complete disagreement with this manuscript, nevertheless the authors should cite this manuscript for first demonstrating that open syntaxin bypasses the requirement for CAPs (paragraph two of the subsection “A post-docking role for CAPS-1 in DCVs Fusion” in the Discussion). The authors citeLin et al., 2010, for demonstrating that CAPS is required for docking DCVs at C. elegans NMJs (“CAPS/UNC-31 deletion […] between control and CAPS DKO” and “Hence, with the exception of C. elegans unc-31 null mutants that do show docking defects […] CAPS-1 functions as a post-docking priming protein for DCV release”); in fact this manuscript describes a TIRF assay in cultured neurons, moreover the correct reference for the TIRF assay isZhou, et al. 2007. The reference for DCV docking at C. elegans neuromuscular junctions in CAPS mutants isHammarlund, et al., 2008. Finally, the authors should citeImig et al., 2014, which demonstrates that there is no defect in docking DCVs in CAPS DKOs by electron tomography, which is a more sensitive assay for docking than TEM.
The reviewers identified several errors in citations. We fully agree and apologize for these mistakes. We corrected them and added Imig et al., 2014, to the discussion on DCV docking.
4) Cargo loading (mentioned in the subsection “DCV secretion is severely reduced upon CAPS deletion in hippocampal neurons”): The authors claim that they demonstrate that loading of cargo is not affected in DCVs in hippocampal neurons in conflict from the results ofSpeidel, 2005). Speidel et al. were looking at catecholamine loading in chromaffin cells; these authors did not claim that there were defects in loading peptide cargo, and the current results are not incompatible with these results. The authors should make this clear. Nevertheless, it is worth stating clearly which results support or contradict existing literature.
The reviewers identified an ambiguous statement about cargo loading. The reviewers are correct. We changed the sentence to better describe that protein loading is not affected in hippocampal neurons whereas catecholamine loading in CAPS-1 KO chromaffin cells is.
5) Abstract: What is “synaptic retention of CAPS”? Maybe “localization of CAPS at synapses” would be better.
We changed “synaptic retention of CAPS” to “synaptic localization of CAPS ” as suggested.
6)Figure 2C: Is “field of view” a legitimate denominator, given that the objective could be parked over a region devoid of axons? That should be dealt with in the Methods, “Fields were selected for presence of labeled sema-positive cells”. A more informative Y-axis would note that these were stimulated samples, “fusions/24 seconds stimulation”. The pause interval should be noted in the legend “16 bursts 50AP at 50Hz every 0.5 seconds”.
The reviewers question whether “field of view” is a legitimate denominator (Figure 2 C). We added a better description in the Methods section and changed y-axis labels in Figure 2 to “fusion events/per field of view” and added “16 bursts 50AP at 50Hz every 0.5 seconds” to the legend and Methods section, as suggested.
7) In CAPS DKO release occurs “only after the second burst of stimulation” and “delaying the onset of DCV release upon stimulation.” It appears from the curves to be half the rate for the CAPS DKO at all time points, even within the first burst.
The reviewers question our conclusion that release onset is delayed in CAPS DKO. We re-analysed the time to fusion in CAPS DKO vs controls and did not find statistically significant differences and removed this conclusion in all parts of the manuscript.
8)Figure 4F: A figure with larger electron micrographs should be included to show context. Additionally, an enlargement showing what the authors score as docked and close but undocked DCVs should be included.
We have included new figures with larger electron micrographs, as requested by the reviewers (Figure 4F) and an enlargement showing how docked and close but undocked DCVs were scored (Figure 4–figure supplement 1).
9) In the Discussion the authors discuss the different roles of CAPS and Munc13 for the release of DCVs at extrasynaptic versus synaptic sites. In addition there seems to be a third component mediating DCV fusion, since some release at extrasynaptic sites and secretion of mobile vesicles remains in CAPS mutants. If that is correct, it is worth stating in the Discussion.
The reviewers point out that an additional third component may mediate extrasynaptic release as both in Munc13 and CAPS DKOs some fusion remains. We agree and discuss this possibility in the Discussion on the subsection headed “A post-docking role for CAPS-1 in DCVs fusion”.
10) “DCV secretion was reduced by 70% in CAPS-1/CAPS-2 double null mutant (DKO) neurons and remaining fusion events required prolonged stimulation”. This was stated in the Abstract and at several points in the manuscript corresponding toFigures 2D, 2F, 3C, 3F, 3L. Any differences between control and CAPS DKO time courses of cumulative events were not at all evident to this reviewer.
This point is in fact the same issue as addressed at point 7. We agree and removed these statements throughout the manuscript.
11) “CAPS deletion specifically reduced secretion of stationary DCVs”. This was stated in the Abstract and at several points in the manuscript corresponding toFigure 4B. Whereas fusion events for stationary DCVs were substantially decreased in DKO, the number of events for moving DCVs is very small and it was not clear that a decrease would be observable above background.
The reviewers argue that fusion events of mobile vesicles are rare and effects on CAPS loss might therefore remain unnoticed. We changed the text to “CAPS deletion strongly reduced secretion of stationary DCVs” and removed the term “specifically” throughout the manuscript.
12) “… but CAPS-1-EYFP and DCVs rarely traveled together” (in the Abstract). This is only anecdotally documented. In fact, the text noted such an occurrence that contrasts with this statement. Because the authors suggest that some CAPS-1 associates with some DCVs (see below), this becomes an important point but it is not documented.
The reviewers state that documentation of CAPS-1-EYFP and DCV co-trafficking is too anecdotal and deserves better documentation. We agree and added a new, quantitative analysis of DCV and CAPS-1-EYFP dynamics to Figure 7 (new Figure 7 B and C) to support our conclusion that co-trafficking of CAPS-1 with mobile DCVs is very rare.
13) “Synaptic retention of CAPS-1-EYFP in DKO neurons was calcium-dependent…” (in the Abstract). This corresponds toFigure 6, which is the most interesting observation in this study. However, it shows a stimulation-dependent decrease of CAPS-1-EYFP from synapses. So the loss of CAPS-1-EYFP from synapses, not its retention, was calcium-dependent or calcium-stimulated. While the authors comment on this observations from a physiological perspective of adaptation, there is no suggestion about underlying mechanism for CAPS-1 or other proteins in which similar calcium-dependent synaptic loss has been observed.
The reviewers argue that our data in Figure 6 argue for Ca2+-dependence of synaptic dispersion, not retention and notes that there are no suggestions about underlying mechanisms. We agree and changed the text to “Synaptic dispersion of CAPS-1-EYFP in DKO neurons was calcium-dependent”. We have recently shown that synaptic dispersion and reclustering of Munc18-1 allows neurons to modulate synaptic strength upon network activity (Cijsouw et al., 2014). Reclustering of Munc18-1 required PKC activation and was syntaxin 1 independent. CAPS1 membrane association requires its PIP2-binding pleckstrin homology (PH) domain. Robust Ca2+ influx in our experiments likely activates phospholipase C (PLC) which hydrolysis PIP2 and may trigger CAPS1 dispersion from synapses. Recent work on CAPS2 shows that the PH-domain is required to prime vesicles (Nguyen Truong et al, 2014). Calcium-dependent PIP2 hydrolysis may therefore act as negative feedback mechanism reducing CAPS1 availability at the synapse upon robust activity. In addition to direct membrane interaction, CAPS1 also binds syntaxin via its MUN domain. Syntaxin also disperses from synapses, which may result in CAPS1 co-dispersion. We added this to the Discussion ” This is also an attractive explanation[…]after high frequency stimulation”.
14) “Deletion of CAPS-1 and CAPS-2 did not affect DCV… docking…”. Docking corresponds to a putative functional state of DCVs. It has been measured by conventional fixation EM, rapid freeze EM, and TIRF microscopy for which there is frequent disagreement. The authors used the first of these techniques and note in the text that their results (lack of effect of CAPS KO on docking) differ from results in C. elegans. Importantly, the C. elegans results utilized rapid freeze EM. Because of known shrinkage and fixation artifacts in conventional EM, the authors should refrain from using the term docking but describe their results as DCVs close to the membrane by EM. It may well be the case when re-examined using other techniques, that there are effects of CAPS deletion on DCV docking as has been shown for Munc13-1.
The reviewers argue that chemical fixation used in our study might produce artifacts that change the precise distance between vesicles and the membrane. As a consequence, vesicles touching the membrane in fixed tissue may in fact have been at a short distance from the membrane. This is true and we have considered it common knowledge, but we have now added a statement on this in the text. We feel that this fact does not preclude the use of the term ‘docking’ and that the suggested alternative ‘DCVs close to the membrane’ is not better, because this might give the impression that DCVs are positioned accidentally near a membrane. The term ‘docking’, like many words in shipping, has a Dutch origin, in this case the word ‘dok’, which describes a situation in which a vessel can be loaded or unloaded. Whether or not the vessel touches the quay is not defined and typically there is some space between vessel and quay. The essential thing is the functional state (to permit (un)loading), very similar to the reviewer’s definition for vesicles in synapses (‘docking corresponds to a putative functional state’).
Although earlier reports using chemical fixation of mouse synapses did not find differences in DCV localization (Jockusch et al., 2007). The reviewers are right that future studies using cryo-fixation might unmask (subtle) docking defects upon CAPS loss. In fact, Imig et al., using cryo-fixation, found a reduction in the number of docked SVs and a lower, albeit not statistically significant number of DCVs within 200 nm of the active zone between CAPS DKO and controls. We added text to the Discussion to point this out.
15) In the single cell studies, the authors use the CAPS2 KO as controls to compare with CAPS DKO cells. While they citeJockusch, 2007, as the basis for this choice, the Jockusch et al. study did not view DCV exocytosis so the rationale for this is unclear. Labeling the figures as “control” provides confusion and should be labeled CAPS2 KO. The authors should compare cells in their assay from all genotypes to rationalize this choice.
We agree with the reviewers that although synaptic transmission is not affected in CAPS 2 KO neurons, DCV release might. We can indeed not exclude this and we therefore might underestimate the effect of CAPS DKO in Figure 3. However, we used wild-type controls in Figure 2 and CAPS DKO leads to a 70% reduction of release events compared to either control. We renamed “control” to “CAPS 2KO” to avoid confusion.
[Editors' note: further revisions were requested prior to acceptance, as described below.]
We have changed Figures 2 and 3 as suggested by reviewer 1 and added Figure 2–figure supplement 2 to better explain the different fusion modes reported by Semaphluorin. Please find below the point-by-point answers to the reviewers’ comments.
Comments by reviewer 1:
1)Figure 2: In the rebuttal (point #2) the authors state that they are showing NPY-pHluorin inFigure 2, but they probably mean semaphorin-pHluorin.
The reviewer is correct and we are sorry for this mistake, Figure 2 indeed shows semaphorin-pHluorin.
2) InFigure 2B, the plasma membrane for the docked vesicle on the left is broken. What happened to the membrane between the TM domains of syntaxin? Is this meant to illustrate a hemi-fusion?
No, this appears to be caused by conversion between Illustrator versions, we changed this so that the membrane is continuous in Figure 2B. We thank the reviewer for pointing this out.
3) The cartoon shows a vesicle opening and the contents rushing from it. In the bottom row, three events are shown. One appears to be a full fusion that does not result in reacidification of the cargo. Two more appear to indicate that the vesicle reacidified without any release of cargo. These seem to have no relationship with the cartoon. There is no explanation of these two kinds of events in the figure legend or in the text (Results section).
The reviewer is correct; the cartoon does not represent the different fusion modes reported by Semaphluorin. The cartoon and fluorescence traces in Figure 2B illustrate the definition of fusion events counted in Figure 2C-F: a sudden increase in fluorescence two standard deviations above initial fluorescence (F0). Hence, all 3 events were counted as fusion events. This was not clear in the original figure and we changed the figure adding a grey dotted line representing the 2xSD, and added text to the figure legend and main text (Results section) to make this clearer. We realize that the different traces may raise confusion. Therefore, we added Figure 2–figure supplement 2 to better explain the different fusion modes reported by SemapHluorin.
4) In the legend it says that this sudden increase in fluorescence is counted as a fusion event in panels C and E. Does that mean only the one with a permanent increase was scored as fusion events? Or were any increases counted as fusion events? Should these be labeled as “complete release” or “incomplete release”, as described for NPY in the rebuttal? The legend and text needs to be less ambiguous, for example: “Both complete and incomplete release of Semaphorin as labeled inFigure 3Bwere counted as fusion events in panels C and E.”
We are sorry for the confusion; all 3 events are fusion events. In Figure 2C-F, we counted all events with an increase in fluorescence two standard deviations above F0. The onset of fusion was defined as the first frame with an increase two times SD above F0 (added to the Materials and methods section). In line with comment #3 we changed Figure legend and text (Results section) to make this clearer.
5)Figure 3. The different assays scored in the top row versus second row inFigure 3are not clear from looking at the figure. The authors should label the top row 'NPY-pHluorin' and the second row 'NPY-mCherry'.
We changed the figure as suggested.
6) The events inFigure 3B and Care all fusions, regardless of whether they were complete or incomplete release. Can the authors indicate what the complete and incomplete fusions look like for NPY-pHluorin. The authors indicate what NPY-mCherry complete release events look like InFigure 3D, but there is no indication what the events in the top row might look like. This is the exact same problem as inFigure 2. It is just not clear what is being scored and how these events are assembled into the panels in the figure.
Similar to our analysis of fusion events in Figure 2, in Figure 3B,C all sudden increases of fluorescence two standard deviations above F0 were counted as fusion events. We added this to figure legend and text. We also added a typical NPY-pHluorin trace to Figure 3B, which reports a fusion event counted in 3B. After the increase in fluorescence, the decline in fluorescence may represent either NPY-Phluorin cargo diffusion (‘complete release’) or vesicle resealing and re-acidification (‘incomplete release’). In Figure 3B we did not discriminate between the two and scored all events > 2xSD above F0.
7)Figure 3Bis labeled 'fusion events', andFigure 3Eis labeled 'events'. One assumes from looking at the figure that they are the same events. Using 'fusions / cell' and 'complete release / cell' would be clearer.
We agree and changed the figure as suggested.
8)Figure 3Gshows that 50% of the DKO cells show no release. The authors should make it clear that these cells were excluded from the analyses showing the 50% decrease in the rate of fusions in the top rows. (Otherwise the data can be explained simply by the fact that a fraction of cells failed to respond, not that their rates of fusion were affected).
We fully agree and added this statement to the figure legend and main text. We would like to thank reviewer 1 for comments and suggestions. They significantly improved the legibility of our manuscript.
Comments by reviewer 3:
9) There remains some confusion about studies inFigure 3and their re-interpretation. This issue can be considered minor but some further clarification would be useful. In the original version of the manuscript, increased fluorescence of a pHluorin construct (Figure 3B) versus complete loss of fluorescence of an mCherry construct (Figure 3E) were interpreted to indicate fusion pore opening and full fusion, respectively. The revised manuscript more appropriately describes what these data indicate but it remains unclear whether these data support the interim conclusion that “incomplete and complete cargo release events were affected (by CAPS KO) to the same extent”. It depends on how these experiments were done. If they were done as suggested in the rebuttal letter: “we compared fusion pore opening (step-like increases in the green channel) to full cargo release (complete loss of signal in the red channel) and concluded that fusion pore openings occur more often than full cargo release” using both channels to detect single events, then the authors' conclusion would stand because they would show pHluorin brightening events that were accompanied by complete or incomplete mCherry loss. If this is the case, the figure legend should indicate this and the authors could consider showing traces pHluorin brightening that were accompanied by either complete or incomplete mCherry loss. If, however, the number of pHluorin brightening’s and mCherry complete losses were determined independently, then it could be the case that sensitivity of event detection favors pHluorin brightening events. From what is currently described in the manuscript, it is unclear.
In Figure 3 the number of pHluorin brightening’s and mCherry complete losses were determined independently and the reviewer is correct that we therefore cannot formally exclude that sensitivity of event detection favors pHluorin brightening events. Which would explain why we see more pHluorin events than mCherry disappearances. To exclude any misinterpretation, we changed the concluding sentence of this paragraph to “Thus, deletion of CAPS-1 resulted in a strong reduction of DCV fusion events reported by NPY-Phluorin or NPY-mCherry”.
10) In the revised manuscript, there is an expanded discussion of how CAPS functions with an emphasis on its PH domain and PIP2. This section needs editing and greater attention to the references cited (see the Nguyen Truong paper for references). What is PLC16? See Kabachinski et al. for discussion of PLCs in CAPS function.
We thank the reviewer for pointing this out. We apologize and corrected citations as suggested. PLC16 is a typo and should state PLC.
Sedimentary rocks host a vast reservoir of organic carbon, such as 2-methylhopane biomarkers, whose evolutionary significance we poorly understand. Our ability to interpret this molecular fossil record is constrained by ignorance of the function of their molecular antecedents. To gain insight into the meaning of 2-methylhopanes, we quantified the dominant (des)methylated hopanoid species in the membranes of the model hopanoid-producing bacterium Rhodopseudomonas palustris TIE-1. Fluorescence polarization studies of small unilamellar vesicles revealed that hopanoid 2-methylation specifically renders native bacterial membranes more rigid at concentrations that are relevant in vivo. That hopanoids differentially modify native membrane rigidity as a function of their methylation state indicates that methylation itself promotes fitness under stress. Moreover, knowing the in vivo (2Me)-hopanoid concentration range in different cell membranes, and appreciating that (2Me)-hopanoids' biophysical effects are tuned by the lipid environment, permits the design of more relevant in vitro experiments to study their physiological functions.
DOI:http://dx.doi.org/10.7554/eLife.05663.001
10.7554/eLife.05663.002eLife digest
The cell membrane that separates the inside of a cell from its outside environment is not a fixed structure. A cell can change the amount and type of different molecules in its membrane, which can alter the rigidity and permeability of the membrane and allow the cell to adapt to changing conditions.
The cell membranes of many bacteria contain molecules called hopanoids. Hopanes are the fossilized forms of these molecules and many hopanes are found extensively in sedimentary rocks. For example, 2-methylated hopanes—the fossilized forms of hopanoids that have a methyl group added to a particular carbon atom—have been found in ancient rocks that formed up to 1.6 billion years ago.
Many researchers have suggested that 2-methylated hopanes (and other molecular fossils) in sedimentary rocks could act as ‘biomarkers’ and be used to deduce what primitive life and ancient living conditions were like. Millions of years ago, several periods occurred where the Earth's oceans lost almost all of their oxygen; this likely placed all life on Earth under great stress. A greater proportion of the hopanes found in rocks formed during those periods are methylated than those seen in rocks from other time periods. However, it was difficult to interpret this observation about the fossil record, as the role of 2-methylated hopanoids in living bacterial cells was unknown.
Wu et al. have now investigated the role of 2-methylated hopanoids by performing experiments on bacterial membranes and found that 2-methylated hopanoids help the other molecules that make up the membrane to pack more tightly together. This makes the membrane more rigid, and the extent of this stiffening depends on the length of the 2-methylated hopanoid and on the other molecules that are present in the membrane. A more rigid membrane would protect the bacteria more in times of stress; therefore, rock layers containing an increased amount of 2-methylhopane are likely to indicate times when the bacteria living at that time were under a great deal of stress.
DOI:http://dx.doi.org/10.7554/eLife.05663.002
Author keywordsRhodospeudomonas palustris TIE-12-methyl hopanoidsphysiological functionbiophysical propertiesevolutionary interpretationResearch organismotherhttp://dx.doi.org/10.13039/100000104National Aeronautics and Space Administration (NASA)NNX12AD93GWuChia-HungBialecka-FornalMajahttp://dx.doi.org/10.13039/100000001National Science Foundation (NSF)1224158WuChia-HungBialecka-FornalMajahttp://dx.doi.org/10.13039/100000011Howard Hughes Medical Institute (HHMI)WuChia-HungNewmanDianne KThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementMethylation specifically enhances the ability of hopanoids to rigidify membranes under physiologically relevant conditions, which impacts the current interpretation of the 2-methylhopane fossil record.Introduction
Lipids play essential roles in compartmentalizing cells for specific functions and creating barriers that are selectively permeable to the environment. The composition of lipids in cell membranes varies significantly and the basic biophysical properties of membranes, such as rigidity and permeability, can be adjusted based on growth conditions and environmental stressors (Lipowsky and Sackmann, 1995; Los and Murata, 2004; Neubauer et al., 2014). One well-studied example is cholesterol in eukaryotic membranes. This essential sterol plays diverse roles in maintaining membrane structural integrity, modifying membrane rigidity, serving as a biosynthetic precursor for steroid hormones, vitamin D, and bile acids, or acting as a protein modifier for signaling pathways (Hanukoglu, 1992; Gallet, 2011; Song et al., 2014). In addition to their important biological functions, lipids are of interest because they are more geostable than other biomolecules. For example, hopanoid molecular fossils, ‘hopanes’, date back over a billion years (Brocks et al., 2005) and are so abundant that the global stock of hopanoids that can be extracted from sedimentary rocks is estimated to be 1013 or 1014 tons, more than the estimated 1012 tons of organic carbon in all living organisms (Ourisson et al., 1984). In contrast to steroids, hopanoids are a less well studied but evolutionarily significant and chemically diverse class of lipids that are thought to be sterol surrogates in bacteria (Figure 1) (Rohmer et al., 1979; Ourisson et al., 1987).10.7554/eLife.05663.003
Structures of selected hopanoids, cholesterol, and squalene.
DOI:http://dx.doi.org/10.7554/eLife.05663.003
The rich record of ancient lipids, including fossil hopanoids, has long been recognized to hold clues into the early history of life and past environments (Ourisson et al., 1984; Summons and Walter, 1990; Brocks and Pearson, 2005; Knoll et al., 2007). But being able to confidently interpret the meaning of any ancient molecular fossil poses considerable challenges. First, we must be able to identify potential sources for these compounds, demanding unambiguous chemical parity between modern and ancient structures. Once this is achieved, understanding whether particular environmental conditions regulate the production of specific hopanoid variants becomes important. But arguably, the most critical goal in advancing our understanding of ancient lipids is being able to identify specific biological functions for their counterparts in cells today. Myriad hopanoid structures are known to exist (Ourisson et al., 1984), yet we only poorly understand the significance of this chemical diversity. For meaningful linkages to be made between modern compounds and ancient biomarkers, we must (1) study those hopanoids that leave a specific trace in sedimentary rocks (e.g., their chemical modifications are geostable), (2) identify their in vivo function(s), and (3) evaluate whether the roles played by these lipids in modern organisms have been conserved over the course of evolution.
Following the recognition in the early 1970s that hopanes are ubiquitous in sedimentary rocks, the occurrence of hopanoids in diverse organisms was documented, and insights were gained into their biosynthesis, biophysical properties, and cellular functions (Ourisson et al., 1987; Ourisson and Rohmer, 1992; Pearson, 2013). For example, studies using hopanoid-deficient mutants have shown that hopanoids promote resistance to antibiotics, detergents, extreme pH, and high osmolarity (Welander et al., 2009; Sáenz, 2010; Schmerk et al., 2011; Malott et al., 2012; Kulkarni et al., 2013). Biophysical studies using mixtures of hopanoids and model lipids have demonstrated that, like cholesterol, bacteriohopanetetrol cyclitol ether can condense membranes at high temperatures but fluidize membranes at low temperatures (Poralla et al., 1980). Similarly, bacteriohopanetetrol (BHT) and bacteriohopanemonol can condense model membranes, and diplopterol (Dip) can form liquid ordered microdomains (Kannenberg et al., 1983; Nagumo et al., 1991; Ourisson and Rohmer, 1992; Sáenz et al., 2012). A recent molecular modeling study pointed out different behaviors between Dip and BHT in their specific location within lipid bilayers and their capacity to condense membranes, suggesting complex roles of hopanoids due to their structural diversity (Poger and Mark, 2013). These biophysical studies have provided insights into the physical capabilities of these hopanoids. However, whether hopanoids play the same roles in vivo has been unclear due to the differences in lipid composition and concentration between model and cellular membranes.
Among various hopanoid modifications, methylation of C-2 on the A-ring has drawn attention from Earth scientists because this modification is preserved episodically in ancient sedimentary rocks dating back to 1.6 billion years ago; accordingly, it has been suggested that 2-methylated hopanes (2Me-hopanes), the molecular fossils of 2Me-hopanoids, could potentially serve as biomarkers to interpret events in the early history of life (Brocks et al., 2005; Rasmussen et al., 2008). For a time, it was thought that 2Me-hopanes were biomarkers of cyanobacteria, and hence the process of oxygenic photosynthesis (Summons et al., 1999), but we now know this not to be the case (Welander et al., 2010; Ricci et al., 2015). Intriguingly, spikes in the C30 2Me-hopane index (ratio of methylated short hopanes to total short hopanes) through geologic time are correlated with episodes of oceanic anoxic events (OAEs), which are thought to have imposed heightened stress on the biosphere (Knoll et al., 2007).
Towards the goal of finding a robust interpretation for 2Me-hopanes, we have elucidated the biosynthetic pathway of 2Me-hopanoids (Welander et al., 2010, 2012), linked 2Me-hopanoids to specific environments and producers through (meta)genomic studies (Ricci et al., 2014), and identified the stress-responsive pathway regulating transcription of the 2-methylase (hpnP) in the model hopanoid-producing bacterium Rhodospeudomonas palustris TIE-1 (Kulkarni et al., 2013). A major challenge in understanding the function of 2Me-hopanoids has been the lack of a clear phenotype in vivo, despite dedicated attempts to find one for the hpnP mutant (Kulkarni et al., 2013). Recent data suggest this phenotypic silence results from changes to the lipidome that compensate for the loss of 2Me-hopanoids (unpublished). To infer a specific in vivo function for 2Me-hopanoids, in vitro studies that mimic cellular composition are therefore necessary.
We hypothesized that methylation at C-2 would change hopanoids' packing with other lipids and proteins and affect membrane biophysical properties such as rigidity. Furthermore, we reasoned that such an effect would depend on the specific lipid composition of the membrane. To test these hypotheses, we took advantage of the existence of specific hopanoid-mutant strains (Welander et al., 2010, 2012), and recently developed protocols allowing quantitative analysis and the purification of hopanoids and their 2-methylated species in large quantities (Wu et al., 2015). This experimental foundation set the stage for what we now report: in vitro membrane studies to examine how 2Me-hopanoids affect membrane rigidity in the context of different lipid environments of relevance to the cell. Our finding that hopanoid methylation enhances membrane rigidity supports the interpretation that past intervals of heightened 2Me-hopane abundance record a history of stress.
ResultsWhole cell membrane fluidity
To test whether 2-methylation changes membrane rigidity, we measured the membrane rigidity of specific hopanoid biosynthetic mutants (Table 1) (Welander et al., 2012) using fluorescence polarization. Figure 2 shows the whole cell membrane rigidity measured at 25°C and 40°C. As expected, at higher temperature, the cell membrane became less rigid across all strains. At 25°C, the Δshc mutant, which lacks all hopanoids, had the least rigid membrane. This could be caused by both the absence of hopanoids and the accumulation of the hopanoid biosynthetic precursor, squalene. The result is also consistent with the observation that in model lipid vesicles, hopanoids make the membrane more rigid (Kannenberg et al., 1985). Interestingly, the production of only short-chain hopanoids (ΔhpnH) is sufficient to recover the rigidity level close to that of the WT. However, when an adenosine molecule is attached to short hopanoids and accumulates in the ΔhpnG mutant, rigidity decreases to Δshc levels. Furthermore, in ΔhpnN where hopanoids are unable to be transported to the outer membrane (Doughty et al., 2011), membrane rigidity is similar to ΔhpnG. These results have two implications. First, the hydrophobic reporter dye we used for fluorescence polarization measurements, diphenyl hexatriene (DPH), reflects mostly the rigidity of the outer membrane. Second, the types of hopanoids and their respective localization between the inner and outer membranes directly impact membrane rigidity.10.7554/eLife.05663.004
Mutant strains used for the whole cell membrane rigidity measurements
DOI:http://dx.doi.org/10.7554/eLife.05663.004
Gene
Function
Deletion effect
shc (hpnF)
Cyclization of squalene to form C30 hopanoids (diploptene and diplopterol)
No hopanoid production and accumulation of squalene
hpnH
Addition of adenosine to diploptene to generate adenosylhopane, a precursor for extended hopanoid production
No extended hopanoid production, accumulation of C30 hopanoids
hpnG
Removal of adenine from adenosyl hopane
No BHT and aminoBHT production, accumulation of adenosylhopane
hpnN
An IM transporter that transports hopanoids to the outer membrane
Absence of hopanoids in the OM and accumulation of hopanoids in the IM
hpnO
Production of aminoBHT
No aminoBHT production
hpnP
Methyl transfer to A ring at C-2
No hopanoid methylation
The function of the gene and the effect of its deletion are listed.
10.7554/eLife.05663.005
Whole cell membrane fluidity.
Error bars represent the standard deviation from three biological replicates (total 21–26 technical replicates).
DOI:http://dx.doi.org/10.7554/eLife.05663.005
Interestingly, no obvious impact on rigidity was found in the absence of either bacteriohopane aminotriol (ΔhpnO) or 2-methylhopanoid (ΔhpnP) or both (ΔhpnOP) compared to WT (Figure 2). This observation could have two possible explanations. One is that these specific hopanoids do not affect membrane rigidity in cells. The other explanation is that the cells might synthesize other lipids that compensate for their effects so that no phenotype is observed. We can distinguish between these two scenarios by measuring membrane rigidity using vesicles made of model lipids and purified hopanoids. However, to design experiments that are physiologically relevant, we first needed to quantify hopanoids distribution in both the inner (IM) and outer membrane (OM) of R. palustris TIE-1.
Quantification of hopanoids in the inner and outer membranes of <italic>R. palustris</italic> TIE-1
It is well appreciated that lipids can have different subcellular localization, even in bacteria (Matsumoto et al., 2006). To understand the biological roles of hopanoids, especially for 2-methylhopanoids, we measured the amounts of different hopanoids in the IM and OM of R. palustris TIE-1 WT and ΔhpnP using a previously described protocol (Figure 3) (Morein et al., 1994; Wu et al., 2015). Table 2 shows that the total yield of the fractionated membranes was about 10% weight of dried cells, which is comparable to 9.1% in Escherichia coli (Neidhardt and Curtiss, 1996). The yields of the total lipid extract (TLE) from the lyophilized fractionated membranes were ∼12–15% from the IM and ∼5% for the OM. This yield is reproducible and could be due to a larger proportion of membrane proteins in R. palustris TIE-1, lower recovery after lyophilization, or loss of certain lipid classes that were not quantitatively extracted by procedures optimized for hopanoids. The TLE yield in the OM was at least 50% lower than that in the IM, which is expected because the outer leaflet of the OM consists of more hydrophilic lipid A and lipopolysaccharides that are not extractable by the hydrophobic Bligh-Dyer lipid extraction method we employed.10.7554/eLife.05663.006
Membrane fractionation and hopanoids analysis using GC–MS.
(A) Three distinct bands were formed after ultracentrifugation in a Percoll gradient. (B) The bands were recovered and resuspended. (C) Samples were ultracentrifuged to pellet down the purified membranes, which sat on top of a transparent solid Percoll layer. (D) GC–MS of fractionated membranes of R. palustris TIE-1 WT and ΔhpnP.
DOI:http://dx.doi.org/10.7554/eLife.05663.006
10.7554/eLife.05663.007
Purification yields of membrane fractionation using Percoll gradient
DOI:http://dx.doi.org/10.7554/eLife.05663.007
Weight % membranes in dry cells
Total
Weight % TLE in membranes
WT
inner
4.7 ± 0.5
12.4 ± 0.8
mix
3.7 ± 1.2
5.6 ± 0.1
outer
3.3 ± 0.4
11.6 ± 1.4
4.9 ± 0.6
ΔhpnP
inner
4.2 ± 1.3
15 ± 2.3
mix
2.3 ± 0.8
8.3 ± 2.3
outer
2.9 ± 0.6
9.4 ± 2.2
5.3 ± 1.4
The yields in wt% of membrane fractionation. Errors represent standard deviation from three biological replicates.
To quantify hopanoids, TLE from IM and OM was analyzed by GC-MS using androsterone as an internal standard and the differences in ionization efficiencies between androsterone and hopanoids were calibrated by external standards using purified hopanoids (Figure 3). Such calibration was recently shown to be essential for accurate hopanoid quantification (Wu et al., 2015). Using this approach, the exact wt% of each hopanoid in TLE was obtained. However, to put the numbers in context and compare the value in mol%, we assumed the average molecular weight of the total lipids is 786 g/mol, the same as dioleoyl phosphatidylcholine (DOPC) and E. coli polar lipid extract (PLE). Because we could only confidently quantify (2Me)-Dip and (2Me)-BHT, we focused our analyses on these four hopanoids. Figure 4 shows the hopanoid quantification results. In both WT and ΔhpnP, each type of hopanoid is enriched in the OM compared to the IM. The total of these four hopanoids in the IM is ∼2.6 mol% of TLE, whereas in the OM, the value can reach 8–11 mol%. For individual hopanoids in WT, the mol% in IM and OM are 1% and 2% for Dip, 1% and 2.4% for 2Me-Dip, 0.4% and 3.4% for BHT, and 0.1% and 0.3% for 2Me-BHT, respectively. In ΔhpnP, the IM and OM values are 2% and 7.5% for Dip, and 0.5% and 4.3% for BHT, respectively (Figure 4). This quantitative measurement of hopanoid content within the IM and OM can be used to evaluate the impact of 2-methylation upon hopanoid subcellular distribution.10.7554/eLife.05663.008
Molar percentage of hopanoids in the inner membrane (IM) and outer membrane (OM) of WT and Δ<italic>hpnP</italic> determined by GC–MS.
Error bars represent the standard deviation from three biological replicates. Total hopanoids = sum of (2Me)-Dip and (2Me)-BHT.
DOI:http://dx.doi.org/10.7554/eLife.05663.008
The ratio of total hopanoids in the OM vs IM is 3.1 ± 0.4 and 4.4 ± 0.6 for WT and ΔhpnP, respectively, which is a significant difference (p = 0.038). Comparing the ratios between short ((2Me)-Dip) and long ((2Me)-BHT) chain hopanoids revealed an enrichment of long-chain hopanoids in the OM compared to the IM in both WT and ΔhpnP (Figure 5). Interestingly, about equal amounts of 2Me-Dip and Dip were found in both IM and OM, whereas 2Me-BHT is 16% and 8% of BHT in the IM and OM, respectively, suggesting hopanoid 2-methylation has neither strong nor consistent effects on the partitioning of short and long species between the IM and OM (Figure 5). However, our data do indicate that 2-methylation impacts that total amount of hopanoid enrichment in the OM.10.7554/eLife.05663.009
Partitioning of hopanoids in the inner membrane (IM) and outer mebraane (OM) of <italic>R. palustris</italic> TIE-1.
(A) Molar ratio between short (Dip and 2Me-Dip) and long (BHT and 2Me-BHT) hopanoids in WT and ΔhpnP. (B) Molar ratio between methylated and desmethylated hopanoid in WT. Error bars represent the standard deviation from three biological replicates. *p = 0.015; **p < 0.01.
DOI:http://dx.doi.org/10.7554/eLife.05663.009
Membrane rigidity measurements using model lipid SUVs
To put these numbers in context and gain a deeper understanding on how the amount, chain-length, and 2-methylation of hopanoids impact membrane biophysical properties, we performed membrane rigidity measurements. Small unilamellar vesicles (SUVs) are commonly used for such measurements because it is straightforward to control their lipid composition (Hope et al., 1985). We used fluorescence polarization to measure the fluorescence of DPH, a reporter dye, in the presence of model lipids and purified hopanoids. Figure 6 shows how the membrane rigidity of model lipids, DOPC and E. coli PLE, responds to the presence of cholesterol, squalene, (2Me)-Dip, and (2Me)-BHT. Because the total hopanoids in cell membranes are ∼10 mol% (Figure 4), we varied the concentration of hopanoids between 5 and 20 mol% in SUVs, which our quantification results suggest are physiologically relevant concentrations. Starting with the simplest single lipid background, DOPC, the addition of cholesterol increases membrane rigidity and the magnitude of change is proportional to the amounts of cholesterol added (black line, Figure 6A). This observation is consistent with the literature and validates our technical approach (Vanblitterswijk et al., 1987). Interestingly, the addition of the hopanoid precursor, squalene, has the opposite effect as cholesterol and makes the membrane less rigid in a concentration-dependent fashion (magenta line, Figure 6A). However, when Dip is added, the membrane rigidity is unaffected (blue line, Figure 6A). Surprisingly, given the structural similarity between cholesterol and Dip, Dip does not seem to rigidify DOPC as extensively as does cholesterol. However, when Dip is further processed by the cell to produce BHT, it rigidifies DOPC vesicles in the same manner as cholesterol (red line, Figure 6A).10.7554/eLife.05663.010
Membrane rigidity measurements at 25°C using model lipids.
(A) Dioleoyl phosphatidylcholine (DOPC) and (B) E. coli polar lipid extract (PLE) were mixed with different mol% of cholesterol, squalene, and hopanoids. Error bars represent the standard deviation from three biological replicates (total 21 technical replicates).
DOI:http://dx.doi.org/10.7554/eLife.05663.010
10.7554/eLife.05663.011
Membrane rigidity measurements at 25°C and 40°C using model lipids.
(A) DOPC and (B) E. coli polar lipid extract (PLE) were mixed with different mol% of cholesterol, squalene, and hopanoids. Error bars represent standard deviation from three biological replicates (total 21 technical replicates).
DOI:http://dx.doi.org/10.7554/eLife.05663.011
2-methylation of both Dip and BHT has striking effects on their ability to rigidify DOPC. Not only does 2-methylation increase DOPC membrane rigidity, but at 20 mol%, 2Me-Dip even outperforms cholesterol and BHT (cyan line, Figure 6A). 2Me-BHT also increases membrane rigidity similar to cholesterol, even though the difference from BHT is much smaller than that between Dip and 2Me-Dip (orange line, Figure 6A). This result demonstrates methylation itself changes the biophysical properties of hopanoids, which directly impacts membrane fluidity. This observation reinforces the interpretation that the lack of phenotype in the ΔhpnP strain may be due to cells synthesizing other lipids that functionally complement the absence of 2Me-hopanoids, rather than due to a lack of an impact at the molecular level per se.
To determine whether the effects on membrane rigidity by these hopanoids hold in more physiologically relevant environments, we repeated the experiment using E. coli PLE as the main component to form SUVs (Figure 6B). E. coli PLE from Avanti Polar Lipids consists of 67% phosphatidylethanolamine (PE), 23.2% phosphatidylglycerol (PG), and 9.8% cardiolipin (CL) (in wt%) and has an average molecular weight identical to DOPC (786 g/mol). Compared to DOPC alone, the SUVs from E. coli PLE are more rigid, probably because the difference in fatty acid chain saturation (green square, Figure 6A,B). Even though E. coli PLE SUVs are more rigid than DOPC SUVs to start with, we observe the same trend, where adding cholesterol rigidifies the vesicles and squalene fluidizes them. However, the concentration dependence on the rigidity change is more dramatic for cholesterol compared to squalene (black line, Figure 6B). Unlike in DOPC, Dip has a small rigidifying effect in the E. coli PLE background, yet similar to DOPC, no concentration dependence is observed (blue line, Figure 6B).
Addition of 2Me-Dip also rigidifies E. coli PLE, but in contrast to what was observed for DOPC, 2Me-Dip shows lower capacity to rigidify the membrane than cholesterol (cyan line, Figure 6B). Interestingly, the extended hopanoid, BHT, also rigidifies the membrane, but seemed to saturate at 10 mol% (red line, Figure 6B). Surprisingly, in sharp contrast to the DOPC background, 2Me-BHT strongly rigidifies the E. coli PLE membranes and has a concentration dependence (orange line, Figure 6B). This result shows that the impact of 2-methylation and hopanoid extension on a membrane biophysical property depends on the specific lipid environment. As expected, when these experiments were repeated at 40°C, membranes were less rigid overall. However, similar trends compared to 25°C were observed in both DOPC and E. coli PLE background (Figure 6—figure supplement 1).
Quantification of phospholipid composition in <italic>R. palustris</italic> TIE-1
Given that 2-methylation of hopanoids differentially impacts membrane rigidity based on the lipid context, to understand the physiological effects of 2-methylhopanoids in the IM and OM of R. palustris TIE-1, we must characterize the composition of phospholipids in native membranes and perform biophysical experiments using these membranes. To determine the exact quantity of each phospholipid in R. palustris TIE-1, we purified the IM and OM as described above and analyzed the lipid composition by LC-MS using electron spray ionization (Malott et al., 2014). The elution profiles between strains and membranes look similar (Figure 7). A total of 33 major phospholipids were identified, including 10 PC, 9 PE, 7 PG, and 7 CL (Table 3). To determine the absolute quantity of each phospholipid, exogenous standards of PC (17:0/17:0), PE (17:0/17:0), and PG (17:0/17:0) that are absent in R. palustris TIE-1 were added as internal standards for LC-MS analyses. Although we can detect cardiolipins, we unfortunately are unable to quantify them due to the low solubility of the cardiolipin standard in the LC-MS solvent.10.7554/eLife.05663.012
LC-MS profiles of the inner membrane (IM) and outer membrane (OM) of <italic>R. palustris</italic> TIE-1 WT and Δ<italic>hpnP</italic>.
DOI:http://dx.doi.org/10.7554/eLife.05663.012
10.7554/eLife.05663.013
Annotation of phospholipids identified by LC-MS analyses (see Figure 7)
DOI:http://dx.doi.org/10.7554/eLife.05663.013
Compound
RT (min)
[M+H]+
[M−C3H7O2HPO4]+
[M+NH4]+
PG32:1
4.39
549.4895
PG34:2
4.89
575.505
PC34:2
5.94
758.5694
PG34:1
6.11
577.5208
PG36:2
6.2
603.5364
PE34:2
6.38
716.5225
PC_cyc35:1
6.76
772.5851
PG_cyc35
6.95
591.5364
PG_cyc37:1
7.05
617.5521
PC(35:2)
7.06
772.5851
PE_cyc35:1
7.28
730.5381
PC34:1
7.51
760.5851
PC36:2
7.65
786.6007
PG(17:0/17:0)
7.86
579.5364
PG36:1
7.99
605.5521
PE34:1
8.14
718.5381
PE36:2
8.27
744.5538
PC_cyc35
8.58
774.6007
PC_cyc37:1
8.72
800.6164
PE_cyc35
9.28
732.5538
PE_cyc37:1
9.42
758.5694
PC(17:0/17:0)
9.75
762.6007
PC36:1
9.91
788.6164
PE(17:0/17:0)
10.53
720.5538
PE36:1
10.71
746.5694
PC_cyc37
11.23
802.632
PE_cyc37
12.1
760.5851
PC36:0
12.79
790.634
PE36
12.98
748.5851
PC36:4
13.39
782.569
CL70:4
15.09
1447.0373
CL68:3
15.1
1421.0217
CL72:4
15.31
1475.0686
CL70:3
15.34
1449.053
CL68:2
15.35
1423.0373
CL72:3
15.58
1477.0843
CL70:2
15.6
1451.0686
The types of lipids (PC: phosphatidylcholine, PE: phosphotidylethanolamine, PG: phosphatidylglycerol, CL: cardiolipin, cyc: cyclopropanation; the first number indicates the total number of carbon of the fatty acid chains and the second number indicates the number of double bonds in these chains) and their retention time (RT, min) and m/z value of the base peak are shown. For PC and PE, the base peak is the proton adduct and for CL, the base peak is the ammonium adduct. For PG, the base peak indicates a loss of glycerophosphate (−171 m/z).
Table 4 shows the wt% of each identified phospholipid in the TLE. Among the identified phospholipids, the IM in WT and ΔhpnP has ∼50%, ∼36%, and ∼12–14% of PC, PE, and PG, respectively, whereas the OM in WT and ΔhpnP has ∼56–58%, ∼30–33%, and ∼11–12% of PC, PE, and PG, respectively. However, when we calculate the total identified phospholipid amount, it only accounts for ∼26–34 wt% of the original TLE samples (Table 4). This low value could be due to (1) low solubility of cardiolipins and other unidentifiable lipids in LC-MS, (2) differences in the ionization efficiency of the phospholipids with different chain length or saturation (Myers et al., 2011), and (3) the ionization suppression effects occur from co-eluting lipids (Brugger et al., 1997; Furey et al., 2013). Nevertheless, the average molecular weight of the phospholipids without cardiolipin is 738 g/mol. Considering the higher molecular weight of cardiolipin, our calculation for the mol% of hopanoids in TLE using an average molecular weight of 786 g/mol may be very close to the real value in R. palustris TIE-1.10.7554/eLife.05663.014
Phospholipid compositions in the inner membrane (IM) and outer membrane (OM) of R. palustris TIE-1 WT and ΔhpnP analyzed by LC-MS
DOI:http://dx.doi.org/10.7554/eLife.05663.014
Weight % of TLE
Compound
RT (min)
WT IM
WT OM
ΔhpnP IM
ΔhpnP OM
PC36:2
7.65
5.15 ± 0.47
5.34 ± 0.26
5.44 ± 0.57
5.94 ± 0.48
PC_cyc37:1
8.72
4.33 ± 0.45
4.01 ± 0.48
4.07 ± 1.13
3.96 ± 1.24
PC36:1
9.91
2.90 ± 0.3
1.84 ± 0.52
2.59 ± 0.93
2.15 ± 1.08
PC34:1
7.51
2.66 ± 0.22
2.23 ± 0.33
2.53 ± 0.58
2.38 ± 0.57
PC34:2
5.94
1.56 ± 0.12
1.24 ± 0.44
1.41 ± 0.51
1.44 ± 0.48
PC_cyc35
8.58
0.23 ± 0.01
0.14 ± 0.05
0.20 ± 0.07
0.17 ± 0.1
PC_cyc35:1
6.76
0.12 ± 0.01
0.07 ± 0.02
0.11 ± 0.04
0.09 ± 0.06
PC(35:2)
7.06
0.09 ± 0.01
0.05 ± 0.03
0.08 ± 0.03
0.06 ± 0.05
PC_cyc37
11.23
0.07 ± 0.01
0.04 ± 0.01
0.06 ± 0.02
0.05 ± 0.03
PC36:0
12.79
0.02 ± 0
0.01 ± 0
0.02 ± 0.01
0.01 ± 0.01
Sum
17.12
14.96
16.50
16.25
PE_cyc37:1
9.42
4.92 ± 0.36
3.36 ± 0.94
4.68 ± 1.5
4.18 ± 1.64
PE34:1
8.14
2.49 ± 0.19
1.44 ± 0.49
2.28 ± 0.84
1.82 ± 0.92
PE36:1
10.71
2.29 ± 0.28
1.08 ± 0.38
2.05 ± 0.85
1.43 ± 0.95
PE36:2
8.27
1.88 ± 0.21
1.22 ± 0.46
1.74 ± 0.65
1.59 ± 0.72
PE34:2
6.38
0.43 ± 0.02
0.23 ± 0.08
0.42 ± 0.16
0.33 ± 0.21
PE_cyc35
9.28
0.20 ± 0.02
0.11 ± 0.04
0.19 ± 0.07
0.15 ± 0.09
PE_cyc35:1
7.28
0.12 ± 0.01
0.06 ± 0.02
0.11 ± 0.05
0.09 ± 0.06
PE_cyc37
12.1
0.07 ± 0.01
0.04 ± 0.01
0.07 ± 0.02
0.05 ± 0.03
PE36
12.98
0.04 ± 0
0.03 ± 0.01
0.04 ± 0.01
0.03 ± 0.01
Sum
12.44
7.58
11.58
9.67
PG36:2
6.2
2.59 ± 0.24
1.68 ± 0.54
2.25 ± 0.67
1.87 ± 0.8
PG36:1
7.99
1.13 ± 0.08
0.69 ± 0.19
0.98 ± 0.31
0.81 ± 0.35
PG34:1
6.11
0.77 ± 0.17
0.5 ± 0.2
0.6 ± 0.24
0.51 ± 0.19
PG_cyc37:1
7.05
0.11 ± 0.02
0.07 ± 0.05
0.08 ± 0.02
0.07 ± 0.04
PG34:2
4.89
0.05 ± 0
0.03 ± 0.01
0.04 ± 0.01
0.03 ± 0.02
PG_cyc35
6.95
0.02 ± 0
0.01 ± 0
0.02 ± 0.01
0.01 ± 0.01
PG32:1
4.39
0.00 ± 0
0.01 ± 0
0 ± 0
0.01 ± 0
Sum
4.67
3
3.97
3.31
PC + PE + PG
Total % of TLE
34.23
25.54
32.06
29.23
Membrane rigidity measurements using lipid extract from <italic>R. palustris</italic> TIE-1 inner and outer membranes
Due to the limitations we encountered in quantifying native phospholipids, we elected to generate SUVs directly using the TLE from fractionated R. palustris TIE-1 IM and OM rather than reconstituting SUVs from commercially available sources. Because our main focus is on the effect of 2-methylhopanoids, we used the membranes from ΔhpnP and added 5, 10, and 20 mol% purified hopanoids to constitute the SUV lipid mixture. We also included WT IM and OM for comparison. Figure 8 shows membrane rigidity measurements at 25°C using R. palustris TIE-1 membranes. In both WT and ΔhpnP, the OM showed higher membrane rigidity than IM, which could be due to the higher hopanoid content in the OM (Figure 4) and differences in phospholipid content. Compared to WT, the ΔhpnP IM and OM showed decreased rigidity. This contrasts with our whole cell membrane rigidity results in which no difference in rigidity was observed (Figure 2). This discrepancy could be due to the presence of lipid A and outer membrane proteins that may affect the overall membrane rigidity in whole cells.10.7554/eLife.05663.015
Membrane rigidity measurements at 25°C using total lipid extract from <italic>R. palustris</italic> TIE-1 inner membrane (IM) and outer membrane (OM).
The IM (A) or OM (B) from ΔhpnP was mixed with different mol% of hopanoids. Error bars represent the standard deviation from three biological replicates (total 21 technical replicates). *p < 0.001 (relative to ΔhpnP).
DOI:http://dx.doi.org/10.7554/eLife.05663.015
10.7554/eLife.05663.016
Membrane rigidity measurements at 40°C using total lipid extract from <italic>R. palustris</italic> TIE-1 inner membrane (IM) and outer membrane (OM).
The IM (A) or OM (B) from ΔhpnP was mixed with different mol% of hopanoids. Small unilamellar vesicles from the lipid mixtures were prepared, and the membrane rigidity was measured by fluorescence polarization of a reporter dye diphenyl hexatriene. Error bars represent standard deviation from three biological replicates (total 21 technical replicates).
DOI:http://dx.doi.org/10.7554/eLife.05663.016
Similar to our observations in model lipids, the addition of Dip has no effect on membrane rigidity, both in the IM and OM. However, 2Me-Dip rigidifies both IM and OM in a concentration-dependent manner, as seen in the model lipid SUVs. Interestingly, unlike the effects BHT exerts in model lipids, it does not rigidify either the IM or OM. While it is possible that BHT has no effect on native membrane rigidity, given that ΔhpnP membranes contain BHT, it seems more likely that endogenous BHT is saturating its membrane rigidifying capacity, similar to what we observed for E. coli PLE (red line, Figure 6B).
When BHT is methylated, it rigidifies both the IM and OM. However, different trends can be seen between these membranes: in the IM, the more 2Me-BHT present, the higher the membrane rigidity, yet OM membrane rigidity appears to saturate by 5 mol% 2Me-BHT and remains constant between 5–20 mol% of 2Me-BHT (Figure 8B). It is tempting to speculate that the reason less 2-methylation occurs for BHT than Dip in the OM (Figures 4, 5) is because less methylation of BHT is needed to significantly impact OM rigidity.
We repeated these experiments at 40°C and observed similar trends as seen at 25°C (Figure 8—figure supplement 1). However, we had larger standard deviations than in our model lipid experiments, which could be due to higher heterogeneity in the samples extracted from IM and OM. Nevertheless, we find clear physiologically relevant distinctions in the rigidifying effects of both short and long 2Me-hopanoids on the IM and OM of R. palustris TIE-1.
Discussion
Until now, a specific role for 2Me-hopanoids in living cells has evaded experimental detection, yet its identification has been of great interest for interpreting the extensive 2Me-hopane fossil record (Welander et al., 2010; Kulkarni et al., 2013). Our findings that 2Me-hopanoids rigidify membranes to different extents depending both on their specific structure (short or long) and lipid context not only provide a clear biological function for these compounds, but also help rationalize why previous efforts to identify such a function have been challenging. That methylation per se can contribute to rigidifying membranes may also help explain the association of methylated hopanoids in certain modern and ancient environments.
Under what circumstances would adding a methyl group at the 2′ position of hopanoids, which seems a rather small modification, be beneficial? Might there be a mechanistic explanation for the enrichment of 2Me-hopanes during stressful OAEs (Knoll et al., 2007)? Several independent lines of evidence bridge our biophysical findings with the abundance patterns in the rock record, together suggesting that 2-methylhopanoids confer stress resistance: (1) 2Me-hopanoids are enriched in the outer membrane of akinetes, survival cell types of the cyanobacterium Nostoc punctiforme (Doughty et al., 2009), (2) a stress-responsive pathway upregulates the HpnP methylase in R. palusris (Kulkarni et al., 2013), (3) in modern environments, the capacity for 2Me-hopanoid production significantly correlates with organisms, metabolisms (e.g., nitrogen fixation), and environments that support plant–microbe interactions (Ricci et al., 2014). This correlation, together with the observation that (2Me)-hopanoids promote symbiosis (Silipo et al., 2014), tempt us to speculate that 2Me-hopanoids may indirectly facilitate nitrogen fixation by enhancing bacterial survival under the stressful conditions that accompany the establishment of symbiosis (Gibson et al., 2008). Going forward, it is worth critically examining this hypothesis; if correct, such an interpretation would indicate that spikes in the 2Me-hopane index may reflect episodes of particular environmental stresses favoring the growth of organisms capable of withstanding it using (2Me)-hopanoids.
How might the 2-methylation of hopanoids permit such an adventitious adaptation? Molecular dynamic simulations of 2Me-hopanoid within a relevant lipid context are required to understand the interactions on an atomic scale. However, we may speculate about the mechanism of rigidification through lessons learned from studies of cholesterol, in which the number of methyl groups control its optimal molecular packing to geometrically complement phospholipid chains (Bloch, 1979). The addition or removal of methyl groups over the evolution of the cholesterol lipid family is thought to have optimally tuned cholesterol's ability to order or condense phospholipid membranes (Miao et al., 2002; Rog et al., 2007). We suggest that when a methyl group is added onto the 2′ position of A-ring in hopanoid, the stearic hindrance between the 2-methyl group and the methyl groups at the 4′ and 10′ position of the A-ring could transform the ring from a chair to a twisted conformation. The two additional 1,3-diaxial interactions elicited by hopanoid 2-methylation could mimic the smoothing and/or tilting effect known for cholesterol, thus rationalizing how 2-methylation may improve the ability of hopanoids to rigidify membranes (Rog et al., 2007).
The differences in (2Me)-hopanoid distribution between the IM and OM pose many interesting questions about their role in maintaining membrane integrity and homeostasis. Compared to (2Me)-Dip, the addition of a hydrophilic tail to form (2Me)-BHT or even BHT-glucosamine (Figure 1) may favor a stronger interaction with the outer leaflet of the OM, in which the lipid head group is heavily modified with hydrophilic molecules. The relative enrichment of (2Me)-BHT in the OM is consistent with such a scenario. The hydrophilic tail of 2Me-BHT could also affect the vertical position of the 2-methyl group in the membrane compared to 2Me-Dip (Rog et al., 2007; Poger and Mark, 2013), which may explain the difference between 2Me-BHT and 2Me-Dip in rigidifying membranes with different compositions. Future research will illuminate whether there are additional interactions between hopanoids and other membrane constituents (e.g., proteins or cell wall components) that facilitate survival under stress.
Finally, it is important to keep in mind that (2Me)-hopanoids may act locally rather than globally with respect to influencing membrane biophysical properties. In our in vitro experiments, we did not observe a significant difference in membrane rigidity when less than 10 mol% of (2Me)-hopanoids were used (Figure 6). This may be due to a critical concentration needed to trigger an effect, which is consistent with molecular dynamic simulations that demonstrate cholesterol starts to self-organize within membranes at concentration above 10 mol% (Martinez-Seara et al., 2010). However, we hasten to point out that the mol% of (2Me)-hopanoids in the native membrane experiment (Figure 8) are not directly comparable to those using model lipids (Figure 6) due to the presence of endogenous hopanoids in ΔhpnP membranes. The existence of BHT in ΔhpnP may explain why the membrane rigidifying effect of exogenously added BHT is saturated at 10 mol% in the E. coli PLE but has no impact on R. palustris TIE-1 IM and OM. In this context, it is noteworthy that cardiolipin significantly increases in the absence of all hopanoids (unpublished). In E. coli, cardiolipin localizes to negatively curved regions of the cell (Renner and Weibel, 2011). Looking beyond the function of methylation, it is possible that certain hopanoid types could fulfill the geometry requirements of curved membranes and facilitate cell division or vesicle formation, consistent with both the microdomain features observed in prior subcellular hopanoid localization studies (Doughty et al., 2014) and the strong cell division defect displayed by a mutant lacking the ability to transport hopanoids to the OM (Doughty et al., 2014). Going forward, consideration of other roles for structurally diverse hopanoids, including the possibility that some might influence membrane protein function (Phillips et al., 2009), modify specific proteins or cell wall components through covalent linkages (Jeong and McMahon, 2002; Silipo et al., 2014), or even play a role in signaling pathways in analogy to cholesterol and phosphatidylcholine (Kuwabara and Labouesse, 2002; Aktas et al., 2010), will enhance our appreciation for this ancient lipid class.
Materials and methodsBacterial strains and chemicals
R. palustris TIE-1 wild type (WT) and mutant strains were grown as previously described (Welander et al., 2012). Purified hopanoids ((2Me)-diplopterol, (2Me)-bacteriohopanetetrol [BHT]) were obtained by following the purification protocols (Wu et al., 2015). E. coli polar lipid extract and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-diheptadecanoyl-sn-glycero-3-phosphocholine (PC[17:0/17:0]), 1,2-diheptadecanoyl-sn-glycero-3-phosphoethanolamine (PE[17:0/17:0]), 1,2-diheptadecanoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (PG[17:0/17:0]) were from Avanti Polar Lipids (Alabaster, AL). Squalene, cholesterol, pyridine, acetic anhydride, morpholinepropanesulfic acid (MOPS), 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), sodium succinate, 1,6-diphenyl-1,3,5-hexatriene (DPH), Percoll, and tetrahydrofuran (THF) were from Sigma–Aldrich (Milwaukee, WI). Yeast extract was from HIMEDIA (Mumbai, India). Peptone was from BD Biosciences (San Jose, CA). Methanol and dichloromethane (DCM) were HPLC grade from Alfa Aesar (Ward Hill, MA).
Whole cell membrane fluidity measurements
To prepare bacterial cells for measurements of membrane fluidity, single colonies of R. palustris TIE-1 WT and mutants were inoculated into 10 ml YPMS (0.3% yeast extract, 0.3% peptone, 50 mM MOPS, 5 mM succinate, pH 7.0) and grown at 30°C, 250 rpm for ∼72 hr to reach late stationary phase (OD600 ∼1.0). The cells (250 μl) were spun down and the cell pellets were washed once with 50 mM HEPES, 50 mM NaCl, pH 7.0 (buffer A). Pellets were resuspended in different amounts of the same buffer to adjust the final OD600 to ∼0.2. To measure membrane fluidity, 4.3 μl of DPH (736 μM stock solution in ethanol; the concentration was determined by ε350 nm = 88 cm−1 mM−1 in methanol) was added into 400 μl of the cell suspension. Samples were incubated in a 25°C or 40°C water bath without light for 30 min, followed by measurements of fluorescence polarization (Fluorolog, HORIBA Instruments (Edison, NJ). Instrument parameter: ex 358 nm, slit = 3 mm; em 428 nm, slit = 7 mm; integration time = 1 s) (Lin et al., 2011). Three biological replicates were measured, each containing 6–14 technical replicates. To reduce bias from the stability of the instrument and the samples, especially at 40°C, we randomized our data acquisition sequence. p value in this manuscript represents t-test using two-tailed equal variance.
Membrane fractionation using Percoll gradient
To prepare cell cultures for membrane fractionation, single colonies of R. palustris TIE-1 WT or ΔhpnP mutant were inoculated into 10 ml YPMS and grown for ∼4 days at 30°C, 250 rpm. The culture (0.5 ml) was then inoculated into 1 l of YPMS in a 2-l flask and grown at 30°C, 250 rpm for 4 days before harvesting by centrifugation at 12,000×g for 20 min at RT. The typical yield was ∼1.8 g of wet cell paste per 1 l culture. To estimate the yield in dried cells, a small aliquot of the wet cell paste was lyophilized until there was no further change in weight. On average the weight of dried cells was one third of wet cells. The wet cell pastes were stored at −80°C before cell lysis.
To lyse the cells, 19 ml of buffer A was added into ∼3.6 g of cells (from 2 l culture) and passed through a French Press twice at 14,000 psi, followed by sonication (Sonic Dismembrator 550, Fisher Scientific (Waltham, MA), 1/8 inch tip, power output 3.5, 1 s on, 4 s off, total on time 5 min at 4°C). The cell debris was spun down at 20,000×g, 20 min at 4°C. The supernatant containing cell membranes was transferred into 4-ml ultracentrifugation tubes (∼3 ml sample per tube) and centrifuged at 80,000 rpm in a TLA-100.3 rotor for 1 hr at 4°C (Optima MAX Ultracentrifuge, Beckman Coulter (Brea, CA)). The resulting membrane pellets in each tube were resuspended in 300 μl buffer A by pipetting while being sonicated in a bath sonicator (VWR (Radnor, PA) B2500A-DTH, 42 kHz, RF Power 85 W). The suspension was combined into one single tube and sonicated again using the probe sonicator (power level 2.5, 1 s on, 4 s off, total on time 2.5 min, 4°C).
To separate inner and outer membranes, ∼320 μl of membrane samples were laid on top of 3 ml 18% Percoll (vol/vol in buffer A), followed by ultracentrifugation at 30,000 rpm in a TLA 100.3 rotor at 4°C for 15 min (Morein et al., 1994). Three visually distinct bands were formed and a pipetman was used to take in sequence of the top band (1 ml), bottom band (∼200–250 μl), and the middle band (0.7–1 ml) (Figure 3). The top and bottom bands constituted the IM and OM, determined by the presence and absence of NADH-oxidase activity, respectively. The band on top of the OM was less defined and exhibited some NADH-oxidase activity, which may be an artifact from sonication steps that mixed the IM and OM, and we therefore discarded it in our subsequent studies. To remove Percoll, samples from the same band were combined and centrifuged in a TLA 100.3 rotor at 50,000 rpm at 4°C for 1.5 hr. After centrifugation, a layer of the fractionated membrane was formed on top of a transparent Percoll layer (Figure 3). The membrane layers were collected by pipetting gently in water and/or scraped gently using a metal spatula. The membrane samples were then frozen at −20°C before being lyophilized. The total lipid extractions (TLE) from the lyophilized membranes were obtained by modified Bligh–Dyer extraction according to published protocols (Kulkarni et al., 2013).
Quantification of lipid compositions of inner and outer membranes of <italic>R. palustris</italic> TIE-1
GC-MS (Thermo Scientific (Waltham, MA) Trace-GC/ISQ mass spectrometer with a Restek Rxi-XLB column [30 m × 0.25 mm × 0.10 μm]) was used to quantify hopanoids from the TLE from the IM and OM. An internal standard, androsterone (750 ng) was air dried with 100 μg of the TLE overnight at RT and derivatized with 50 μl acetic anhydride and 50 μl pyridine at 60°C for 30 min, followed by GC-MS analyses as described (Welander et al., 2009; Kulkarni et al., 2013). To account for the difference in ionization efficiencies between androsterone and hopanoids, calibration curves using androsterone and purified hopanoids were generated to quantify hopanoids (Wu et al., 2015).
LC-MS (Waters (Milford, MA) Acquity UPLC/Xevo G2-S time-of-flight mass spectrometer with a CSH C18 column [2.1 × 100 mm × 1.7 μm]) was used to quantify both phospholipids and hopanoids. Phospholipid internal standards (PC[17:0/17:0], PE[17:0/17:0], PG[17:0/17:0], 1 μg each) were mixed with 100 μg of the TLE from fractionated membranes and air-dried overnight at RT. LC solvent (200 µl, Isopropanol:acetonitrile:water = 2:1:1) was then added into the samples, followed by sonication before analyses by LC-MS as described earlier (Malott et al., 2014; Wu et al., 2015). The column temperature was maintained at 55°C. A binary solvent system containing solvent A (acetonitrile:water; 60:40) and solvent B (isopropanol:acetonitrile; 90:10), both with 10 mM ammonium formate and 0.1% formic acid was used. The flow rate was set at 400 μl/min and the elution program started at 40% B, increased linearly to 43% B in 2 min, then to 50% B in 0.1 min, followed by a linear increase to 54% B over 9.9 min, a jump to 70% B in 0.1 min, another linear increase to 99% B over 5.9 min, a subsequent decrease to 40% B in 0.1 min, and then maintained at the same level for 1.9 min.
The eluents from the column were ionized by electrospray ionization (ESI). MSE data from 100 to 1500 m/z were collected in either the positive or negative ion mode. MSE consisting of both low energy and high energy scans were obtained simultaneously. During data analysis product ions can be associated with parent ions if they are coincident in chromatographic time. Electrospray conditions were capillary voltage 2.0 kV, cone voltage 30 V, source offset 60 V, source temperature 120°C, desolvation temperature 550°C, cone gas 20 l/hr, and desolvation gas 900 l/hr. The TOF-MS was run in resolution mode, typically 32,000 m/Δm. The mass axis was calibrated with sodium formate clusters. Leucine enkephalin was used as a mass reference during acquisition. The data were collected in continuum mode, and then converted to centroid mode for quantitative analysis using the Quanlynx program (Waters Corporation, Milford, MA) (Wu et al., 2015).
Membrane fluidity measurements in small unilamellar vesicles (SUVs)
Cholesterol, squalene, and hopanoid stock solutions were prepared at 1 mg/ml in THF and the E. coli PLE and DOPC were prepared at 10 mg/ml in DCM. To prepare lipid mixture, a total of 1 μmol of lipid was added into 0.5 ml of DCM and dried in a rotary evaporator. Because any residual solvents can cause high errors in fluidity measurements, the samples were placed under vacuum overnight to ensure complete removal of organic solvents.
To prepare SUVs, 1 ml of buffer A was added into the glass vials containing dried lipid mixtures. Samples were suspended by sonication for 1 hr at RT in a bath sonicator (VWR B2500A-DTH, 42 kHz, RF Power 85 W). The suspended lipids (murky giant multilamellar vesicles) were transferred into 1.5-ml eppendorf and flash frozen in liquid nitrogen for 3 min, followed by thawing in a 37°C water bath for 3 min. This freeze–thaw cycle, which breaks down the giant vesicles into smaller ones, was repeated two more times. SUVs were prepared by passing the samples through 0.1-μm polycarbonate membranes (Whatman) using Avanti mini-Extruder at RT (Avanti Polar Lipids). The extrusion was performed a total of 11 times and the vesicle suspension became clear during the process. The sizes and stability of the SUVs was determined by dynamic light scattering (Wyatt (Santa Barbara, CA) DynaPro NanoStar. Instrument parameters: acquisition time 5 s, number of acquisition 10, laser wavelength 659 nm, laser power 10%, 25°C). The average size distribution of the SUVs was between 80 and 90 nm and remained stable for at least 4 hr at RT.
SUVs after extrusion were diluted 1:1 in buffer A to reach a final concentration of 0.5 mM (400 μl total volume). DPH (1.8 μl of 44.5 μM stock solution in ethanol) was added into the sample and vortexed immediately. The SUV-DPH samples were incubated in 25°C or 40°C water bath without light for at least 30 min before the fluorescence polarization was measured using the parameters described above. The concentration of the fluorescence reporter dye DPH and the instrument parameters were optimized to have a strong and linear signal output.
Membrane fluidity measurements using total lipid extract from inner and outer membranes of <italic>R. palustris</italic> TIE-1
Different amounts of purified hopanoids (5, 10, and 20 nmol) were added to 100 nmol of the inner or outer membrane extracts of ΔhpnP (assuming average molecular weight is 786 g/mol). The same procedures as described above were followed for the preparation of SUVs. Buffer A (600 μl) was used to suspend the dried lipids so that the final lipid concentrations before the addition of DPH were between 0.088 and 0.1 mM (400 μl sample volume). To measure fluorescence polarization, DPH (1.8 μl of 7.4 μM stock solution in ethanol) was used (the final concentration of DPH was 0.03 μM, which was ∼0.03 mol% of the total lipids in the sample). Controls of membranes from WT or ΔhpnP only without addition of hopanoids were included.
Acknowledgements
We thank the members of the Newman lab for critical comments on the manuscript. We thank Dr Nathan Dalleska for help with UPLC-TOF-MS. The UPLC-TOF-MS equipment in the California Institute of Technology's Environmental Analysis Center was used in the work. We thank Dr Heun Jin Lee and Dr Eva Schmid for vesicle preparation advice. This work was supported by grants from NASA (NNX12AD93G), the National Science Foundation (1224158), and the Howard Hughes Medical Institute (HHMI) to DKN. DKN is an HHMI Investigator.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
C-HW, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
MB-F, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
DKN, Conception and design, Analysis and interpretation of data, Drafting or revising the article
ReferencesAktasMWesselMHackerSKlüsenerSGleichenhagenJNarberhausF2010Phosphatidylcholine biosynthesis and its significance in bacteria interacting with eukaryotic cellsEuropean Journal of Cell Biology8988889410.1016/j.ejcb.2010.06.013BlochKE1979Speculations on the evolution of sterol structure and functionCRC Critical Reviews in Biochemistry71510.3109/10409237909102566BrocksJJLoveGDSummonsREKnollAHLoganGABowdenSA2005Biomarker evidence for green and purple sulphur bacteria in a stratified Palaeoproterozoic seaNature43786687010.1038/nature04068BrocksJJPearsonA2005Building the biomarker tree of lifeReviews in Mineralogy & Geochemistry5923325810.2138/rmg.2005.59.10BruggerBErbenGSandhoffRWielandFTLehmannWD1997Quantitative analysis of biological membrane lipids at the low picomole level by nano-electrospray ionization tandem mass spectrometryProceedings of the National Academy of Sciences of USA942339234410.1073/pnas.94.6.2339DoughtyDMColemanMLHunterRCSessionsALSummonsRENewmanDK2011The RND-family transporter, HpnN, is required for hopanoid localization to the outer membrane of Rhodopseudomonas palustris TIE-1Proceedings of the National Academy of Sciences of USA108E1045E105110.1073/pnas.1104209108DoughtyDMDieterleMSessionsALFischerWWNewmanDK2014Probing the subcellular localization of hopanoid lipids in bacteria using NanoSIMSPLOS ONE9e8445510.1371/journal.pone.0084455DoughtyDMHunterRSummonsRENewmanDK20092-Methylhopanoids are maximally produced in akinetes of Nostoc punctiforme: geobiological implicationsGeobiology752453210.1111/j.1472-4669.2009.00217.xFureyAMoriartyMBaneVKinsellaBLehaneM2013Ion suppression; a critical review on causes, evaluation, prevention and applicationsTalanta11510412210.1016/j.talanta.2013.03.048GalletA2011Hedgehog morphogen: from secretion to receptionTrends in Cell Biology2123824610.1016/j.tcb.2010.12.005GibsonKEKobayashiHWalkerGC2008Molecular determinants of a symbiotic chronic infectionAnnual Review of Genetics4241344110.1146/annurev.genet.42.110807.091427HanukogluI1992Steroidogenic enzymes - structure, function, and role in regulation of steroid-hormone biosynthesisThe Journal of Steroid Biochemistry and Molecular Biology4377980410.1016/0960-0760(92)90307-5HopeMJBallyMBWebbGCullisPR1985Production of large unilamellar vesicles by a rapid extrusion procedure: characterization of size distribution, trapped volume and ability to maintain a membrane potentialBiochimica Et Biophysica Acta812556510.1016/0005-2736(85)90521-8JeongJMcmahonAP2002Cholesterol modification of Hedgehog family proteinsThe Journal of Clinical Investigation11059159610.1172/JCI0216506KannenbergEBlumeAGeckelerKPorallaK1985Properties of hopanoids and phosphatidylcholines containing mega-Cyclohexane fatty-acid in Monolayer and Liposome experimentsBiochimica Et Biophysica Acta81417918510.1016/0005-2736(85)90434-1KannenbergEBlumeAMcelhaneyRNPorallaK1983Monolayer and calorimetric studies of phosphatidylcholines containing branched-chain fatty-acids and of their iteractions with cholesterol and with a bacterial hopanoid in model membranesBiochimica Et Biophysica Acta73311111610.1016/0005-2736(83)90096-2KnollAHSummonsREWaldbauerJRZumbergeJE2007The geological succession of primary producers in the oceansFalkowskiPKnollAHIn the evolution of primary producers in the seaBostonAcademic Press133164KulkarniGWuCHNewmanDK2013The general stress response factor EcfG regulates expression of the C-2 hopanoid methylase HpnP in Rhodopseudomonas palustris TIE-1Journal of Bacteriology1952490249810.1128/JB.00186-13KuwabaraPELabouesseM2002The sterol-sensing domain: multiple families, a unique role?Trends in Genetics1819320110.1016/S0168-9525(02)02640-9LinZFLiuNLinGZPengCL2011Factors altering the membrane fluidity of spinach thylakoid as determined by fluorescence polarizationActa Physiologiae Plantarum331019102410.1007/s11738-011-0737-5LipowskyRSackmannE1995Structure and dynamics of membranes I. from cells to vesicles / II. generic and specific interactionsElsevierLosDAMurataN2004Membrane fluidity and its roles in the perception of environmental signalsBiochimica Et Biophysica Acta166614215710.1016/j.bbamem.2004.08.002MalottRJSteen-KinnairdBRLeeTDSpeertDP2012Identification of hopanoid biosynthesis genes involved in Polymyxin resistance in Burkholderia multivoransAntimicrobial Agents and Chemotherapy5646447110.1128/AAC.00602-11MalottRJWuCHLeeTDHirdTJDalleskaNFZlosnikJENewmanDKSpeertDP2014Fosmidomycin decreases membrane hopanoids and potentiates the effects of colistin on Burkholderia multivorans clinical isolatesAntimicrobial Agents and Chemotherapy585211521910.1128/AAC.02705-14Martinez-SearaHRógTKarttunenMVattulainenIReigadaR2010Cholesterol induces specific spatial and orientational order in cholesterol/phospholipid membranesPLOS ONE5e1116210.1371/journal.pone.0011162MatsumotoKKusakaJNishiboriAHaraH2006Lipid domains in bacterial membranesMolecular Microbiology611110111710.1111/j.1365-2958.2006.05317.xMiaoLNielsenMThewaltJIpsenJHBloomMZuckermannMJMouritsenOG2002From lanosterol to cholesterol: structural evolution and differential effects on lipid bilayersBiophysical Journal821429144410.1016/S0006-3495(02)75497-0MoreinSHenricsonDRilforsL1994Separation of inner and outer-membrane vesicles from Escherichia coli in self-generating percoll gradientsAnalytical Biochemistry216475110.1006/abio.1994.1006MyersDSIvanovaPTMilneSBBrownHA2011Quantitative analysis of glycerophospholipids by LC-MS: acquisition, data handling, and interpretationBiochimica Et Biophysica Acta181174875710.1016/j.bbalip.2011.05.015NagumoASatoYSuzukiY1991Electron-spin-resonance studies of phosphatidylcholine interacted with cholesterol and with a hopanoid in liposomal membraneChemical & Pharmaceutical Bulletin393071307410.1248/cpb.39.3071NeidhardtFCCurtissR1996Escherichia coli and salmonella: cellular and molecular biologyWashington, D.CASM PressOurissonGAlbrechtPRohmerM1984The Microbial origin of fossil-fuelsScientific American251445110.1038/scientificamerican0884-44OurissonGRohmerM1992Hopanoids. 2. biohopanoids - a novel class of bacterial lipidsAccounts of Chemical Research2540340810.1021/ar00021a004OurissonGRohmerMPorallaK1987Prokaryotic hopanoids and other polyterpenoid sterol surrogatesAnnual Review of Microbiology4130133310.1146/annurev.mi.41.100187.001505PearsonA2013Lipidomics for geochemistryLondonElsevierPhillipsRUrsellTWigginsPSensP2009Emerging roles for lipids in shaping membrane-protein functionNature45937938510.1038/nature08147PogerDMarkAE2013The relative effect of sterols and hopanoids on lipid bilayers: when comparable is not identicalThe Journal of Physical chemistry B117161291614010.1021/jp409748dPorallaKKannenbergEBlumeA1980A glycolipid containing hopane isolated from the acidophilic, thermophilic Bacillus acidocaldarius, has a cholesterol-like function in membranesFEBS Letters11310711010.1016/0014-5793(80)80506-0RasmussenBFletcherIRBrocksJJKilburnMR2008Reassessing the first appearance of eukaryotes and cyanobacteriaNature4551101110410.1038/nature07381RennerLDWeibelDB2011Cardiolipin microdomains localize to negatively curved regions of Escherichia coli membranesProceedings of the National Academy of Sciences of USA1086264626910.1073/pnas.1015757108RicciJNColemanMLWelanderPVSessionsALSummonsRESpearJRNewmanDK2014Diverse capacity for 2-methylhopanoid production correlates with a specific ecological nicheThe ISME Journal867568410.1038/ismej.2013.191RicciJNMichelAJNewmanDK2015Phylogenetic analysis of HpnP reveals the origin of 2-methylhopanoid production in AlphaproteobacteriaGeobiology10.1111/gbi.12129RogTPasenkiewicz-GierulaMVattulainenIKarttunenM2007What happens if cholesterol is made smoother: Importance of methyl substituents in cholesterol ring structure on phosphatidylcholine-sterol interactionBiophysical Journal923346335710.1529/biophysj.106.095497RohmerMBouvierPOurissonG1979Molecular evolution of biomembranes: structural equivalents and phylogenetic precursors of sterolsProceedings of the National Academy of Sciences of USA7684785110.1073/pnas.76.2.847SáenzJP2010Hopanoid enrichment in a detergent resistant membrane fraction of Crocosphaera watsonii: Implications for bacterial lipid raft formationOrganic Geochemistry4185385610.1016/j.orggeochem.2010.05.005SáenzJPSezginESchwillePSimonsK2012Functional convergence of hopanoids and sterols in membrane orderingProceedings of the National Academy of Sciences of USA109142361424010.1073/pnas.1212141109SchmerkCLBernardsMAValvanoMA2011Hopanoid production is required for low-pH Tolerance, antimicrobial resistance, and Motility in Burkholderia cenocepaciaJournal of Bacteriology1936712672310.1128/JB.05979-11SilipoAVitielloGGullyDSturialeLChaintreuilCFardouxJGarganiDLeeHIKulkarniGBussetNMarchettiRPalmigianoAMollHEngelRLanzettaRPaduanoLParrilliMChangWSHolstONewmanDKGarozzoDD'erricoGGiraudEMolinaroA2014Hopanoid-Lipid A, a new component improving bacterial membrane resistanceNature Communications(in press)SongYLKenworthyAKSandersCR2014Cholesterol as a co-solvent and a ligand for membrane proteinsProtein Science2312210.1002/pro.2385SummonsREJahnkeLLHopeJMLoganGA19992-Methylhopanoids as biomarkers for cyanobacterial oxygenic photosynthesisNature40055455710.1038/23005SummonsREWalterMR1990Molecular fossils and microfossils of prokaryotes and protists from proterozoic sedimentsAmerican Journal of Science290A212244VanblitterswijkWJVandermeerBWHilkmannH1987Quantitative contributions of cholesterol and the individual classes of phospholipids and their degree of fatty acyl (un)saturation to membrane fluidity measured by fluorescence polarizationBiochemistry261746175610.1021/bi00380a038WelanderPVColemanMLSessionsALSummonsRENewmanDK2010Identification of a methylase required for 2-methylhopanoid production and implications for the interpretation of sedimentary hopanesProceedings of the National Academy of Sciences of USA1078537854210.1073/pnas.0912949107WelanderPVDoughtyDMWuCHMehaySSummonsRENewmanDK2012Identification and characterization of Rhodopseudomonas palustris TIE-1 hopanoid biosynthesis mutantsGeobiology1016317710.1111/j.1472-4669.2011.00314.xWelanderPVHunterRCZhangLCSessionsALSummonsRENewmanDK2009Hopanoids play a role in membrane integrity and pH homeostasis in Rhodopseudomonas palustris TIE-1Journal of Bacteriology1916145615610.1128/JB.00460-09WuCHKongLBialecka-FornalMParkSThompsonALKulkarniGConwaySJNewmanDK2015Quantitative hopanoid analysis enables robust pattern detection and comparison between laboratories.GeobiologyIn press10.7554/eLife.05663.017Decision letterClardyJonReviewing editorHarvard Medical School, United States
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Methylation at the C–2 position of hopanoids increases rigidity in native bacterial membranes” for consideration at eLife. Your article has been favorably evaluated by Richard Losick (Senior editor) and 4 reviewers, one of whom is a member of our Board of Reviewing Editors.
All the individuals responsible for the peer review of your submission have agreed to reveal their identity: Jon Clardy (Reviewing editor); Andrew Knoll (peer reviewer), Martin Burke (peer reviewer), and Alex Brown (peer reviewer).
The Reviewing editor and the other reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
The reviewers were uniformly enthusiastic about the suitability of a revised version of this study for eLife. The suggested areas for revision were:
1) The study makes a valuable contribution to the wide but sporadic distribution of 2Me-hopanes in sedimentary rocks. However, the jump from experimental results to geological explication is a bit abrupt, and the motivating issue of geological occurrence should be expanded. The discussion of nitrogen fixation by R. palustris, and possibly other bacteria, could also be expanded.
2) Lipid quantification forms an essential part of this study, and some parts need to be clarified. The citation in the Results section to an as yet unpublished manuscript (Neubauer et al., 2014) doesn't make clear what method was used, and a brief mention of ionization mode, chromatography conditions, and other relevant information should be added. In the same section, some comments on possible sources of error for low values are mentioned. In this regard, it is known that PC, PE, and PG have very different head groups and ionize very differently in MS; PG ionizes well in negative mode, PE ionizes well in both positive and negative, and PC ionizes better in positive mode. In addition, chain length and degree of unsaturation also affect ionization. Previous work in the area of lipid analysis using mass spectrometry should be included. In particular, the well documented effects on ion suppression should be mentioned (see Brugger et al., 1997, in PNAS). General lipidomic methodologies and reviews on phospholipids (H. A. Brown), neutral lipids (Robert Murphy), and sterols (David Russell) might be useful.
3) The manuscript clearly establishes that the introduction of a methyl group at C–2 of the hopane skeleton has a dramatic effect on membrane rigidity. The speculation that the structural basis of the effect could arise from the twisting of the A–ring from chair to boat is fundamentally sound, and as the manuscript notes, forms a promising area for future work. The authors might want to mention that, although 1,3-diaxial interactions of the C4 and C6 methyl groups in sterols (e.g. lanosterol) do not appear to distort the A–ring, in the context of the hopane scaffold (lacking an equatorial C3 hydroxyl) these interactions themselves could be sufficient to distort the A–ring into a twist boat and/or the methyl at C–2 would alter packing by changing interactions in the plane of the hopane structure.
10.7554/eLife.05663.018Author response
1) The study makes a valuable contribution to the wide but sporadic distribution of 2Me-hopanes in sedimentary rocks. However, the jump from experimental results to geological explication is a bit abrupt, and the motivating issue of geological occurrence should be expanded. The discussion of nitrogen fixation by R. palustris, and possibly other bacteria, could also be expanded.
We have significantly revised the Introduction and Discussion to make the linkages between the geological context for this work and the significance of our biophysical discoveries more clear.
2) Lipid quantification forms an essential part of this study, and some parts need to be clarified. The citation in the Results section to an as yet unpublished manuscript (Neubauer et al., 2014) doesn't make clear what method was used, and a brief mention of ionization mode, chromatography conditions, and other relevant information should be added. In the same section, some comments on possible sources of error for low values are mentioned. In this regard, it is known that PC, PE, and PG have very different head groups and ionize very differently in MS; PG ionizes well in negative mode, PE ionizes well in both positive and negative, and PC ionizes better in positive mode. In addition, chain length and degree of unsaturation also affect ionization. Previous work in the area of lipid analysis using mass spectrometry should be included. In particular, the well documented effects on ion suppression should be mentioned (seeBrugger et al., 1997, in PNAS). General lipidomic methodologies and reviews on phospholipids (H. A. Brown), neutral lipids (Robert Murphy), and sterols (David Russell) might be useful.
Detailed LC-MS conditions have been added into the Materials and methods section. We also rephrased our discussion of the possible cause for the low calculated quantities of phospholipids for clarity. Additional references have been added.
3) The manuscript clearly establishes that the introduction of a methyl group at C–2 of the hopane skeleton has a dramatic effect on membrane rigidity. The speculation that the structural basis of the effect could arise from the twisting of the A–ring from chair to boat is fundamentally sound, and as the manuscript notes, forms a promising area for future work. The authors might want to mention that, although 1,3-diaxial interactions of the C4 and C6 methyl groups in sterols (e.g. lanosterol) do not appear to distort the A–ring, in the context of the hopane scaffold (lacking an equatorial C3 hydroxyl) these interactions themselves could be sufficient to distort the A–ring into a twist boat and/or the methyl at C–2 would alter packing by changing interactions in the plane of the hopane structure.
We have modified the text according to the reviewers’ comments.
diff --git a/elife05789.xml b/elife05789.xml
new file mode 100644
index 0000000..1a761e4
--- /dev/null
+++ b/elife05789.xml
@@ -0,0 +1 @@
+elifeeLifeeLifeeLife2050-084XeLife Sciences Publications, Ltd0578910.7554/eLife.05789Research articleDevelopmental biology and stem cellsCellular interpretation of the long-range gradient of Four-jointed activity in the Drosophila wingHaleRosalind12BrittleAmy L12†FisherKatherine H12†MonkNicholas A M3StruttDavid12*Bateson Centre, University of Sheffield, Sheffield, United KingdomDepartment of Biomedical Science, University of Sheffield, Sheffield, United KingdomSchool of Mathematics and Statistics, University of Sheffield, Sheffield, United KingdomLawrencePeter AReviewing editorUniversity of Cambridge, United KingdomFor correspondence: d.strutt@sheffield.ac.uk
To understand how long-range patterning gradients are interpreted at the cellular level, we investigate how a gradient of expression of the Four-jointed kinase specifies planar polarised distributions of the cadherins Fat and Dachsous in the Drosophila wing. We use computational modelling to test different scenarios for how Four-jointed might act and test the model predictions by employing fluorescence recovery after photobleaching as an in vivo assay to measure the influence of Four-jointed on Fat-Dachsous binding. We demonstrate that in vivo, Four-jointed acts both on Fat to promote its binding to Dachsous and on Dachsous to inhibit its binding to Fat, with a bias towards a stronger effect on Fat. Overall, we show that opposing gradients of Fat and Dachsous phosphorylation are sufficient to explain the observed pattern of Fat–Dachsous binding and planar polarisation across the wing, and thus demonstrate the mechanism by which a long-range gradient is interpreted.
DOI:http://dx.doi.org/10.7554/eLife.05789.001
10.7554/eLife.05789.002eLife digest
Epithelial cells form sheets that line the body surfaces and internal cavities of animals—such as the skin and the lining of the gut. Certain structures on the surface of epithelial cell sheets—for example scales, hair, and feathers—are often all orientated in a particular direction. Epithelial cells with structures organised like this are described as being ‘planar polarised’.
Different proteins work together to set up planar polarity in a sheet of epithelial cells. Dachsous and Fat are two proteins that are found in the cell membranes of epithelial cells, including in the wings of the fruit fly Drosophila. These proteins bind to each other and link a cell to its neighbour. Dachsous and Fat accumulate on opposing sides of each cell: Fat accumulates on the side closest to the fly's body, and Dachsous builds up on the side closest to the wing tip. This pattern provides directional cues that help orientate surface structures, and the pattern is established, in part, by the activity of an enzyme called Four-jointed.
Four-jointed adds phosphate groups onto Dachsous and Fat. The activity of the Four-jointed enzyme forms a gradient along a developing wing: levels are low near the fly's body, and high at the wing tip. Previous experiments performed on cells grown in the laboratory showed that when Four-jointed adds phosphate groups to Fat and Dachsous, it prevents Dachsous from binding to Fat. However, it also makes Fat more able to bind to Dachsous. These opposing effects are thought to cause the proteins to accumulate on opposing sides of each cell. However, this has yet to be demonstrated in real tissue, not least because of the technical difficulty of measuring whether Fat-Dachsous binding has occurred in living organisms.
Here, Hale et al. overcome this challenge using a method called ‘fluorescence recovery after photobleaching’ (or FRAP) to measure Fat and Dachsous binding in the epithelial cells in the developing Drosophila wing. Combining these experimental results with a computational model confirmed the findings of previous laboratory studies: that Four-jointed makes it easier for Fat to bind to Dachsous, and makes it more difficult for Dachsous to bind to Fat. The opposing effects on the activity of Fat and Dachsous that result from the Four-jointed gradient in the developing wing are able to fully explain the observed patterns of Fat-Dachsous binding and of planar polarisation across the wing.
Overall, Hale et al. demonstrate how a gradient of protein activity that spans many cells is sensed and interpreted by individual cells to establish planar polarity. However, exactly how the phosphate groups added to Dachsous and Fat by Four-jointed modifies how they bind to each other remains a question for future work.
DOI:http://dx.doi.org/10.7554/eLife.05789.002
Author keywordsPCPplanar polaritygradientsFour-jointedResearch organismD. melanogasterhttp://dx.doi.org/10.13039/100004440Wellcome Trust100986StruttDavidhttp://dx.doi.org/10.13039/100004440Wellcome Trust089717BrittleAmy LStruttDavidThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementOpposing gradients of Fat and Dachsous phosphorylation are sufficient to explain the observed pattern of Fat-Dachsous planar polarisation across the Drosophila wing.Introduction
Planar polarity (also known as planar cell polarity [PCP]) refers to the coordinated polarisation of cells within the plane of a tissue such as an epithelium. How epithelia are planar polarised and how planar polarisation is co-ordinated across a tissue has intrigued researchers for decades. In the late 1950's, Locke presented evidence that orienting gradients could control epithelial tissue patterning (Locke, 1959), but despite years of research, the mechanisms by which graded cues might mediate the coordinated polarisation of individual cells remain incompletely understood (for review see Strutt, 2009).
Of the known molecular systems regulating planar polarity, only for the Drosophila Fat-Dachsous-Four-jointed (Ft-Ds-Fj) pathway is there strong evidence for a primary role of graded activity in providing orienting cues (Zeidler et al., 1999, 2000; Casal et al., 2002; Strutt and Strutt, 2002; Yang et al., 2002; Ma et al., 2003; Matakatsu and Blair, 2004; Simon, 2004; Ambegaonkar et al., 2012; Brittle et al., 2012). Ft and Ds are large atypical cadherins known to bind to each other heterophilically (Ma et al., 2003; Matakatsu and Blair, 2004), and Fj is a kinase shown to be active in the Golgi (Strutt et al., 2004; Ishikawa et al., 2008). Complementary graded expression patterns of Ds and Fj (Zeidler et al., 1999, 2000; Yang et al., 2002; Ma et al., 2003) result in the planar polarisation of Ft and Ds across cells, with (in the developing wing) Ds accumulating distally and Ft accumulating proximally (Ambegaonkar et al., 2012; Bosveld et al., 2012; Brittle et al., 2012). How the complementary expression patterns of Fj and Ds result in the accumulation of Ft and Ds on opposing cell edges is still under investigation, however, the data suggest that higher proximal expression of Ds and the opposing gradient of Fj expression leads to a gradient of Ft–Ds dimer formation (Casal et al., 2006; Lawrence et al., 2008; Strutt, 2009).
Waddington (1943) first reported a genetic interaction between fj and ds, and subsequent work revealed that Fj was able to regulate the localisation and function of both Ft and Ds (Casal et al., 2002; Strutt and Strutt, 2002; Yang et al., 2002; Ma et al., 2003; Cho and Irvine, 2004; Strutt et al., 2004; Casal et al., 2006). Fj is able to phosphorylate several cadherin repeats of both Ft and Ds (Ishikawa et al., 2008), raising the possibility that this is a mechanism by which Fj regulates Ft–Ds binding. This led to examination of the result of Fj phosphorylation on Ft and Ds binding in vitro (Brittle et al., 2010; Simon et al., 2010).
Using a cell aggregation assay, based on Drosophila S2 cells co-transfected with Fj, Ds, or Ft, Brittle et al. (2010) deduced that Fj was able to act on three previously identified serines in Ds to inhibit the binding of Ds to Ft (Ishikawa et al., 2008; Brittle et al., 2010). Using an alkaline phosphatase-based cell surface binding assay, Simon et al. (2010) similarly found that Fj inhibited the ability of Ds to bind to Ft and were also able to demonstrate an improvement in Ft binding to Ds. In addition, in vivo assays examining the propagation of polarity from over-expression clones also found that over-expressed Fj promoted Ft activity and inhibited Ds activity (Brittle et al., 2010).
So far there is no direct in vivo evidence that Fj normally does act on both Ft and Ds to modulate their binding during tissue patterning, or that any such activity is directly dependent on the mapped phosphorylation sites in the cadherin domains. Furthermore, it is unclear how the observed asymmetric subcellular distributions of Ft and Ds are related to the proposed differences in binding affinities between neighbouring cells and across the tissue.
To address these issues, we have carried out in vivo studies of Ft and Ds behaviour, making use of both normal protein forms and variants mutated at the mapped phosphorylation sites, and using protein mobility as measured using fluorescence recovery after photobleaching (FRAP) as a novel assay for in vivo binding activity. Furthermore, to better understand the possible effects of Fj phosphorylation on Ft–Ds binding and how this might lead to their asymmetric subcellular distributions, we have combined our experimental approach with computational modelling.
ResultsModelling Ft and Ds interactions and the emergence of cellular asymmetry
As an aid to understand the possible consequences for the generation of cellular asymmetry of Fj acting on either Ft or Ds or both, we generated a computational model reflecting a one-dimensional line of cells each with two compartments, to simulate Ft–Ds binding between cells according to a Fj gradient (see ‘Materials and methods’). In the model, Fj is allowed to phosphorylate either Ft or Ds in proportion to its concentration, and Ft and Ds are then permitted to freely bind at cell edges until they reach equilibrium (see Figure 1A). To produce a representative model, we measured the gradient of Fj expression across Drosophila larval wing discs (Figure 1C), using fj null clones to determine the appropriate background correction (Figure 1C′′). A gradient of 2.5–3.5% between adjacent cells (Figure 1B) was observed along the proximo-distal axis. We also allowed redistribution of Ft and Ds to occur in the model, in accordance with our experimental observations of protein mobility (see Figure 2C,D,L). A hierarchy of binding affinities was used, based on the in vitro experiments (Brittle et al., 2010; Simon et al., 2010), with phosphorylated Ft (FtP) and unphosphorylated Ds (Ds) producing the strongest partnership, and unphosphorylated Ft (Ft) and phosphorylated Ds (DsP) producing the weakest partnership (see ‘Materials and methods’).10.7554/eLife.05789.003
Computational modelling of the effect of a Fj gradient on Ft–Ds binding.
(A) Cartoon illustrating Ft–Ds binding and the predicted effect of a Fj gradient on the distribution of Ft–Ds heterodimers in a single cell. In the absence of any gradients (i.e., uniform Ft, Ds, and Fj expression across the tissue), Ft (blue) and Ds (purple) freely associate at junctions, with a certain proportion binding to form heterodimers (indicated by double black bars). We predict that by adding a Fj gradient (yellow bar) an asymmetric distribution of Ft–Ds binding across a cell will occur. If Fj inhibits Ds binding and promotes Ft binding more strongly to the right, Ft molecules in each cell prefer to bind to Ds in the next left-most cell (with which they have a stronger binding interaction), and so preferentially accumulate at left cell edges; similarly Ds molecules prefer to associate with Ft in the next right-most cell and in turn accumulate at right cell edges. Hence, the overall consequence of a Fj gradient and free movement of Ft and Ds within cells is the generation of cellular asymmetries of Ft and Ds distributions. (B) Graph illustrating the proximal-distal measurements of the Fj gradient across Drosophila third instar larval wing discs, showing a typical gradient of around 3% (dashed lines indicate 95% confidence intervals, error bars indicate standard deviation (SD), n = 5). Cell number increases from 0 to 30 moving from proximal to distal. (C) Confocal images of wing discs from animals with EGFP knocked-in to the endogenous ds locus, stained for Fj (red) and observed for native GFP fluorescence (green). A yellow line indicates where Fj gradient measurements were taken from proximal [P] to distal [D]. (C′) An example zoomed-in image showing the distribution of Fj and Ds in this region. (C″) fjd1 mutant clones (absence of blue ß-gal staining, cell junctions also labelled in blue) demonstrate that Fj is expressed throughout the tissue. Mutant clones were used to subtract background from the gradient measurements. (D and E) A computational model of Ft–Ds binding at cellular junctions. Each bar represents one cell and the scale indicates the amount of bound Ds or Ft in arbitrary units at junctions between cells (note scale is cut off at 20 units), also see Figure 1—figure supplement 1. Sloped top edges represent the difference between proximal and distal cell edges within one cell. A more graded slope within a cell indicates an increase in asymmetry. 23 cells are represented and the observed Fj gradient of 3% (B) has been used. Looking at both bound Ds (D) and bound Ft (E) shows Ds is preferentially localised distally and Ft proximally. If Fj acts on Ds only (D and E), weak asymmetry is seen within cells (3.8% increase) and a tissue gradient of higher to lower Ft–Ds binding is seen as Fj increases. If Fj acts on Ft only (D′ and E′) similar weak asymmetry (3.6% increase) is seen within cells, however, a tissue gradient of lower to higher Ft–Ds binding is observed as Fj increases. If Fj can act on both Ft and Ds (D″ and E″) stronger asymmetry is seen within cells (7.9% increase) and the tissue gradient is much reduced.
DOI:http://dx.doi.org/10.7554/eLife.05789.003
10.7554/eLife.05789.004
Representation of mathematical model.
(A) one-dimensional row of cells with proximal (P) and distal (D) membranes in cells 1 to i.
DOI:http://dx.doi.org/10.7554/eLife.05789.004
10.7554/eLife.05789.005
Ft and Ds show mutually dependent stable fractions at junctions.
Wing discs containing clones of (A) ds-EGFP/+ and (B) ft-EGFP/+ labelled for EGFP (green) and Ds or Ft, respectively (red), show that addition of an EGFP tag does not affect protein localisation. Quantification of (A′) Ds or (B′) Ft junctional protein confirms expression levels of tagged proteins are similar to wild-type. FRAP analysis of (C) Ds-EGFP and (D) Ft-EGFP puncta (bright regions, solid line) and non-puncta (dashed line) reveals a large stable fraction of protein in puncta (see Figure 2—figure supplement 1 for an example of a puncta and non-puncta region and Figure 2—figure supplement 2 for FRAP graphs with individual data points). Multiplying the intensity of FRAP regions prior to bleaching (C′ and D′) by the stable fraction plateau (C and D) reveals a final ‘stable amount’ of protein at junctions (C″ and D″) and a large significant difference between puncta and non-puncta regions (both p = 0.0001, error bars denote standard error (SEM) here and in all remaining figures). (E) Apical confocal sections of a V5-Ft antibody internalisation assay in pre-pupal wing, with time points of 0, 10, and 20 min, reveal persistent Ft puncta over time while quantification of total apical levels (right panel) shows protein internalisation is occurring (p = 0.01 at 10 min and p = 0.001 at 20 min) (F) Sub-apical confocal sections of pre-pupal wing show V5-Ft previously at the cell surface is seen in discrete intracellular puncta at t = 20, where it co-localises with early endosome marker Hrs (magenta) (F′), confirming antibody internalisation rather than antibody drop off. FRAP analysis of Ds-EGFP in ftG-rv clones (G) and Ft-EGFP in a dsUA071 background (H) resulted in more rapid and increased protein recovery. Dashed lines denote 95% confidence intervals. As recovery did not reach 100%, a small stable population could remain. Also see Figure 2—figure supplement 4 for FRAP graphs with individual data points. Re-bleaching experiments suggest that this is not an experimental artefact (see Figure 2—figure supplement 5). Live images of Ds-EGFP (I) and Ds-EGFP with a ftG-rv clone (I′; blue dots indicate first row of mutant cells) show a loss of puncta and diffuse distribution of Ds-EGFP when Ft is not present. Similar distribution of Ft-EGFP (J) is seen in a dsUA071 mutant (J′). (K) Comparison of Ds-EGFP and Ft-EGFP protein levels taken at the same confocal settings suggests there is almost twice as much Ft-EGFP (n = 4 wings) as Ds-EGFP (n = 2 wings) in puncta regions (p = 0.01). Intensity measurements were taken from manually selected puncta using 1 µm2 ROIs. Around 50 ROI measurements were taken per wing and average intensities were plotted and analysed using an unpaired t-test. (L) Bar chart representing time taken for 50% protein recovery to occur during FRAP experiments. Both Ft-EGFP and Ds-EGFP mobile fractions exhibit slower recovery in puncta compared to non-puncta and mutants.
DOI:http://dx.doi.org/10.7554/eLife.05789.005
10.7554/eLife.05789.006
Description of FRAP method.
A cartoon flowchart describing the steps involved in a FRAP experiment. Initially, puncta and non-puncta regions are selected and their intensity measured (pre-bleach intensity). The regions are bleached for a few seconds by increasing the laser power. Bleached regions are imaged over time and recovery of the unstable population is measured for around 5 min. Measurements are plotted and a one-phase exponential curve is fitted to the data. The plateau of this curve gives us the unstable fraction and to get the stable fraction we minus the unstable fraction from 1, for example, 1–0.78 = 0.22. This figure is multiplied by the pre-bleach intensity to give the stable amount. So if, for example, 78% unbleached protein has recovered, 22% of the bleached protein in the region of interest is then the stable fraction. We multiply this by the original pre-bleach intensity. So, a starting intensity of 500 units is multiplied by 0.22 to get a final stable amount of 110 units. However, if a mutant has a starting intensity of 250 and the same stable fraction of 22% we would say the stable amount of the mutant is 55. The mutant has therefore much less stable protein at junctions despite having a similar stable fraction. An example of puncta and non-puncta regions in ds-EGFP is also shown.
DOI:http://dx.doi.org/10.7554/eLife.05789.006
10.7554/eLife.05789.007
Individual data points for FRAP experiments.
FRAP graphs containing individual data points with SEM for each experiment. (A) ds-EGFP homozygous puncta and non-puncta. (B) ft-EGFP homozygous puncta and non-puncta.
DOI:http://dx.doi.org/10.7554/eLife.05789.007
10.7554/eLife.05789.008
FRAP on Ds-EGFP in the pre-pupal wing.
(A) FRAP on distal pre-pupal wings demonstrates a difference between puncta and non-puncta in the same wing region as the internalisation assay, confirming the presence of stable and unstable protein populations. (A′) Intensity of puncta and non-puncta regions in the pre-pupal wing. (A″) Stable amount of puncta and non-puncta in the pre-pupal wing confirms a large difference in stability between regions.
DOI:http://dx.doi.org/10.7554/eLife.05789.008
10.7554/eLife.05789.009
Individual data points for FRAP experiments.
FRAP graphs containing individual data points with SEM for each experiment. (A) ds-EGFP ftG-rv clones (no puncta observed). (B) ft-EGFP dsUA071 (no puncta observed).
DOI:http://dx.doi.org/10.7554/eLife.05789.009
10.7554/eLife.05789.010
Control re-bleach FRAP experiments.
(A) Re-bleach experiment performed on ft-EGFP dsUA071. The re-bleach experiment confirms there are no obvious bleaching artefacts in our FRAP experiments. A FRAP region is selected and bleached as normal (see ‘Materials and methods’), recovery is allowed for 2 min prior to a second bleaching event which is performed in a smaller region within the first bleached region (A′). Recovery is measured and reaches around 100%. If there were any bleaching artefacts (such as photodamage-induced stabilisation of a protein population), recovery after the second bleach would not reach 100%.
DOI:http://dx.doi.org/10.7554/eLife.05789.010
Using these starting parameters, our modelling predicts that even if Fj acts on only one of either Ft or Ds, then a weak asymmetry of both Ft and Ds is produced across cells (Figure 1D,D′,E,E′). However, in both cases there is also a shallow gradient of bound protein across the tissue. If, however, Fj acts on both Ft and Ds, there is improved cellular asymmetry and a negligible tissue gradient (Figure 1D″,E″). Notably, neither situation predicts the approximate two-fold asymmetry in Ds distribution across the cell axis that is experimentally observed (Brittle et al., 2012, see ‘Discussion’).
Thus, in our model, Fj acting on only Ft, or only Ds, or both results in asymmetric subcellular distributions of Ft and Ds. Therefore, in order to resolve the role of Fj in Ft–Ds asymmetry generation, we sought to understand its effects on Ft–Ds binding in vivo.
Ft and Ds exhibit stable populations at the cell junctions
As there is no method for directly measuring the strength of binding of two proteins in the in vivo context of a living tissue, we instead employed FRAP to assay the mobility of Ft and Ds in the developing Drosophila wing. During FRAP, flies with an EGFP-tagged protein (e.g., Ds-EGFP) are used, and a fluorescent region of interest is selected and bleached. The same region is imaged over time and recovery of fluorescent protein is measured. Recovery is due to movement of unbleached protein into the bleached area from elsewhere in the cell and any protein that has been bleached remains bleached and is not reactivated. Therefore, if after bleaching and imaging, over time we see, for example, fluorescence recovering to 30% of the starting intensity we assume that 70% of the bleached protein remains and is therefore stably bound at the junction. We perform all of our experiments using the same microscope settings (unless otherwise stated) and since protein levels (and thus fluorescent intensity) can vary between genotypes, we must take this into account during our analysis. To do this, we multiply the starting intensity by the stable fraction to get a final stable amount (see Figure 2—figure supplement 1). We make the assumption that a decrease in the speed or size of the mobile population provides a readout of increased binding interactions. For instance, if the size of the mobile population (unstable fraction) increases in a particular genotype and, once the intensity is taken into account, a smaller stable amount of overall protein is calculated, we assume that binding affinity has decreased. Additionally, if the speed of protein recovery after bleaching increases in a particular mutant, we infer that binding is not as strong as ‘wild-type’ and the protein can therefore relocate more quickly. FRAP was performed as previously described (Strutt et al., 2011 see also ‘Materials and methods’ and Figure 2—figure supplement 1: individual data points and confidence intervals are also provided in the relevant figure supplements and Supplementary file 1) in the same proximal region of the wing disc (unless otherwise stated) using a strain of flies expressing the Ds protein endogenously tagged with EGFP (Brittle et al., 2012), and a newly generated strain in which EGFP was inserted at the C-terminus of the endogenous Ft coding region (see ‘Materials and methods’). Both transgenes were found to be expressed and localised similarly to wild-type (Figure 2A,B) and insertion of the tag did not affect wing size or shape suggesting protein function is normal.
Live imaging of both Ft-EGFP and Ds-EGFP (Figure 2I,J) revealed that the junctional populations exhibit a punctate distribution, as previously seen by immunolabelling in fixed tissue (Ma et al., 2003; Brittle et al., 2012) and also for components of the ‘core’ planar polarity pathway (Strutt et al., 2011). In our FRAP experiments, puncta and non-puncta regions were bleached and fluorescence recovery was measured over time (Figure 2C,D, Figure 2—figure supplement 2). The Ft-EGFP signal was almost twice the level of Ds-EGFP (Figure 2K), therefore laser power had to be adjusted accordingly between experiments meaning the final stable amounts for Ds-EGFP cannot be compared directly to those for Ft-EGFP. For both puncta and non-puncta, there was some recovery of fluorescence demonstrating that the proteins were mobile at junctions. Recovery was not complete indicating that there was also a population of stable Ds-EGFP and Ft-EGFP that did not recover during the time period used. Measuring protein recovery revealed a significant difference in stability between puncta and non-puncta regions (Figure 2C,D and Table 1) with puncta showing an increased amount of stable protein (Figure 2C″,D″, both p = 0.0001) when taking pre-bleach intensity levels into account (Figure 2C′,D′). Overall, the FRAP assays indicated that there were both stable and unstable populations of Ft-EGFP and Ds-EGFP present at junctions, with stable material concentrated into bright punctate regions.10.7554/eLife.05789.011
Comparison of rate and stabilisation data
DOI:http://dx.doi.org/10.7554/eLife.05789.011
Genotype
Stable amount
SEM of stable amount
Half time
Confidence interval
Ds-EGFP Homozygous Puncta
292.9
±18.5
38.5
30.8 to 51.3
Ds-EGFP Homozygous Non-Puncta
72.1
±21.8
27.9
22.13 to 37.6
Ds-EGFP ftG-rv Homozygous
n/a
n/a
8.3
5.8 to 14.5
Ds-EGFP fjd1 Homozygous Puncta
167.2
±21.6
33.89
25.5 to 50.5
Ds-EGFP fjd1 Homozygous Non-Puncta
26.4
±20.5
5.6
3.8 to 10.3
Ft-EGFP Homozygous Puncta
148.0
±9.5
24.2
19.28 to 32.3
Ft-EGFP Homozygous Non-Puncta
52.0
±10.6
9.5
7.2 to 13.9
Ft-EGFP dsUA071 Homozygous
n/a
n/a
8.5
5.7 to 16.6
Ft-EGFP fjd1 Homozygous Puncta
114.1
±1.4
16.8
13.3 to 22.8
Ft-EGFP fjd1 Homozygous Non-Puncta
52.6
±19.0
11.4
8.7 to 16.6
Ds-EGFP Homozygous Puncta Distal (High Fj)
378.0
±17.6
93.0
59.1 to 217.6
Ds-EGFP fjd1 Homozygous Puncta Distal
142.7
±28
69.7
51.3 to 108.6
As an independent assay to confirm the presence of stable and unstable protein populations, we performed an antibody internalisation assay in the pre-pupal wing (see ‘Materials and methods’ and Strutt et al., 2011) to assess the endocytic turnover of Ft, using a Ft transgene tagged extracellularly with a V5 epitope (Feng and Irvine, 2009). The internalisation assay is performed in pre-pupal wings as the peripodial membrane prevents the assay from being effective in the wing disc. Live pre-pupal wings were dissected in Schneider's medium and incubated with antibody against V5 at 0°C followed by washing and chasing at room temperature before fixation. The amount of V5-Ft at the cell surface was determined via incubation with secondary antibody in the absence of detergent. When endocytosis was allowed by moving the tissue to room temperature the total amount of apical cell surface protein decreased over time suggesting protein internalisation was occurring and indicating the presence of an unstable cell surface population of V5-Ft (Figure 2E). To confirm that protein loss was not due to antibody drop off, V5-Ft previously found at the surface was traced to internal vesicles (Figure 2F) and found to co-localise with the early endosome marker Hrs (Figure 2F′). Populations of V5-Ft appeared to be resistant to endocytic turnover with persistent puncta of cell surface V5-Ft observed after 20 min (Figure 2E). This was consistent with the puncta stability observed in FRAP experiments both in wing discs (Figure 2D″) and also in the pre-pupal wing (Figure 2—figure supplement 3). Note however, that although we assume that these remaining populations in each assay are the same, we cannot be certain that this is the case. During FRAP, we are unable to observe the fate of the bleached protein. Furthermore, in the internalisation assay, protein might be internalised and return to the same sites via exocytosis.
Ft and Ds require each other for junctional stability
To understand whether the observed protein stability at junctions was due to the presence of heterophilic binding between Ft and Ds, FRAP was performed on both Ft-EGFP and Ds-EGFP in ds and ft null mutant backgrounds, respectively (Figure 2G–J, Figure 2—figure supplement 4). Removal of the putative binding partner resulted in the loss of obvious puncta and the almost complete recovery of fluorescence after bleaching, indicating that the majority of protein had become unstable at junctions. Thus, the observed stability of each protein is dependent on the presence of the binding partner.
Comparing data for the rate of recovery of fluorescence of Ds-EGFP (Table 1 and Figure 2L) further revealed that the mobile populations of Ds-EGFP show slower movement, either in puncta or non-puncta, in the presence of Ft, compared to that in the absence of Ft (where no puncta are observed). This suggests that in addition to the immobile population of Ds bound to Ft which is concentrated in puncta as described in the previous section, there is also a mobile population of Ds-EGFP present in both puncta and non-puncta regions which is bound to Ft and shows less rapid movement than free Ds-EGFP (i.e., Ds-EGFP in the absence of Ft). The reduction in the rate of movement of these mobile Ds-Ft heterodimers in puncta as opposed to non-puncta might indicate the presence of a cis-dimerisation mechanism that promotes the clustering of heterodimers into the puncta. Alternatively, mechanisms such as physical interactions with the cytoskeleton similar to those seen for E-Cadherin clustering (Cavey et al., 2008) may be responsible for the reduced mobility.
Our data on the rate of recovery of Ft-EGFP (Table 1 and Figure 2L) also support these conclusions. However, in this case there is a negligible difference in the rate of movement of Ft-EGFP in non-punctate regions in the presence of Ds, as compared to the rate of movement of Ft-EGFP in the absence of Ds, and also a faster rate of Ft-EGFP mobility within puncta as compared to Ds-EGFP in puncta. We surmise that these differences are due to there being an excess of Ft-EGFP present over Ds (see Figure 2K), resulting in an increased proportion of the population of Ft-EGFP being unbound and thereby free to move into bleached regions.
Overall, we conclude from our FRAP experiments using ‘wild-type’ Ft-EGFP and Ds-EGFP, that Ft and Ds bind to each other across cell membranes resulting in the production of immobile fractions of Ft and Ds at cell junctions, and a reduction in mobility of Ft and Ds at junctions.
The in vivo effect of Fj on Ft/Ds stability
Having thus demonstrated that Ft–Ds binding can be measured in vivo by virtue of the effects that binding has on protein mobility and stability, we next sought to understand the effects of Fj on Ft–Ds binding. Fj is able to phosphorylate both Ft and Ds (Ishikawa et al., 2008) and modify the binding affinities between the two proteins in vitro (Brittle et al., 2010; Simon et al., 2010); however, the functional in vivo result of these modifications is unknown. To investigate this, we performed FRAP on endogenously expressed Ft-EGFP and Ds-EGFP in a fj mutant background, measuring fluorescence recovery to infer the stability of Ft–Ds dimers (Figure 3, Figure 3—figure supplement 1). As Ft–Ds binding across junctions is mutually dependent, we infer that if one protein gains or loses stability, so will the other. Moreover, as Fj is thought to act on Ft and Ds in opposite ways, if we remove Fj, we might not expect to see any difference in overall stability of Ft–Ds dimers as the positive and negative effects could cancel each other out. If a difference is observed, it suggests that phosphorylating one protein has a stronger effect than phosphorylating the other.10.7554/eLife.05789.012
Ft and Ds stability upon loss of Fj.
Stable fraction of (A) Ds-EGFP and (B) Ft-EGFP in a fjd1 background (yellow lines). Solid lines represent puncta and dashed lines represent non-puncta. ‘Wild-type’ values are plotted for comparison. Intensity of Ds-EGFP decreases (A′) and Ft-EGFP increases (B′), in a fjd1 background, whereas stable amount of protein in puncta falls in both (A″ (p = 0.004) and B″ (p = 0.04)). The reduction in Ds-EGFP levels at junctions in the absence of Fj is consistent with less Ds being bound to Ft in this situation and excess unbound Ds being removed from junctions. The reason for an increase in Ft levels is unknown, as presumably less Ft is bound to Ds, but suggests that unbound Ft is not removed from junctions. See Figure 3—figure supplement 1 for FRAP curves with individual data points. (C) Live images of wing discs taken during a FRAP experiment show the recovery of puncta over time in ds-EGFP and ds-EGFP in a fjd1 background. Protein recovery after 70(sec) is increased in a fjd1 background and overall protein levels appear reduced.
DOI:http://dx.doi.org/10.7554/eLife.05789.012
10.7554/eLife.05789.013
Individual data points for FRAP experiments.
FRAP graphs containing individual data points with SEM for each experiment. (A) ds-EGFP fjd1 puncta and non-puncta. (B) ft-EGFP fjd1 puncta and non-puncta.
DOI:http://dx.doi.org/10.7554/eLife.05789.013
The co-expression of Fj with Ds in a cell culture assay has been shown to reduce the ability of Ds to bind to Ft (Brittle et al., 2010). Therefore, we hypothesise that removing Fj might increase the stability of any Ft–Ds dimer (corresponding to a decrease in the height of the FRAP recovery plateau). However, we found that Ds-EGFP became less stable upon removal of Fj (Figure 3A″, p = 0.004), with a reduction in both the stable fraction (i.e., an increase in the height of the plateau) and the stable amount at junctions (Figure 3A,A″). Stills of live FRAP experiments demonstrate the increased recovery of Ds-EGFP in a fj mutant background after 70 s (Figure 3C). Co-expression of Fj with Ft has been shown to increase the affinity of Ft for Ds (Simon et al., 2010); therefore, we should expect a loss of stability of Ft-EGFP in a fj background. In this case analysis of Ft-EGFP recovery in a fj background did result in a reduction in stable fraction and stable amount at junctions (Figure 3B,B″, p = 0.04). The half time rate of recovery of Ft-EGFP and Ds-EGFP was also decreased in the absence of Fj, consistent with a reduction of binding affinity allowing more rapid mobility (Table 1).
Thus, the overall result of removing Fj was a reduction in stability of the Ft–Ds dimer. A loss of stability, rather than a gain, suggests that, at least in the wing disc, the effect Fj has on Ft is dominant to any which it might have on Ds.
Fj acts on both Ft and Ds in vivo altering Ft/Ds dimer stability
Using endogenously tagged Ft and Ds proteins, we have uncovered a dominant effect of Fj on Ft; however, the overall result may mask any potential consequence of Fj action on Ds. As the in vitro evidence suggests that Fj modifies Ft–Ds binding via phosphorylation of their extracellular cadherin repeats (Ishikawa et al., 2008; Brittle et al., 2010; Simon et al., 2010), we turned to mutations in the mapped phosphorylation sites to separate effects on each molecule. We have previously described fly strains expressing EGFP-tagged Ds phosphorylation mutants and mimetics uniformly under control of the Actin5C promoter (Brittle et al., 2010). Under these expression conditions, protein levels are higher than endogenous and there is a loss of visible puncta meaning that FRAP experiments are performed solely in junctions and puncta and non-puncta are not distinguished. Microscope settings were also altered to take the increased intensity levels into account. All of the phosphomutant experiments were performed in relevant mutant backgrounds so the Actin5C construct was the only form of expression of the protein in question. For example, if we were looking at Act-Ds-EGFP, we performed the experiment in a ds background.
If Fj does have an independent effect on Ds, we might expect that preventing Fj from phosphorylating Ds (Ds phosphomutant) whilst not affecting Ft phosphorylation (by leaving Fj cellular activity intact) should allow us to see this. Based on previous experiments where phosphorylating Ds reduces the binding affinity between Ft and Ds (Brittle et al., 2010), we are able to hypothesise that mutating the phosphorylation sites in Ds, thus preventing Fj phosphorylation, should improve binding affinity and therefore increase the stable amount of bound protein. We also hypothesise that mutating the phosphorylation sites of Ft (so Ft cannot be phosphorylated by Fj) will result in a reduction in binding affinity and therefore a reduced stable amount.
When comparing the stability of ‘wild-type’ Ds-EGFP (expressed from the Act-ds-EGFP transgene) and a form of Ds with the phosphorylation sites mutated to alanine to block phosphorylation (Act-ds-S>Ax3-EGFP; Figure 4A, Figure 4—figure supplement 1), there was a strong increase in the amount of stable protein present in the mutant (Figure 4A″, 196.3 ± 18 and 428.9 ± 35 intensity units respectively, p = 0.0004). As Fj was still expressed and presumably able to act normally on Ft in this experiment, the increase in stable amount must be solely due to the inability of Fj to phosphorylate Ds. Additional loss of Fj, to remove potential Ft phosphorylation from the experiment, resulted in the expected decrease in stability, back to a level similar to ‘wild-type’ (Figure 4B,E, Figure 4—figure supplement 1), again consistent with the effect of Fj on Ft being dominant to that on Ds. A bar chart representing all of the different stable amounts of Act-DsEGFP is provided in Figure 4E.10.7554/eLife.05789.014
Effects of mutating the phosphorylation sites in Ds and Ft on in vivo stability.
(A) Comparison of Act-ds-EGFP with Act-ds-EGFP phosphomutant (Act-ds-S>Ax3-EGFP). FRAP is performed on junctions as no puncta are visible. Ds-S>Ax3-EGFP has a larger remaining stable fraction after bleaching (A), is more highly expressed at junctions (A′) and has a significantly larger stable amount at junctions (A″ p = 0.0004) than ‘wild-type’. (B) When Fj is removed from the phosphomutant background (fj−Act-ds-S>Ax3-EGFP) intensity levels drop (B′) and the stable amount at junctions returns to below ‘wild-type’ levels (B″ and E). (C) Ft-EGFP phospho-mutant (Act-ft-S/T>Ax5-EGFP) has a smaller stable fraction than Ft-EGFP (Act-ft-EGFP), shows reduced levels at junctions (C′), and has a significantly reduced stable amount at junctions (C″, p = 0.006). (D, D′, D″) In a fjd1 mutant background, the phosphomutant protein (fj-Act-ft-S/T>Ax5-EGFP) does not become any less stable suggesting any loss of stability in (C) is caused primarily by a loss of Ft phosphorylation. See Figure 4—figure supplement 1 for FRAP graphs with individual data points. (E–F) An overview of stable amounts in (E) Act-ds-EGFP and (F) Act-ft-EGFP shows that despite there being a significant increase in stable Ds when its phosphorylation sites are mutated, the removal of Fj results in a loss of stable Ds (E; p = 0.02) and also Ft (F; p = 0.02). Thus the phosphorylation state of Ds only appears to significantly affect Ft–Ds binding if Ft is already phosphorylated by Fj.
DOI:http://dx.doi.org/10.7554/eLife.05789.014
10.7554/eLife.05789.015
Individual data points for FRAP experiments.
FRAP graphs containing individual data points with SEM for each experiment. (A) Act-ds-EGFP and Act-ds-S>Ax3-EGFP (no puncta available). (B) Act-ds-S>Ax3-EGFP and fj-Act-ds-EGFPS>Ax3-EGFP (no puncta observed). (C) Act-ft-EGFP and Act-ft-S/T>Ax5-EGFP (no puncta observed). (D) Act-ft-S/T>Ax5-EGFP and fj-Act-ft-S/T>Ax5-EGFP (no puncta observed). (E) Comparison of Act-ds-EGFP and Act-ds-EGFP phosphomimetic (Act-ds-S>Dx3-EGFP). The expected result upon expression of a ds phosphomimetic might be a reduction in stable fraction and stable amount. However, the Act-ds-S>Dx3-EGFP stable fraction is the same as Act-ds-EGFP (E), intensity levels are similar (E′) and stable amounts are not significantly different (E″) suggesting that mutating the phosphorylation sites has not been successful or does not have any additional effect (see text). (F) Comparison of Act-ft-EGFP and Act-ft-EGFP phosphomimetic (Act-ft-S/T>Dx4-EGFP). Mimicking the phosphorylation sites in Act-ft-EGFP might be expected to increase both the stable fraction and stable amount of protein when compared to ‘wild-type’ Act-ft-EGFP. However, the stable fraction remains the same (F), and although intensity levels of the phosphomimetic are higher (F″), the stable amount is not significantly increased (F″).
DOI:http://dx.doi.org/10.7554/eLife.05789.015
We were unable to detect any decrease in stability using our previously described Ds phosphomimetic (Act-ds-S>Dx3-EGFP) fly strain (Figure 4—figure supplement 1). We suspect that this is due to the mutation of serines to aspartates having only a modest phosphomimetic effect caused by the lower negative charge of aspartic acid compared to a phosphate group. The effect of mimicking the phosphorylation of Ds is therefore no greater than the normal effect of Fj phosphorylation on Ds in the region of the wing being assayed. It is also likely that our FRAP assay is unable to detect subtle differences in binding affinity.
We also constructed Ft phosphomutants and mimics (see ‘Materials and methods’) and tested them in a similar manner. Consistent with previous results, a Ft phosphomutant (Act-ft-S/T>Ax5-EGFP) showed a decreased stable fraction (Figure 4C, Figure 4—figure supplement 1) and a large reduction in stable amount when compared to ‘wild-type’ (Figure 4C″, 97.9 ± 15 and 194.5 ± 24 intensity units respectively, p = 0.006). Additionally, a Ft phosphomimetic (Act-ft-S/T>Dx4-EGFP) showed slightly improved stability compared to ‘wild-type’ (Figure 4—figure supplement 1), however, this increase was not statistically significant. Removal of Fj, and therefore Ds phosphorylation, from the phosphomutant background (Figure 4D, Figure 4—figure supplement 1) did not result in a significant change in stable amount at the junctions confirming that the observed reduction in stability of the phosphomutant (Figure 4C″) was primarily due to a loss of Fj phosphorylation of Ft. A bar chart showing the stable amounts of Act-ft-EGFP in each condition is provided in Figure 4F.
Importantly, the phosphomutant experiments have revealed in vivo affects of Fj phosphorylation on both Ft and Ds. Mutating the Ds phosphorylation sites resulted in a significantly improved binding ability, however, this improved ability was lost when Ft was also not phosphorylated. Additionally, the phosphomutant experiments have confirmed that there is a dominant effect of Fj phosphorylation on Ft.
Modelling Ft–Ds interactions with a Ft phosphorylation bias
Previously, our model of Ft–Ds interactions produced consistent cellular asymmetry with a negligible tissue gradient when Fj acted upon both Ft and Ds (Figure 1D,E). However, our data have shown that, although Fj can act on both Ft and Ds, the phosphorylation of Ft to increase Ft–Ds binding is dominant. Furthermore, in the absence of Ft phosphorylation (in a Ft phosphomutant), the phosphorylation of Ds (‘wildtype’ vs fj-) has no significant effect on the overall amount of bound protein. In order to understand how this might affect the establishment of cellular asymmetry and binding across a tissue, we modified our model parameters as follows: first, we reduced the degree by which phosphorylated Ds inhibits Ft–Ds binding in complexes where Ft is phosphorylated (i.e., complexes A and B); second, we made the binding affinities for complexes formed from unphosphorylated Ft (i.e., complexes C and D) the same (see ‘Materials and methods’). We then ran the model with Fj acting on both Ds and Ft (Figure 5A). We consequently still see the establishment of cellular asymmetries in Ft–Ds distribution within each cell, but also see a tissue gradient of Ft–Ds binding that follows the Fj gradient. The model therefore predicts that as you move from a region of low Fj to high Fj, Ft–Ds dimer stability and thus the stable amount at cell junctions should increase.10.7554/eLife.05789.016
Ft–Ds binding varies across the wing in response to the Fj expression gradient.
(A) Output of the revised computational model in which Fj acts more strongly on Ft than Ds, and in which unphosphorylated Ft binds with equal affinity to both phosphorylated and unphosphorylated Ds. When Fj can act on both Ft and Ds with a Ft bias, an intermediate outcome between acting on Ft only and acting on Ft and Ds without a bias is seen and a cellular asymmetry of 4.1% is observed. A tissue gradient of 4.7% in the model when acting on both Ft and Ds, compared to 10% when Fj acts on Ft alone (Figure 1D′,E′) suggests that Ds phosphorylation is required to counter the strong graded effects of Ft phosphorylation. (B) FRAP analysis of endogenously expressed Ds-EGFP reveals a larger stable fraction in regions of high Fj (distal) despite showing slightly less overall fluorescence (B′). Overall, a significantly increased stable amount is seen in high Fj regions (B″ p = 0.01, 23% difference). (C) FRAP analysis in a fj mutant shows that this difference is lost when taking pre-bleach intensity levels into account (C′, C″). See Figure 5—figure supplement 1 for FRAP graphs with individual data points. (D) Images taken during a FRAP experiment in a region of high Fj demonstrate the low level of recovery over time when compared to Ds-EGFP recovery in a region of low Fj (as seen in Figure 3C). (E) FRAP analysis of junctionally localised Ds-EGFP expressed from the Act-ds-S>Ax3-EGFP transgene show overall reduced stable fractions, however, increased distal intensity levels of Ds-EGFP (E′) mean the overall stable amount is significantly increased in regions of high Fj (E″, p = 0.01, 53% difference). In a fjd1 mutant (F) any observed difference in stable amount across the tissue is lost (F″) when taking pre-bleach intensity levels into account (F′) (proximal equates to low Fj, distal equates to high Fj). This is again consistent with Fj acting primarily on Ft. See Figure 5—figure supplement 1 for FRAP curves with individual data points.
DOI:http://dx.doi.org/10.7554/eLife.05789.016
10.7554/eLife.05789.017
Individual data points for FRAP experiments.
FRAP graphs containing individual data points with SEM for each experiment. (A) ds-EGFP homozygous puncta at low (proximal) and high (distal) Fj regions. (B) ds-EGFP homozygous puncta at proximal and distal regions in a fj mutant. (C) Act-ds-S>Ax3-EGFP puncta at low (proximal) and high (distal) Fj regions. (D) Act-ds-S>Ax3-EGFP puncta at proximal and distal Fj regions in a fj mutant.
DOI:http://dx.doi.org/10.7554/eLife.05789.017
Ft–Ds binding varies across the Fj gradient
We next endeavoured to test in vivo the prediction of a gradient of Ft–Ds binding across the tissue. We have already shown that Fj protein levels are high in the distal wing pouch and lower proximally (Strutt et al., 2004; Figure 1B). We therefore would expect there to be an increase in stability of Ft and Ds as we move from regions of low Fj (proximal) to regions of high Fj (distal). To test this, we used FRAP analysis on endogenously expressed Ds-EGFP (Figure 5B, Figure 5—figure supplement 1) and were able to detect a statistically significant increase in stable junctional amount in more distal regions (Figure 5B″, 292.9 ± 19 proximally and 378 ± 18 distally, p = 0.01, 23% difference [see also Supplementary file 1]). This difference in stable amount was lost when Fj was removed (Figure 5C″ 167.2 ± 22 proximally and 142.7 ± 28.6 distally, NS [see also Supplementary file 1]), implying that the Fj gradient normally produced the difference across the tissue.
We repeated this experiment using the Act-ds-S>Ax3-EGFP phosphomutant, to assess the result of Fj only acting on Ft. Despite relatively small differences between stable fractions (Figure 5E, Figure 5—figure supplement 1), we saw a greater difference in stable Ds-S>Ax3-EGFP at junctions between high and low Fj regions (Figure 5E″, 109.2 ± 21 proximally and 230.4 ± 37 distally p = 0.01, 53% difference) which was again lost upon removal of Fj (Figure 5F″, 97.8 ± 9 proximally and 78.2 ± 25 distally, NS). These results directly confirm the predictions of the model, showing not only that a slight gradient of binding strength exists across the tissue, but also that an opposing gradient of Ds phosphorylation usually acts to counter the effects of a gradient of Ft phosphorylation, resulting in a relatively even distribution of bound Ft–Ds complexes at junctions across the tissue axis from low to high Fj.
Discussion
A long-standing problem in developmental biology is understanding how long-range patterning gradients are interpreted at the cellular level. More specifically, with regard to understanding how cell polarity is coordinated across sheets of cells, a major goal is to determine the mechanism by which a gradient of transcription across a tissue (produced for instance in response to a morphogen) can be sensed by individual cells to result in each cell adopting a uniform polarisation. A particular challenge for such a sensing mechanism is that the difference in levels of transcription between adjacent cells may be very small (e.g., only a few percent or less of the peak expression levels), and at the high end of the gradient this difference needs to be read against the background of a high overall expression level.
The Ft-Ds-Fj pathway in Drosophila represents an excellent system for addressing this problem. The Ft and Ds cadherins bind heterophilically between adjacent cells, and the visible readout of polarity is the asymmetric distribution of these dimers across the cell axis, such that approximately two-fold higher Ds is found on one cell edge, bound to Ft on the apposing edge of the neighbouring cell (Ambegaonkar et al., 2012; Bosveld et al., 2012; Brittle et al., 2012). However, the measured Ft asymmetry across the cell axis is weaker than that of Ds, most likely because there is a larger population of Ft present at junctions, including a presumably significant unbound fraction which is symmetrically distributed.
Importantly, the asymmetric distribution of Ft–Ds dimers is a result of the patterns of transcription of the ds and fj genes as specified by upstream morphogens (Ambegaonkar et al., 2012; Bosveld et al., 2012; Brittle et al., 2012). In this study, we focus specifically on the mechanisms determining Ft–Ds subcellular polarity in the Drosophila third instar wing imaginal disc. At this stage, Fj is expressed in a gradient, high at the putative distal end of the wing (i.e., the centre of the wing pouch) and low proximally towards the wing hinge (Strutt et al., 2004), whereas Ds is relatively uniformly expressed in the visible region of the wing pouch, but higher in hinge regions of the wing (Strutt and Strutt, 2002). It is important to point out that much of the proximal wing blade is folded at the larval stage and most likely also has high levels of Ds expression and this region of high Ds expression could therefore also be promoting Ft–Ds asymmetry. Although we believe that both the Fj and Ds expression patterns are important cues for specifying Ft–Ds asymmetry in the wing pouch, the evidence suggests that boundaries of Ds expression may only be able to act as a patterning cue over a few cell diameters (Ambegaonkar et al., 2012; Brittle et al., 2012), and therefore throughout much of the wing pouch the Fj gradient is likely to be a dominant cue.
In in vitro assays, Fj is able to mediate the phosphorylation of extracellular cadherin repeats of Ft and Ds, and phosphorylation of Ft has been shown to promote its binding to Ds, whereas phosphorylation of Ds appears to inhibit its binding to Ft (Ishikawa et al., 2008; Brittle et al., 2010; Simon et al., 2010). Thus, the gradient of Fj in the wing pouch is predicted to produce opposing gradients of Ft–Ds binding affinities, with Ft–Ds binding favoured between cells moving down a Fj gradient, and Ds-Ft binding favoured between cells moving up a Fj gradient (Figure 6A). It has been previously proposed that these opposing binding gradients might play a role in producing uniform Ft–Ds interactions across the tissue (Simon et al., 2010).10.7554/eLife.05789.018
Models of Ft–Ds interactions.
(A) Cartoon illustrating our model of Ft–Ds binding across a tissue according to a Fj gradient. Ft-P (dark blue) and Ds-P (dark purple) increase, and Ft (light blue) and Ds (light purple) decrease as you move up the Fj gradient. Weak binding (no line) between heterodimers can occur between all populations, intermediate binding (double lines) can occur when Ft-P is available and the strongest bond (solid line) occurs only between Ft-P and Ds. As there is a bias towards Ft phosphorylation increasing binding, more intermediate binding occurs in regions of high Fj even though less favourable Ds (light purple) is available. In regions of low Fj, although there is more favourable Ds available, Ft-P is less abundant meaning fewer intermediate bonds are able to form. As Ft-P allows for stronger binding in cells with higher Fj, Ds in the next left-most cell preferentially accumulates to the right cell edge meaning the Fj gradient produces cellular asymmetry. A gradient of tissue binding is also produced with cells in the left-most cell producing fewer intermediate bonds than those in the right-most cell. (B) Conformational changes caused by the phosphorylation of Ft and Ds could provide an explanation for the differences in binding affinity caused by apparently similar phosphorylation events. The ‘open’ conformation of Ft and the ‘closed’ conformation of Ds may bind preferably such that if Ft is phosphorylated and Ds is not, binding is strongest (+++). If Ft and Ds are both phosphorylated, binding strength is intermediate (++). ‘Closed’ conformation of non-phosphorylated Ft produces the weakest binding and is not affected by Ds (+).
DOI:http://dx.doi.org/10.7554/eLife.05789.018
In this study, we use a mixture of computational and experimental methods to address the mechanism of action of Fj. Initially, we constructed a simple one-dimensional computational model, in which Fj activity either promotes Ft binding activity or inhibits Ds binding activity, and in which Ft–Ds dimers are then allowed to freely bind at either cell edge until they reach an equilibrium state. Using this we show that a Fj gradient can lead to cellular asymmetry of Ft–Ds if Fj acts on either Ft or Ds or both. Our predictions differ from those of a previous study (Simon et al., 2010) which suggested that Fj acting only on Ft would not result in a cellular asymmetry of activity: the origin of this difference is that in our model (in accordance with our experimental observations) the Ft and Ds populations are mobile and able to redistribute to the most favourable cell edge. However, in agreement with the same study, our model does predict that only in the situation that Fj acts on both Ft and Ds is the amount of Ft–Ds dimers bound at the junctions approximately the same across the entire axis of the Fj gradient.
Using FRAP to measure the levels of stable amounts and mobility of Ft and Ds at cell junctions, combined with either removal of fj activity or mutation of the Fj phosphorylation sites in Ft and Ds, we directly demonstrate that Ft and Ds stably associate at cell junctions in vivo, that Fj modulates this binding with a net positive effect in promoting binding, and that as predicted from the in vitro assays, Fj acts independently on both Ft and Ds to promote or inhibit their binding, respectively.
An interesting experimental observation is that the effects of Fj on Ft–Ds binding and the total amounts stably localised to cell junctions are relatively modest. For instance, blocking the phosphorylation of either Ft or Ds only decreases the stable junctional amounts by about twofold. Thus, all the molecular species (i.e., Ds and DsP and Ft and FtP) contribute to the final population of bound Ft and Ds at cell junctions.
Our experimental results reveal that the effect of Fj on Ft is stronger than the effect of Fj on Ds. A simple prediction of this observation, confirmed by our computational model, is that the ability of the gradient of phosphorylated Ds to oppose the gradient of phosphorylated Ft is reduced, and therefore a gradient of Ft–Ds binding is expected to be observed across the tissue (Figures 5A and 6A). This prediction was confirmed experimentally. We cannot rule out that under normal circumstances the in vivo effects of Fj on Ds are in fact negligible compared to those on Ft, as our experiments make use of Ds transgenes and may not fully represent events occurring when wild-type Ds is expressed from its endogenous locus. Nevertheless, the simplest interpretation of both our observations and previous published work is that Ds phosphorylation contributes to normal patterning. In sum, our findings demonstrate an in vivo mechanism for how a Fj expression gradient is converted into cellular asymmetries via phosphorylation of both Ds and Ft.
We note that our computational simulations result in relatively modest cellular asymmetries of Ft and Ds distributions (around 10% of the total cellular levels), whereas in vivo, observed Ds asymmetry can be as high as twofold (Ambegaonkar et al., 2012; Brittle et al., 2012). However, our model is only intended to make simple predictions regarding the effects of binding and redistribution of Ft and Ds molecules in a simple one-dimensional system and does not capture the full complexity of a three-dimensional cell environment and changes that occur in protein production and degradation and other cell properties over time. Furthermore, it is generally believed that relatively weak asymmetries generated by expression gradients might subsequently be amplified by feedback mechanisms (Ambegaonkar et al., 2012; Brittle et al., 2012), and some molecular mechanisms that might contribute to amplification have recently been identified (Bosch et al., 2014; Rodrigues-Campos and Thompson, 2014).
With regard to possible amplification mechanisms, an intriguing observation is that the majority of stable Ft–Ds at junctions is concentrated in bright regions which we refer to as ‘puncta’. Furthermore, even mobile junctional populations of Ft and Ds show reduced mobility within these brighter regions. This might suggest that in addition to trans-interactions between Ft and Ds in neighbouring cells, there may also be cis-interactions between Ft and Ds molecules in cell junctions, a view supported by previous experimental reports (Matakatsu and Blair, 2006; Sopko et al., 2009). Such a clustering mechanism could provide the molecular basis of a positive feedback interaction. Previous observations have suggested that Ft–Ds puncta do not co-localise with ‘core’ planar polarity protein puncta at junctions (Ma et al., 2003). We are unsure as to the reasons for these distinct puncta populations, they could correspond to specialised membrane domains that favour binding or they could be random regions of protein clustering driven by cis-dimerisation or cytoskeleton tethering.
A final conundrum is the observation, first made in vitro, which we have now confirmed in vivo, that modification of analogous serine residues in cadherin repeats of Ft and Ds by Fj phosphorylation leads to opposite effects on their binding activity. The simplest model would be that phosphorylation either promoted or inhibited binding to a partner, but this is evidently not the mechanism in play here. A possible working hypothesis is that phosphorylation events lead to changes in the intramolecular conformation of the extracellular regions of Ft and Ds, by analogy to the way in which phosphorylation frequently acts to cause other classes of molecules to enter ‘open’ or ‘closed’ conformations (Xu and Carpenter, 1999; Potter et al., 2005; Bertocchi et al., 2012). A possible scenario is that phosphorylation of each molecule causes it to enter an ‘open’ conformation, and that ‘open’ Ft binds most favourably to Ds, but ‘closed’ Ds binds most favourably to Ft (Figure 6B). A recent publication analysing the configuration of mammalian Ft4 and Ds1 revealed multiple hairpin-like bends in their C-terminal regions caused by the loss of calcium binding linkers (Tsukasaki et al., 2014). It is possible that Fj phosphorylation results in conformational changes between cadherin repeat domains near to these calcium binding sites resulting in the regulation of binding strength, as suggested previously for E-cadherin (Petrova et al., 2012). Further studies of the structures and mode of heterophilic interactions between Ft and Ds will be required to resolve this question.
Materials and methodsAntibodies, immunolabelling and imaging
Wing discs were dissected from wandering third instar larvae, fixed in 4% paraformaldehyde and washed in PBS containing 0.1% Triton-X-100 prior to immunolabelling. Primary antibodies used for histology were rabbit anti-Ds (Strutt and Strutt, 2002), rabbit anti-Ft (Brittle et al., 2012), guinea pig anti-Hrs (Lloyd et al., 2002), mouse anti-βGAL (Promega, Wisconsin, USA), and mouse anti-Armadillo (DSHB, Iowa City, USA). A rabbit serum against Fj was generated using a fusion protein corresponding to amino acids 111–433, affinity purified and used at 1/100 for immunostaining. Secondary antibodies used were anti-Rb RRX and Cy2 (Jackson, Pennsylvania, USA), anti-guinea pig A568 and anti-mouse Cy5 (Molecular Probes, Oregon, USA). Images are averages of three confocal microscope sections taken on an Olympus FV1000 (Pennsylvania, USA), a Leica SP1 (Solms, Germany), or a Nikon A1R (Tokyo, Japan) confocal and processed using ImageJ (NIH, USA) and Adobe Photoshop (California, USA). ImageJ was used to measure the mean intensity of endogenously tagged ds-EGFP and ft-EGFP at junctions using at least nine wings of each genotype.
For Fj gradient measurements, immunolabellings were carried out on fixed wild-type wing discs using the rabbit-Fj antibody. Images were taken as 0.2-μm confocal slices throughout the disc and averages of the full stack were taken. Regions of interest were hand drawn per cell using a membrane marker as a guide and average intensity per cell was measured. Discs containing null clones were used for antibody background subtraction. Distal cells containing maximum measured signal were normalised to 100% and gradient of a proximal row of cells was calculated.
Antibody internalisation assay
Antibody internalisation assays were carried out on 5.5 hr APF pupal wings as previously described (Strutt et al., 2011). To detect V5-Ft, wings from flies of the genotype ftG-rv/ft8; P[acman] V5-ft (Feng and Irvine, 2009) were dissected and incubated with anti-V5 antibody (Novus Biologicals, Colorado, USA). For detection of extracellular V5-Ft, tissue was incubated in secondary antibody in the absence of detergent, and post-fixed before adding other antibodies with detergent. Internalised V5-Ft was co-stained with guinea pig anti-Hrs (Lloyd et al., 2002). For quantitation of extracellular staining at least five wings at each time point were imaged from at least two experiments taking 0.15-µm sections and using constant confocal settings. An average intensity of the three brightest confocal slices at the level of the apical junctions was measured in ImageJ. Laser-off background was subtracted, and the readings were normalized to 1.0 at t0.
To visualise internalised V5-Ft, wings were imaged in apical and subapical regions. Statistical analysis was carried out using ordinary one-way ANOVA with Tukey's test for multiple comparisons.
Molecular biology
Constructs were generated using standard molecular biology techniques and mutagenised and PCR-amplified regions verified by sequencing. Full-length ft is a 22-kb genomic fragment from BACR11D14 (BACPAC Resources, California, USA), containing the entire coding sequence and tagged with EGFP subcloned into pAttB-FRT-polyA-FRT (derived from pAct-FRT-polyA-FRT [Strutt, 2001]). Point mutations in the cadherin domains were introduced using QuikChange Multi-Site Directed Mutagenesis kit (Stratagene, California, USA). 5 serines or threonines identified as phosphorylation sites in Ft (CAD3 (S273), CAD5 (S497), CAD10 (T1052), CAD11 (S1156) and CAD13 (S1387)) (Ishikawa et al., 2008) were mutated to alanine to generate Ft−S/T>Ax5-EGFP. A phosphomimetic form of Ft (Ft S/T>Dx4-EGFP) contains point mutations to aspartate of the following residues (CAD3 (S273), CAD10 (T1052), CAD11 (S1156), and CAD13 (S1387)).
Fly strains
Alleles used are described in FlyBase. Homologous recombination was used to target EGFP into the C-terminus of the ft gene at its endogenous locus using the pRK2 targeting vector (Huang et al., 2008). ds-EGFP (Brittle et al., 2012) and ft-EGFP flies were recombined with fjd1, ftG-rv, or dsUA071. All endogenous FRAP experiments were performed using homozygous ds-EGFP or ft-EGFP. Clones were generated using the FLP/FRT system (Xu and Rubin, 1993) and marked with arm-lacZ (Vincent et al., 1994). Transgenes containing ft-EGFP and point mutants were integrated into the same landing site (attP2 68A4) (Groth et al., 2004) by Genetivision (Texas, USA). Act-ds-EGFP and point mutations used are described previously (Brittle et al., 2010).
Genotypes used were:
y w Scer\FLP1Ubx.hs; ds38k/ dsUA071; AttB{w+ActP-FRT-polyA-FRT-dsX-EGFP} /+
y w Scer\FLP1Ubx.hs; dsUA071fjP1/ ds38kfjP1; AttB{w+ActP-FRT-polyA-FRT-dsX-EGFP} /+
y w Scer\FLP1Ubx.hs; ft8/ ftG-rv; AttB{w+ActP-FRT-polyA-FRT-ftX-EGFP} ftG-rv/+
y w Scer\FLP1Ubx.hs; ft8fjd1/ ftG-rvfjP1; AttB{w+ActP-FRT-polyA-FRT-ftX-EGFP} /+
where dsX and ftX refers to wild-type ds or ft or one of the phosphomutants.
Live imaging and fluorescence recovery after photobleaching (FRAP)
Prior to dissection, an imaging chamber was built using a 22 × 50 mm cover glass as a base slide (Thermo Scientific, Massachusetts, USA). Sellotape was placed smoothly on to the centre of the slide and a 7-mm2 area was cut out using a razor blade. Wandering stage larvae were collected and cleaned by rinsing in PBS. Wing discs were dissected in Shields and Sang M3 media (Sigma [#S3652], Missouri, USA) with 2% added foetal bovine serum (M3FBS). Discs were transferred by pipette to the cut out area in the imaging chamber with around 4 µl of media and arranged with their apical surface facing towards the base cover glass. The M3FBS was spread evenly around the cut out area. A 13-mm circular cover glass (Thermo Scientific) was carefully placed over the top allowing extra media to spread to the edges. Quick drying nail varnish was used to seal the cover glass after leaving to settle for a few minutes.
For FRAP, samples were imaged on an inverted Nikon A1R GaAsP confocal using a Nikon 60× oil objective lens at 11.76× zoom producing a FRAP region of 256 × 256 pixels with a pixel size of 0.07 µm. For pre- and post-bleach images, a 448 Argon laser was used at an output of 0.5% with varying gain settings depending upon phenotype. Eight 1 µm2 regions of interest (ROIs) were selected per wing and bleached using the 488 argon laser at 50% power, passing 1–3 times for between 0.5 and 1.5 s depending upon experiment. Two pre-bleach images were captured with no delay as well as an immediate post-bleach image, 10 images were then captured every 5 s, followed by 10 images every 10 s and 10 images every 30 s. The initial rapid imaging was done in order to capture adequate rate information. For analysis, ROIs were individually reselected in ImageJ at each time point and acquisition bleaching was measured in non-bleached regions. Data were corrected for acquisition bleaching and normalised against pre-bleach values. An XY graph was plotted for each wing in PRISM (v.6 GraphPad). A one-phase exponential association curve was fitted for each ROI and an average plateau value recorded. A one-phase exponential curve was generated as we observe a single mode of recovery that reaches a plateau. We note that should the mode of recovery be more complex than this a two-phase exponential curve may be more appropriate; however, we do not have relevant experimental data to support this. Pre-bleach values were averaged per wing and multiplied by their associated plateau giving a stable amount. Stable amounts were then averaged across wings producing a final figure for each genotype. Stable amounts were analysed using unpaired t-tests or ordinary one-way ANOVAs with Tukey's test for multiple comparisons. Final stable fraction graphs were produced using the average plot for each wing and intensity was averaged across ROIs.
Modelling
We developed a computational model of Ft–Ds binding at cellular junctions, described by a set of ordinary differential equations. In this framework, we define a one-dimensional row of cells, whose proximal and distal membranes contain the same initial amount of Ft and Ds (Figure 1—figure supplement 1). These molecules were then phosphorylated according to a linear gradient of Fj activity, applied to reflect in vivo measurements of Fj quantity. Phosphorylated and unphosphorylated molecules were allowed to bind across junctions resulting in formation of four possible complexes, each in two orientations, namely, FtP—Ds (A), FtP—DsP (B), Ft—Ds (C), and Ft—DsP (D). This consideration of all four possible complexes between phosphorylated and non-phosphorylated forms of Ft and Ds is a key distinguishing feature of our model, when compared to previous computational approaches which only considered a single binding species essentially equivalent to our complex A (Abley et al., 2013; Mani et al., 2013; Jolly et al., 2014). Reaction rates were derived from mass action, giving the following equations for proximally oriented complexes:dAPidt=kaonFtPPiDsDi−1−kaoffAPi,dBPidt=kbonFtPPiDsPDi−1−kboffBPi,dCPidt=kconFtPiDsDi−1−kcoffCPi,dDPidt=kdonFtPiDsPDi−1−kdoffDPi.
Subscripts denote the proximal (P) membrane in cell i neighbouring distal membrane (D) in cell i−1. Equivalent equations were derived for distally oriented complexes. Each reaction is parameterised by rate constants kon and koff for binding and unbinding reactions of each complex. Equations for unbound molecules were derived in a similar fashion, with the addition of a simple term allowing redistribution within a cell, parameterised by the coefficient, Diff:dDsPidt=−kaonFtPDi−1DsPi−kconFtDi−1DsPi+kaoffADi−1+kcoffCDi−1+Diff(DsDi−DsPi),dDsPPidt=−kbonFtPDi−1DsPPi−kdonFtDi−1DsPPi+kboffBDi−1+kdoffDDi−1+Diff(DsPDi−DsPPi),dFtPidt=−kconFtPiDsDi−1−kdonFtPiDsPDi−1+kcoffCPi+kdoffDPi+Diff(FtDi−FtPi),dFtPPidt=−kaonFtPPiDsDi−1−kbonFtPPiDsPDi−1+kaoffAPi+kboffBPi+Diff(FtPDi−FtPPi).
Binding rates between molecules are parameterised by the association constant (kon/koff) for each complex, since a higher ‘on’ rate constant will result in an increased concentration of complexed molecules. Relative binding strengths of different molecule combinations reflect findings from in vivo and in vitro studies (Brittle et al., 2010; Simon et al., 2010), such that phosphorylation of Ft (FtP) promotes its binding and conversely phosphorylation of Ds (DsP) inhibits its binding. The following hierarchy of relevant association constants was used initially, A > B = C > D, such that the complex containing the two favoured molecules, FtP and Ds making complex A, had a faster ‘on’ rate than other combinations. Initial values of kon/koff for A, B, C, and D were chosen as 1, 1/4, 1/4 and 1/16, respectively, to reflect the relative differences in stable amounts measured experimentally.
Binding was allowed to continue over time until convergence. Simulations were run using an in-built ode solver (ode23s) in MATLAB (R2013a; MathWorks, Massachusetts, USA) and final concentrations of bound molecules were plotted.
We went on to adapt this model to reflect our experimental finding that Fj appears to have a dominant effect on Ft. To achieve this, we altered association constants of binding reactions to allow phosphorylation of Ds to have a less significant effect on binding strengths, giving A > B > C = D. Thus association constants for each complex A, B, C, and D, were given as 1, 1/2, 1/4, and 1/4, respectively. Further explanation can be found in the ‘Results’.
Acknowledgements
We thank Ken Irvine for the P[acman]V5-ft fly stock, Hugo Bellen for the anti-Hrs antibody, John Walker for generating the ft-EGFP stock, and Katrina Hofstra for assistance producing the anti-Fj antibody. DNA clones were obtained from BacPac Resources and antibodies from the DSHB. This work was supported by a Wellcome Trust Senior Fellowship to DS Confocal facilities were provided by the Wellcome Trust, MRC and Yorkshire Cancer Research.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
RH, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
ALB, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
KHF, Conception and design, Acquisition of data, Analysis and interpretation of data
NAMM, Conception and design
DS, Conception and design, Drafting or revising the article
Additional files10.7554/eLife.05789.019
Table contains numbers of wings, plateau data, and 95% confidence of plateau data for each FRAP experiment.
DOI:http://dx.doi.org/10.7554/eLife.05789.019
ReferencesAbleyKDe ReuillePBStruttDBanghamAPrusinkiewiczPMareeAFGrieneisenVACoenE2013An intracellular partitioning-based framework for tissue cell polarity in plants and animalsDevelopment1402061207410.1242/dev.062984AmbegaonkarAAPanGManiMFengYIrvineKD2012Propagation of Dachsous-Fat planar cell polarityCurrent Biology221302130810.1016/j.cub.2012.05.049BertocchiCVaman RaoMZaidel-BarR2012Regulation of adherens junction dynamics by phosphorylation switchesJournal of Signal Transduction201212529510.1155/2012/125295BoschJASumabatTMHafeziYPellockBJGandhiKDHariharanIK2014The Drosophila F-box protein Fbxl7 binds to the protocadherin fat and regulates Dachs localization and Hippo signalingeLife3e0338310.7554/eLife.03383BosveldFBonnetIGuiraoBTliliSWangZPetitalotAMarchandRBardetPLMarcqPGranerFBellaïcheY2012Mechanical control of morphogenesis by Fat/Dachsous/Four-jointed planar cell polarity pathwayScience33672472710.1126/science.1221071BrittleAThomasCStruttD2012Planar polarity specification through asymmetric subcellular localization of Fat and DachsousCurrent Biology2290791410.1016/j.cub.2012.03.053BrittleALRepisoACasalJLawrencePAStruttD2010Four-jointed modulates growth and planar polarity by reducing the affinity of dachsous for fatCurrent Biology2080381010.1016/j.cub.2010.03.056CasalJLawrencePAStruhlG2006Two separate molecular systems, Dachsous/Fat and Starry night/Frizzled, act independently to confer planar cell polarityDevelopment1334561457210.1242/dev.02641CasalJStruhlGLawrencePA2002Developmental compartments and planar polarity in DrosophilaCurrent Biology121189119810.1016/S0960-9822(02)00974-0CaveyMRauziMLennePFLecuitT2008A two-tiered mechanism for stabilization and immobilization of E-cadherinNature45375175610.1038/nature06953ChoEIrvineKD2004Action of fat, four-jointed, dachsous and dachs in distal-to-proximal wing signalingDevelopment1314489450010.1242/dev.01315FengYIrvineKD2009Processing and phosphorylation of the Fat receptorProceedings of the National Academy of Sciences of USA106119891199410.1073/pnas.0811540106GrothACFishMNusseRCalosMP2004Construction of transgenic Drosophila by using the site-specific integrase from phage phiC31Genetics1661775178210.1534/genetics.166.4.1775HuangJZhouWWatsonAMJanYNHongY2008Efficient ends-out gene targeting in DrosophilaGenetics18070370710.1534/genetics.108.090563IshikawaHOTakeuchiHHaltiwangerRSIrvineKD2008Four-jointed is a Golgi kinase that phosphorylates a subset of cadherin domainsScience32140140410.1126/science.1158159JollyMKRizviMSKumarASinhaP2014Mathematical modeling of sub-cellular asymmetry of fat-dachsous heterodimer for generation of planar cell polarityPLOS ONE9e9764110.1371/journal.pone.0097641LawrencePAStruhlGCasalJ2008Do the protocadherins Fat and Dachsous link up to determine both planar cell polarity and the dimensions of organs?Nature Cell Biology101379138210.1038/ncb1208-1379LloydTEAtkinsonRWuMNZhouYPennettaGBellenHJ2002Hrs regulates endosome membrane invagination and tyrosine kinase receptor signaling in DrosophilaCell10826126910.1016/S0092-8674(02)00611-6LockeM1959The Cuticular pattern in an insect, Rhodnius-Prolixus StalJournal of Experimental Biology36459476MaDYangCHMcNeillHSimonMAAxelrodJD2003Fidelity in planar cell polarity signallingNature42154354710.1038/nature01366ManiMGoyalSIrvineKDShraimanBI2013Collective polarization model for gradient sensing via Dachsous-Fat intercellular signalingProceedings of the National Academy of Sciences of USA110204202042510.1073/pnas.1307459110MatakatsuHBlairSS2004Interactions between Fat and Dachsous and the regulation of planar cell polarity in the Drosophila wingDevelopment1313785379410.1242/dev.01254MatakatsuHBlairSS2006Separating the adhesive and signaling functions of the Fat and Dachsous protocadherinsDevelopment1332315232410.1242/dev.02401PetrovaYISpanoMMGumbinerBM2012Conformational epitopes at cadherin calcium-binding sites and p120-catenin phosphorylation regulate cell adhesionMolecular Biology of the Cell232092210810.1091/mbc.E11-12-1060PotterMDBarberoSChereshDA2005Tyrosine phosphorylation of VE-cadherin prevents binding of p120- and beta-catenin and maintains the cellular mesenchymal stateThe Journal of Biological Chemistry280319063191210.1074/jbc.M505568200Rodrigues-CamposMThompsonBJ2014The ubiquitin ligase FbxL7 regulates the Dachsous-Fat-Dachs system in DrosophilaDevelopment1414098410310.1242/dev.113498SimonMA2004Planar cell polarity in the Drosophila eye is directed by graded Four-jointed and Dachsous expressionDevelopment1316175618410.1242/dev.01550SimonMAXuAIshikawaHOIrvineKD2010Modulation of fat:dachsous binding by the cadherin domain kinase four-jointedCurrent Biology2081181710.1016/j.cub.2010.04.016SopkoRSilvaEClaytonLGardanoLBarrios-RodilesMWranaJVarelasXArbouzovaNIShawSSaburiSMatakatsuHBlairSMcNeillH2009Phosphorylation of the tumor suppressor fat is regulated by its ligand Dachsous and the kinase discs overgrownCurrent Biology191112111710.1016/j.cub.2009.05.049StruttD2009Gradients and the specification of planar polarity in the insect cuticleCold Spring Harbor Perspectives in Biology1a00048910.1101/cshperspect.a000489StruttDI2001Asymmetric localization of frizzled and the establishment of cell polarity in the Drosophila wingMolecular Cell736737510.1016/S1097-2765(01)00184-8StruttHMundyJHofstraKStruttD2004Cleavage and secretion is not required for Four-jointed function in Drosophila patterningDevelopment13188189010.1242/dev.00996StruttHStruttD2002Nonautonomous planar polarity patterning in Drosophila: dishevelled-independent functions of frizzledDevelopmental Cell385186310.1016/S1534-5807(02)00363-5StruttHWarringtonSJStruttD2011Dynamics of core planar polarity protein turnover and stable assembly into discrete membrane subdomainsDevelopmental Cell2051152510.1016/j.devcel.2011.03.018TsukasakiYMiyazakiNMatsumotoANagaeSYonemuraSTanoueTIwasakiKTakeichiM2014Giant cadherins Fat and Dachsous self-bend to organize properly spaced intercellular junctionsProceedings of the National Academy of Sciences of USA111160111601610.1073/pnas.1418990111VincentJPGirdhamCHO'FarrellPH1994A cell-autonomous, ubiquitous marker for the analysis of Drosophila genetic mosaicsDevelopmental biology16432833110.1006/dbio.1994.1203WaddingtonCH1943The development of some 'leg genes' in DrosophilaJournal of Genetics45294310.1007/BF02982772XuTRubinGM1993Analysis of genetic mosaics in developing and adult Drosophila tissuesDevelopment11712231237XuYCarpenterG1999Identification of cadherin tyrosine residues that are phosphorylated and mediate Shc associationJournal of Cellular Biochemistry7526427110.1002/(SICI)1097-4644(19991101)75:23.0.CO;2-BYangCHAxelrodJDSimonMA2002Regulation of Frizzled by fat-like cadherins during planar polarity signaling in the Drosophila compound eyeCell10867568810.1016/S0092-8674(02)00658-XZeidlerMPPerrimonNStruttDI1999The four-jointed gene is required in the Drosophila eye for ommatidial polarity specificationCurrent Biology91363137210.1016/S0960-9822(00)80081-0ZeidlerMPPerrimonNStruttDI2000Multiple roles for four-jointed in planar polarity and limb patterningDevelopmental Biology22818119610.1006/dbio.2000.994010.7554/eLife.05789.020Decision letterLawrencePeter AReviewing editorUniversity of Cambridge, United Kingdom
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Interpreting a long-range gradient at the cellular level: the mechanism of action of Four-jointed in the Drosophila wing” for consideration at eLife. Your article has been evaluated by K VijayRaghavan (Senior editor), and four reviewers, one of whom, Peter Lawrence, is serving as a guest Reviewing editor.
The Reviewing editor and the other reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
I have now assembled four reviews and all agree that a revised version of your paper will be published in eLife. We are impressed with the directness of your approach to an important and challenging problem that, up to now, most have preferred to simulate rather than investigate. No additional experiments are asked for but taken together there are a large number of criticisms that in my view arise mainly from a lack of clarity in your manuscript. The techniques you use, the way you present the data and some of your descriptions are complicated, and have caused confusion. So I think the simplest way to proceed is for me to send you these criticisms more or less undiluted.
The only suggestion we all agree on is that you should provide more details of your theoretical model, its assumptions, equations, and parameters so readers can understand it better and compare it with other similar models. There are other points that come up in more than one review. But my opinion is based on the fact that your paper is yours, not ours, your names will go forward with the paper and therefore I believe you, not us, should decide how to revise it. But I urge you to consider the points raised, try to make it clearer. In this case I have decided not to ask you to provide a reasoned explanation for all you have decided to do and not to do. This would take a huge amount of time and effort and the result might well be longer than the paper and written only really to satisfy the curiosity of we reviewers. However I know that as a responsible scientist you will want to take these criticisms seriously and do your best to make your paper as lucid as you can.
Regarding the Jolly model, if you go to PLoS ONE you will see that we already raised a query with Jolly about their model.
Here is a list of all the comments for you to consider:
Reviewer 1:
This is an interesting and accomplished article. It addresses in a new and concrete way a long-standing question that asks how a shallow multicellular gradient can be converted into a consistently oriented and asymmetric distribution of molecules within single cells. This question has often been explored by modelling but the models are not usually checked out by observation. However this paper is based on the Ft/Ds system for planar cell polarity and combines molecular measurements with modelling. The amount of bound Ds and Ft is estimated by an apparently ingenious use of FRAP. The paper has interesting findings: for the first time the authors have weighed the relative contributions of the Ds, Ft and Fj distributions to the final pattern and tried to quantify how and why the effects of Fj on Ds and Ft differ in strength. They have found evidence for a central tenet of the Ds/Ft model, as proposed earlier, that there should be a gradient of Ds/Ft heterodimers across the tissue. They have estimated the steepness of the Fj gradient from measurements based on Fj antibody. In our opinion the paper clearly deserves publication in eLife but only after revision, the main purpose of which would be to make its scientific strategy and results clearer for the reader.
Results:
There are problems with the Results section in that the details of some experiments confuse us, possibly because the authors have not presented their results as clearly as they should. Below we list some of the problems we have encountered in trying to understand both the data and the model. The results of the authors' model are presented in Figure 1DE. We have tried to work out how the model is set up and what the parameters, rules and equations are (as shown in the first paragraph), but have not succeeded. Perhaps more could be explained in a legend or in the Methods?
In the first paragraph of the Results section, the expression “a hierarchy of binding affinities”: what does this mean in terms of the model and how is this hierarchy calculated?
Fourth paragraph of the Results: “the speed… of the mobile population” ?
Figures 2C and 2D are not explained well, e.g. in the fifth paragraph of the Results section, the unstable fraction is estimated by a height of the curve, while the stable fraction is the area above the line, we can work out how the area is calculated but it is not well explained how a height is compared with an area.
In making these exponential curves the data is “fitted” to the curve but it is not clear how this is done, how objective it is, how good the fit is. After all the data could also be “fitted” to a straight line; presumably the fit would not be so good.
Seventh paragraph of the Results section onwards: it might be better to tell us at the beginning, that in ft- and in ds-, no puncta are seen (although we are not sure if that is so of ds-?)
Ninth paragraph of the Results section: this is an interesting finding and consequently we ask has the excess of Ft over Ds been put in the model?
Eleventh and twelfth paragraphs of the Results section and Figure 3: this is a difficult section, particularly as the effects of Fj on Ds are going to be declared insignificant, when compared to the effects of Fj on Ft, and therefore one is led down two paths and one seems to go nowhere. It is hard on the brain and I think the authors should do more work to make the data behind Figure 3 more accessible. For example consider Figures 3 A and B and Figures 3A and A’. Are these results in accord with Figure 2, G and H?
Thirteenth and fourteenth paragraphs of the Results section: This is also a difficult section as one is trying to deduce the effects on heterodimers from the effect of Fj on Ft or Ds in conditions in which one of these effects is blocked by a mutation in the site of Fj's action. We get the impression that the results again show that the effect of Fj on Ds is insignificant. If this is so, then couldn't the whole be presented more simply and the effect on Fj ignored in these presentations (there is still evidence in vivo published in Brittle et al. that Fj can effect Ds ability to bind to Ft, but this was with overexpression of Fj, so this matter should be addressed in the Discussion). Figure 4D' would seem to show almost no effect, barely at the level of significance. If this is so, then the headline for the thirteenth paragraph of the Results section and the statement in the eighteenth paragraph could mislead us?
The histograms in Figure 4, we are confused by them. They do show large effects, but are these differences caused more by the original intensities of levels in the puncta and non-puncta than by the variable that is pertinent (which is the difference in stability due to phosphorylation)? Consider Figure 4C for instance where we seem to be misled by the p=0.006. An example to explain (if we understand correctly): if you have an intensity of 1000 in the puncta and a recovery of 0.8 it gives 800 as a stable amount. If in the non-puncta there is an intensity of 200 and a recovery also of 0.8 the result would be 160; and course the difference between 160 and 800 is great but those two figures are not the figures one should be comparing as the recovery is 0.8 in both cases.
Another example for illustration: imagine you have, in the puncta, an intensity of 1000 and a recovery of 0.4, giving a stable amount of 400 and in the non-puncta an intensity of 200 and a recovery of 0.9, giving a stable amount of 180. In this case 400 is clearly higher than 180, but can we not be misled if we use these figures? The differences between the two fitted curves are really what matters here, is that not so? Therefore should the histograms be left out as they confuse and give an impression of a difference that is not the difference that matters?
In the legend of Figure 2, calculation of “stable amount” could be better explained and does this data fit with results shown in Figures 5C and C'.
Thinking about Figure 4 and the disparity of the effects of Fj on Ds and Ft: the effects measured would depend on locality in the gradient (Figure 5). Indeed in vivo results depended on where the clone was made (Brittle et al.). And we see here the effect of Fj varies across the wing. Thus where in the wing these measurements are made is likely to be an important factor but it is not controlled here?
Where Fj is high shouldn't Ds be low, therefore if Fj is removed, stability should surely be greater where Fj is normally high, not as in Figure 4C where we see the reverse?
Does Figure 4C show a significant change, i.e. a significant effect on Ds?
Eighteenth paragraph of the Results section: We don't know the parameters used to set up the model, a problem for the reader of this section.
Paragraph 19 of the Results section: where are these measurements made, puncta or non-puncta?
Figure 6: the model presented in A is virtually identical to that presented in Casal et al., 2006, is that not so?
It might be useful for the authors to consider the possibility that phosphorylation may not be all or none (per Ds or Ft molecule) but could be partial. Would that possibility change interpretations?
Reviewer 2:
The authors are attempting to understand the long-standing developmental biology problem of how global tissue wide patterning gradients can lead to local cellular asymmetries of protein interactions and localization. This is a highly important question that should be of interest to a wide range of audiences. The Fat-Dachsous-Four-jointed pathway in Drosophila presents an ideal system to investigate the mechanisms of planar cell polarity establishment. The authors are attempting to understand the problem quantitatively by developing a 1D computational model. Using experimentally measured values for a Four-jointed (Fj) gradient, and relative binding affinities between phosphorylated and unphosphorylated Fat and Dachsous, they claim to be able to computational explain the observed in vivo patterns of Fat-Dachsous binding and planar polarization across the wing.
Major concern:
The attempt to measure in vivo protein stabilities and deduce relative binding affinities is to be applauded. However, even with these new in vivo parameters, their model still did not produce in vivo like cellular asymmetries of Fat and Dachsous distribution, which is their main readout for the model. The main interest in this subject area is how relatively mild patterning gradients across tissues can be interpreted to give relatively high cellular asymmetries, such as the 2-fold differences in Ds that has been experimentally measured previously. I feel the authors have not been able to address this issue fully, and merely show a mild cellular asymmetry as a result of their new proposed model, which is only slightly better than their first model, where they did not have the new in vivo parameters. Admittedly, the authors do acknowledge this problem, and discuss several possible mechanisms for generating a higher cellular asymmetry, such as feedback and amplification mechanisms. Unless the authors can attempt to include these mechanisms and show that they can be significant factors in generating the high cellular asymmetries, the mystery, in my opinion, still exists. As a result, in its current state, this work does not represent a significant enough advance on the field to be suitable for publication in eLife.
Other points to address:
1) The authors only measure the Fj gradient in one quadrant of the wing pouch (Figure 1C). What about the other regions (quadrants) of the wing pouch? What are the gradients of Fj? If these are used in the model, can these gradients produce the same asymmetries in Fat and Ds (using same parameters for everything else)?
2) To go with Figure 2C and D, please provide images to show what is a 'puncta' bleach and a 'non-puncta' bleach. It is hard to see how one would define and select these regions.
3) For figures 1D-E and 5A it would be useful to show the actual values of the bound Ds and bound Ft on either side of each cell. The current graphs can only show asymmetries, but not how much, and whether that changes across the tissue. For example, for bound Ds in Figure 5A, the asymmetry seems greater more distally in the tissue. Is that true in vivo?
4) The authors should provide more details for the set-up of the model, such as the equations for the ODE and actual parameter values.
Reviewer 3:
The manuscript is an informative and valuable extension of the authors' 2010 study on the role of Fj in the regulation of Ft-Ds interactions, informed by more recent findings on the polarization of Ds and Ft localization within single cells. The former study was largely limited to measurements of Ft-Ds binding in vitro, and the current work extends this to in vivo work using late third instar wing imaginal discs.
The work looks pretty straightforward, although the authors have had to make some assumption about the relationship between stability, recovery and binding; it is always possible that some of the changes are caused by something other than Ft-Ds binding, such as trafficking or protein stability. Nonetheless, the results largely match the expectations based on previous work, with only a few exceptions. These include the finding that the Fj effect on Ft seems to predominate in vivo over its effects on Ds, and that this creates a slight gradient of Ft-Ds binding strength across the wing that depends on distal Fj. The authors also find that a phosphomimetic version of Ds does not show the expected decrease in stable binding. The authors argue that the mutations do not perfectly mimic phosphorylation, and indeed that matches the reduced effect this version had in their previous in vitro assays, where they made the same argument.
Other than the concerns about less direct causes of the results, I had only a couple major comments:
1) The authors' data indicates that Ft-Ds binding is less stable in the absence of Fj. The one counter-example I found in the literature was rather less direct, but in Matakatsu and Blair (2012) the Hippo signaling boundary effects created by Ft overexpression extended further distally in the absence of Fj than its presence, suggesting stronger Ft-Ds binding in the absence of Fj. Is it possible that the result depends as well on the levels of Ft or Ds being expressed? That might make one a bit cautious about results at anything other than endogenous expression levels.
2) If the authors think that the polarization of Ft, Ds and Dachs in the distal wing/pouch is due entirely to Fj, has anyone demonstrated that? Brittle et al. (2012) only looked at dachs in fj ds double mutants, I believe. Bosveld looked at fj mutants but in the notum, not wing, and to my eye there was still some Ds polarization.
Reviewer 4:
This manuscript focuses on how long range gradients contribute to planar cell polarity. I focused primarily on assessing the modeling and FRAP analysis used in this work. In general, the interpretation of the FRAP data is consistent with the conceptual model presented by the authors, and the experiments appear to be well performed and thorough. However, I have several concerns regarding the computational model. Other points also require further clarification.
1) The details of the mathematical model are not presented here. This makes it almost impossible to evaluate the soundness of the model. The authors are advised to present the complete underlying assumptions and governing PDE equations, boundary conditions, initial conditions, and the parameters used for the modeling. Some of these parameters (such as the diffusion coefficients of Ft and Ds) could be extracted from the experiments presented. In addition, the authors indicate that model is in PDE (diffusion was included), but the numerical scheme presented here was determined using an ODE solver. This point needs to be clarified.
2) The way the kinetic equations (hierarchy of relevant associations) are presented in the Method is confusing. They should be represented using standard notations.
3) The manuscript fails to cite and discuss a recent study using mathematical modeling to examine mechanisms contributing to the sub-cellular asymmetry of Ft-Ds heterodimers (Jolly et al., 2014, Mathematical Modeling of Sub-Cellular Asymmetry of Fat-Dachsous Heterodimer for Generation of Planar Cell Polarity, PLoS ONE, 9(5):e97641). This is very surprising given that the work of Jolly et al. addresses very similar questions as the current study, albeit solely from a theoretical perspective. It is essential that the authors elaborate on how their current findings relate to this published work.
4) In this study, FRAP is used to monitor changes in putative interactions between proteins that occur between two cells and thus in separate membranes. Under these specialized conditions, one might intuitively expect diffusion of both binding partners to be slowed, as the authors suggest. However, it seems likely that other types of interactions would be required to cause immobilization per se. The authors suggest that cis-interactions may be important (eighth paragraph of the Discussion section). Other types of events, such as interactions of proteins with the actin cytoskeleton might be important as an immobilization mechanism and/or the establishment of “stable” populations. It appears that such a mechanism was previously suggested to be important for the case of E-cadherin in puncta (Cavey et al., Nature, 2008, 453:751-6, A two-tiered mechanism for stabilization and immobilization of E-cadherin). This possibility should be discussed.
5) On a related note, the authors report that Ft and Ds localize in part to puncta where the proteins appear to be immobilized but do not say very much about the possible identity of the puncta or their relationship to other puncta previously identified. These points should be further developed in the Discussion.
6) The authors use the word “stable” in several different contexts, beginning in the second headline of the Results section. For the case of the FRAP measurements, they use this to refer to an immobile fraction of proteins. They use it again to refer to a pool of proteins that do not undergo internalization in the endocytosis assay. These two “stable” pools are not necessarily identical. As such, the exact meaning of “stable” needs to be better qualified in each case.
10.7554/eLife.05789.021Author response
We have now revised our manuscript in response to the comments from the four expert reviewers. The key comments were regarding general clarity in the main text, particularly regarding our analysis and interpretation of FRAP experiments, as well as requesting more detail on the model set up and equations. We have addressed these concerns in our revised draft, adding three new supplementary figures as well as more detailed explanations to aid understanding and clarity. Further specific points are addressed individually below.
Reviewer 1:
There are problems with the Results section in that the details of some experiments confuse us, possibly because the authors have not presented their results as clearly as they should. Below we list some of the problems we have encountered in trying to understand both the data and the model. The results of the authors' model are presented inFigure 1DE. We have tried to work out how the model is set up and what the parameters, rules and equations are (as shown in the first paragraph), but have not succeeded. Perhaps more could be explained in a legend or in the Methods?
In the first paragraph of the Results section, the expression “a hierarchy of binding affinities”: what does this mean in terms of the model and how is this hierarchy calculated?
Further explanation has been added to the Methods. In short, the “hierarchy” reflects the previously published findings that FtP binds more strongly than Ft and DsP binds less strongly than Ds (Brittle et al. 2010; Simon et al., 2010). A complex that contains both favoured molecules (complex A with FtP and Ds) will bind better than one which contains both least favoured molecules (complex D with Ft and DsP). The remaining combinations of molecules making up complexes B and C will bind somewhere in the middle. We have given B and C association constants and began by making them equal to one another since we had no evidence to suggest one was better at binding than the other. The values broadly reflect the differences in stable amount we have measured in our experiments. For example, complex A and C both contain unphosphorylated Ds, but differ in the phosphorylation of Ft. We have set association constants as 1 and ¼ for A and C respectively, which is similar to the differences in stable amount seen in Figure 4E comparing Ds phosphomutant in the presence (complex A) or absence (complex C) of Fj.
Fourth paragraph of the Results section: “the speed… of the mobile population”?
We have extensively rewritten this part of the text. In particular, see additional text in the section “Ft and Ds exhibit stable populations at the cell junctions” beginning: “Additionally, if the speed of protein recovery…”
Figures 2C and 2Dare not explained well, e.g. in the fifth paragraph of the Results section, the unstable fraction is estimated by a height of the curve, while the stable fraction is the area above the line, we can work out how the area is calculated but it is not well explained how a height is compared with an area.
Further explanation of FRAP and its interpretation has been added to the section headed “Ft and Ds exhibit stable populations at the cell junctions”. A supplementary figure has also been added explaining this point. In essence our statement “area above the line” was confusing: it is in fact the height above the line which is important.
In making these exponential curves the data is “fitted” to the curve but it is not clear how this is done, how objective it is, how good the fit is. After all the data could also be “fitted” to a straight line; presumably the fit would not be so good.
All FRAP data points and curves have been added to Figure 2–figure supplements 1-5, along with the 95% confidence intervals of the curve fit in Table 1. This illustrates how a one-phase exponential curve fits very well in each case.
Seventh paragraph of the Results section onwards: it might be better to tell us at the beginning, that in ft- and in ds-, no puncta are seen (although we are not sure if that is so of ds-?)
This has been added (see the sentence beginning with “Removal of the putative binding partner…”, in the subsection “Ft and Ds require each other for junctional stability”, in Results).
Ninth paragraph of the Results section: this is an interesting finding and consequently we ask has the excess of Ft over Ds been put in the model?
Simply adding a 2-fold excess of Ft into our current model has only a moderate effect on cellular asymmetry measured in terms of bound Ft and Ds (however the “excess” unbound Ft does of course obscure observation of the bound fraction at junctions, hence the need to measure it using FRAP). A 10x increase in Ft still only moderately alters cellular asymmetry (from 7.8% to 6.5%). We plan to publish a full exploration of the model elsewhere, which will include changes in levels.
The histograms inFigure 4, we are confused by them. They do show large effects, but are these differences caused more by the original intensities of levels in the puncta and non-puncta than by the variable that is pertinent (which is the difference in stability due to phosphorylation)? ConsiderFigure 4Cfor instance where we seem to be misled by the p=0.006. An example to explain (if we understand correctly): if you have an intensity of 1000 in the puncta and a recovery of 0.8 it gives 800 as a stable amount. If in the non-puncta there is an intensity of 200 and a recovery also of 0.8 the result would be 160; and course the difference between 160 and 800 is great but those two figures are not the figures one should be comparing as the recovery is 0.8 in both cases.
Another example for illustration: imagine you have, in the puncta, an intensity of 1000 and a recovery of 0.4, giving a stable amount of 400 and in the non-puncta an intensity of 200 and a recovery of 0.9, giving a stable amount of 180. In this case 400 is clearly higher than 180, but can we not be misled if we use these figures? The differences between the two fitted curves are really what matters here, is that not so? Therefore should the histograms be left out as they confuse and give an impression of a difference that is not the difference that matters?
The FRAP experiments in Figure 4 are carried out on transgenic proteins under the control of the actin promoter and are thus expressed at high levels with no distinguishable puncta. We have added the sentence “FRAP is performed on junctions as no puncta are visible” into the figure legend for Figure 4.
Since we are comparing two different fluorescent proteins, or indeed in other figures we are comparing different genetic backgrounds, we must take their differing expression levels into account. We are using FRAP to infer the amount of protein stably localised (and thus presumed bound) at junctions and comparing this in different conditions. If the recovery is 0.8 in two different conditions, this is in fact not the value we are interested in, since we want to get an idea of the amount of stable protein present in the junction. We have added Figure 1B to clarify this.
In the legend ofFigure 2, calculation of “stable amount” could be better explained and does this data fit with results shown inFigures 5Cand C'.
Thinking aboutFigure 4and the disparity of the effects of Fj on Ds and Ft: the effects measured would depend on locality in the gradient (Figure 5). Indeed in vivo results depended on where the clone was made (Brittle et al.). And we see here the effect of Fj varies across the wing. Thus where in the wing these measurements are made is likely to be an important factor but it is not controlled here?
FRAP experiments are always carried out in the same region of the wing for this reason and are thus controlled. This has now been made clear in the text.
It might be useful for the authors to consider the possibility that phosphorylation may not be all or none (per Ds or Ft molecule) but could be partial. Would that possibility change interpretations?
This is an interesting idea and in terms of the model could significantly increase the number of possible complexes we need to consider. For example, if Ds has three phosphorylation sites and we look at all possible combinations of none, 1, 2 or 3 sites being phosphorylated, there are 8 possible Ds configurations. Since we don’t believe that this would change the outcomes of the model and there is no experimental evidence to give us an idea of binding strengths of these molecules in various combinations, this is beyond the scope of our current work and not adding value given the degree of uncertainty involved. We have added a discussion point to our manuscript (see the sentence beginning with “An interesting experimental observation…”, in the Discussion).
Reviewer 2:
1) The authors only measure the Fj gradient in one quadrant of the wing pouch (Figure 1C). What about the other regions (quadrants) of the wing pouch? What are the gradients of Fj? If these are used in the model, can these gradients produce the same asymmetries in Fat and Ds (using same parameters for everything else)?
Our study attempts to address the general question of how a gradient can be turned into a cellular asymmetry rather than a study of the wing overall. Thus we have chosen to use a region where the asymmetries are clear in order to approach this issue. Other areas of the wing clearly have different asymmetries/gradients, but are usually weaker and thus would give less robust results. In the modeling framework a steeper gradient does give a stronger cellular asymmetry as one might expect.
Reviewer 3:
1) The authors' data indicates that Ft-Ds binding is less stable in the absence of Fj. The one counter-example I found in the literature was rather less direct, but in Matakatsu and Blair (2012) the Hippo signaling boundary effects created by Ft overexpression extended further distally in the absence of Fj than its presence, suggesting stronger Ft-Ds binding in the absence of Fj. Is it possible that the result depends as well on the levels of Ft or Ds being expressed? That might make one a bit cautious about results at anything other than endogenous expression levels.
Ft overexpression in clones will recruit Ds from neighbouring wildtype cells thus causing inversions on the distal side of clones by relocalising Ds to proximal cell edges. This relocalisation of Ds is against the effect of the Fj gradient (which is promoting distal localisation of Ds), so removal of Fj will increase Ft-Ds asymmetry in tissue on the distal side of Ft overexpression clones: see for instance the increased non-autonomy around Ft and Ds over-expression clones in a fj background in the abdomen in Casal et al., 2006. We therefore infer that increased Ft asymmetry, rather than stability of binding, is the driving factor for Dachs regulation and Hippo activation.
2) If the authors think that the polarization of Ft, Ds and Dachs in the distal wing/pouch is due entirely to Fj, has anyone demonstrated that?Brittle et al. (2012)only looked at dachs in fj ds double mutants, I believe. Bosveld looked at fj mutants but in the notum, not wing, and to my eye there was still some Ds polarization.
In the proximal wing we agree that Fj is not the only mechanism, in part (as pointed out by the reviewer) because we've previously reported residual Dachs asymmetry in a fj background in this region. In more distal regions we do not have any definitive data. Therefore we have expanded our discussion to reinforce the view that a Ds boundary may also be important, particularly in the proximal wing.
Reviewer 4:
In this study, FRAP is used to monitor changes in putative interactions between proteins that occur between two cells and thus in separate membranes. Under these specialized conditions, one might intuitively expect diffusion of both binding partners to be slowed, as the authors suggest. However, it seems likely that other types of interactions would be required to cause immobilization per se. The authors suggest that cis interactions may be important (eighth paragraph of the Discussion section). Other types of events, such as interactions of proteins with the actin cytoskeleton might be important as an immobilization mechanism and/or the establishment of “stable” populations. It appears that such a mechanism was previously suggested to be important for the case of E-cadherin in puncta (Cavey et al., Nature, 2008, 453:751-6, A two-tiered mechanism for stabilization and immobilization of E-cadherin). This possibility should be discussed.
On a related note, the authors report that Ft and Ds localize in part to puncta where the proteins appear to be immobilized but do not say very much about the possible identity of the puncta or their relationship to other puncta previously identified. These points should be further developed in the Discussion.
This is an intriguing point and we have expanded our discussion of possible origins of puncta and how they differ from other puncta observed at cell junctions (see the sentence, in the Results, beginning with “Additionally, mechanisms such as…”, as well as the sentence, in the Discussion, “Previous observations have suggested that Ft-Ds puncta…”.
We were also asked to include the few items of data originally referred to as “data not shown” (which we omitted in order to avoid a long Supplemental Data section that might try the patience of readers). This is now in Figure 3.
A number of other comments were made by reviewers regarding the mathematical model presented here. We would like to clarify that the model was used to simulate different possible scenarios of the Fj mechanism, giving qualitative outcomes of Ft-Ds binding. Several other models have been previously published to investigate Ft-Ds polarity and Reviewer 4 specifically mentions Jolly et al., 2014. Other papers also include Mani et al., PNAS, 2012 and Abley et al., Development, 2013, all three of which only consider one possible complex forming, essentially equivalent in our model to complex A (FtP–Ds). This means that in each study the authors needed to implement feedback mechanisms to avoid non-uniform levels of bound material across the tissue. We have added citations to all three publications and a description of this key difference (see Materials and methods). Reviewer 2 noticed subtle differences in the modelled cellular asymmetries across the tissue presented in Figure 5A. Upon further examination we found that this was an artifact of the presentation of the model implementation, and we have thus replaced the figure, which we hope appears clearer. Regardless, there is no significant difference in asymmetries across the tissue in any of our modeling figures, furthermore, we now indicate the average cellular asymmetry for each graph in the figure legends. There was some confusion over whether this was an ODE or PDE model. Since we have stated that we included a diffusion term in our equations, one might presume that we are considering how molecule concentrations vary over space as well as time, thus requiring PDEs. We have now renamed this a redistribution term, since we have simply redistributed unbound molecules across the two membranes of each cell within our ODE model.
Finally, other mechanisms such as amplification, production/degradation and trafficking have not been explored in the current version of the model. Also, since this is not presented as a modelling paper, we have not presented a full exploration of parameter space. We plan to address these issues in a separate theoretical work at a later date.
Sustainable biofuel production from renewable biomass will require the efficient and complete use of all abundant sugars in the plant cell wall. Using the cellulolytic fungus Neurospora crassa as a model, we identified a xylodextrin transport and consumption pathway required for its growth on hemicellulose. Reconstitution of this xylodextrin utilization pathway in Saccharomyces cerevisiae revealed that fungal xylose reductases act as xylodextrin reductases, producing xylosyl-xylitol oligomers as metabolic intermediates. These xylosyl-xylitol intermediates are generated by diverse fungi and bacteria, indicating that xylodextrin reduction is widespread in nature. Xylodextrins and xylosyl-xylitol oligomers are then hydrolyzed by two hydrolases to generate intracellular xylose and xylitol. Xylodextrin consumption using a xylodextrin transporter, xylodextrin reductases and tandem intracellular hydrolases in cofermentations with sucrose and glucose greatly expands the capacity of yeast to use plant cell wall-derived sugars and has the potential to increase the efficiency of both first-generation and next-generation biofuel production.
DOI:http://dx.doi.org/10.7554/eLife.05896.001
10.7554/eLife.05896.002eLife digest
Plants can be used to make ‘biofuels’, which are more sustainable alternatives to traditional fuels made from petroleum. Unfortunately, most biofuels are currently made from simple sugars or starch extracted from parts of plants that we also use for food, such as the grains of cereal crops.
Making biofuels from the parts of the plant that are not used for food—for example, the stems or leaves—would enable us to avoid a trade-off between food and fuel production. However, most of the sugars in these parts of the plant are locked away in the form of large, complex carbohydrates called cellulose and hemicellulose, which form the rigid cell wall surrounding each plant cell.
Currently, the industrial processes that can be used to make biofuels from plant cell walls are expensive and use a lot of energy. They involve heating or chemically treating the plant material to release the cellulose and hemicellulose. Then, large quantities of enzymes are added to break these carbohydrates down into simple sugars that can then be converted into alcohol (a biofuel) by yeast.
Fungi may be able to provide us with a better solution. Many species are able to grow on plants because they can break down cellulose and hemicellulose into simple sugars they can use for energy. If the genes involved in this process could be identified and inserted into yeast it may provide a new, cheaper method to make biofuels from plant cell walls.
To address this challenge, Li et al. studied how the fungus Neurospora crassa breaks down hemicellulose. This study identified a protein that can transport molecules of xylodextrin—which is found in hemicellulose—into the cells of the fungus, and two enzymes that break down the xylodextrin to make simple sugars, using a previously unknown chemical intermediate. When Li et al. inserted the genes that make the transport protein and the enzymes into yeast, the yeast were able to use plant cell wall material to make simple sugars and convert these to alcohol.
The yeast used more of the xylodextrin when they were grown with an additional source of energy, such as the sugars glucose or sucrose. Li et al.'s findings suggest that giving yeast the ability to break down hemicellulose has the potential to improve the efficiency of biofuel production. The next challenge will be to improve the process so that the yeast can convert the xylodextrin and simple sugars more rapidly.
DOI:http://dx.doi.org/10.7554/eLife.05896.002
Author keywordsxylosehemicellulosebiofuelxylodextrincofermentationxylosyl-xylitolResearch organismB. subtilisE. coliN. crassaS. cerevisiaeOtherhttp://dx.doi.org/10.13039/100006978University of California BerkeleyEnergy Biosciences InstituteLiXinYuVivian YaciLinYupingChomvongKulikaEstrelaRaíssaParkAnnseaLiangJulie MZnameroskiElizabeth AFeehanJoannaKimSoo RinJinYong-SuGlassN LouiseCateJamie HDhttp://dx.doi.org/10.13039/501100003593Conselho Nacional de Desenvolvimento Científico e TecnológicoEstrelaRaíssaThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementParallel pathways of xylose metabolism identified in fungi that grow on plants can be used in yeast to promote hemicellulose conversion to biofuels.Introduction
The biological production of biofuels and renewable chemicals from plant biomass requires an economic way to convert complex carbohydrate polymers from the plant cell wall into simple sugars that can be fermented by microbes (Carroll and Somerville, 2009; Chundawat et al., 2011). In current industrial methods, cellulose and hemicellulose, the two major polysaccharides found in the plant cell wall (Somerville et al., 2004), are generally processed into monomers of glucose and xylose, respectively (Chundawat et al., 2011). In addition to harsh pretreatment of biomass, large quantities of cellulase and hemicellulase enzyme cocktails are required to release monosaccharides from plant cell wall polymers, posing unsolved economic and logistical challenges (Lynd et al., 2002; Himmel et al., 2007; Jarboe et al., 2010; Chundawat et al., 2011). The bioethanol industry currently uses the yeast Saccharomyces cerevisiae to ferment sugars derived from cornstarch or sugarcane into ethanol (Hong and Nielsen, 2012), but S. cerevisiae requires substantial engineering to ferment sugars derived from plant cell walls such as cellobiose and xylose (Kuyper et al., 2005; Jeffries, 2006; van Maris et al., 2007; Ha et al., 2011; Hong and Nielsen, 2012; Young et al., 2014).
Results
In contrast to S. cerevisiae, many cellulolytic fungi including Neurospora crassa (Tian et al., 2009) naturally grow well on the cellulose and hemicellulose components of the plant cell wall. By using transcription profiling data (Tian et al., 2009) and by analyzing growth phenotypes of N. crassa knockout strains, we identified separate pathways used by N. crassa to consume cellodextrins (Galazka et al., 2010) and xylodextrins released by its secreted enzymes (Figure 1A and Figure 1—figure supplement 1). A strain carrying a deletion of a previously identified cellodextrin transporter (CDT-2, NCU08114) (Galazka et al., 2010) was unable to grow on xylan (Figure 1—figure supplement 2), and xylodextrins remained in the culture supernatant (Figure 1—figure supplement 3). As a direct test of transport function of CDT-2, S. cerevisiae strains expressing cdt-2 were able to import xylobiose, xylotriose, and xylotetraose (Figure 1—figure supplement 4). Notably, N. crassa expresses a putative intracellular β-xylosidase, GH43-2 (NCU01900), when grown on xylan (Sun et al., 2012). Purified GH43-2 displayed robust hydrolase activity towards xylodextrins with a degree of polymerization (DP) spanning from 2 to 8, and with a pH optimum near 7 (Figure 1—figure supplement 5). The results with CDT-2 and GH43-2 confirm those obtained independently in Cai et al. (2014). As with cdt-1, orthologues of cdt-2 are widely distributed in the fungal kingdom (Galazka et al., 2010), suggesting that many fungi consume xylodextrins derived from plant cell walls. Furthermore, as with intracellular β-glucosidases (Galazka et al., 2010), intracellular β-xylosidases are also widespread in fungi (Sun et al., 2012) (Figure 1—figure supplement 6).10.7554/eLife.05896.003
Consumption of xylodextrins by engineered <italic>S. cerevisiae</italic>.
(A) Two oligosaccharide components derived from the plant cell wall. Cellodextrins, derived from cellulose, are a major source of glucose. Xylodextrins, derived from hemicellulose, are a major source of xylose. The 6-methoxy group (blue) distinguishes glucose derivatives from xylose. R1, R2 = H, cellobiose or xylobiose; R1 = β-1,4-linked glucose monomers in cellodextrins of larger degrees of polymerization; R2 = β-1,4-linked xylose monomers in xylodextrins of larger degrees of polymerization. (B) Xylose and xylodextrins remaining in a culture of S. cerevisiae grown on xylose and xylodextrins and expressing an XR/XDH xylose consumption pathway, CDT-2, and GH43-2, with a starting cell density of OD600 = 1 under aerobic conditions. (C) Xylose and xylodextrins in a culture as in (B) but with a starting cell density of OD600 = 20. In both panels, the concentrations of xylose (X1) and xylodextrins with higher DPs (X2–X5) remaining in the culture broth after different periods of time are shown. All experiments were conducted in biological triplicate, with error bars representing standard deviations.
DOI:http://dx.doi.org/10.7554/eLife.05896.003
10.7554/eLife.05896.004
Transcriptional levels of transporters expressed in <italic>N. crassa</italic> grown on different carbon sources.
Transcript levels reported in fragments per kilobase per million reads (FPKM) are derived from experiments published in Coradetti et al. (2012); Sun et al. (2012). *CBT-1 transports cellobionic acid, the product of lytic polysaccharide monooxygenases (LPMOs, or CaZy family AA9 and AA10) (Xiong et al., 2014).
DOI:http://dx.doi.org/10.7554/eLife.05896.004
10.7554/eLife.05896.005
Growth of <italic>N. crassa</italic> strains on different carbon sources.
(A) Wild-type (WT) N. crassa, or N. crassa with deletions of transporters cdt-1 (Δcdt-1) or cdt-2 (Δcdt-2), were grown on M. giganteus plant cell walls, or purified plant cell wall components. Avicel is a form of cellulose derived from plant cell walls. The black box shows the severe growth phenotype of the Δcdt-2 strain grown on xylan medium. (B) N. crassa biomass accumulation after 3 days of growth on xylan.
DOI:http://dx.doi.org/10.7554/eLife.05896.005
10.7554/eLife.05896.006
Xylodextrins in the xylan culture supernatant of the <italic>N. crassa</italic> Δ<italic>cdt-2</italic> strain.
25 µl of 1:200 diluted N. crassa xylan culture supernantant was analyzed by HPAEC on a CarboPac PA200 column. While no detectable soluble sugars were found in the culture supernatant of the wild-type strain (magenta line), the Δcdt-2 strain (blue line) left a high concentration of unmodified and modified xylodextrins in the culture supernatant. Little xylose was found, indicating xylose was transported by means of different transporters.
DOI:http://dx.doi.org/10.7554/eLife.05896.006
10.7554/eLife.05896.007
Transport of xylodextrins into the cytoplasm of <italic>S. cerevisiae</italic> strains expressing <italic>N. crassa</italic> transporters.
The starting xylodextrin concentration for each purified component was 100 µM. The remaining xylose (X1) and xylodextrins in the culture media are shown for experiments with S. cerevisiae harboring an empty expression plasmid (vector), or with S. cerevisiae individually expressing transporters CDT-1 or CDT-2. Xylodextrins used include xylobiose (X2), xylotriose (X3), xylotetraose (X4), and xylopentaose (X5). Error bars indicate standard deviations of biological triplicates.
DOI:http://dx.doi.org/10.7554/eLife.05896.007
10.7554/eLife.05896.008
Xylobiase activity of the predicted β-xylosidase GH43-2.
(A) GH43-2 hydrolysis of xylodextrins with degrees of polymerization from at least 2–8 (X2–X8). The 30 min chromatogram is offset for clarity. (B) The pH optimum of GH43-2, determined by measuring the extent of hydrolysis of xylobiose to xylose. The HPAEC chromatogram peak area change for xylose is shown.
DOI:http://dx.doi.org/10.7554/eLife.05896.008
10.7554/eLife.05896.009
Phylogenetic distribution of predicted intracellular β-xylosidases GH43-2 in filamentous fungi.
Homologs of GH43-2 (NCU01900) were found with BLAST (Altschul et al., 1997) queries of respective sequence against NCBI protein database. Representative sequences from a diversified taxonomy were chosen and aligned with the MUSCLE algorithm (Edgar 2004). A maximum likelihood phylogenetic tree was calculated based on the alignment with the Jones-Taylor-Thornton model by using software MEGA v6.05 (Tamura et al., 2013). Xylan-induced extracellular GH43-3 (NCU05965) was used as an outgroup. The NCBI GI numbers of the sequences used to build the phylogenetic tree were indicated besides the species names. 1000 bootstrap replicates were performed to calculate the supporting values shown on the branches. The scale bar indicates 0.2 substitutions per amino acid residue.
DOI:http://dx.doi.org/10.7554/eLife.05896.009
10.7554/eLife.05896.010
Xylodextrin consumption profiles of <italic>S. cerevisiae</italic> strains lacking the xylodextrin pathway.
Shown are the concentrations of the remaining sugars in the culture broth after different periods of time of (A) the WT D452-2 strain with starting cell density at OD600 = 1, (B) D452-2 with a S. stipitis xylose utilization pathway (plasmid pLNL78, Table 1) with a starting cell density at OD600 = 1, (C) WT D452-2 strain with a starting cell density at OD600 = 20, and (D) D452-2 with a S. stipitis xylose utilization pathway (plasmid pLNL78) with a starting cell density at OD600 = 20. In all panels, xylose (X1) and xylodextrins of higher DPs (X2–X5) are shown. Error bars represent standard deviations of biological triplicates (panels A–D).
DOI:http://dx.doi.org/10.7554/eLife.05896.010
Cellodextrins and xylodextrins derived from plant cell walls are not catabolized by wild-type S. cerevisiae (Matsushika et al., 2009; Galazka et al., 2010; Young et al., 2010). Reconstitution of a cellodextrin transport and consumption pathway from N. crassa in S. cerevisiae enabled this yeast to ferment cellobiose (Galazka et al., 2010). We therefore reasoned that expression of a functional xylodextrin transport and consumption system from N. crassa might further expand the capabilities of S. cerevisiae to utilize plant-derived xylodextrins. Previously, S. cerevisiae was engineered to consume xylose by introducing xylose isomerase (XI), or by introducing xylose reductase (XR) and xylitol dehydrogenase (XDH) (Jeffries, 2006; van Maris et al., 2007; Matsushika et al., 2009). To test whether S. cerevisiae could utilize xylodextrins, a S. cerevisiae strain was engineered with the XR/XDH pathway derived from Scheffersomyces stipitis—similar to that in N. crassa (Sun et al., 2012)—and a xylodextrin transport (CDT-2) and consumption (GH43-2) pathway from N. crassa. The xylose utilizing yeast expressing CDT-2 along with the intracellular β-xylosidase GH43-2 was able to directly utilize xylodextrins with DPs of 2 or 3 (Figure 1B and Figure 1—figure supplement 7).
Notably, although high cell density cultures of the engineered yeast were capable of consuming xylodextrins with DPs up to 5, xylose levels remained high (Figure 1C), suggesting the existence of severe bottlenecks in the engineered yeast. These results mirror those of a previous attempt to engineer S. cerevisiae for xylodextrin consumption, in which xylose was reported to accumulate in the culture medium (Fujii et al., 2011). Analyses of the supernatants from cultures of the yeast strains expressing CDT-2, GH43-2 and the S. stipitis XR/XDH pathway surprisingly revealed that the xylodextrins were converted into xylosyl-xylitol oligomers, a set of previously unknown compounds rather than hydrolyzed to xylose and consumed (Figure 2A and Figure 2—figure supplement 1). The resulting xylosyl-xylitol oligomers were effectively dead-end products that could not be metabolized further.10.7554/eLife.05896.011
Production and enzymatic breakdown of xylosyl-xylitol.
(A) Structures of xylosyl-xylitol and xylosyl-xylosyl-xylitol. (B) Computational docking model of xylobiose to CtXR, with xylobiose in yellow, NADH cofactor in magenta, protein secondary structure in dark green, active site residues in bright green and showing side-chains. Part of the CtXR surface is shown to depict the shape of the active site pocket. Black dotted lines show predicted hydrogen bonds between CtXR and the non-reducing end residue of xylobiose. (C) Production of xylosyl-xylitol oligomers by N. crassa xylose reductase, XYR-1. Xylose, xylodextrins with DP of 2–4, and their reduced products are labeled X1–X4 and xlt1–xlt4, respectively. (D) Hydrolysis of xylosyl-xylitol by GH43-7. A mixture of 0.5 mM xylobiose and xylosyl-xylitol was used as substrates. Concentration of the products and the remaining substrates are shown after hydrolysis. (E) Phylogeny of GH43-7. N. crassa GH43-2 was used as an outgroup. 1000 bootstrap replicates were performed to calculate the supporting values shown on the branches. The scale bar indicates 0.1 substitutions per amino acid residue. The NCBI GI numbers of the sequences used to build the phylogenetic tree are indicated beside the species names. (F) Activity of two bacterial GH43-7 enzymes from B. subtilis (BsGH43-7) and E. coli (EcGH43-7).
DOI:http://dx.doi.org/10.7554/eLife.05896.011
10.7554/eLife.05896.012
Xylosyl-xylitol oligomers generated in yeast cultures with xylodextrins as the sole carbon source.
(A) Carbohydrates from culture supernatants of strain SR8U expressing CDT-2 and GH43-2 (plasmid pXD8.4), resolved by HPAEC, abbreviated as follows: X1, xylose; X2, xylobiose; X3, xylotriose; X4, xylotetraose; xlt, xylitol; xlt2, xylosyl-xylitol; xlt3, xylosyl-xylosyl-xylitol. (B) LC-MS and LC-MS/MS spectra for xylosyl-xylitol. High-resolution MS spectra show m/z ratios for the negative ion mode. The deprotonated and formate adduct ions were determined with an accuracy of 0.32 and 0.33 ppm, respectively. The MS/MS spectrum in the lower panel shows the product ion matching the predicted fragment. The parental ion, [xylosyl-xylitol + H]−, is denoted with the black diamond mark. (C) LC-MS and LC-MS/MS spectra for xylosyl-xylosyl-xylitol. The deprotonated and formate adduct ions were determined with an accuracy of 0.51 and 0.37 ppm, respectively. The MS/MS spectrum in the lower panel shows the product ions matching the predicted fragments. The parental ion, [xylosyl-xylosyl-xylitol + H]−, is denoted with the black diamond mark.
DOI:http://dx.doi.org/10.7554/eLife.05896.012
10.7554/eLife.05896.013
Xylodextrin metabolism by a co-culture of yeast strains to identify enzymatic source of xylosyl-xylitol.
A mixture of a xylose utilizing strain (SR8) with a cell density at OD600 = 1.0 and a xylodextrin hydrolyzing strain (D452-2 expressing CDT-2 and GH43-2 from plasmid pXD8.4) with a cell density at OD600 = 20 was co-cultured in a medium containing 2% xylodextrin. Xylobiose (X2) and xylotriose (X3) decreased, whereas xylose (X1) initially increased. Subsequent X1 consumption correlated with production of xylitol. Notably, xylosyl-xylitol oligomers were not detected, suggesting that the xylodextrin reductase activity was present only in the xylose-fermenting strain expressing XR. Error bars represent standard deviations of biological triplicates.
DOI:http://dx.doi.org/10.7554/eLife.05896.013
10.7554/eLife.05896.014
Chromatogram of xylosyl-xylitol hydrolysis products generated by <bold>β</bold>-xylosidases.
Reaction products from the enzymatic assays in Figure 2D were resolved by ion-exclusion HPLC. Peak areas were used to quantify the concentration of substrates and products at the end of the reaction.
DOI:http://dx.doi.org/10.7554/eLife.05896.014
Since the production of xylosyl-xylitol oligomers as intermediate metabolites has not been reported, the molecular components involved in their generation were examined. To test whether the xylosyl-xylitol oligomers resulted from side reactions of xylodextrins with endogenous S. cerevisiae enzymes, we used two separate yeast strains in a combined culture, one containing the xylodextrin hydrolysis pathway composed of CDT-2 and GH43-2, and the second with the XR/XDH xylose consumption pathway. The strain expressing CDT-2 and GH43-2 would cleave xylodextrins to xylose, which could then be secreted via endogenous transporters (Hamacher et al., 2002) and serve as a carbon source for the strain expressing the xylose consumption pathway (XR and XDH). The engineered yeast expressing XR and XDH is only capable of consuming xylose (Figure 1B). When co-cultured, these strains consumed xylodextrins without producing the xylosyl-xylitol byproduct (Figure 2—figure supplement 2). These results indicate that endogenous yeast enzymes and GH43-2 transglycolysis activity are not responsible for generating the xylosyl-xylitol byproducts, that is, that they must be generated by the XR from S. stipitis (SsXR).
Fungal xylose reductases such as SsXR have been widely used in industry for xylose fermentation. However, the structural details of substrate binding to the XR active site have not been established. To explore the molecular basis for XR reduction of oligomeric xylodextrins, the structure of Candida tenuis xylose reductase (CtXR) (Kavanagh et al., 2002), a close homologue of SsXR, was analyzed. CtXR contains an open active site cavity where xylose could bind, located near the binding site for the NADH co-factor (Kavanagh et al., 2002; Kratzer et al., 2006). Notably, the open shape of the active site can readily accommodate the binding of longer xylodextrin substrates (Figure 2B). Using computational docking algorithms (Trott and Olson, 2010), xylobiose was found to fit well in the pocket. Furthermore, there are no obstructions in the protein that would prevent longer xylodextrin oligomers from binding (Figure 2B).
We reasoned that if the xylosyl-xylitol byproducts are generated by fungal XRs like that from S. stipitis, similar side products should be generated in N. crassa, thereby requiring an additional pathway for their consumption. Consistent with this hypothesis, xylose reductase XYR-1 (NCU08384) from N. crassa also generated xylosyl-xylitol products from xylodextrins (Figure 2C). However, when N. crassa was grown on xylan, no xylosyl-xylitol byproduct accumulated in the culture medium (Figure 1—figure supplement 3). Thus, N. crassa presumably expresses an additional enzymatic activity to consume xylosyl-xylitol oligomers. Consistent with this hypothesis, a second putative intracellular β-xylosidase upregulated when N. crassa was grown on xylan, GH43-7 (NCU09625) (Sun et al., 2012), had weak β-xylosidase activity but rapidly hydrolyzed xylosyl-xylitol into xylose and xylitol (Figure 2D and Figure 2—figure supplement 3). The newly identified xylosyl-xylitol-specific β-xylosidase GH43-7 is widely distributed in fungi and bacteria (Figure 2E), suggesting that it is used by a variety of microbes in the consumption of xylodextrins. Indeed, GH43-7 enzymes from the bacteria Bacillus subtilis and Escherichia coli cleave both xylodextrin and xylosyl-xylitol (Figure 2F).
To test whether xylosyl-xylitol is produced generally by microbes as an intermediary metabolite during their growth on hemicellulose, we extracted and analyzed the metabolites from a number of ascomycetes species and B. subtilis grown on xylodextrins. Notably, these widely divergent fungi and B. subtilis all produce xylosyl-xylitols when grown on xylodextrins (Figure 3A and Figure 3—figure supplement 1). These organisms span over 1 billion years of evolution (Figure 3B), indicating that the use of xylodextrin reductases to consume plant hemicellulose is widespread.10.7554/eLife.05896.015
Xylosyl-xylitol and xylosyl-xylosyl-xylitol production by a range of microbes.
(A) Xylodextrin-derived carbohydrate levels seen in chromatograms of intracellular metabolites for N. crassa, T. reesei, A. nidulans and B. subtilis grown on xylodextrins. Compounds are abbreviated as follows: X1, xylose; X2, xylobiose; X3, xylotriose; X4, xylotetraose; xlt, xylitol; xlt2, xylosyl-xylitol; xlt3, xylosyl-xylosyl-xylitol. (B) Phylogenetic tree of the organisms shown to produce xylosyl-xylitols during growth on xylodextrins. Ages taken from Wellman et al. (2003); Galagan et al. (2005); Hedges et al. (2006).
DOI:http://dx.doi.org/10.7554/eLife.05896.015
10.7554/eLife.05896.016
LC-MS/MS multiple reaction monitoring chromatograms of xylosyl-xylitols from cultures of microbes grown on xylodextrins.
Shown are MS/MS transitions for xylosyl-xylitol (in red, m/z 283.1035 → 151.0612 transition) and xylosyl-xylosyl-xylitol (in green, m/z 415.1457 → 151.0612 transition) analyzed from intracellular metabolites of N. crassa, T. reesei, A. nidulans and B. subtilis grown on xylodextrins, after separation by liquid chromatography.
DOI:http://dx.doi.org/10.7554/eLife.05896.016
We next tested whether integration of the complete xylodextrin consumption pathway would overcome the poor xylodextrin utilization by S. cerevisiae (Figure 1) (Fujii et al., 2011). When combined with the original xylodextrin pathway (CDT-2 plus GH43-2), GH43-7 enabled S. cerevisiae to grow more rapidly on xylodextrin (Figure 4A) and eliminated accumulation of xylosyl-xylitol intermediates (Figure 4B–D and Figure 4—figure supplement 1). The presence of xylose and glucose greatly improved anaerobic fermentation of xylodextrins (Figure 5 and Figure 5—figure supplement 1 and Figure 5—figure supplement 2), indicating that metabolic sensing in S. cerevisiae with the complete xylodextrin pathway may require additional tuning (Youk and van Oudenaarden, 2009) for optimal xylodextrin fermentation. Notably, we observed that the XR/XDH pathway produced much less xylitol when xylodextrins were used in fermentations than from xylose (Figure 5 and Figure 5—figure supplement 2B). Taken together, these results reveal that the XR/XDH pathway widely used in engineered S. cerevisiae naturally has broad substrate specificity for xylodextrins, and complete reconstitution of the naturally occurring xylodextrin pathway is necessary to enable S. cerevisiae to efficiently consume xylodextrins.10.7554/eLife.05896.017
Aerobic consumption of xylodextrins with the complete xylodextrin pathway.
(A) Yeast growth curves with xylodextrin as the sole carbon source under aerobic conditions with a cell density at OD600 = 1. Yeast strain SR8U without plasmids, or transformed with plasmid expressing CDT-2 and GH43-2 (pXD8.4), CDT-2 and GH43-7 (pXD8.6) or all three genes (pXD8.7) are shown. (B–D) Xylobiose consumption with xylodextrin as the sole carbon source under aerobic conditions with a cell density of OD600 = 20. Xylosyl-xylitol (xlt2) accumulation was only observed in the SR8U strain bearing plasmid pXD8.4, that is, lacking GH43-7. Error bars represent standard deviations of biological triplicates (panels A–D).
DOI:http://dx.doi.org/10.7554/eLife.05896.017
10.7554/eLife.05896.018
Culture media composition during yeast growth on xylodextrin.
Yeast growth with xylodextrin as the sole carbon source (concentration g/l) under aerobic conditions with a cell density at OD600 = 20. Yeast strain SR8 transformed with plasmid expressing CDT-2 and GH43-2 (pXD8.4), CDT-2 and GH43-7 (pXD8.6), or all three genes (pXD8.7). All growth experiments were performed in biological triplicate, and error bars indicate the standard deviation between experiments.
DOI:http://dx.doi.org/10.7554/eLife.05896.018
10.7554/eLife.05896.019
Anaerobic fermentation of xylodextrins in co-fermentations with xylose or glucose.
(A) Anaerobic fermentation of xylodextrins and xylose, in a fed-batch reactor. Strain SR8U expressing CDT-2, GH43-2, and GH43-7 (plasmid pXD8.7) was used at an initial OD600 of 20. Solid lines represent concentrations of compounds in the media. Blue dotted line shows the total amount of xylose added to the culture over time. Error bars represent standard deviations of biological duplicates. (B) Anaerobic fermentation of xylodextrins and glucose, in a fed-batch reactor. Glucose was not detected in the fermentation broth. Error bars represent standard deviations of biological duplicates.
DOI:http://dx.doi.org/10.7554/eLife.05896.019
10.7554/eLife.05896.020
Anaerobic xylodextrin utilization in the presence of xylose.
Strain carrying the complete xylodextrin pathway (CDT-2, GH43-2, GH43-7, XR/XDH) grown under anaerobic conditions in oMM media (Lin et al., 2014) containing 4% xylose and 3% xylodextrin. The consumption of xylobiose (X2) and xylotriose (X3) stalled when xylose (X1) was depleted and resumed after supplying additional xylose at hour 48. This experiment is representative of those carried out with different xylose to xylodextrin ratios.
DOI:http://dx.doi.org/10.7554/eLife.05896.020
10.7554/eLife.05896.021
Control anaerobic fermentations with <italic>S. cerevisiae</italic> strain expressing the complete xylodextrin utilization pathway.
Strain SR8U with plasmid pXD8.7 expressing CDT-2, GH43-2, and GH43-7 was used at an initial OD600 of 20. Solid lines represent concentrations of compounds in the media. Blue dotted line shows the total amount of xylose added to the culture over time. (A) Fermentation profile of the strain in oMM medium containing 4% xylodextrin in the reactor without feeding xylose. (B) Fermentation profile of the strain in oMM medium without xylodextrin in the reactor but with continuous xylose feeding.
DOI:http://dx.doi.org/10.7554/eLife.05896.021
The observation that xylodextrin fermentation was stimulated by glucose (Figure 5B) suggested that the xylodextrin pathway could serve more generally for cofermentations to enhance biofuel production. We therefore tested whether xylodextrin fermentation could be carried out simultaneously with sucrose fermentation, as a means to augment ethanol yield from sugarcane. In this scenario, xylodextrins released by hot water treatment (Hendriks and Zeeman, 2009; Agbor et al., 2011; Vallejos et al., 2012) could be added to sucrose fermentations using yeast engineered with the xylodextrin consumption pathway. To test this idea, we used strain SR8U engineered with the xylodextrin pathway (CDT-2, GH43-2, and GH43-7) in fermentations combining sucrose and xylodextrins. We observe simultaneous fermentation of sucrose and xylodextrins, with increased ethanol yields (Figure 6). Notably, the levels of xylitol production were found to be low (Figure 6), as observed in cofermentations with glucose (Figure 5B).10.7554/eLife.05896.022
Xylodextrin and sucrose co-fermentations.
(A) Sucrose fermentation. Vertical axis, g/l; horizontal axis, time in hours. (B) Xylodextrin and sucrose batch co-fermentation using strain SR8U expressing CDT-2, GH43-2, and GH43-7 (plasmid pXD8.7). Vertical axis, g/l; horizontal axis, time in hours. The xylodextrins were supplied at 10 g/l which containing xylobiose (4.2 g/l) and xylotriose (2.3 g/l). Not fermented in the timeframe of this experiment, the xylodextrin sample also included xylotetraose and xylopentaose, in addition to hemicellulose modifiers such as acetate.
DOI:http://dx.doi.org/10.7554/eLife.05896.022
Discussion
Using yeast as a test platform, we identified a xylodextrin consumption pathway in N. crassa (Figure 7) that surprisingly involves a new metabolic intermediate widely produced in nature by many fungi and bacteria. In bacteria such as B. subtilis, xylosyl-xylitol may be generated by aldo-keto reductases known to possess broad substrate specificity (Barski et al., 2008). The discovery of the xylodextrin consumption pathway along with cellodextrin consumption (Galazka et al., 2010) in cellulolytic fungi for the two major sugar components of the plant cell wall now provides many modes of engineering yeast to ferment plant biomass-derived sugars (Figure 7). An alternative xylose consumption pathway using xylose isomerase could also be used with the xylodextrin transporter and xylodextrin hydrolase GH43-2 (van Maris et al., 2007). However, the XR/XDH pathway may provide significant advantages in realistic fermentation conditions with sugars derived from hemicellulose. The breakdown of hemicellulose, which is acetylated (Sun et al., 2012), releases highly toxic acetate, degrading the performance of S. cerevisiae fermentations (Bellissimi et al., 2009; Sun et al., 2012). The cofactor imbalance problem of the XR/XDH pathway, which can lead to accumulation of reduced byproducts (xylitol and glycerol) and therefore was deemed a problem, can be exploited to drive acetate reduction, thereby detoxifying the fermentation medium and increasing ethanol production (Wei et al., 2013).10.7554/eLife.05896.023
Two pathways of oligosaccharide consumption in <italic>N. crassa</italic> reconstituted in <italic>S. cerevisiae</italic>.
Intracellular cellobiose utilization requires CDT-1 or CDT-2 along with β-glucosidase GH1-1 (Galazka et al., 2010) and enters glycolysis after phosphorylation by hexokinases (HXK) to form glucose-6-phosphate (Glc-6-P). Intracellular xylodextrin utilization also uses CDT-2 and requires the intracellular β-xylosidases GH43-2 and GH43-7. The resulting xylose can be assimilated through the pentose phosphate pathway consisting of xylose/xylodextrin reductase (XR), xylitol dehydrogenase (XDH), and xylulokinase (XK).
DOI:http://dx.doi.org/10.7554/eLife.05896.023
With optimization, that is, through improvements to xylodextrin transporter performance and chromosomal integration (Ryan et al., 2014), the newly identified xylodextrin consumption pathway provides new opportunities to expand first-generation bioethanol production from cornstarch or sugarcane to include hemicellulose from the plant cell wall. For example, we propose that xylodextrins released from the hemicellulose in sugarcane bagasse by using compressed hot water treatment (Hendriks and Zeeman, 2009; Agbor et al., 2011; Vallejos et al., 2012) could be directly fermented by yeast engineered to consume xylodextrins, as we have shown in proof-of-principle experiments (Figure 6). Xylodextrin consumption combined with glucose or cellodextrin consumption (Figure 7) could also improve next-generation biofuel production from lignocellulosic feedstocks under a number of pretreatment scenarios (Hendriks and Zeeman, 2009; Vallejos et al., 2012). These pathways could find widespread use to overcome remaining bottlenecks to fermentation of lignocellulosic feedstocks as a sustainable and economical source of biofuels and renewable chemicals.
Materials and methods<italic>Neurospora crassa</italic> strains
N. crassa strains obtained from the Fungal Genetics Stock Center (FGSC) (McCluskey et al., 2010) include the WT (FGSC 2489), and deletion strains for the two oligosaccharide transporters: NCU00801 (FGSC 16575) and NCU08114 (FGSC 17868) (Colot et al., 2006).
<italic>Neurospora crassa</italic> growth assays
Conidia were inoculated at a concentration equal to 106 conidia per ml in 3 ml Vogel's media (Vogel, 1956) with 2% wt/vol powdered Miscanthus giganteus (Energy Bioscience Institute, UC-Berkeley), Avicel PH 101 (Sigma-Aldrich, St. Louis, MO), beechwood xylan (Sigma-Aldrich), or pectin (Sigma-Aldrich) in a 24-well deep-well plate. The plate was sealed with Corning breathable sealing tape and incubated at 25°C in constant light and with shaking (200 rpm). Images were taken at 48 hr. Culture supernatants were diluted 200 times with 0.1 M NaOH before Dionex high-performance anion exchange chromatographic (HPAEC) analysis, as described below. N. crassa growth on xylan was also determined by measuring N. crassa biomass accumulation. N. crassa grown on xylan for 3 days was harvested by filtration over a Whatman glass microfiber filter (GF/F) on a Büchner funnel and washed with 50 ml water. Biomass was then collected from the filter, dried in a 70°C oven, and weighed.
Plasmids and yeast strains
Template gDNA from the N. crassa WT strain (FGSC 2489) and from the S. cerevisiae S288C strain was extracted as described in http://www.fgsc.net/fgn35/lee35.pdf (McCluskey et al., 2010). Open reading frames (ORFs) of the β-xylosidase genes NCU01900 and NCU09652 (GH43-2 and GH43-7) were amplified from the N. crassa gDNA template. For biochemical assays, each ORF was fused with a C-terminal His6-tag and flanked with the S. cerevisiae PTEF1 promoter and CYC1 transcriptional terminator in the 2µ yeast plasmid pRS423 backbone. Plasmid pRS426_NCU08114 was described previously (Galazka et al., 2010). Plasmid pLNL78 containing the xylose utilization pathway (xylose reductase, xylitol dehydrogenase, and xylulose kinase) from S. stipitis was obtained from the lab of John Dueber (Latimer et al., 2014). Plasmid pXD2, a single-plasmid form of the xylodextrin pathway, was constructed by integrating NCU08114 (CDT-2) and NCU01900 (GH43-2) expression cassettes into pLNL78, using the In-Fusion Cloning Kit (Clontech). Plasmid pXD8.4 derived from plasmid pRS316 (Sikorski and Hieter, 1989) was used to express CDT-2 and GH43-2, each from the PCCW12 promoter. Plasmid pXD8.6 was derived from pXD8.4 by replacing the GH43-2 ORF with the ORF for GH43-7. pXD8.7 contained all three expression cassettes (CDT-2, GH43-2, and GH43-7) using the PCCW12 promoter for each. S. cerevisiae strain D452-2 (MATaleu2 his3 ura3 can1) (Kurtzman, 1994) and SR8U (the uracil autotrophic version of the evolved xylose fast utilization strain SR8) (Kim et al., 2013) were used as recipient strains for the yeast experiments. The ORF for N. crassa xylose reductase (xyr-1, NcXR) was amplified from N. crassa gDNA and the introns were removed by overlapping PCR. XR ORF was fused to a C-terminal His6-tag and flanked with the S. cerevisiae PCCW12 promoter and CYC1 transcriptional terminator and inserted into plasmid pRS313.
A list of the plasmids used in this study can be found in Table 1.10.7554/eLife.05896.024
S. cerevisiae was grown in an optimized minimum medium (oMM) lacking uracil into late log phase. The oMM contained 1.7 g/l YNB (Sigma-Aldrich, Y1251), twofold appropriate CSM dropout mixture, 10 g/l (NH4)2SO4, 1 g/l MgSO4.7H2O, 6 g/l KH2PO4, 100 mg/l adenine hemisulfate, 10 mg/l inositol, 100 mg/l glutamic acid, 20 mg/l lysine, 375 mg/l serine, and 100 mM 4-morpholineethanesulfonic acid (MES), pH 6.0 (Lin et al., 2014). Cells were then harvested and washed three times with assay buffer (5 mM MES, 100 mM NaCl, pH 6.0) and resuspended to a final OD600 of 40. Substrate stocks were prepared in the same assay buffer at a concentration of 200 μM. Transport assays were initiated by mixing equal volumes of the cell suspension and the substrate stock. Reactions were incubated at 30°C with continuous shaking for 30 min. Samples were centrifuged at 14,000 rpm at 4°C for 5 min to remove yeast cells. 400 μl of each sample supernatant was transferred to an HPLC vial containing 100 μl 0.5 M NaOH, and the concentration of the remaining substrate was measured by HPAEC as described below.
Enzyme purification
S. cerevisiae strains transformed with pRS423_GH43-2, pRS423_GH43-7, or pRS313_NcXR were grown in oMM lacking histidine with 2% glucose until late log phase before harvesting by centrifugation. E. coli strains BL21DE3 transformed with pET302_BsGH43-7 or pET302_EcGH43-7 were grown in TB medium, induced with 0.2 mM IPTG at OD600 of 0.8, and harvested by centrifugation 12 hr after induction. Yeast or E. coli cell pellets were resuspended in a buffer containing 50 mM Tris–HCl, 100 mM NaCl, 0.5 mM DTT, pH 7.4 and protease inhibitor cocktail (Pierce Biotechnology, Rockford, IL). Cells were lysed with an Avestin homogenizer, and the clarified supernatant was loaded onto a HisTrap column (GE Healthcare, Sweden). His-tagged enzymes were purified with an imidazole gradient, buffer-exchanged into 20 mM Tris–HCl, 100 mM NaCl, pH 7.4, and concentrated to 5 mg/ml.
Enzyme assays
For the β-xylosidase assay of GH43-2 with xylodextrins, 0.5 μM of purified enzyme was incubated with 0.1% in-house prepared xylodextrin or 1 mM xylobiose (Megazyme, Ireland) in 1× PBS at 30°C. Reactions were sampled at 30 min and quenched by adding 5 vol of 0.1 M NaOH. The products were analyzed by HPAEC as described below. For pH profiling, acetate buffer at pH 4.0, 4.5, 5.0, 5.5, 6.0, and phosphate buffer at 6.5, 7.0, 7.5, 8 were added at a concentration of 0.1 M. For the β-xylosidase assay of GH43-2 and GH43-7 with xylosyl-xylitol, 10 µM of purified enzyme was incubated with 4.5 mM xylosyl-xylitol and 0.5 mM xylobiose in 20 mM MES buffer, pH = 7.0, and 1 mM CaCl2 at 30°C. Reactions were sampled at 3 hr and quenched by heating at 99°C for 10 min. The products were analyzed by ion-exclusion HPLC as described below.
For the xylose reductase assays of NcXR, 1 μM of purified enzyme was incubated with 0.06% xylodextrin and 2 mM NADPH in 1× PBS at 30°C. Reactions were sampled at 30 min and quenched by heating at 99°C for 10 min. The products were analyzed by LC-QToF as described below.
Oligosaccharide preparation
Xylodextrin was purchased from Cascade Analytical Reagents and Biochemicals or prepared according to published procedures (Akpinar et al., 2009) with slight modifications. In brief, 20 g beechwood xylan (Sigma–Aldrich) was fully suspended in 1000 ml water, to which 13.6 ml 18.4 M H2SO4 was added. The mixture was incubated in a 150°C oil bath with continuous stirring. After 30 min, the reaction was poured into a 2-L plastic container on ice, with stirring to allow it to cool. Then 0.25 mol CaCO3 was slowly added to neutralize the pH and precipitate sulfate. The supernatant was filtered and concentrated on a rotary evaporator at 50°C to dryness. The in-house prepared xylodextrin contained about 30% xylose monomers and 70% oligomers. To obtain a larger fraction of short chain xylodextrin, the commercial xylodextrin was dissolved to 20% wt/vol and incubated with 2 mg/ml xylanase at 37°C for 48 hr. Heat deactivation and filtration were performed before use.
Xylosyl-xylitol was purified from the culture broth of strain SR8-containing plasmids pXD8.4 in xylodextrin medium. 50 ml of culture supernatant was concentrated on a rotary evaporator at 50°C to about 5 ml. The filtered sample was loaded on an XK 16/70 column (GE Healthcare) packed with Supelclean ENVI-Carb (Sigma–Aldrich) mounted on an ÄKTA Purifier (GE Healthcare). The column was eluted with a gradient of acetonitrile at a flow rate of 3.0 ml/min at room temperature. Purified fractions, verified by LC-MS, were pooled and concentrated. The final product, containing 90% of xylosyl-xylitol and 10% xylobiose, was used as the substrate for enzyme assays and as an HPLC calibration standard.
Measurement of xylosyl-xylitol production by fungi and <italic>B. subtilis</italic>
N. crassa strain (FGSC 2489) and Aspergillus nidulans were stored and conidiated on agar slants of Volgel's medium (Vogel, 1956) with 2% glucose. Trichoderma reesei (strain QM6a) was conidiated on potato dextrose agar (PDA) plates. Condia from each fungi were collected by resuspending in water and used for inoculation at a concentration of 106 cells per ml. N. crassa and A. nidulans were inoculated into Volgel's medium with 2% xylodextrin. T. reesei was inoculated into Trichoderma minimal medium (Penttilä et al., 1987) with 2% xylodextrin. N. crassa, A. nidulans, and T. reesei were grown in shaking flasks at 25°C, 37°C, and 30°C respectively. After 40 hr, mycelia from 2 ml of culture were harvested and washed with water on a glass fiber filter and transferred to a pre-chilled screw-capped 2 ml tube containing 0.5 ml Zirconia beads (0.5 mm) and 1.2 ml acidic acetonitrile extraction solution (80% Acetonitrile, 20% H2O, and 0.1 M formic acid, [Rabinowitz and Kimball, 2007]). The tubes were then plunged into liquid nitrogen. The harvest process was controlled within 30 s. Samples were kept at −80°C until extraction, as described below.
B. sublitis was stored on 0.5× LB (1% tryptone, 0.5% yeast extract, and 0.5% NaCl) agar plates. A single colony was inoculated into 0.5× LB liquid medium with 1% glucose and allowed to grow in a 37°C shaker overnight. An inoculum from the overnight culture was transferred to fresh 0.5× LB liquid medium with 1% xylodextrin at an initial OD600 of 0.2. After 40 hr, 2 ml of the culture was spun down and washed with cold PBS solution. Zirconia beads and acidic acetonitrile extraction solution were added to the cell pellet. The tubes were then flash frozen immediately and kept at −80°C until extraction.
For extraction, all samples were allowed to thaw at 4°C for 10 min, bead beat for 2 min, and vortexed at 4°C for 20 min. 50 µl of the supernatant from each sample was analyzed by LC-MS/MS (see ‘Mass spectrometric analyses’ section).
Aerobic yeast cultures with xylodextrins
Yeast strains were pre-grown aerobically overnight in oMM medium containing 2% glucose, washed three times with water, and resuspended in oMM medium. For aerobic growth, strains were inoculated at a starting OD600 of 1.0 or 20 in 50 ml oMM medium with 3% wt/vol xylodextrins and cultivated in 250 ml Erlenmeyer flasks covered with four layers of miracle cloth, shaking at 220 rpm. At the indicated time points, 0.8 ml samples were removed and pelleted. 20 μl supernatants were analyzed by ion-exclusion HPLC to determine xylose, xylitol, glycerol, and ethanol concentrations. 25 μl of 1:200 diluted or 2 μl of 1:100 diluted supernatant was analyzed by HPAEC or LC-QToF, respectively, to determine xylodextrin concentrations.
Fed-batch anaerobic fermentations
Anaerobic fermentation experiments were performed in a 1-L stirred tank bioreactor (DASGIP Bioreactor system, Type DGCS4, Eppendorf AG, Germany), containing oMM medium with 3% wt/vol xylodextrins inoculated with an initial cell concentration of OD600 = 20. The runs were performed at 30°C for 107 hr. The culture was agitated at 200 rpm and purged constantly with 6 l/hr of nitrogen. For xylose plus xylodextrin co-fermentations, xylose was fed continuously at 0.8 ml/hr from a 25% stock. During the fermentation, 3 ml cell-free samples were taken each 4 hr with an autosampler through a ceramic sampling probe (Seg-Flow Sampling System, Flownamics, Madison, WI). 20 μl of the supernatant fraction were analyzed by ion-exclusion HPLC to determine xylose, xylitol, glycerol, acetate, and ethanol concentrations. 2 μl of 1:100 diluted supernatant was analyzed by LC-QToF to determine xylodextrin concentrations. For glucose plus xylodextrin co-fermentations, glucose was fed continuously at 2 ml/hr from a 10% stock. Analytes were detected as described for xylose plus xylodextrin co-fermentations, with the addition of the measurement of glucose concentrations in the culture broth.
Co-fermentation of sucrose plus xylodextrins
Yeast strain SR8U with plasmid pXD8.7 was pre-grown aerobically to late-log phase in oMM medium lacking uracil and containing 2% glucose, washed with water, and resuspended in oMM medium. Media containing 75 g/l sucrose plus or minus 15 g/l xylodextrins were inoculated with 20 OD of the washed yeast seed culture and purged with N2. Fermentations were carried out in 50 ml of oMM medium in 125 ml serum bottles shaking at 220 rpm in a 30°C shaker. At the indicated time points, 1 ml samples were removed and pelleted. 5 μl supernatants were analyzed by ion-exclusion HPLC to determine sucrose, glucose, fructose, xylose, xylitol, glycerol, and ethanol concentrations. 2 μl of 1:100 diluted supernatant was analyzed by LC-QToF, as described below, to determine xylodextrin concentrations.
Ion-exclusion HPLC analysis
Ion-exclusion HPLC was performed on a Prominence HPLC (Shimadzu, Japan) equipped with a refractive index detector. Xylose fermentation samples were resolved on a Rezex RFQ-Fast Fruit H+ 8% column (100 × 7.8 mm, Phenomenex, Torrance, CA) using a flow rate of 1 ml/min at 50°C. Xylodextrin fermentation samples were resolved on Aminex HPX-87H Column (300 × 7.8 mm, Bio-Rad, Hercules, CA) at a flow rate of 0.6 ml/min at 40°C. Both columns used a mobile phase of 0.01 N H2SO4.
HPAEC analysis
HPAEC analysis was performed on a ICS-3000 HPLC (Thermo Fisher, Sunnyvale, CA) using a CarboPac PA200 analytical column (150 × 3 mm) and a CarboPac PA200 guard column (3 × 30 mm) at 30°C. Following injection of 25 μl of diluted samples, elution was performed at 0.4 ml/min using 0.1 M NaOH in the mobile phase with sodium acetate gradients. For xylodextrin and xylosyl-xylitol separation, the acetate gradients were 0 mM for 1 min, increasing to 80 mM in 8 min, increasing to 300 mM in 1 min, keeping at 30 mM for 2 min, followed by re-equilibration at 0 mM for 3 min. Carbohydrates were detected using pulsed amperometric detection (PAD) and peaks were analyzed and quantified using the Chromeleon software package.
Mass spectrometric analyses
All mass spectrometric analyses were performed on an Agilent 6520 Accurate-Mass Q-TOF coupled with an Agilent 1200 LC (Agilent Technologies, Santa Clara, CA). Samples were resolved on a 100 × 7.8 mm Rezex RFQ-Fast Fruit H+ 8% column (Phenomenex) using a mobile phase of 0.5% formic acid at a flow rate of 0.3 ml/min at 55°C.
To determine the accurate masses of the unknown metabolites, 2 µl of 1:100 diluted yeast culture supernatant was analyzed by LC-QToF. Nitrogen was used as the instrument gas. The source voltage (Vcap) was 3000 V in negative ion mode, and the fragmentor was set to 100 V. The drying gas temperature was 300°C; drying gas flow was 7 l/min; and nebulizer pressure was 45 psi. The ESI source used a separate nebulizer for the continuous, low-level introduction of reference mass compounds (112.985587, 1033.988109) to maintain mass axis calibration. Data were collected at an acquisition rate of 1 Hz from m/z 50 to 1100 and stored in centroid mode.
LC-MS/MS was performed to confirm the identity of xylosyl-xylitol and xylosyl-xylosyl-xylitol. The compound with a retention time (RT) of 5.8 min and m/z ratio of 283.103 and the compound with an RT of 4.7 min and m/z ratio of 415.15 were fragmented with collision energies of 10, 20, and 40 eV. MS/MS spectra were acquired, and the product ions were compared and matched to the calculated fragment ions generated by the Fragmentation Tools in ChemBioDraw Ultra v13.
To quantify the carbohydrates and carbohydrate derivatives in the culture, culture supernatants were diluted 100-fold in water and 2 µl was analyzed by LC-QToF. Spectra were imported to Qualtitative Analysis module of Agilent MassHunter Workstation software using m/z and retention time values obtained from the calibration samples to search for the targeted ions in the data. These searches generated extracted ion chromatograms (EICs) based on the list of target compounds. Peaks were integrated and compared to the calibration curves to calculate the concentration. Calibration curves were calculated from the calibration samples, prepared in the same oMM medium as all the samples, and curve fitting for each compound resulted in fits with R2 values of 0.999. 4-morpholineethanesulfonic acid (MES), the buffer compound in the oMM medium with constant concentration and not utilized by yeast, was used as an internal standard (IS) for concentration normalization.
Acknowledgements
We thank L Acosta-Sampson and A Gokhale for helpful discussions, J Dueber for xylose utilization pathway plasmids, Z Baer, J Kuchenreuthe and M Maurer for helps in anaerobic fermentation, and S Bauer and A Ibañez Zamora for help with analytical methods. This work was supported by funding from the Energy Biosciences Institute (JHDC, NLG and YSJ) and by a pre-doctoral fellowship from CNPq and CAPES through the program ‘Ciência sem Fronteiras’ (R E).
Additional informationCompeting interests
XL: A patent application related to some of the work presented here has been filed on behalf of the Regents of the University of California.
JHDC: A patent application related to some of the work presented here has been filed on behalf of the Regents of the University of California.
The other authors declare that no competing interests exist.
Author contributions
XL, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
VYY, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
EAZ, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
JHDC, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
YL, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
KC, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
RE, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
AP, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
JML, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
JF, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
SRK, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents
Y-SJ, Conception and design, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents
NLG, Conception and design, Analysis and interpretation of data, Drafting or revising the article
ReferencesAgborVBCicekNSparlingRBerlinALevinDB2011Biomass pretreatment: fundamentals toward applicationBiotechnology Advances2967568510.1016/j.biotechadv.2011.05.005AkpinarOErdoganKBostanciS2009Production of xylooligosaccharides by controlled acid hydrolysis of lignocellulosic materialsCarbohydrate Research34466066610.1016/j.carres.2009.01.015AltschulSFMaddenTLSchäfferAAZhangJZhangZMillerWLipmanDJ1997Gapped BLAST and PSI-BLAST: a new generation of protein database search programsNucleic Acids Research253389340210.1093/nar/25.17.3389BarskiOATipparajuSMBhatnagarA2008The aldo-keto reductase superfamily and its role in drug metabolism and detoxificationDrug Metabolism Reviews4055362410.1080/03602530802431439BellissimiEvan DijkenJPPronkJTvan MarisAJ2009Effects of acetic acid on the kinetics of xylose fermentation by an engineered, xylose-isomerase-based Saccharomyces cerevisiae strainFEMS Yeast Research935836410.1111/j.1567-1364.2009.00487.xCaiPGuRWangBLiJWanLTianCMaY2014Evidence of a critical role for cellodextrin transporte 2 (CDT-2) in both cellulose and hemicellulose degradation and utilization in Neurospora crassaPLOS ONE9e8933010.1371/journal.pone.0089330CarrollASomervilleC2009Cellulosic biofuelsAnnual Review of Plant Biology6016518210.1146/annurev.arplant.043008.092125ChundawatSPBeckhamGTHimmelMEDaleBE2011Deconstruction of lignocellulosic biomass to fuels and chemicalsAnnual Review of Chemical and Biomolecular Engineering212114510.1146/annurev-chembioeng-061010-114205ColotHVParkGTurnerGERingelbergCCrewCMLitvinkovaLWeissRLBorkovichKADunlapJC2006A high-throughput gene knockout procedure for Neurospora reveals functions for multiple transcription factorsProceedings of the National Academy of Sciences of USA103103521035710.1073/pnas.0601456103CoradettiSTCraigJPXiongYShockTTianCGlassNL2012Conserved and essential transcription factors for cellulase gene expression in ascomycete fungiProceedings of the National Academy of Sciences of USA1097397740210.1073/pnas.1200785109EdgarRC2004MUSCLE: multiple sequence alignment with high accuracy and high throughputNucleic Acids Research321792179710.1093/nar/gkh340FujiiTYuGMatsushikaAKuritaAYanoSMurakamiKSawayamaS2011Ethanol production from xylo-oligosaccharides by xylose-fermenting Saccharomyces cerevisiae expressing beta-xylosidaseBioscience, Biotechnology, and Biochemistry751140114610.1271/bbb.110043GalaganJEHennMRMaLJCuomoCABirrenB2005Genomics of the fungal kingdom: insights into eukaryotic biologyGenome Research151620163110.1101/gr.3767105GalazkaJMTianCBeesonWTMartinezBGlassNLCateJH2010Cellodextrin transport in yeast for improved biofuel productionScience330848610.1126/science.1192838HaSJGalazkaJMKimSRChoiJHYangXSeoJHGlassNLCateJHJinYS2011Engineered Saccharomyces cerevisiae capable of simultaneous cellobiose and xylose fermentationProceedings of the National Academy of Sciences of USA10850450910.1073/pnas.1010456108HamacherTBeckerJGárdonyiMHahn-HägerdalBBolesE2002Characterization of the xylose-transporting properties of yeast hexose transporters and their influence on xylose utilizationMicrobiology14827832788HedgesSBDudleyJKumarS2006TimeTree: a public knowledge-base of divergence times among organismsBioinformatics222971297210.1093/bioinformatics/btl505HendriksATZeemanG2009Pretreatments to enhance the digestibility of lignocellulosic biomassBioresource Technology100101810.1016/j.biortech.2008.05.027HimmelMEDingSYJohnsonDKAdneyWSNimlosMRBradyJWFoustTD2007Biomass recalcitrance: engineering plants and enzymes for biofuels productionScience31580480710.1126/science.1137016HongKKNielsenJ2012Metabolic engineering of Saccharomyces cerevisiae: a key cell factory platform for future biorefineriesCellular and Molecular Life Sciences692671269010.1007/s00018-012-0945-1JarboeLRZhangXWangXMooreJCShanmugamKTIngramLO2010Metabolic engineering for production of biorenewable fuels and chemicals: contributions of synthetic biologyJournal of Biomedicine & Biotechnology201076104276111810.1155/2010/761042JeffriesTW2006Engineering yeasts for xylose metabolismCurrent Opinion in Biotechnology1732032610.1016/j.copbio.2006.05.008KavanaghKLKlimacekMNidetzkyBWilsonDK2002The structure of apo and holo forms of xylose reductase, a dimeric aldo-keto reductase from Candida tenuisBiochemistry418785879510.1021/bi025786nKimSRSkerkerJMKangWLesmanaAWeiNArkinAPJinYS2013Rational and evolutionary engineering approaches uncover a small set of genetic changes efficient for rapid xylose fermentation in Saccharomyces cerevisiaePLOS ONE8e5704810.1371/journal.pone.0057048KratzerRLeitgebSWilsonDKNidetzkyB2006Probing the substrate binding site of candida tenuis xylose reductase (AKR2B5) with site-directed mutagenesisThe Biochemical Journal393515810.1042/BJ20050831KurtzmanCP1994Molecular taxonomy of the yeastsYeast101727174010.1002/yea.320101306KuyperMHartogMMToirkensMJAlmeringMJWinklerAAvan DijkenJPPronkJT2005Metabolic engineering of a xylose-isomerase-expressing Saccharomyces cerevisiae strain for rapid anaerobic xylose fermentationFEMS Yeast Research539940910.1016/j.femsyr.2004.09.010LatimerLNLeeMEMedina-CleghornDKohnzRANomuraDKDueberJE2014Employing a combinatorial expression approach to characterize xylose utilization in Saccharomyces cerevisiaeMetabolic Engineering25202910.1016/j.ymben.2014.06.002LinYChomvongKAcosta-SampsonLEstrelaRGalazkaJMKimSRJinYSCateJH2014Leveraging transcription factors to speed cellobiose fermentation by Saccharomyces cerevisiaeBiotechnology for Biofuels712610.1186/s13068-014-0126-6LyndLRWeimerPJvan ZylWHPretoriusIS2002Microbial cellulose utilization: fundamentals and biotechnologyMicrobiology and Molecular Biology Reviews6650657710.1128/MMBR.66.3.506-577.2002MatsushikaAInoueHKodakiTSawayamaS2009Ethanol production from xylose in engineered Saccharomyces cerevisiae strains: current state and perspectivesApplied Microbiology and Biotechnology84375310.1007/s00253-009-2101-xMcCluskeyKWiestAPlamannM2010The Fungal Genetics Stock Center: a repository for 50 years of fungal genetics researchJournal of Biosciences3511912610.1007/s12038-010-0014-6PenttiläMNevalainenHRättöMSalminenEKnowlesJ1987A versatile transformation system for the cellulolytic filamentous fungus trichoderma reeseiGene6115516410.1016/0378-1119(87)90110-7RabinowitzJDKimballE2007Acidic acetonitrile for cellular metabolome extraction from Escherichia coliAnalytical Chemistry796167617310.1021/ac070470cRyanOWSkerkerJMMaurerMJLiXTsaiJCPoddarSLeeMEDeLoacheWDueberJEArkinAPCateJH2014Selection of Chromosomal DNA Libraries using a Multiplex CRISPR SystemeLife310.7554/eLife.03703SikorskiRSHieterP1989A system of shuttle vectors and yeast host strains designed for efficient manipulation of dna in Saccharomyces cerevisiaeGenetics1221927SomervilleCBauerSBrininstoolGFacetteMHamannTMilneJOsborneEParedezAPerssonSRaabTVorwerkSYoungsH2004Toward a systems approach to understanding plant cell wallsScience3062206221110.1126/science.1102765SunJTianCDiamondSGlassNL2012Deciphering transcriptional regulatory mechanisms associated with hemicellulose degradation in Neurospora crassaEukaryot Cell1148249310.1128/EC.05327-11TamuraKStecherGPetersonDFilipskiAKumarS2013MEGA6: Molecular evolutionary genetics analysis version 6.0Molecular Biology and evolution302725272910.1093/molbev/mst197TianCBeesonWTIavaroneATSunJMarlettaMACateJHGlassNL2009Systems analysis of plant cell wall degradation by the model filamentous fungus Neurospora crassaProceedings of the National Academy of Sciences of USA106221572216210.1073/pnas.0906810106TrottOOlsonAJ2010AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreadingJournal of Computational Chemistry3145546110.1002/jcc.21334VallejosMEZambonMDAreaMCda Silva CurveloAA2012Low liquid-solid ratio (LSR) hot water pretreatment of sugarcane bagasseGreen Chemistry141982198910.1039/c2gc35397kvan MarisAJWinklerAAKuyperMde LaatWTvan DijkenJPPronkJT2007Development of efficient xylose fermentation in Saccharomyces cerevisiae: xylose isomerase as a key componentAdvances in Biochemical Engineering/Biotechnology10817920410.1007/10_2007_057VogelHJ1956A convenient growth medium for neurospora (Medium N)Microbial Genetics Bulletin134243WeiNQuartermanJKimSRCateJHJinYS2013Enhanced biofuel production through coupled acetic acid and xylose consumption by engineered yeastNature Communications4258010.1038/ncomms3580WellmanCHOsterloffPLMohiuddinU2003Fragments of the earliest land plantsNature42528228510.1038/nature01884XiongYCoradettiSTLiXGritsenkoMAClaussTPetyukVCampDSmithRCateJHYangFGlassNL2014The proteome and phosphoproteome of Neurospora crassa in response to cellulose, sucrose and carbon starvationFungal Genetics and Biology72213310.1016/j.fgb.2014.05.005YoukHvan OudenaardenA2009Growth landscape formed by perception and import of glucose in yeastNature46287587910.1038/nature08653YoungELeeSMAlperH2010Optimizing pentose utilization in yeast: the need for novel tools and approachesBiotechnology for Biofuels32410.1186/1754-6834-3-24YoungEMTongABuiHSpoffordCAlperHS2014Rewiring yeast sugar transporter preference through modifying a conserved protein motifProceedings of the National Academy of Sciences of USA11113113610.1073/pnas.131197011110.7554/eLife.05896.025Decision letterBrakhageAxelReviewing editorLeibniz Institute for Natural Product Research and Infection Biology - Hans Knöll Institute, Germany
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Expanding xylose metabolism in yeast for plant cell wall conversion to biofuels” for consideration at eLife. Your article has been favorably evaluated by Detlef Weigel (Senior editor) and two reviewers, one of whom is a member of our Board of Reviewing Editors.
The Reviewing editor and the other reviewer discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
We think your manuscript was significantly improved and has now given some fundamental insights into novel metabolic intermediates. However, there are several points that we ask you to consider for further improvement. Specifically, we would like to see that either the Thevelein strain is used (e.g. Biotechnol Biofuels, 2013, Jun 21; 6(1):89) and the experiments regarding the XI-based pathway is repeated, or that the entire paragraph about XI (paragraph ten of the Introduction, with figure 5) is removed. In addition, please discuss the potential advantages of an XI-based pathway in connection with xylodextrin fermentation.
Minor comments:
Reviewer #1:
There are a couple of points, which I still recommend to reconsider:
a) In the Abstract, the finding of the xylosyl-xylitol intermediates in other organisms should be mentioned, underlining the broader finding of the manuscript.
b) The manuscript of Ryan et al. should be cited.
c) Figure 1–figure supplement 2: Growth of N. crassa. The figure is rather poor, the reduced growth of N. crassa Δcdt-2 on xylan is hardly visible. I suggest you provide a proper biomass measurement.
d) Figure 1–figure supplement 6: It is difficult to believe that the N. crassa GH43-3 is so different from the other enzymes that it appears more closely related to enzymes of basidiomycetes (Schizophyllum) than to other ascomycetes.
e) Figure 5–figure supplement 1: SDs of data are missing.
Reviewer #2:
When you state that “The excess reducing power generated by the XR/XDH pathway, initially deemed a problem…”, the XR/XDH pathway does not generate excess reducing power. The problem of the pathway is rather the cofactor imbalances between NADH and NADPH. This sentence should be omitted.
[Editors’ note: a previous version of this study was rejected after discussions between the reviewers, but the authors submitted for reconsideration. The previous decision letter is shown below.]
Thank you for choosing to send your work entitled “Expanding xylose metabolism in yeast for plant cell wall conversion to biofuels” for consideration at eLife. Your full submission has been evaluated by Detlef Weigel (Senior editor) and two peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the decision was reached after discussions between the reviewers. We regret to inform you that your work will not be considered further for publication.
The reviewers had several concerns about your manuscript. Reviewer #1 indicated that some parts of the findings have already been published and the manuscript is not conceptually new. Also, the reviewer questioned the relevance of the novel oligosaccharides identified. Reviewer #2 was more positive. However, at the end also this reviewer questioned the break-through provided in your manuscript. In summary, we do not feel that the manuscript in its present version has sufficient novelty and quality for eLife. However, we could imagine being interested in a manuscript describing the combination of both pathways for xylose catabolism in a genetically engineered yeast, which should show a superior capacity to degrade xylose and to produce ethanol.
Reviewer #1:
In the current manuscript the authors describe the continuation of their work on the generation genetically engineered Saccharomyces cerevisiae (yeast) strains which are able to concert xylose to ethanol. The authors transferred a xylodextrin transporter, an intracellular β-xylosidase and an intracellular xylosyl-xylitol-specific ecxylosidase from the model fungus Neurospra crassa, which can grow on xylose, to S. cerevisiae. Production of the encoded proteins in an S. cerevisiae strain transformed with an XR/XDH/XK-pathway from S. stipitis resulted in the utilization of xylodextrins as the sole carbon source.
Major comments
1) Although I think in general genetic engineering of yeast to obtain strains with properties to degrade xylose is interesting, unfortunately, a very similar approach was published in 2010, by the same group, for the utilization of cellodextrin by recombinant S. cerevisiae by expressing a cellodextrin transporter and an intracellular β-glucosidase. Now, the authors applied the very same principle to the degradation of xylodextrin.
2) Furthermore, the here analyzed xylodextrin transporter CDT-2 was recently characterized in detail by Cai et al., 2014 (PLOS One 9:e89330).
3) Also, the expression of 93xylosidase in yeast was reported before (Biosci Biotechnol Biochem, 2011, 75,1140-6). Thus, the only novelty of this manuscript is the production of xylosyl-xylitol oligomers as a side-activity of S. stipitis and N. crassa xylose reductases (XR). These oligomers appear to be inhibitory in engineered S. cerevisiae to utilize xylodextrins. The authors solved this problem by the identification and expression of a new hydrolyase, GH43-7, which can cleave the xylosyl-xylitol oligomers. Although the authors claim that these oligomers appear to be widely distributed in nature, evidence for this statement is lacking. By contrast, the production of these oligomers might just result from the strategy applied. In general, S. cerevisiae can be engineered to consume xylose either by expressing a xylose isomerase (XI) or by applying the XR/XDH pathway. Surprisingly, the authors used the XR/XDH pathway, although it is generally accepted that the XI pathway is more favorable for yeast (Biotechnol Biofuels, 2013, 6, 89; Metab Eng., 2012, 14: 611-22). As far as I know, the available xylose fermenting yeast strains from industry (DSM, Lesaffre, Terranol) are all based on the XI strategy. When the authors had used the XI strategy, they never would have encountered the problem of generating the xylosyl-xylitol oligomers by the heterologous XR.
Reviewer #2:
This paper describes expression of a novel pathway for metabolism of xylobiose and other short chain xylans in yeast. The pathway was identified in N. crassa and then transferred to yeast. The pathway involves two transporters and three intracellular hydrolysases, which were all identified in N. crassa based on RNAseq analysis. The xylose consumption pathway relies on the less efficient xylose reductase and xylitol dehydrogenase pathway, but it still seems to work due to the partly reduction of xylodextrins intracellularly. This activity is, however, only conferred by the P. stipidis XR, so the pathway for uptake of xylodextrins could work equally well (and probably better) with the xylose isomerase pathway, which is to be preferred as it does not result in accumulation of xylitol (also here observed by the authors, but not commented).
I think this paper is a fantastic new contribution to the field and it definitely serves consideration. Even though it could be interesting to combine the new pathway with the XI route for xylose catabolism, I do not think it is fair to ask the reviewers to demonstrate this. However, I suggest that they clearly state that the main benefit of their work is actually that they have found a way to overcome one of the most fundamental problems in xylose catabolism, namely an alternative transport of xylose into the cell. Xylose is transported through the HXT transporters and has competitive inhibition by glucose. This pathway may work in concert with glucose consumption, and this may be the real benefit. The data presented by the authors even indicate this. I suggest that this is clearly stated in the paper.
Besides this I do not have any major comments.
10.7554/eLife.05896.026Author response
We think your manuscript was significantly improved and has now given some fundamental insights into novel metabolic intermediates. However, there are several points that we ask you to consider for further improvement. Specifically, we would like to see that either the Thevelein strain is used (e.g. Biotechnol Biofuels, 2013, Jun 21; 6(1):89) and the experiments regarding the XI-based pathway is repeated, or that the entire paragraph about XI (paragraph ten of the Introduction, withfigure 5) is removed. In addition, please discuss the potential advantages of an XI-based pathway in connection with xylodextrin fermentation.
We thank the reviewers for their positive feedback on our manuscript. Although we agree it will be important to compare multiple XR/XDH and XI strains to gauge their performance, we think these experiments would fit better in a separate paper. We have therefore removed the paragraph and experiments regarding the XI-based pathway from our manuscript (paragraph beginning with “It has been proposed”, along with Figure 6, Figure 6–figure supplement 1, Figure 6–figure supplement 2, Table 1 and the associated Methods sections). We moved the observation of decreased xylitol production to the previous paragraph, where Figure 5–figure supplement 2B follows logically.
We also added a sentence and cited van Maris et al. (2007) on the potential use of the XI-based pathway in the penultimate paragraph of the main text.
Minor comments:
Reviewer #1:
There are a couple of points, which I still recommend to reconsider:
a) In the Abstract, the finding of the xylosyl-xylitol intermediates in other organisms should be mentioned, underlining the broader finding of the manuscript.
We thank the reviewer for this suggestion. We have changed the Abstract accordingly.
b) The manuscript of Ryan et al. should be cited.
We added this reference to the final paragraph, as it is indeed something worth pursuing for optimizing this pathway.
c)Figure 1–figure supplement 2: Growth of N. crassa. The figure is rather poor, the reduced growth of N. crassa Δcdt-2 on xylan is hardly visible. I suggest you provide a proper biomass measurement.
We actually have the biomass data on xylan medium, but thought the photos would give the reader a better impression. We have now added the biomass data as panel B and worked to improve the image contrast of the original photos using linear scaling of the colors.
d)Figure 1–figure supplement 6: It is difficult to believe that the N. crassa GH43-3 is so different from the other enzymes that it appears more closely related to enzymes of basidiomycetes (Schizophyllum) than to other ascomycetes.
This is an interesting point. GH43-3 is only 34% identical to the Schizophyllum protein over 85% of latter protein’s length. Using a simple PSI-BLAST search cannot detect N. crassa GH43-2 without significant phylogenetic support. Very likely, GH43-3 and GH43-2 are members of the same glycosyl hydrolase superfamily, but have very different functions.
e)Figure 5–figure supplement 1: SDs of data are missing.
We carried out this experiment with three different xylose to xylodextrin ratios and saw the same pattern of consumption. This figure is a representative experiment, and we have indicated this in the figure legend.
Reviewer #2:
When you state that “The excess reducing power generated by the XR/XDH pathway, initially deemed a problem…”, the XR/XDH pathway does not generate excess reducing power. The problem of the pathway is rather the cofactor imbalances between NADH and NADPH. This sentence should be omitted.
We can see why the reviewer does not like this sentence, as it is not very precise. We have reworded the sentence to more accurately reflect what is going on, i.e. that cofactor imbalance leads to accumulation of reduced byproducts (xylitol and glycerol). We do think it’s worth keeping this sentence in the paper, as realistic hydrolysates will likely have a mixture of xylose and xylodextrins, and we have conclusively demonstrated that the cofactor imbalance of XR/XDH pathway can be exploited to drive reduction of acetate to ethanol (Wei et al., 2013).
[Editors’ note: the author responses to the previous round of peer review follow.]
In the last round of review, the reviewers were not convinced that the xylodextrin consumption pathway we identified in N. crassa and the xylosyl-xylitol metabolic intermediates were widespread in nature. Furthermore, the reviewers were not convinced that this pathway could be competitive with one using xylose isomerase (XI) as opposed to xylose reductase plus xylitol dehydrogenase (XR/XDH). We have now carried out a large number of experiments to address these concerns.
First, we tested whether xylosyl-xylitols are generated intracellularly in a number of fungi, as well as in Bacillus subtilis. As shown in the new Figure 3 and Figure 3–figure supplement 1, we find xylosyl-xylitols present in all of these organisms when they are grown on xylodextrins. These experiments convincingly show that xylosyl-xylitols are in fact widespread in nature, generated by microbes spanning well over 1 billion years of evolution. They are not simply a byproduct of reconstituting xylodextrin consumption in S. cerevisiae.
Second, we carried out a direct comparison between two of the best xylose fermentation strains in the public domain–strain SR8, which uses the XR/XDH pathway (see Kim et al., 2013, PLOS One), and strain SXA-R2P-E which uses the XI pathway and which was kindly provided by Prof. Hal Alper (http://www.biotechnologyforbiofuels.com/content/7/1/122). Using anaerobic conditions, we find that, regardless of starting conditions, cell loading, or media, strain SR8 is far superior to strain SXA-R2P-E in terms of ethanol productivity. Strain SXA-R2P-E has a slightly higher yield of ethanol and produces less xylitol. To our knowledge, this is the first direct experimental comparison between high-performing XI and XR/XDH strains in the literature, and will be of wide interest.
Remarkably, we were surprised to find that, when using xylodextrins, the XR/XDH pathway produces far less xylitol when compared to xylose fermentations (Figure 5 and Figure 5–figure supplement 2B). Thus, not only does the capacity of the XR/XDH pathway for high flux far exceed that of the XI pathway, the propensity of the XR/XDH pathway to produce xylitol can be reduced by using xylodextrins as opposed to xylose in fermentations. Thus, we think it is highly worthwhile to explore the potential of the xylodextrin pathway we’ve identified for the first time. Although the flux through the upstream part of the xylodextrin consumption pathway is not as high as with xylose fermentation, we think future efforts such as directed evolution of the xylodextrin transporter can be used to solve this problem (see Ryan, 2014, eLife 3:e03703).
Finally, we now show that the xylodextrin pathway can be exploited in a range of cofermentation scenarios, including glucose plus xylodextrins (suggested by reviewer #2) and sucrose plus xylodextrins. These are shown in Figure 5B and Figure 7. Cofermentation of xylose equivalents with glucose equivalents is a “holy grail” in the biofuel field, and we have now at least doubled the options available for future exploration (two previously described scenarios include glucose plus xylose and cellobiose plus xylose).
In summary, for both the ecological insights regarding xylodextrin consumption by microbes in nature, and the possible applications of xylodextrin fermentation to produce biofuels, we think this paper should be published in eLife.
Reviewer #1:
In the current manuscript the authors describe the continuation of their work on the generation genetically engineered Saccharomyces cerevisiae (yeast) strains which are able to concert xylose to ethanol. The authors transferred a xylodextrin transporter, an intracellular β-xylosidase and an intracellular xylosyl-xylitol-specific β-xylosidase from the model fungus Neurospra crassa, which can grow on xylose, to S. cerevisiae. Production of the encoded proteins in an S. cerevisiae strain transformed with an XR/XDH/XK-pathway from S. stipitis resulted in the utilization of xylodextrins as the sole carbon source.
Major comments
1) Although I think in general genetic engineering of yeast to obtain strains with properties to degrade xylose is interesting, unfortunately, a very similar approach was published in 2010, by the same group, for the utilization of cellodextrin by recombinant S. cerevisiae by expressing a cellodextrin transporter and an intracellular β-glucosidase. Now, the authors applied the very same principle to the degradation of xylodextrin.
We agree that the approach we have used here is somewhat similar to that used in 2010 for the utilization of cellobdextrins. However, we note that this is the approach evolved in N. crassa for optimal growth on the plant cell wall, and involves an entirely surprising metabolic intermediate that had not been identified before. Furthermore, we now show this to be a very general pathway for consuming xylose. It is used in fungi spanning evolutionary time to the likely advent of the general architecture of the plant cell wall now extant in plants. Furthermore, the pathway is even present in bacteria, as we have now shown using B. subtilis and phylogenetic analysis of other bacteria (see Figure 3 and Figure 3–figure supplement 1). Thus, this pathway has arisen in organisms that span well over a billion years of evolutionary history. From a fundamental science perspective, we think this is certainly conceptually new and worth publishing in eLife.
2) Furthermore, the here analyzed xylodextrin transporter CDT-2 was recently characterized in detail byCai et al., 2014(PLOS One 9:e89330).
As noted above, xylodextrin transport is only a small part of the story. Furthermore, without identifying the xylosyl-xylitol intermediates, the PLOS ONE paper is far from complete, and misses a major aspect of how microbes grow on plants. Again, the discovery that organisms as diverse as B. subtilis and many fungi produce these intermediates is entirely new, and was missed by every one who has worked on xylose metabolism.
3) Also, the expression of β-xylosidase in yeast was reported before (Biosci Biotechnol Biochem, 2011, 75,1140-6). Thus, the only novelty of this manuscript is the production of xylosyl-xylitol oligomers as a side-activity of S. stipitis and N. crassa xylose reductases (XR). These oligomers appear to be inhibitory in engineered S. cerevisiae to utilize xylodextrins. The authors solved this problem by the identification and expression of a new hydrolyase, GH43-7, which can cleave the xylosyl-xylitol oligomers. Although the authors claim that these oligomers appear to be widely distributed in nature, evidence for this statement is lacking. By contrast, the production of these oligomers might just result from the strategy applied.
As we have now shown, xylosyl-xylitol oligomers are generated by a wide variety of microbes spanning over a billion years of evolution. We think this is a fundamental discovery worth publishing in eLife.
In general, S. cerevisiae can be engineered to consume xylose either by expressing a xylose isomerase (XI) or by applying the XR/XDH pathway. Surprisingly, the authors used the XR/XDH pathway, although it is generally accepted that the XI pathway is more favorable for yeast (Biotechnol Biofuels, 2013, 6, 89; Metab Eng., 2012, 14: 611-22). As far as I know, the available xylose fermenting yeast strains from industry (DSM, Lesaffre, Terranol) are all based on the XI strategy. When the authors had used the XI strategy, they never would have encountered the problem of generating the xylosyl-xylitol oligomers by the heterologous XR.
We strongly disagree that the XI pathway has been shown to be superior to the XR/XDH pathway. There have been few if any direct comparisons of these pathways in the literature, with what could be described as state-of-the-art strains. To overcome this deficiency we have used two of the best strains available in the public domain in head-to-head comparisons. For the XR/XDH pathway, we used strain SR8, which is one of the best reported strains for xylose fermentation. For the XI pathway, we used strain SXA-R2P-E from Prof. Hal Alper’s lab, which he kindly provided us for these comparisons. This strain was recently reported in Biotechnology for Biofuels, and is one of the two best XI strains in the public domain (the one from Metab. Eng., 2012, 14: 611-22 is not available, although we requested it).
We used a number of conditions, all anaerobic, to compare the two strains head-to-head in batch fermentations. We find that, regardless of starting OD (or g cells/L culture), the SR8 strain expressing the XR/XDH pathway is far superior to the XI strain, with roughly twice the ethanol productivity. By contrast the XI strain is slightly better in terms of ethanol yield for the XR/XDH strain (see Figure 6, Figure 6–figure supplement 1, Figure 6–figure supplement 2 and Table 1 in the new manuscript). Thus, even prior to considering the use of xylodextrins, the XR/XDH pathway is highly competitive, if not outright superior, to the XI pathway. Thus the rationale put forward by reviewer #1 that the XI pathway is preferred is not obvious.
We have also made the surprising observation that using xylodextrins minimizes xylitol byproduct formation by the XR/XDH pathway, which was previously seen as a fundamental weakness of the pathway. See Figure 5 and Figure 5–figure supplement 2B. This suggests that, with further optimization of the upstream part of the pathway, i.e. through directed evolution of the transporter (see our recent work in eLife, Ryan et al.), the XR/XDH pathway could achieve both high ethanol productivity and yield, eliminating the rationale for the use of XI.
In addition, we discussed with Dr. Amit Gokhale from BP scenarios that could be of immediate use in improving biofuel production. He noted that, as we indicated in our original manuscript, xylodextrin fermentation could be used as an add-on to existing sugarcane ethanol production processes. In short, hot water pretreatment could be used to release xylodextrins from bagasse, and the resulting water could be used as the diluent for the cane juice used in ethanol fermentations. This would have the benefit of increasing ethanol yield beyond that possible from using cane juice alone. Indeed, we find this to be the case (see Figure 7 in the new manuscript). Thus, cofermentation of sucrose and xylodextrins is a new and promising direction for producing biofuels.
Reviewer #2:
This paper describes expression of a novel pathway for metabolism of xylobiose and other short chain xylans in yeast. The pathway was identified in N. crassa and then transferred to yeast. The pathway involves two transporters and three intracellular hydrolysases, which were all identified in N. crassa based on RNAseq analysis. The xylose consumption pathway relies on the less efficient xylose reductase and xylitol dehydrogenase pathway, but it still seems to work due to the partly reduction of xylodextrins intracellularly. This activity is, however, only conferred by the P. stipidis XR, so the pathway for uptake of xylodextrins could work equally well (and probably better) with the xylose isomerase pathway, which is to be preferred as it does not result in accumulation of xylitol (also here observed by the authors, but not commented).
As described above in response to reviewer #1, the XI pathway is inferior to the XR/XDH pathway with respect to productivity, by a wide margin. We also note that, to our surprise, the XR/XDH pathway does not generate nearly as much xylitol when xylodextrins are used as a carbon source rather than xylose. This could in fact reflect one of the key advantages of using xylodextrins in preference to xylose, both in nature and in an industrial setting.
I think this paper is a fantastic new contribution to the field and it definitely serves consideration. Even though it could be interesting to combine the new pathway with the XI route for xylose catabolism, I do not think it is fair to ask the reviewers to demonstrate this. However, I suggest that they clearly state that the main benefit of their work is actually that they have found a way to overcome one of the most fundamental problems in xylose catabolism, namely an alternative transport of xylose into the cell. Xylose is transported through the HXT transporters and has competitive inhibition by glucose. This pathway may work in concert with glucose consumption, and this may be the real benefit. The data presented by the authors even indicate this. I suggest that this is clearly stated in the paper.
We thank the reviewer for the positive feedback. As we describe above in response to reviewer #1, we have directly compared XR/XDH and XI pathways and found the XR/XDH pathway to be superior in terms of productivity. We also show that the use of xylodextrins reduces the amount of xylitol byproduct produced by the strain expressing XR/XDH. As suggested by the reviewer, we have now shown that glucose plus xylodextrin cofermentation is indeed possible, and produces very little xylitol as a byproduct (Figure 5B). We have also used xylodextrins in combination with sucrose, and see increased ethanol yields that are very promising in terms of an application to enhance sugarcane ethanol production (Figure 7).
An outstanding question is how receptor tyrosine kinases (RTKs) determine different cell-fate decisions despite sharing the same signalling cascades. Here, we uncovered an unexpected mechanism of RTK trafficking in this process. By quantitative high-resolution FRET microscopy, we found that phosphorylated epidermal growth factor receptor (p-EGFR) is not randomly distributed but packaged at constant mean amounts in endosomes. Cells respond to higher EGF concentrations by increasing the number of endosomes but keeping the mean p-EGFR content per endosome almost constant. By mathematical modelling, we found that this mechanism confers both robustness and regulation to signalling output. Different growth factors caused specific changes in endosome number and size in various cell systems and changing the distribution of p-EGFR between endosomes was sufficient to reprogram cell-fate decision upon EGF stimulation. We propose that the packaging of p-RTKs in endosomes is a general mechanism to ensure the fidelity and specificity of the signalling response.
DOI:http://dx.doi.org/10.7554/eLife.06156.001
10.7554/eLife.06156.002eLife digest
Molecules called growth factors can stimulate cells to grow, divide, or differentiate into more specialised cell types. Cells detect these molecules via proteins called receptor tyrosine kinases that span their surface membrane. The growth factor binds to the portion of the receptor outside the cell, which makes the receptor send signals to the cell's nucleus that change how the cell grows, divides, or specialises.
Different growth factors and receptor tyrosine kinases affect cell development in different ways. However, it was unclear how this occurred, as the receptors all send signals via the same signalling pathways. Some researchers proposed that specific responses could be triggered if some receptor tyrosine kinases activated these pathways more strongly than other receptors, or if they activated the pathways for different lengths of time.
Now, Villaseñor et al. have looked at a receptor tyrosine kinase for a growth factor called EGF. Activated EGF receptors are marked with a phosphate group (or ‘phosphorylated’) and are then removed from the surface membrane and packaged into structures within the cell called endosomes. Villaseñor et al. found that different endosomes contain the same mean amount of phosphorylated EGF receptor. When exposed to higher EGF concentrations, the cells respond by increasing the number of endosomes, and so the average number of phosphorylated EGF receptors in each endosome remains almost constant. Villaseñor et al. used mathematical modelling to show that this mechanism, which they refer to as an ‘analogue-to-digital conversion’, ensures a robust signal, and can regulate the signalling of the activated receptors in both space and time.
Different growth factors either increase or decrease the number and size of endosomes in various cell types. Moreover, Villaseñor et al. found that changing how the phosphorylated EGF receptor is distributed between endosomes alters how the cells interpret the signal and differentiate when they are exposed to EGF. These findings mean that the signalling activity of a cell could be predicted from the number and size of its endosomes. Moreover, the findings also suggest that interfering with this mechanism could change cell behaviour, for example, it could stop cancer cells proliferating and force them to differentiate instead.
DOI:http://dx.doi.org/10.7554/eLife.06156.002
Author keywordssignal transductionendocytosismembrane transportResearch organismHumanmouserathttp://dx.doi.org/10.13039/501100002347Bundesministerium für Bildung und ForschungVirtual Liver InitiativeVillaseñorRobertoNonakaHidenoriDel Conte-ZerialPerlaKalaidzidisYannisZerialMarinohttp://dx.doi.org/10.13039/501100004189Max-Planck-GesellschaftVillaseñorRobertoNonakaHidenoriDel Conte-ZerialPerlaKalaidzidisYannisZerialMarinohttp://dx.doi.org/10.13039/501100001659Deutsche ForschungsgemeinschaftVillaseñorRobertoNonakaHidenoriDel Conte-ZerialPerlaKalaidzidisYannisZerialMarinohttp://dx.doi.org/10.13039/501100002974Daimler und Benz StiftungVillaseñorRobertoThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementCells package active receptors in endosomes at fairly constant amounts and can determine different cell-fate decisions by regulating the number and lifetime of receptor packages.Introduction
Cells respond to various signals by activating different types of RTKs and committing to specific cell-fate decisions (Katz et al., 2007). A remarkable property of this system is that different RTKs can elicit distinct cellular responses through the same signal transduction machinery (Marshall, 1995; Kholodenko et al., 2006). In several cases, signalling specificity results from differences in amplitude and duration of the intracellular signalling cascades (Marshall, 1995; Maroun et al., 2000; Nagashima et al., 2007). For example, in PC12 cells, EGF stimulation of EGFR leads to transient Erk phosphorylation and cell proliferation, whereas NGF binding to TrkA leads to sustained Erk phosphorylation and cell differentiation (Marshall, 1995). Differences in signalling amplitude and duration can arise from positive or negative feedback loops within the same signalling pathway (Santos et al., 2007) or activation of additional signalling components (York et al., 1998). To explain such differences, it has been proposed that both EGF and NGF stimulation induce a specific ‘molecular context’ that determines the topology of the signal transduction network (Santos et al., 2007). How such a topology is determined for different RTKs and whether it is the sole determinant of signal specificity is unclear (Kholodenko, 2007).
Insights into this problem may be provided by the spatio-temporal distribution of RTKs along the endosomal system. The detection of phosphorylated receptors and signalling adaptors in endosomes (Di Guglielmo et al., 1994; Vieira et al., 1996; Sorkin, 2001; Teis et al., 2002; Lampugnani et al., 2006; Galperin and Sorkin, 2008; Schenck et al., 2008; Coumailleau et al., 2009) led to the concept that signalling is initiated at the plasma membrane but continues in endosomes (Di Guglielmo et al., 1994). Indeed, inhibition of endocytosis by blocking Dynamin function causes significant alterations in signalling specificity (Vieira et al., 1996). However, recent studies challenged this concept arguing that EGFR signalling occurs primarily at the plasma membrane (Damke et al., 1994; Brankatschk et al., 2012; Sousa et al., 2012). Interestingly, a recent systems survey of endocytosis (Collinet et al., 2010) revealed an unexpected tight control in the number, size, and cargo content for EGF-positive endosomes, raising the question of why is EGF packaging in endosomes so accurately controlled? Here, we hypothesized that the tight control of the endosomal distribution of EGF could serve to regulate signal transmission. We tested this hypothesis by quantitatively analysing the endosomal distribution of EGFR as an RTK model system in endosomes and evaluating its impact on cell-fate decisions.
Results
To measure the content of p-EGFR in individual endosomes, we used two independent assays (for a detailed description see ‘Materials and methods’ and Figure 1—figure supplement 1). First, we modified a FRET-FLIM microscopy assay previously used to measure the spatial distribution of p-EGFR at the plasma membrane (Wouters and Bastiaens, 1999; Verveer et al., 2000). The assay measured the FRET signal between EGFR-GFP and an anti-phospho-tyrosine antibody (p-Tyr-ab) labelled with AlexaFluor 555. Since FLIM microscopy lacks the spatial resolution to analyse the receptor activation at a sub-cellular level, we modified the assay into a high-resolution FRET microscopy assay. However, instead of the total cell signal, we measured the distribution of EGFR and p-EGFR at the level of individual endosomes resolved by high-resolution confocal microscopy and quantitative automated image analysis (Rink et al., 2005; Collinet et al., 2010) (Figure 1—figure supplement 2). To avoid artefacts of overexpression, we used HeLa cells transfected with a bacterial artificial chromosome (BAC) transgene stably expressing EGFR-GFP under its endogenous promoter (Poser et al., 2008). In these cells (Figure 1—figure supplement 3A), the uptake of EGF was only ∼twofold higher compared to endogenous (Figure 1—figure supplement 3B). However, the transport kinetics were similar (Figure 1—figure supplement 3C). Second, we measured p-EGFR with an antibody against a specific phospho-tyrosine residue (Tyr1068). Both assays gave very similar results (Figure 1—figure supplement 4). As the FRET assay is not restricted to a single phosphorylation site that can change over time (Morandell et al., 2008), we used it as a primary assay in further experiments. Under the fixation conditions used, we observed no significant difference in the morphology (Figure 1—figure supplement 5A) or area (Figure 1—figure supplement 5B) of EGFR-positive endosomes (Video 1). For every time point, ∼15,000 endosomes from over 200 cells were analysed.10.7554/eLife.06156.003
Live-cell imaging of EGFR endocytosis.
HeLa EGFR-GFP BAC cells were imaged with a spinning disk microscope after 1 minute of EGF stimulation with 10 ng/ml EGF. Movie shows maximal projection of 3 z-slices of 0.8 mm thickness.
DOI:http://dx.doi.org/10.7554/eLife.06156.003
Continuous stimulation with EGF triggered the internalization of EGFR into endosomes (Figure 1 and Figure 1—figure supplement 2). The total amount of endosomal EGFR peaked after 15 min and decreased, reflecting (1) down-regulation of surface receptors (Wiley et al., 1991) and (2) their degradation (Dunn and Hubbard, 1984) over time (Figure 1A, green curve). On the other hand, the total p-EGFR levels reached a maximum already at 10 min, followed by a phase of decay (Figure 1A, red curve). Comparison of decay kinetics for both curves after 15 min showed that de-phosphorylation of p-EGFR occurred faster than degradation (τdecay EGFR = 88.13 ± 14.49, τdecay p-EGFR = 30.97 ± 1.69, for details see ‘Materials and methods’). Our FRET measurements are thus consistent with previously reported EGFR transport and phosphorylation kinetics determined by biochemical and microscopic methods (Di Guglielmo et al., 1994; Burke et al., 2001).10.7554/eLife.06156.004
Cells keep a constant amount of p-EGFR in endosomes.
(A) Time course of total integral intensity of EGFR (green) and p-EGFR (red) in endosomes measured by a FRET microscopy assay in HeLa EGFR BAC cells after continuous stimulation with 10 ng/ml EGF. The total integral intensity is defined as the sum of integral intensities of all endosomes in an image normalized by the area covered by the cells (for details see ‘Materials and methods’ and Supplementary information). (B) Time course of mean integral intensity per endosome for total EGFR (green curve) and p-EGFR (red curve) as in (A). Intensity curves (A–B) were normalized to the intensity value at 10 min. Crosses show the corresponding values after 1 min of EGF stimulation and incubation in ligand-free medium for 10 or 30 min (pulse-chase). (C) Time course of histogram distributions of the total EGFR integral intensity per endosome upon EGF stimulation as in (A). (D) Time course of histogram distributions of the p-EGFR integral intensity per endosome upon EGF stimulation as in (A). In both graphs, receptors in CCVs are responsible for the width of the distribution at 3 min (red curves in C and D). For comparison, histogram amplitude in B and C were normalized by each curve integral. In each graph, the integral intensity values were scaled by the mode of the histogram at 10 min. The experimental points from all histograms were fitted with a log-normal distribution. (E–F) Distribution of p-EGFR in endosomes as a function of EGF concentration after continuous stimulation for 30 min. Mean number of endosomes with EGFR (green curve) and p-EGFR (red curve) per 1000 μm2 of the area covered by cells (E) and mean integral intensity of EGFR (green curve) and p-EGFR (red curve) per endosome (F). On panel (F) curves were normalized to the intensity value at 10 ng/ml EGF. Lines are hyperbolic fits (E) or least square fits (F) to the experimental points. In both cases insets show the same graphs in linear scale. The different magnitude of the error bars in (E) and (F) is due to the averaging by the total number of images (E) or the total number of endosomes (F). In all cases, points show mean ± SEM. All measurements were done in three independent experiments with a total of ∼150 cells per time point or condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.004
10.7554/eLife.06156.005
Bleed-through correction for p-EGFR detection by FRET microscopy.
(A) Representative image of HeLa cells with no EGFR-GFP expression stained with an anti-p-Tyr antibody directly labelled with AlexaFluor 555 used to quantify the amount of fluorescence bleed-through. Scale bars, 10 μm. (B) Distribution of the ratio of FRET-p-Tyr maximum intensities of individual colocalized objects in the FRET and p-Tyr channel. Since there is no GFP fluorescence, these objects give an estimation of fluorescence bleed-through (filled circles). The continuous black line is the fit of three Gaussian components (shown in coloured dashed lines). The mean and variance of the red and blue curves were used for image corrections (see ‘Materials and methods’ for details). (C) Mean FRET-p-Tyr intensity distribution before (black curve) and after correction (red curve).
DOI:http://dx.doi.org/10.7554/eLife.06156.005
10.7554/eLife.06156.006
EGFR and p-EGFR measurements by FRET microscopy.
(A) Representative images of HeLa EGFR-GFP BAC cells after continuous stimulation with 10 ng/ml EGF for the indicated time points. EGFR-GFP fluorescence is shown in green, the corrected p-EGFR intensity is shown in red, and DAPI-stained nuclei are shown in blue. Measurements from each individual endosome were used for all quantifications. Scale bars, 10 μm. (B–C) Time course of histogram distributions of the total p-EGFR (B) or EGFR (C) integral intensity per endosome upon EGF stimulation as in Figure 1. The histogram shows the number of vesicles per 1000 μm2 of the area covered by cells. Intensity values were scaled by the mode of the histogram at 10 min. In all graphs, experimental points were fitted with a log-normal distribution. Points show the mean from three independent experiments with a total of ∼150 cells per time point or condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.006
10.7554/eLife.06156.007
BAC expression of EGFR-GFP does not change EGF transport kinetics.
(A) Representative Western blot of comparing the expression of EGFR and EGFR-GFP in HeLa Kyoto and HeLa EGFR-GFP BAC cells. The lower band corresponds to the untagged receptor, whereas the upper band corresponds to EGFR-GFP, which is absent in HeLa Kyoto cells. (B) Time course of EGF integral intensity in endosomes in HeLa Kyoto (black curve) and HeLa EGFR-GFP BAC cells (red curve). Intensity curves were normalized to the intensity value at 10 min for HeLa Kyoto cells. (C) Comparison of both time courses after dividing the HeLa EGFR-GFP BAC cells by 2. Squares show the difference between both curves. Experimental points show mean ± SEM from one representative experiment with a total of ∼150 cells per time point and condition. Time courses were fitted as in Figure 1.
DOI:http://dx.doi.org/10.7554/eLife.06156.007
10.7554/eLife.06156.008
Validation of FRET measurements with a specific anti-Tyr1068 antibody.
Representative images of p-EGFR staining by an antibody against a single phospho-tyrosine residue of EGFR after 0, 10, and 30 min of continuous stimulation with 10 ng/ml EGF. Scale bars, 10 μm. Time course of the p-EGFR mean integral intensity per endosome measured by standard immunofluorescence (blue) and by FRET assay (red). For comparison, both curves were normalized to the value at 10 min. Experimental points were fitted as in Figure 1A.
DOI:http://dx.doi.org/10.7554/eLife.06156.008
10.7554/eLife.06156.009
PFA fixation does not significantly change endosome EGFR endosome morphology.
(A) Representative images of HeLa BAC cells expressing EGFR-GFP after 20 min of stimulation with 10 ng/ml EGF before (left panel) or after fixation (right panel). Scale bars, 10 μm. (B) Histogram distribution of EGFR endosome area before (blue) and after fixation (red). The histogram shows the number of EGFR endosomes per 1000 μm2 of the area covered by cells. Measurements were taken from ∼500 cells from one representative experiment.
DOI:http://dx.doi.org/10.7554/eLife.06156.009
10.7554/eLife.06156.010
The total amount of p-EGFR in endosomes decays with the same kinetics as the number of endosomes with p-EGFR.
Time course of total integral p-EGFR intensity in endosomes (red) and endosomes with p-EGFR (black) per 1000 μm2 of the area covered by cells (black) after stimulation with 10 ng/ml EGF as in Figure 1. Points show mean ± SEM. All measurements were done in three independent experiments with a total of ∼150 cells per time point or condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.010
10.7554/eLife.06156.011
p-EGFR has a narrower integral intensity per endosome distribution than the total EGFR at late time points.
(A–E) Histogram distributions for the p-EGFR (red) or total EGFR (green) integral intensity per endosome at 5 (A), 10 (B), 15 (C), 30 (D), and 60 (E) min of continuous stimulation with 10 ng/ml EGF. For comparison, the amplitude of all histograms was normalized by the curve integral and the integral intensity was scaled by the mode of each histogram. In all graphs, experimental points were fitted with a log-normal distribution. Points show the mean from three independent experiments with a total of ∼150 cells per time point or condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.011
10.7554/eLife.06156.012
The mean amount of p-EGFR per endosome increases at high concentrations of EGF.
(A) Mean integral intensity of EGFR (green curve) and p-EGFR (red curve) per endosome upon stimulation with different EGF concentrations for 30 min (B) Time course of mean integral intensity of p-EGFR per endosome after continuous stimulation with 10 ng/ml (black curve) or 100 ng/ml (red curve) EGF. Curves were normalized by the intensity value at 10 min for 10 ng/ml EGF. Experimental points were fitted as in Figure 1A.
DOI:http://dx.doi.org/10.7554/eLife.06156.012
10.7554/eLife.06156.013
The mean p-EGFR amount per endosome does not correlate with endosome area at late time points after EGF stimulation.
(A) Mean p-EGFR integral intensity per endosome as a function of endosome area upon 10 (black curve) or 30 min (red curve) of EGF stimulation. Both curves were normalized to the intensity value at 1 μm2 for 10 min stimulation. (B) Histogram distribution of endosome area upon 10 (black curve) or 30 min (red curve) of EGF stimulation. Each histogram was normalized by its respective curve integral. Points show mean ± SEM. All measurements were done in three independent experiments with a total of ∼150 cells per time point or condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.013
We next determined the distribution of EGFR and p-EGFR in individual endosomes. The number of endosomes with p-EGFR decayed with similar kinetics as the total p-EGFR signal (τdecay N-p-EGFR = 45.24 ± 11.39 vs τdecay p-EGFR = 30.97 ± 1.69; compare red with black curve in Figure 1—figure supplement 6). The mean content of total EGFR per endosome increased over time and then rapidly decayed reaching steady state, due to the balance of continuous EGF uptake and degradation (Figure 1B, green curve). After a rapid increase, the mean content of p-EGFR in each endosome stabilized to a fairly constant level after ∼20 min (Figure 1B, red curve). Similar results were obtained when EGF was pulsed for 1 min and chased for different periods of time (Figure 1B, blue and black points).
To determine how the endosomal content of p-EGFR originates over time, we compared the distributions of EGFR and p-EGFR content per endosome. The width of distribution of total EGFR increased with time (Figure 1C), due to the fact that, as EGF continues to flow in, it first enters small early endosomes and progressively accumulates in larger ones (Rink et al., 2005). In contrast, the p-EGFR distribution first widened like that of total EGFR but then became almost twofold narrower than that of EGFR (Figure 1—figure supplement 7A–E compare the red and green curves) and stabilized after 30 min (Figure 1D). These results suggest an unexpected behaviour of p-EGFR, which over time stabilizes at a constant mean level per endosome.
Surprisingly, the mean amount of p-EGFR in endosomes was not ligand dependent. We stimulated cells with different concentrations of EGF for 30 min, when the amount of p-EGFR per endosome reached its steady state (Figure 1B,D). Once again, we found that EGFR and p-EGFR behaved very differently. The number of endosomes containing EGFR saturated already at low concentrations of EGF (0.5–1.0 ng EGF; Figure 1E, green curve) whereas the amount of total EGFR per endosome increased almost linearly (Figure 1F, green curve). This is expected because the higher the concentration of EGF, the higher the internalization of EGFR, whereas the number of receiving endosomes does not change significantly. In contrast, the number of endosomes with p-EGFR augmented with increasing EGF concentrations (Figure 1E, see red curve in semi-logarithmic scale, inset in linear scale). Strikingly, the mean amount of p-EGFR per endosome remained fairly constant, despite the EGF concentration varying almost over three orders of magnitude (Figure 1F, red curve). Therefore, increasing concentrations of EGF resulted in an increase in the number of endosomes with the same mean package of p-EGFR. Importantly, such a packaging is saturable because at high EGF concentrations the mean p-EGFR content per endosome was no longer constant with time (Figure 1—figure supplement 8).
The finding that endosomes contain a constant mean level of p-EGFR is striking. We performed several control experiments to verify that this is not an artefact caused by the FRET method or the assay. First, the mean amount of p-EGFR per endosome did increase at EGF concentrations higher than 10 ng/ml (Figure 1—figure supplement 8), indicating that the value measured is not artificially fixed, for example, by limited antigen accessibility. Second, a similar constant mean value of p-EGFR per endosomes was estimated with an independent method using the Tyr1068 antibody (Figure 1—figure supplement 4). Third, the narrow distribution of p-EGFR per endosome may simply reflect the sorting into endosomes of regular size. Whereas at 10 min the p-EGFR amount per endosome increased with the endosome area (Figure 1—figure supplement 9A, black curve), at 30 min (steady state, Figure 1B), the same mean amount was present in small and large endosomes alike (Figure 1—figure supplement 9A, red curve). Therefore, the amount of activated receptors per endosome is independent of endosome area. Finally, we verified that it is not a phenomenon peculiar to HeLa cells but also occurring in non-immortalized, non-cancer cell lines. Using the anti-phosphoTyr1068 antibody, we found that in primary mouse hepatocytes upon EGF stimulation the mean amount of p-EGFR per endosome saturated at ∼20 min whereas the mean amount of EGF continued to grow (data not shown), indicating that the packaging of p-EGFR in endosomes is not peculiar to a signalling-aberrant cancerous cell line.
In which endocytic compartment was p-EGFR packaged so uniformly? Nearly 80% of p-EGFR colocalized with the early endosomal marker EEA1 throughout the time course (Figure 2—figure supplement 1A). Less than 10% of p-EGFR colocalized with APPL1 after 15 min showing that it passed this endosomal compartment (Miaczynska et al., 2004) (Miaczynska et al., submitted). Very little p-EGFR colocalized with LAMP-1, a marker of late endosomes and lysosomes (Figure 2—figure supplement 1C). One possibility is that the packages of p-EGFR may reflect incorporation into intra-luminal vesicles (ILV) of multi-vesicular bodies (MVB). This possibility was ruled out using a previously described differential detergent solubilisation method (Malerød et al., 2007). We could determine that a large fraction of EGFR was not accessible to antibodies upon digitonin permeabilization, reflecting sequestration into ILVs (Malerød et al., 2007; Piper and Katzmann, 2007) (see Suppl. information and Figure 2A,B). In contrast, p-EGFR was always detectable suggesting that it was not within ILVs.10.7554/eLife.06156.014
The constant mean amount of p-EGFR per endosome corresponds to receptor clusters that are regulated by Hrs and PTPN11.
(A) Representative images of total EGFR and p-EGFR after staining with saponin or digitonin permeabilization methods. Immunofluorescence staining of LBPA is shown as a control marker for ILVs in MVBs. Scale bars, 10 μm. (B) Integral intensity of EGFR, p-EGFR, and LBPA (mean ± SEM) after permeabilization with digitonin or saponin. **p < 0.005 by a two-tailed t-test. Measurements were done in three independent experiments with a total of ∼150 cells per condition. (C) Time course of mean integral intensity per endosome for ub-EGFR (blue curve) upon EGF stimulation as in Figure 1A. p-EGFR is included for comparison. (D) Representative STORM images of p-EGFR (red) stained using a rabbit monoclonal anti-p-EGFR (Tyr 1068) antibody overlaid on top of a high magnification confocal image of EGFR (green). Left panels show clusters of p-EGFR upon stimulation with EGF for 10 or 30 min. Right panels show clusters of p-EGFR upon stimulation with EGF for 30 min in Hrs down-regulation or mock treatment. (E) Time course of the mean p-EGFR integral intensity per endosome in Hrs (red), Snf8 (blue), Vps24 (green), or mock-treated cells (black) (using three different siRNA oligonucleotides per gene). All curves were normalized by the intensity value at 10 min for the mock sample. Points show mean ± SEM from three different siRNAs per gene. Scale bar, 1 μm. (F) Integral intensity distribution of p-EGFR per endosome after down-regulation for 72 hr of PTPN11 (red) or in mock treatment (black) after continuous stimulation with 10 ng/ml EGF for 30 min. Red points show the average distribution of three different siRNAs. Experimental points were fitted as in Figure 1A.
DOI:http://dx.doi.org/10.7554/eLife.06156.014
10.7554/eLife.06156.015
The majority of the p-EGFR is located in EEA1-positive endosomes.
(A) Time course of integral intensity of total EGFR (green curve) and p-EGFR (red curve) colocalizing with EEA1 after continuous stimulation with 10 ng/ml EGF for different time points. Intensity curves were normalized to the intensity value at 10 min (B–C) Fraction of the total integral intensity of EGFR (green curve) or p-EGFR (red curve) colocalized with EEA1 (B) or LAMP-1 (C). In all cases, points show mean ± SEM. Measurements were done in three independent replicates with a total of ∼150 cells per time point or condition. Time courses were fitted to obtain the decay half-time τ (for details on the fitting procedure see ‘Materials and methods’).
DOI:http://dx.doi.org/10.7554/eLife.06156.015
10.7554/eLife.06156.016
ub-EGFR measurements by FRET microscopy.
(A) Representative images of HeLa EGFR-GFP BAC cells after continuous stimulation with 10 ng/ml EGF for the indicated time points. EGFR-GFP fluorescence is shown in green, the corrected ub-EGFR intensity is shown in red. Measurements from each individual endosome were used for all quantifications. Scale bars, 10 μm. (B–C) Heat map of the 2D co-distribution of ub-EGFR and EGFR (B) or ub-EGFR and p-EGFR (C) integral intensity per endosome upon EGF stimulation as in Figure 1 after 10, 30, or 60 min. ub-EGFR and EGFR are well correlated, whereas the distribution of p-EGFR is significantly narrower than ub-EGFR at 30 and 60 min. Heat maps show the result of one representative experiment.
DOI:http://dx.doi.org/10.7554/eLife.06156.016
10.7554/eLife.06156.017
Quantification of number of EGFR and pEGFR molecules per endosome.
(A) Distribution histogram of the differences in intensity between individual endosomes in consecutive frames during sequential photo-bleaching of EGFR-GFP (see ‘Materials and methods’ for details). (B) Difference between the number of positive and negative events in (A) for each ΔIntensity value. (C) Distribution histogram of the differences in intensity between individual endosomes in consecutive frames during sequential photo-bleaching of p-EGFR (see ‘Materials and methods’ for details). (D) Difference between the number of positive and negative events in (C) for eachΔIntensity value. The local amplitude maxima of the periodic function in (B) and (D) give an estimate of the change in intensity values when 1,2,3, …, n number of molecules are bleached. (E) Distribution of the number of molecules of EGFR-GFP and p-EGFR in individual endosomes after stimulation with 10 ng/ml EGF for 10 min.
DOI:http://dx.doi.org/10.7554/eLife.06156.017
10.7554/eLife.06156.018
Hrs, but not ESCRT-II or ESCRT-III components, increases the mean p-EGFR amount per endosome.
(A–B) Histogram distributions of the total EGFR (green) or p-EGFR (red) integral intensity per endosome after 30 min of EGF stimulation as in Figure 5 for Hrs (A) or mock-treated cells (B). (C) Representative images of HeLa EGFR BAC cells after continuous stimulation with 10 ng/ml EGF for 30 min and Hrs, Snf8, or Vps24 knock-down. Scale bars, 10 μm. (D) Changes in EGFR at the endosomal surface after knock-down of different ESCRT components measured by the differential permeabilization assay shown in Figure 2A Bar graphs show mean ± SEM. Measurements were done in three independent experiments using three different siRNA oligonucleotides with a total of ∼150 cells. In all graphs, experimental points were fitted with a log-normal distribution. Points show the mean from three independent experiments with a total of ∼150 cells per time point or condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.018
10.7554/eLife.06156.019
Kinetics of Shc1 recruitment to endosomes.
(A) Representative images of HeLa EGFR-GFP BAC cells before (left panel) and after 10 min stimulation with 10 ng/ml EGF (right panel). EGFR fluorescence is shown in green and Shc1 is shown in red. (B) Time course of mean integral intensity of Shc1 per endosome (black curve). The mean intensity of p-EGFR per endosome (red curve) is included for comparison. (C) Mean integral intensity of Shc1 per endosome as a function of EGF concentration. The curve was normalized to the intensity value at 10 ng/ml EGF. The solid line is a least square fits to the experimental points. In all cases, points show mean ± SEM. Measurements were done in three independent replicates with a total of ∼150 cells per time point or condition. Time courses were fitted as in Figure 1.
DOI:http://dx.doi.org/10.7554/eLife.06156.019
10.7554/eLife.06156.020
Pharmacological inhibition of EGFR kinase rapidly decreases the total p-EGFR in endosomes only at high but not low EGF concentrations.
Time course of the total p-EGFR integral intensity upon stimulation with 10 (black and green curves) or 100 ng/ml (red and blue curves) of EGF. AG1478 (green and blue curves) was added 10 min after stimulation with EGF and remained in the medium throughout the time course. All curves were normalized by the intensity value at 10 min for the DMSO—10 ng/ml sample. Experimental points were fitted as in Figure 1. Points show mean ± SEM of ∼150 cells per time point and condition from one representative experiment.
DOI:http://dx.doi.org/10.7554/eLife.06156.020
10.7554/eLife.06156.021
Pharmacological inhibition of EGFR kinase activity increases the mean p-EGFR amount per endosomes.
(A) Representative images of HeLa EGFR BAC cells after continuous stimulation with 10 ngl/ml EGF for 30 min. Inhibitors were added 10 min after stimulation with EGF and remained in the medium throughout the time course. Scale bars, 10 μm. (B) Time course of the mean p-EGFR integral intensity per endosome in AG1478 (red), Gefitinib (blue), Lapatinib (green), or DMSO-treated cells (black). All curves were normalized by the intensity value at 10 min for the mock sample. Experimental points were fitted as in Figure 1. Measurements for AG1478 were done in three independent experiments; measurements from Gefitinib and Lapatinib show a representative experiment with a total of ∼150 cells per time point and condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.021
10.7554/eLife.06156.022
Phosphatases can control the p-EGFR packaging in endosomes.
(A) Total p-EGFR integral intensity after continuous stimulation with 10 ng/ml EGF in HeLa EGFR-GFP BAC cells and down-regulation of the indicated phosphatase or treatment with transfection reagent only. (B) Mean p-EGFR integral intensity per endosome after continuous stimulation with 10 ng/ml EGF in HeLa EGFR-GFP BAC cells and down-regulation of the indicated phosphatase or treatment with transfection reagent only. In both cases, bars are colour coded according to their respective phosphatase. All phosphatases were down-regulated with at least three oligos. Bars show mean ± SEM of all images for each oligo.
DOI:http://dx.doi.org/10.7554/eLife.06156.022
How do the kinetics of p-EGFR endosomal packaging compare with the kinetics of receptor dephosphorylation and ubiquitylation? After 10 min of EGF stimulation, the pool of p-EGFR in EEA1-positive endosomes (Figure 2—figure supplement 1A, red curve) decayed faster than total EGFR (Figure 2—figure supplement 1A, green curve; τdecay EGFR = 56.03 ± 5.72, τdecay p-EGFR = 37.08 ± 4.19), possibly due to de-phosphorylation or preferential removal of p-EGFR from early endosomes. The latter can be excluded since the fraction of p-EGFR in EEA1-positive endosomes remained almost constant throughout the time course (Figure 2—figure supplement 1B). EGFR ubiquitylation is required for its internalization into endosomes, and this is dependent on EGFR phosphorylation and the recruitment of the c-Cbl E3 ligase (Sigismund et al., 2013). To compare the levels of ubiquitylated EGFR (ub-EGFR) with those of p-EGFR within the endosomal system, we modified the FRET assay using an anti-ubiquitin antibody (Figure 2—figure supplement 2A). The kinetics of ub-EGFR were significantly different from those of p-EGFR. Whereas the levels of p-EGFR peaked at 15 min, ub-EGFR reached its maximum at 30 min after stimulation (Figure 2C, compare red with blue curves) and decreased more slowly than p-EGFR at later times, probably reflecting deubiquitylation prior to receptor sequestration into ILVs (Piper and Katzmann, 2007). These results are consistent with the fact that the appearance of p-EGFR precedes that of ub-EGFR (Umebayashi et al., 2008). Moreover, ub-EGFR had a similar distribution to that of EGFR (Figure 2—figure supplement 2B) but significantly wider than p-EGFR (Figure 2—figure supplement 2C), suggesting that the mechanisms responsible for stabilizing the mean levels of p-EGFR per endosome are not correlated with receptor ubiquitylation.
Our data suggest the existence of a saturable mechanism adjusting the amount of p-EGFR in each individual endosome. Such a constant mean amount may be due to the formation of small clusters within early endosomes. To test this possibility, we imaged the spatial distribution of p-EGFR in endosomes using the anti-EGFR phosphoTyr1068 antibody by super-resolution microscopy. Using direct Stochastic Optical Reconstruction Microscopy (dSTORM) (Lampe et al., 2012), we could indeed visualize clusters of p-EGFR (Figure 2D, left panel) that decreased in size between 10 and 30 min of EGF internalization, in agreement with the narrowing of p-EGFR distribution over time (Figure 1D). To determine the number of molecules in the clusters, we used two methods. First, we developed a new method to estimate the number of fluorescent molecules in light microscopy images by measuring the intensity fluctuations during photo-bleaching over time (for details see ‘Materials and methods’ and Figure 2—figure supplement 3). Based on the fluorescence signal from the anti-phosphoTyr1068 antibody and EGFR-GFP, we estimated an average of 102 ± 38 and 76 ± 29 (Mean ± SEM) molecules of EGFR and p-EGFR per endosome 30 min after EGF (10 ng/ml) internalization (Figure 2—figure supplement 3), corresponding to 707 ± 265 and 527 ± 202 molecules per μm3 of endosomal volume (apparent, assessed by light microscopy), respectively. A hundred EGFR molecules would require ∼12 clathrin-coated vesicles for delivery to endosomes (see ‘Materials and methods’). We also estimated the total number of GFP-EGFR per cell and found values (29,000) well in agreement with previous estimates for HeLa cells (see ‘Materials and methods’ and Figure 2—figure supplement 1A,B). Second, based on the size of receptor from the PDB database (structure ID: 3NJP), we calculated that 83 ± 25 (Mean ± SEM, N = 1456) receptors could fit in the apparent area of p-EGFR visualized by dSTORM, a value which is remarkably in agreement with the fluorescence intensities estimates.
To further validate that the constant mean amount of p-EGFR per endosome corresponds to receptor clusters, we performed a focused RNAi screen on established components of the endosomal receptor sorting machinery (CHLCb, CHLCa, Hip1, Hip1R, Htt, Tom1, Tollip, Tom1L1, Tom1L2, Hrs, Snf8, Vps24). We found that only Hrs depletion resulted in a continuous accumulation of p-EGFR in endosomes with time (Figure 2E). At the same time, the p-EGFR intensity distribution widened similar to that of total EGFR (Figure 2—figure supplement 4A,B). The silencing of Hrs also caused an increase in the size of p-EGFR clusters within each endosome as revealed by dSTORM (Figure 2D, right panel). Interestingly, this is not due to the inhibition of ILV formation, as down-regulation of Snf8 and Vps24, members of the ESCRT-II and ESCRT-III complexes, respectively (Piper and Katzmann, 2007), reduced the sequestration of EGFR into the endosomal lumen (Figure 2—figure supplement 4D) but did not have significant effects on the amount of p-EGFR per endosome (Figure 2E, blue and green curves). Thus, the constant mean amount of p-EGFR in endosomes likely corresponds to the receptor clusters observed by super-resolution microscopy. This raises the question of whether p-EGFR can be accessible to downstream signalling components. Therefore, we measured the recruitment of a direct downstream effector of p-EGFR, Shc1. The kinetics of Shc1 recruitment to endosomes precisely mimic the kinetics of p-EGFR (Figure 2—figure supplement 5) arguing that the p-EGFR clusters are signalling competent.
Upon internalization, EGF enters the early endosomal network and, similar to LDL (Rink et al., 2005), following endosome homotypic fusion and fission reactions, accumulates in few large endosomes prior to transfer to late endosomes. A mechanism must exist that prevents the continuous accretion of p-EGFR upon endosome fusion. A simple mechanism could be that the de-phosphorylation rate increases with the increase in p-EGFR per endosome. When two endosomes fuse, the resulting endosome should contain the sum of EGFR and p-EGFR of the original endosomes. However, given such de-phosphorylation rate dependency, the amount of p-EGFR would return to the level prior to fusion, thus stabilizing the mean amount of p-EGFR per endosome. A prediction of this hypothesis is that the kinase activity of EGFR in endosomes controls its own dephosphorylation. To test this, we inhibited the EGFR kinase activity pharmacologically with AG1478, lapatinib or gefitinib 10 min after EGF stimulation (to prevent alterations on receptor internalization) and determined the effects on the receptors already internalized and phosphorylated. We compared low with high concentrations of EGF, that is, under conditions of saturation of p-EGFR packaging in endosomes (Figure 1—figure supplement 8). At low EGF concentrations, when the packaging mechanism is not saturated, the total amount of p-EGFR was not significantly reduced by the inhibitors (Figure 2—figure supplement 6 compare black and green curves). This behaviour argues that the packages of p-EGFR in endosomes are protected from the phosphatases. In addition, the inhibitors caused a continuous accumulation of p-EGFR in fewer and larger endosomes over time (Figure 2—figure supplement 7). In contrast, adding the inhibitor after stimulation with high concentrations of EGF caused a sharp reduction in the total amount of p-EGFR (Figure 2—figure supplement 6 compare red and blue curves), as observed previously (Kleiman et al., 2011). This means that the kinase activity of EGFR is necessary to maintain the levels of p-EGFR in individual endosomes. These results support the idea that the dephosphorylation of p-EGFR in endosomes indeed depends on the EGFR activity within the endosomal packages.
Which phosphatases are responsible for controlling p-EGFR packaging in endosomes? To identify them, we performed a focused RNAi screen against 21 protein tyrosine phosphatases (PTP) expressed in HeLa cells (Tarcic et al., 2009). Hits were defined if silencing satisfied three conditions: (1) it increased the total amount of p-EGFR in endosomes and (2) increased the mean amount of p-EGFR per endosome, and (3) the phenotype was observed with at least two siRNAs per gene. Five phosphatases, PTP4A1, PTPN11, PTPN9, PTPN18, and PTPRK, increased the amount of p-EGFR in individual endosomes (Figure 2F and Figure 2—figure supplement 8). Interestingly, PTPN11 is an EGFR interactor (Deribe et al., 2009) whose activity is enhanced upon tyrosine phosphorylation (Agazie and Hayman, 2003), suggesting a molecular mechanism whereby p-EGFR could regulate its own de-phosphorylation in endosomes.
What are the consequences of such mechanism for signal transduction? To address these questions and generate testable predictions, we developed a mathematical model that describes the amount of total intracellular p-EGFR over time. Previously, excellent models have been developed that quantitatively describe EGFR endocytosis and signalling (Felder et al., 1992; French et al., 1994; Kholodenko et al., 1999; Kholodenko, 2002; Resat et al., 2003). However, although all these models described in detail the dynamics of ligand binding, dimer formation and endocytosis, recycling and degradation of the receptor, they did not consider the trafficking dynamics of the phosphorylated receptors with respect to the dynamics of the endosomal network because these data were not available. Our new experimental data brought two new concepts. First, dephosphorylation and degradation of p-EGFR occur sequentially but are uncoupled. Second, the amount of p-EGFR is controlled at the level of individual endosomes. These new concepts require further development of the existing EGFR mathematical models. Our model was formulated as a set of ordinary differential equations (ODE, see ‘Materials and methods’ and Figure 3) describing (1) the total amount of EGFR and p-EGFR at the plasma membrane as a function of ligand binding, (2) endocytosis of p-EGFR and its indirect effects on EGFR endocytosis, and (3) distribution of cargo between early endosomes at different stages of maturation (e.g., formation of MVB). For this, we considered the processes of receptor internalization, dephosphorylation, degradation, recycling, endosome fusion and fission. As in previous models (French et al., 1994; Resat et al., 2003), we described time course kinetics of total cellular p-EGFR, surface and endosomal EGFR and p-EGFR. Importantly, our model also describes the total number of p-EGFR-positive endosomes and mean amount of p-EGFR per endosome (see ‘Materials and methods’ for details). To account for the observed stabilization of the mean amount of p-EGFR per endosome over time (Figure 1), the dependency of p-EGFR dephosphorylation on EGFR kinase activity (Figure 2—figure supplement 6,7) and the fact that the mechanism is saturable (Figure 1—figure supplement 8), we included a sigmoidal dependency of the p-EGFR dephosphorylation rate on the amount of p-EGFR per endosome. The model was then fitted to the experimental data from the p-EGFR time course (Figure 3A,B). Figure 3C shows that this simple theoretical model can reproduce our observations of a constant mean amount of p-EGFR per endosome in a wide range of EGF concentrations when fitted to the experimental data. Importantly, a model without this non-linear dephosphorylation dependency could correctly describe the total amount of EGFR and p-EGFR in endosomes (Figure 3—figure supplement 1A,B) but did not agree with the measurements for the mean amount of p-EGFR per endosome (Figure 3—figure supplement 1C), thus supporting the sigmoidal dependency of the p-EGFR de-phosphorylation rate on the amount of p-EGFR per endosome (Figure 3). Previous models did not include this non-linear term because data on the distribution of p-EGFR in individual endosomes was not available.10.7554/eLife.06156.023
Mathematical model of p-EGFR predicts signalling amplitude and duration depends on early endosome fusion/fission rate.
Parameters of the mathematical model were fitted to the experimentally measured number of p-EGFR endosomes, total integral intensity of p-EGFR, mean integral intensities of p-EGFR per endosome and total vesicular EGFR. The experimental data were obtained in a time course of EGF stimulation at four concentrations (0.5, 1.0, 5.0, and 10 ng/ml, colour coded as indicated). The fit results are presented on panels (A–C). The experimental data and model predictions are drawn as filled circles and solid curves, respectively. (A) Number of p-EGFR endosomes per 1000 μm2 of cell area. (B) Total integral intensity of p-EGFR measured by FRET. The scaling factors that convert arbitrary numbers of the model to the experimental data were found by the least square procedure (see ‘Materials and methods’). (C) Comparison of mean integral intensity of p-EGFR per endosome measured experimentally (filled circles) and mathematical model (solid curves) of the time course of p-EGFR upon EGF stimulation as in Figure 1A. The concentration of EGF is colour coded as presented. (D) Model predictions of the total amount of p-EGFR in endosomes as a function of EGF concentration and in the presence of different homotypic early endosome fusion rates (colour coded as indicated).
DOI:http://dx.doi.org/10.7554/eLife.06156.023
10.7554/eLife.06156.024
A mathematical model without the non-linear phosphorylation dependency cannot describe the mean amount of p-EGFR per endosome.
Parameters of a mathematical model with a first-order dephosphorylation rate were fitted to the experimental data as in Figure 3. (A) Total integral intensity of EGFR. (B) Total integral intensity of p-EGFR measured by FRET. (C) Mean integral intensity of p-EGFR per endosome. The experimental data and model predictions are drawn as filled circles and solid curves, respectively.
DOI:http://dx.doi.org/10.7554/eLife.06156.024
An unexpected prediction of our model is that the total de-phosphorylation rate, and thus the total amount of p-EGFR, is dependent on the fusion/fission rate of the endosomes (Figure 3D). If so, could this have an effect on signal transduction? To test these hypotheses, we reduced early endosome homotypic fusion by lowering the intracellular concentration of established components of the endosome tethering and fusion machinery, EEA1, Rabenosyn5, Vps45 (Christoforidis et al., 1999; Ohya et al., 2010), Syntaxin-6 and Syntaxin-13 (Brandhorst et al., 2006) that play no direct role in signalling. These genes were down-regulated by RNAi in combinations and only partially (∼50–70% depletion for each protein, Figure 4A) to achieve a significant inhibition of endosome fusion and yet prevent or reduce cell toxicity. This procedure caused a mild redistribution of EGFR to endosomes of smaller size (<0.5 μm2 cross-section area, for details see ‘Materials and methods’ and Figure 6—figure supplement 2) (Figure 4C,D). Similar results were obtained upon depletion of a second combination of genes (EEA1, Stx13, Stx6, not shown, see below). Note that these treatments generated a pattern of endosomes similar to that observed in different cell types and under different culture conditions (see below, Figure 6) and neither altered the surface levels of EGFR (Figure 4—figure supplement 1) nor its kinetics of uptake (Figure 4B) and exit from endosomes, that is, recycling and degradation (Figure 4—figure supplement 2). We also excluded potential effects on endosome acidification, because blocking it with bafilomycin did increase both p-EGFR and total EGFR (Figure 4—figure supplement 3). Remarkably, under our experimental conditions of mild down-regulation of the early endosomal fusion machinery the packaging of active receptors was unaffected as shown by both the time course and the steady-state mean (constant) amount of p-EGFR per endosome (Figure 4E; see above, Figure 1B). In contrast, the total number of endosomes with p-EGFR and their life-time augmented (Figure 4F), resulting in a net increase in the total amount and life-time of p-EGFR (Figure 4G). Notably, reduction of the endosome fusion rate in the mathematical model (∼40%, in line with the depletion of tethering proteins, Figure 4A) is sufficient to reproduce fairly well the experimental increase in p-EGFR endosomes observed (Figure 4F). These results support the hypothesis that EGFR activation can be modulated by the endosomal system. Since p-EGFR de-phosphorylation precedes EGFR degradation (see above, Figure 1A, Figure 2—figure supplement 2) and EGFR degradation is unaffected (Figure 4B, Figure 4—figure supplement 2), we deduce that the effect on the life-time of p-EGFR caused by reduced endosomal fusion is primarily due to reduced de-phosphorylation.10.7554/eLife.06156.025
Increasing the number and life-time of p-EGFR endosomes results in prolonged EGFR activation.
(A) Protein down-regulation of EEA1 and Rabenosyn5 72 hr after siRNA transfection. RT-PCR showed an 80% reduction in Vps45 mRNA levels (data not shown). (B) Time course of EGFR integral intensity in endosomes after partial protein depletion of EEA1, Rabenosyn5, and Vps45 (red curve) or mock treatment (black curve). Cells were given a 1-min pulse of 10 ng/ml EGF, washed and chased for the indicated time points before fixation. (C) Representative images of HeLa EGFR BAC cells after EEA1, Rabenosyn5, and Vps45 knock-down or treatment with transfection reagent only (mock). Scale bars, 10 μm. (D) Shift in the EGFR-endosome area distribution toward smaller endosomes after EEA1, Rabenosyn5, and Vps45 knock-down. The values of the histograms of endosome area distribution for the control and knock-down conditions were normalized and subtracted. The curve shows the relative increase (above zero) or reduction (below zero) in the number of endosomes for each area bin (in logarithmic scale) (for details see ‘Materials and methods’ and Figure 6—figure supplement 2). Experimental points were fitted with two log-normal distributions. (E–G) Changes in p-EGFR endosomes in EEA1, Rabenosyn5, and Vps45 knock-down (red curve) or mock-treated (black curve) cells after continuous stimulation with 10 ng/ml EGF. Time courses of the mean integral intensity of p-EGFR per endosome (E), mean number of p-EGFR endosomes determined experimentally (squares) or predicted by the mathematical model (solid curves) for a 37% endosomes fusion rate (red curve) compared to control (black curve) (F), and total p-EGFR integral intensity in endosomes (G) measured as in Figure 1. Intensity curves were normalized to the intensity value at 10 min for mock-treated cells. Experimental points show mean ± SEM. All measurements were done in three independent experiments with a total of ∼150 cells per time point or condition. Time courses were fitted as in Figure 1.
DOI:http://dx.doi.org/10.7554/eLife.06156.025
10.7554/eLife.06156.026
Knock-down of fusion machinery does not change EGFR distribution at the plasma membrane in HeLa cells.
Cells were stimulated with 100 ng/ml EGF-AlexaFluor 488 for 10 min on ice to prevent receptor endocytosis. The AlexaFluor 488 signal was enhanced by detection with a specific antibody to detect the amount of EGFR at the plasma membrane. Scale bar, 10 μm. (A) Representative images of HeLa EGFR BAC cells after EEA1, Rabenosyn5, and Vps45 knock-down or treatment with transfection reagent only (mock). (B) Total intensity of EGF-AlexaFluor 488 (Mean ± SEM) in knock-down or control cells. The total intensity was normalized to the fraction of the area covered by cells. Measurements were done in three independent replicates with a total of ∼150 cells per time point or condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.026
10.7554/eLife.06156.027
Knock-down of fusion machinery does not change EGFR degradation in HeLa cells.
(A–B) Time course of EGFR degradation after partial protein depletion of the three endosomal fusion components EEA1, Rabenosyn5, and Vps45 or mock treatment and continuous stimulation with 10 ng/ml EGF for the indicated times in the presence of 10 μg/ml cyclohexamide in HeLa EGFR BAC cells. (A) Representative EGFR and γ-Tubulin Western blots and (B) their quantification for EEA1, Rabenosyn5, and Vps45 knock-down (red curve) or mock-treated (black curve) samples. Points show mean ± SEM from three independent experiments. Lines are linear fits to the experimental points.
DOI:http://dx.doi.org/10.7554/eLife.06156.027
10.7554/eLife.06156.028
Blocking endosome acidification with Bafilomycin increases both total EGFR and p-EGFR, but not the mean amount of p-EGFR per endosome.
(A) Time course of p-EGFR integral intensity in endosomes after incubation with 50 nM BafilomycinA1 (red curve) or 1% DMSO (blue curve) for 30 min and during the remaining of the time course. (B) Time course of EGFR integral intensity in endosomes after incubation with 50 nM BafilomycinA1 (red curve) or 1% DMSO (blue curve) for 30 min and during the remaining of the time course. (C) Time course of the mean p-EGFR integral intensity per endosome after incubation with 50 nM BafilomycinA1 (red curve) or 1% DMSO (blue curve) for 30 min and during the remaining of the time course. Experimental points show mean ± SEM. All measurements were done in three independent experiments with a total of ∼150 cells per time point and condition. Time courses were fitted as in Figure 1.
DOI:http://dx.doi.org/10.7554/eLife.06156.028
Increased EGFR phosphorylation results in sustained Erk signalling (Sasagawa et al., 2005; Nakakuki et al., 2010) and this leads to the phosphorylation and stabilization of the immediate early gene product c-Fos (Nakakuki et al., 2010). We asked whether the redistribution of endosomal EGFR could be sufficient to induce sustained Erk activation and c-Fos phosphorylation. Indeed, upon EGF stimulation, both the amplitude and duration of Erk1/2 phosphorylation were increased in the depleted cells compared to control (Figure 5A,B). Consistently, c-Fos phosphorylation was also higher after 30 min of EGF stimulation (Figure 5C,D). The fact that the amount and life-time of total EGFR in endosomes remained unvaried in these experiments (Figure 4B) eliminates the trivial possibility that the observed changes are due to modulation of receptor degradation.10.7554/eLife.06156.029
Redistribution of endosomal EGFR increases the amplitude and duration of MAPK signalling.
(A–B) Time course of Erk1/2 phosphorylation after partial protein depletion of the three endosomal fusion components EEA1, Rabenosyn5, and Vps45 or mock treatment and continuous stimulation with 10 ng/ml EGF for the indicated times in HeLa EGFR BAC cells. (A) Representative phospho-Erk1/2 and Erk1/2 Western blots and (B) their quantification for EEA1, Rabenosyn5, and Vps45 knock-down (red curve) or mock-treated (black curve) samples. Points show mean ± SEM from three independent experiments. The time course was fitted as in Figure 1. (C–D) Nuclear c-Fos phosphorylation in EEA1, Rabenosyn5, and Vps45 knock-down or mock-treated cells as in (A) after 30 min of EGF stimulation. (C) Representative images of EEA1 and phospho-c-Fos immunostaining in EEA1, Rabenosyn5, and Vps45 knock-down or mock-treated cells. Scale bars, 20 μm. (D) Total intensity of nuclear phospho-c-Fos in EEA1, Rabenosyn5, and Vps45 knock-down or mock-treated cells. Bar graph shows mean ± SEM. Measurements were done in three independent experiments from a total of ∼1000 cells per condition. *p < 0.05 by a 2-tailed t-test.
DOI:http://dx.doi.org/10.7554/eLife.06156.029
The experiments on HeLa cells and the theoretical analysis raise the question of whether modulation of early endosome homotypic fusion is a general mechanism to regulate signal amplitude and duration. If this were the case, we would predict that growth factors with different signalling outputs (amplitude and duration) differentially modulate the endosomal distribution (i.e., endosome number, size, and cargo content). To test this prediction, we examined different growth factors and cellular systems. First, we used primary mouse hepatoblasts where HGF promotes their proliferation (Tanimizu et al., 2003). In these cells, HGF but not EGF elicits a sustained Erk response (Figure 6—figure supplement 1). Indeed as predicted, stimulation of hepatoblasts with HGF caused a strong shift in the distribution of early endosomes toward smaller sizes (Figure 6A, red curve Figure 6B), whereas EGF had the opposite effect (Figure 6A, green curve, Figure 6B). Second, we turned to an in vitro model of reference for cell-fate decisions, PC12 cells. In PC12 cells, EGF stimulation leads to transient Erk phosphorylation and cell proliferation, whereas NGF leads to sustained Erk phosphorylation and cell differentiation (Marshall, 1995). Consistent with our results in primary mouse hepatoblasts, NGF stimulation in PC12 cells caused a significant shift in the distribution of early endosomes toward smaller sizes compared with EGF (Figure 6C,D). Moreover, NGF itself was distributed to a larger number of smaller endosomes in comparison with EGF (Figure 6E,F). Altogether, these data argue that the modulation of endosome fusion, reflected by the changes in endosome number and size, is a general property of growth factors. These data further suggest that signalling amplitude and duration can be regulated by changes in the fusion rate of endosomes (see Table 1).10.7554/eLife.06156.030
Growth factors differentially shift the distribution of the number and size of endosomes.
(A) Representative images of primary mouse hepatoblasts after stimulation with 10 ng/ml EGF or HGF for 30 min (B) Shift in the EEA1-positive endosome area distribution after stimulation with HGF (red curve) or EGF (green curve). The values of the histograms of endosome area distribution for growth factor stimulated and non-stimulated cells were normalized and subtracted. The curve shows the relative increase (above zero) or reduction (below zero) in the number of endosomes for each area bin (in logarithmic scale). HGF stimulation increased while EGF decreased the proportion of endosomes smaller than 0.2 μm2. (C–F) PC12 cells after stimulation for 30 min with 100 ng/ml EGF or 50 ng/ml NGF. (C) Representative images of EEA1-positive endosomes. (D) Shift in the EEA1-positive endosome area distribution after stimulation with NGF (red curve) or EGF (green curve) measured as in (B). NGF stimulation increased while EGF slightly decreased the proportion of endosomes smaller than 0.2 μm2. (E) Representative images of EGF or NGF. (F) Differences in the area distribution of endosomal NGF and EGF measured as in (B). NGF is enriched in endosomes smaller than 0.2 μm2 relative to EGF. For all graphs points show the mean ± SEM of experimental distributions. Measurements were done in three independent experiments with n ∼150 cells per condition. In all graphs, experimental points were fitted with two log-normal distributions. Image scale bars, 10 μm.
DOI:http://dx.doi.org/10.7554/eLife.06156.030
10.7554/eLife.06156.031
HGF triggers sustained Erk1/2 activation in primary mouse hepatoblasts.
(A–B) Time course of Erk1/2 phosphorylation after continuous stimulation with 10 ng/ml HGF or EGF for the indicated times in mouse primary hepatoblasts. (A) Representative phospho-Erk1/2 and Erk1/2 Western blots and (B) its quantification for HGF (red curve) or EGF (black curve) stimulation.
DOI:http://dx.doi.org/10.7554/eLife.06156.031
10.7554/eLife.06156.032
Quantification of the difference between two area distributions.
The differences between two endosome area distributions are measured as follows: (1) The binned histograms of the endosome area are built from the measurements of individual vesicles with bins linear in a logarithmic scale. (2) The histograms are normalized on their integrals, i.e., histograms are scaled to have the sum of values in all bins equal to one. (3) The histogram from the control condition is subtracted from the respective histograms of interest. The relative enrichment (red lines) or depletion (black lines) in the population of vesicles is calculated by the integral over a particular area interval.
DOI:http://dx.doi.org/10.7554/eLife.06156.032
10.7554/eLife.06156.033
Changes in endosome number and area
DOI:http://dx.doi.org/10.7554/eLife.06156.033
Cell type
Endosome marker or cargo
Growth factor
Endosome number*
Endosome area (μm2)
Increase in number of smaller vesicles
HeLa#
EGFR
EGF
22 ± 9
0.518 ± 0.023 (control = 0.629 ± 0.029)
9.53% ± 0.014 (<0.4 μm2)
E14.5 hepatoblast
EEA1
EGF
−6 ± 18
0.286 ± 0.02 (control = 0.294 ± 0.02)
−0.91% ± 0.003 (<0.3 μm2)
E14.5 hepatoblast
EEA1
HGF
18 ± 12
0.276 ± 0.02 (control = 0.294 ± 0.02)
2.05% ± 0.002 (<0.3 μm2)
PC12
EEA1
EGF
3 ± 10
0.471 ± 0.05 (control = 0.461 ± 0.05)
−1.03% ± 0.01 (<0.3 μm2)
PC12
EEA1
NGF
23 ± 16
0.454 ± 0.05 (control = 0.461 ± 0.05)
2.2% ± 0.01 (<0.3 μm2)
PC12
EGF
EGF
316 ± 46
0.276 ± 0.003
–
PC12
NGF
NGF
341 ± 5
0.245 ± 0.007
5.15% ± 0.01 (<0.3 μm2, difference from EGF-endosomes)
Endosome number is expressed as the difference from the control or non-stimulated cells. The value shows the number of endosomes per 1000 μm2 of area covered by cells.
HeLa cells after knock-down of EEA1, Rabenosyn5, and Vps45. All values show mean ± SEM.
Finally, we tested whether the differences in endosomal distribution can be, at least in part, causative of the different cell-fates triggered by EGF and NGF in PC12 cells. If so, we would predict that redistributing EGF to a larger number of small endosomes as seen in PC12 stimulated with NGF would be sufficient to switch signalling specificity and induce differentiation of PC12 cells. Therefore, we applied the same protocol of partial protein depletion previously used for HeLa cells (Figure 4) in PC12 cells and consistently observed a mild redistribution of EGF into smaller endosomes (Figure 7—figure supplement 1A,C,D). Also in this case, the partial depletion did not result in major changes in EGF transport kinetics in PC12 cells (Figure 7—figure supplement 1B), but increased the phosphorylation of both Erk (Figure 7—figure supplement 2A,B) and c-Fos (Figure 7—figure supplement 2C,D) upon stimulation with EGF. Next, we stimulated PC12 cells with EGF or NGF for 24 hr and analysed for neurite formation and β-III tubulin expression as markers of differentiation (Ohuchi et al., 2002) and for EdU incorporation as a measure of proliferation (Figure 7A). Stimulation with NGF increased the number of cells with neurites (Figure 7B, quantification in Figure 7C) and positive for β-III tubulin (Figure 7B, quantification in Figure 7D), and reduced cell proliferation (Figure 7A, quantification in Figure 7E), the opposite of the stimulation with EGF. Remarkably, upon redistribution of endosomes, EGF increased process formation (Figure 7B, quantification in Figure 7C), β-III tubulin expression (Figure 7B, quantification in Figure 7D), and reduced cell proliferation (Figure 7A, quantification in Figure 7E). The type of response was therefore similar to that of NGF, although the efficacy was lower. Nevertheless, these results show that a mild reduction of homotypic early endosome fusion was sufficient to modify cell fate and induce neuronal differentiation of PC12 cells.10.7554/eLife.06156.034
Redistribution of endosomal EGF is sufficient to trigger neuronal differentiation in PC12 cells.
(A–B) Representative images of PC12 cells after partial protein depletion of either EEA1, Rabenosyn5, and Vps45 or EEA1, Syntaxin-6, and Syntaxin-13, or mock treatment and stimulation with 100 ng/ml EGF or 50 ng/ml NGF for 24 hr. Scale bars, 50 μm. (B) A high-resolution image of single cells to highlight the changes in β-III tubulin expression and neurite formation. β-III tubulin is shown in green, nuclei are shown in blue, and EdU-positive nuclei are shown in pink. Scale bars, 10 μm. Note that in Figure 6C,E, the short incubation times did not permit neurite outgrowth. (C) Increase in the number of cells with β-III tubulin-positive processes longer than 1 μm compared to mock-treated cells after EGF stimulation. (D) Increase in β-III tubulin expression measured by the total intensity of the cytoplasmic β-III tubulin immunostaining. The total intensity per image was normalized by the image area covered by cells. (E) Number of proliferating cells measured by EdU incorporation. The number of EdU-positive nuclei was divided by the total number of nuclei. In all cases, data show mean ± SEM. For each parameter, pair-wise comparisons were done against EGF-stimulated mock-treated cells. *p < 0.05, **p < 0.005 by Fisher's LSD test. All measurements were done in three independent experiments with a total of ∼15000 cells per condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.034
10.7554/eLife.06156.035
Knock-down of fusion machinery redistributes endosomal EGF in PC12 cells.
(A) Partial protein depletion of Syntaxin-6 and Syntaxin-13 72 hr after electroporation. Protein reduction of EEA1 and Rabenosyn5 was similar to that in HeLa cells (not shown). (B) Time course of EGF integral intensity in endosomes after EEA1, Rabenosyn5, and Vps45 (red curve) or EEA1, Syntaxin-6, and Syntaxin-13 knock-down (blue curve) or mock treatment (black curve). Cells were given a 1-min pulse of 100 ng/ml of EGF-AlexaFluor 555, washed and chased for the indicated time points before fixation. Curves were normalized to the intensity value at 10 min for mock-treated cells. Points show mean ± SEM. All measurements were done in three independent replicates with a total of ∼150 cells per time point or condition. Time courses were fitted as in Figure 1 (C–D) Shift in the EGF-endosome area distribution after EEA1, Rabenosyn5, and Vps45 (C) or EEA1, Syntaxin-6, and Syntaxin-13 (D) knock-down measured as in Figure 3. Endosomes smaller than 0.2 μm2 (cross-sectional area) are increased after EEA1, Rabenosyn5, and Vps45 (C) or EEA1, Syntaxin-6, and Syntaxin-13 knock-down (D). Points show the mean ± SEM. All measurements were done in four independent replicates with a total of ∼200 cells per time point or condition. Experimental points were fitted with two lognormal distributions.
DOI:http://dx.doi.org/10.7554/eLife.06156.035
10.7554/eLife.06156.036
Redistribution of endosomal EGF is sufficient to increase MAPK activation in PC12 cells.
(A–D) Analysis of MAPK activation in PC12 cells after partial protein depletion of either EEA1, Rabenosyn5, and Vps45 or EEA1, Syntaxin-6, and Syntaxin-13, or mock treatment and stimulation with 100 ng/ml EGF or 50 ng/ml NGF for 30 min (A) Representative images of Erk1/2 activation by immunofluorescence in PC12 cells. phospho-Erk1/2 is shown in green and nuclei are shown in blue. Scale bars, 10 μm. (B) Increase in phospho-Erk1/2 intensity compared to EGF-treated control cells. The total intensity was normalized by the fraction of the area covered by cells. (C) Representative images of c-Fos phosphorylation by immunofluorescence in PC12 cells. phospho-c-Fos is shown in green. Scale bar, 25 μm. (D) Increase in nuclear phospho-c-Fos intensity compared to EGF-treated control cells. In all cases, data show mean ± SEM. For each parameter, pair-wise comparisons were done against EGF-stimulated mock-treated cells. *p < 0.05, **p < 0.005 by Fisher's LSD test. All measurements were done in three independent experiments with a total of ∼500 cells per condition.
DOI:http://dx.doi.org/10.7554/eLife.06156.036
Discussion
Genomic studies have revealed that signalling pathways exert a profound effect on the endosomal system (Pelkmans et al., 2005; Stasyk et al., 2007; Collinet et al., 2010). Parameters such as number of endosomes and size are tightly controlled in the case of EGF endocytosis (Collinet et al., 2010). Our results provide a rationale for such modulation and a novel framework for interpreting and predicting the signalling response of phosphorylated RTKs. In homogeneous assays (e.g., by Western blot), the total levels of active RTKs can be observed to rapidly decay with time in most signalling systems (Dunn and Hubbard, 1984; Burke et al., 2001; Sousa et al., 2012). These methods, however, measure the average steady state of an entire cell population and lack the spatial information. Here, we employed quantitative high- and super-resolution microscopy to resolve details of this process with sub-cellular resolution and high sensitivity. We discovered that the mean amount of p-EGFR per endosome was fairly constant over time and p-EGFR was found in small clusters in early endosomes.
The endosomal network is shaped by the balance of endosome fusion and fission (Foret et al., 2012) and this balance is also necessary for the formation of the endosomal clusters of p-EGFR. Modulation of the endosomal fusion/fission machinery manifests itself as a change in the size of endosomes (Sigismund et al., 2008) (Figure 6). Shifting the balance toward smaller endosomes through inhibition of fusion increased the number and reduced the size of endosomes, consequently expanding the number and life-time of p-EGFR clusters. Although the inhibition of endosome fusion was very mild, we cannot exclude the possibility that it may alter the recruitment and/or activity of signalling components by yet unknown mechanisms. On the other hand, for the interpretation of phenotypes upon perturbations on signalling, it is also important to consider the impact they have on the endosomal network (Collinet et al., 2010).
By analogy with synaptic transmission (Edwards, 2007), the packages of p-EGFR in early endosomes could be considered as quanta of signalling molecules. The concept of phosphorylated RTK quanta is reminiscent of analogue-to-digital communication systems, where a continuous variable (e.g., extracellular growth factor concentration) is transformed into a sequence of binary levels (e.g., phosphorylated RTK quanta in endosomes). An analogue-to-digital switch was described for Ras nanoclusters at the plasma membrane (Tian et al., 2007). In the case of endosomal digital signalling, our mathematical model predicts that it could serve two functions. First, it provides a mechanism to regulate signal amplitude and duration following RTK internalization. As a consequence, the total de-phosphorylation rate becomes dependent on the fusion/fission rate of the endosomes. This is interesting in view of the specific modulation of the endosome fusion/fission rates by growth factors (Figure 6, see below). Second, it acts as a noise dampening system (Ladbury and Arold, 2012), suppressing the noise due to, for example, fluctuations of EGF in the extracellular medium, expression levels of EGFR on the cell surface, etc. An increase in the amount of p-EGFR would result in faster de-phosphorylation rates. In contrast, low concentrations of EGF or EGFR would result in low de-phosphorylation rates. The middle point between the two extremes is the hallmark of signalling resilience. In addition, such a digital system may facilitate the integration of signalling information from different RTKs into a single, correct cell-fate decision. Our results highlight the importance of measuring the spatio-temporal distribution of signalling molecules using quantitative image analysis approaches to gain a deeper understanding of signal transduction regulation.
What is the molecular machinery responsible for the formation of the clusters and how is the number of p-EGFR molecules regulated? Clearly, the clustering mechanism is saturable (Figure 2A,B), as very high concentrations of EGF above some threshold suppress the correct endosomal packaging in addition to changes in the entry routes and signal output (Sigismund et al., 2008). We found that both Hrs and a few phosphatases, notably PTPN11 (SHP2), specifically regulate the amount of receptors within the p-EGFR clusters and their size. Hrs is known to interact with EGFR and regulate its degradation together with other components of the ESCRT machinery (Umebayashi et al., 2008). However, the effect of Hrs on the size of the p-EGFR clusters appears to be independent of the formation of ILVs, as suggested by the fact that Snf8 and Vps24 down-regulation does not produce the same effect.
Our mathematical model revealed that a correlation between p-EGFR dephosphorylation rate and p-EGFR amount per endosome can explain the mean constant size of p-EGFR quanta. We can envisage various non-exclusive mechanisms that can account for this correlation. One possible mechanism is a scaffold with a characteristic size that binds to p-EGFR and protects it from phosphatases. This hypothesis correlates higher total EGFR kinase activity to higher p-EGFR dephosphorylation, but only indirectly. Increasing the concentration of EGF in the medium would lead to a higher rate of delivery of p-EGFR to endosomes through vesicles which have no scaffold. If scaffold formation were rate limiting, the increased flux of p-EGFR into endosomes would reduce the fraction of protected p-EGFR thus exposing it to dephosphorylation. A caveat of this model is that, as the fusion of endosomes proceeds over time, multiple quanta would be expected to be brought together, increasing the mean amount of p-EGFR per endosome. This expectation is in contradiction with our experimental data (Figure 1B,D). With this model, additional factors must thus be taken into account to explain why multiple quanta cannot co-exist on the same endosomes.
The finding that Hrs knock-down increases the levels of p-EGFR suggests a different scaffold-based model. Instead of acting as a p-EGFR protective scaffold (or part of a scaffold), Hrs could exert the opposite function and stabilize the unphosphorylated EGFR, preventing its re-phosphorylation (Kleiman et al., 2011). Since the activity of Hrs is negatively regulated by p-EGFR (Row et al., 2005; Bache et al., 2002), this model is compatible with the data showing loss of quanta and increase in endosomal p-EGFR levels upon Hrs knock-down (Figure 2D,E). However, this hypothesis alone can neither explain the formation of quanta nor the finding that blocking p-EGFR kinase activity does not change the total levels of p-EGFR over time (Figure 2—Figure supplement 6).
Another mechanism is based on Turing Instability (Turing, 1952) (a reaction-diffusion mechanism). This mechanism is perhaps less intuitive but widely spread in biological processes, such as symmetry breaking and pattern formation in morphogenesis (Kondo and Miura, 2010). It is based on the observation that p-EGFR recruits and phosphorylates PTPN11 (SHP2) in a phosphor-tyrosine dependent manner (Deribe et al., 2009), thus enhancing its phosphatase activity (Agazie and Hayman, 2003). Briefly, p-EGFR would recruit and activate the phosphatase SHP2, forming a negative feedback loop. The phosphatase would diffuse on the surface of endosomes, dephosphorylating p-EGFR molecules before being itself inactivated in the absence of further interactions with p-EGFR. Such reaction-diffusion mechanism within a specific parameter range is known to form spatially restricted clusters of active molecular species (Turing Instability) (Turing, 1952), in this particular case quanta of p-EGFR on endosomes. A transient increase in p-EGFR after an endosome fusion event would increase the recruitment and/or activity of SHP2, re-establishing the p-EGFR quanta through dephosphorylation. If the characteristic length of Turing Instability is larger than endosome size, then multiple quanta cannot co-exist within a single endosome. The Turing Instability hypothesis explains the observed increase in p-EGFR quanta size after EGFR kinase inhibition, keeping the total p-EGFR levels unchanged (Figure 2—figure supplement 6,7), as well as the increase in total endosomal p-EGFR upon inhibition of endosome fusion (Figure 4G). However, it does not explain the effect of Hrs knock-down. A combination of the Turing instability and Hrs-mediated (negative) scaffold mechanisms is more consistent with our observations.
The regulation of endosomal packing reported in our study is likely not restricted to EGFR alone but is a general property, as different growth factors affect the endosomal network according to their specific signal output and cellular context (Figure 6). Hrs and SHP2 are also recruited by other RTKs (Agazie and Hayman, 2003; Row et al., 2005). The relative affinity of SHP2 to different receptors could lead to larger or smaller quanta, thus tuning the specificity of the signalling response. RTK quanta with different sizes could also result from differential phosphorylation of Hrs by RTKs (Row et al., 2005), given that the relative amount of Hrs on endosomes depends on its phosphorylation state (Urbé et al., 2000).
By which mechanisms can RTKs regulate the endosomal network? It has been shown that RTKs can modulate the activity of the transport machinery. For example, activation of p38 MAP kinase causes phosphorylation of the Rab5 effectors EEA1 and Rabenosyn-5, enhancing their recruitment to endosomes and consequently stimulating early endosome fusion (Macé et al., 2005; Cavalli et al., 2001). RTK stimulation also modulates the nucleotide cycle of Rab5 via activation of the Rab5 GEF RIN1 (Tall et al., 2001) or inactivation of the Rab5 GAP RN-tre (Lanzetti et al., 2000). Therefore, we predict that in general RTK ligands that stimulate the endosomal fusion machinery (such as EGF) will have a short phosphorylation half-life, whereas ligands that change the fusion/fission balance in favour of smaller endosomes (such as NGF) will have a long phosphorylation half-life. The combined effects of quanta size regulation through Hrs and SHP2 and modulation of fusion/fission will give a specific signalling amplitude and duration in different cell types stimulated with different ligands. We propose that the shape of distribution of the endosomal network can serve as a predictive parameter of the signalling status of the cell.
Our results support the concept of endosomes as signalling platforms (Di Guglielmo et al., 1994), a view recently shared for the β2-adrenoceptor (Irannejad et al., 2013) but opposed by other studies (Brankatschk et al., 2012; Sousa et al., 2012). This apparent contradiction can be explained by the fact that, under normal conditions of endocytosis, only the small fraction of p-RTK in endosomes are protected from inactivation and degradation, and can thus contribute to signal propagation. A feature of the p-EGFR clusters is that, with the increase in the local concentration, the stability of the active EGFR dimer (Chung et al., 2010) and signalling properties (Verveer et al., 2000) would also be increased. By blocking endocytosis, the levels of active receptors are artificially increased at the cell surface (Sousa et al., 2012), bypassing the normal requirement for endosomal regulation.
Our observations raise many more questions concerning the molecular mechanisms of quanta formation and their impact on cell fate decision. Clearly, the variety of models on quanta formation requires future experimental tests to determine the correct mechanism and reveal its molecular details. In addition, it will be important to validate our observations in an in vivo animal model to demonstrate that the dynamics of the endosomal network reflect the signalling activity by RTK under physiological conditions.
Materials and methodsp-EGFR FRET microscopy assay
To reduce the consequences of EGFR overexpression, we used HeLa Kyoto cells transfected with a bacterial artificial chromosome (BAC) transgene stably expressing EGFR-GFP under its endogenous promoter (Poser et al., 2008). Cells were incubated for different times in serum-free medium with 10 ng/ml EGF (Invitrogen, California, USA) or for 30 min with 0.05, 0.1, 0.25, 0.5, 1, 2.5, 5, 7.5, or 10 ng/ml EGF. Cells were then fixed and processed for immunofluorescence as previously described (Collinet et al., 2010) using a mouse monoclonal anti-phospho-tyrosine 4G10 antibody (Millipore, California, USA) directly labelled with AlexaFluor 555 (Molecular Probes, Invitrogen). For colocalization measurements, samples were also incubated with rabbit polyclonal anti-EEA1 (Rink et al., 2005) or mouse monoclonal anti-LAMP-1 (BD Biosciencies, California, USA) antibodies. Images were acquired using a laser-scanning confocal microscope (Duoscan, Zeiss) with a 63×/1.4 oil objective. Multicolour images were acquired in three sequential scans: GFP fluorescence and AlexaFluor 555 fluorescence were detected simultaneously with two different detectors using 488 and 561 nm laser light and a 505/530 band-pass filter or a 593 nm long-pass spectral range in a META detector (Zeiss); FRET signal was detected with 458 nm excitation and a 593 nm long-pass spectral range in a META detector (Zeiss). 10 images per time point were collected, and each image was the maximum projection of four confocal sections of ∼1 μm thickness with 0.5 μm step. For the comparison between live and fixed cells, images were acquired with an automated spinning-disk confocal microscope (OPERA, Evotec Technologies-PerkinElmer) with a 40×/0.9 NA water immersion objective. EGFR-GFP was excited with a 488 nm laser and detected with a 520/35 nm filter. DAPI for nuclei identification was excited in a separate exposure with a 405 nm laser and detected a 450/50 nm filter. Eighty images per condition were acquired. Every image contained on average 20 cells.
Image analysis was performed using custom designed image analysis software (MotionTracking) as previously described (Rink et al., 2005; Collinet et al., 2010). The ‘integral intensity’ corresponds to the integral of fluorescent marker intensity per endosome. The ‘total integral intensity’ is defined as the sum of integral intensities of all endosomes in an image normalized by the area covered by the cells. The ‘endosome cross-sectional area’ was measured as the apparent fluorescent area (in µm2) of an endosome (above the half-maximum value of fluorescence intensity of each structure). Since MotionTracking approximates real image intensity by a sum of analytical functions (Rink et al., 2005), the resulting area and intensity have no pixel granularity.
High-resolution microscopy FRET-based assay
p-EGFR or ub-EGFR was first identified on the basis of triple colocalization between objects detected by the EGFR (488 nm laser excitation and 505/530 nm bandpath emission filter), anti-p-Tyr antibody p-Tyr-ab for p-EGFR or anti-mono and polyubiquitynilated conjugates (FK2) (Enzo Biosciences, New York, USA) (561 nm laser excitation and a 593 nm long-pass filter), and FRET (458 nm laser excitation and a 593 nm long-pass filter) channels (Figure 1—figure supplement 1A). Colocalization was scored by cross-sectional overlap >30%. The FRET signal was corrected for spectral bleed-through (SBT). Two major processes contribute to the SBT: (1) the GFP fluorescence bleed-through in the FRET channel and (2) direct excitation of AlexaFluor 555 by the 458 nm laser. We performed control experiments to estimate SBT for subsequent correction. To estimate GFP fluorescence bleed-through, we imaged EGFR-GFP BAC HeLa cells in the FRET channel (excitation 458 nm) without p-Tyr-ab staining. The signal in the FRET channel was below our detection limit and, therefore, we omitted correction in the subsequent analysis. The SBT by direct excitation of AlexaFluor 555 was estimated by quantification of FRET vesicles that colocalized with p-Tyr-ab or ub-ab, but not with EGFR (bleed-through control). Following the approach of Gordon et al., (1998), the correction in this case will be,F=I−k·T,where F is the corrected intensity in the FRET channel, I is the raw intensity in the FRET channel, T is the intensity in the p-Tyr-ab channel, k=<Icontrol><Tcontrol> is the bleed-through coefficient (ratio of means) calculated from control vesicles. Unfortunately, this correction method provided a good estimation of the average FRET signal, but when applied to individual endosomes it gave negative intensities for a substantial (30–40%) number of cases, thus precluding the estimation of mean intensity per endosome. In order to identify the source of negative intensities, we calculated the distribution of ratios of intensities in the FRET channel to the intensities in the p-Tyr-ab channel per endosome (Figure 1—figure supplement 1B). This distribution is broad and one can conclude that correction by Equation 1 will inevitably produce negative values in some cases. We fitted the distribution by the sum of three Gaussian components (Figure 1—figure supplement 1B, red, green, and blue dashed lines). By using control cells that did not express EGFR-GFP, we tested that the first two components correspond to SBT (direct excitation of AlexaFluor 555 by the 458 nm laser). Next, we developed a probabilistic model to find the expected FRET signal, given the p-Tyr-ab signal and the constants (µ, σ) of Gaussian distribution of the intensities ratios. The distribution of ratios is P(m)dm=12πσe−(m−μ)22σ2dm. We denoted m=I−FT the ratio of bleed-through signal in the FRET channel to the signal in the p-Tyr-ab channel for individual endosomes. After this substitution, the probability to obtain a FRET signal F is P(F)dF=P(m(F))dmdFdF=12πσTexp(−(F−(I−μT))22σ2T2)dF. Since the bleed-through cannot be higher than the measured signal I, we can calculate the expectation of the FRET signal as: 〈F〉=∫0IF·P(F)dF∫0IP(F)dF. After substitution and integration, we get:〈F〉=I−μT+2πσT(1−e−12(μσ)2)−sgn(I−μT)·(1−e−12(I−μTσT)2)erf(μ2σ)+erf(I−μT2σT).
One can see that (a) if I − µT > 0 (i.e., SBT is small relative to the true FRET signal), then the last term in the formula is very small and 〈F〉≈I−μT in agreement with Gordons' formula; (b) even if I − µT < 0 (i.e., SBT is large relative to the true FRET signal or the FRET signal is absent), the Equation 2 always gives small, but positive values. As such, Equation 2 provides a good estimation of the expected FRET intensity given the measured intensities in the FRET and p-Tyr-ab channels.
Next, we developed this approach further by taking into account that the real bleed-through distribution is the sum of two Gaussians with mean values µ1, µ2, standard deviations σ1, σ2 and their contribution in the total distribution a1, a2. Following the same approach as above we get:〈F〉=I−(a1μ1+a2μ2)T+2π(a12σ12+a22σ22)T(1−e−M)−sgn(N)·(1−e−N2)erf(M)+erf(N),where M=(μ12σ1)2+(μ22σ2)2 and N=I−(a1μ1+a2μ2)T2(a12σ12+a22σ22)T.
The example of FRET correction by Equation 3 is presented on Figure 1C.
To validate the FRET measurements, cells were treated with EGF, stained using a rabbit monoclonal anti-p-EGFR (Tyr 1068) antibody, and imaged using a laser-scanning confocal microscope using the same protocols as described above.
EGFR and p-EGFR single molecule quantification
EGFR-GFP BAC cells were incubated with 10 ng/ml EGF for 30 min, fixed and stained with the rabbit monoclonal anti-p-EGFR (Tyr 1068) antibody as described above. One field of view was sequentially acquired to record bleaching of GFP and fluorescently labelled secondary antibodies. The resulting time series was segmented with MotionTracking as described above, individual objects were tracked for consecutive frames, and the fluorescence intensity of every endosome between two consecutive frames was subtracted to build the ΔIntensity distribution (Figure 2—figure supplement 3A,C). The width of distribution is mostly determined by the fluctuations of intensities. However, due to bleaching, the distribution is slightly skewed toward negative values. First, the ΔIntensity was binned. Then the difference between frequencies of negative and positive ΔIntensity of equal absolute values was plotted as function of ΔIntensity (Figure 2—figure supplement 3B,D). We named it neg-double-difference function. Since every bin of neg-double-difference function in the vicinity of the first peaks contained ∼2500 events, random fluctuations were strongly suppressed and the averaging revealed the discrete structure of bleaching, that is, bleaching of individual molecules. The local amplitude positive maxima correspond to discrete intensity changes when 1, 2, 3, …, n molecules are bleached (see e.g., arrows on peaks at 280, 560, and 800 integral intensity units in Figure 2—figure supplement 3B). We estimated that one molecule of GFP and alexa555-antibody corresponds to the integral intensity units at the first peak of the neg-double-difference function (280 and 190 integral intensity units for EGFR-GFP and alexa555, respectively). This method estimated directly the number of EGFR-GFP molecules. The total number of EGFR molecules per endosome was corrected for the ratio of endogenous and BAC EGFR-GFP expressions (1.29 ± 0.07, based on WB quantification). Since the method estimated the number of fluorophores, in the case of antibodies with unknown labelling stoichiometry and epitope accessibility, the result has to be corrected for a scaling factor. The scaling factor of antibody labelling was estimated as 1.9 by comparison of EGFR and p-EGFR distributions at 3 and 5 min of EGF stimulation (10 ng/ml), when most internalized EGFR are still phosphorylated (Sorkin and Goh, 2009), with distributions at 30 min.
We used this fluorescence intensity-based method also to estimate the number of EGFR molecules in both clathrin-dependent and independent vesicles. From geometrical calculations, assuming that an uncoated CCV has a diameter of 90 nm and an EGFR dimer has a diameter of ∼15 nm and luminal domain of ∼10 nm, we estimate that a CCV can contain up to 70 EGFR molecules. However, this calculation does not take into account that vesicles contain multiple types of transmembrane proteins and thus, the value can only be an upper limit. Therefore, we estimated the number of EGFR molecules per diffraction-limited, EEA1-negative vesicle that can be observed following 5 min of 10 ng/ml EGF stimulation. Using this method, we estimated 8.5 ± 3.5 molecules/vesicle. Based on this value, we calculated that ∼12 vesicles are required to deliver the 102 EGFR molecules/EEA1-positive endosome.
Fitting of time course kinetics
Time courses were fitted with the sum of two exponential terms: one for growth and one for decline.Ae−tτ1+Be−tτ2
The constant of the decline exponent τ2 was used as an estimation of the decay time of the corresponding process. The fitting of the experimental data was done according to an optimization scheme previously described (Press et al., 1992).
p-EGFR detection in MVBs
To discriminate p-EGFR exposed on the surface of endosomes from p-EGFR sequestered into ILVs, we used a differential detergent solubilisation method as previously described for protease protection assays (Malerød et al., 2007). Cells were fixed and permeabilized with saponin 0.1% for 10 min or digitonin 0.001% for 1 min. After permeabilization, cells were washed with PBS and stained with a mouse monoclonal anti-phospho-tyrosine-AlexaFluor 555 antibody (Millipore), a mouse monoclonal anti-LBPA (a gift by J Gruenberg, University of Geneva) antibody, or a mouse monoclonal anti-GFP (Roche, Switzerland) antibody together with a goat anti-mouse-AlexaFluor 555 antibody (Molecular Probes, Invitrogen) to reveal the antigen signal.
Membrane permeabilization with saponin allows access of antibodies both to the cytosol and the luminal content of endosomes, whereas digitonin only to the cytosol. Upon digitonin permeabilization, the staining of LBPA, a marker of ILVs and EGF or EGFR was strongly reduced in comparison with saponin permeabilization (Figure 2A), consistent with their localization predominantly within the endosomal lumen. After 30 min of EGF stimulation, the endosomal, but not the plasma membrane EGFR staining was strongly reduced in cells permeabilized with digitonin compared with saponin, probably reflecting the internalization of receptors into ILVs (Figure 2A,B). In contrast, the p-EGFR levels were only moderately reduced upon permeabilization with digitonin compared with saponin extraction (Figure 2A,B), suggesting that the majority of p-EGFR faced the cytosolic surface of endosomes and was not within ILVs. To measure Shc1 recruitment to endosomes, cells were permeabilized with saponin and stained with a rabbit polyclonal anti-Shc1 antibody (BD Biosciences). Image acquisition, correction, and analysis proceeded as described above.
dSTORM microscopy
Cells were stimulated for different times with 10 ng/ml EGF and fixed as described above. To detect p-EGFR in endosomes, cells were stained using a rabbit monoclonal anti-p-EGFR (Tyr 1068) antibody. For dSTORM microscopy, the samples were mounted on medium optimized for enhanced switching between fluorescent and non-fluorescent states as previously described (van de Linde et al., 2011; Lampe et al., 2012). Imaging was performed using a H3 Andor spinning disk microscope with a 100× objective as previously described (van de Linde et al., 2011; Lampe et al., 2012).
Calculation of changes in endosome area distributions
First, the binned histograms of endosome area were built with bin widths linear in a logarithmic scale. Then, the histograms were normalized on their integrals, that is, histograms were scaled to have the sum of values in all bins equal to one. Finally, the histogram from the control condition was subtracted from the respective histograms of the different conditions (Figure 6—figure supplement 2).
Mathematical model of p-EGFR propagation through the endosomal network
To describe the time course of the formation of a mean constant amount of p-EGFR per endosome during endocytosis, we postulated a sigmoidal dependency of the dephosphorylation rate on the amount of p-EGFR per endosome. The rationale for this is that if the amount of p-EGFR per endosome is above a critical value, dephosphorylation is significantly increased, whereas if the amount is lower, dephosphorylation is decreased. The delay between EGF stimulation and onset of internalization of p-EGFR into early endosomes is well documented (Burke et al., 2001; Wiley, 2003). This delay includes EGF binding to receptor (∼3 min), CCV formation (∼1–2 min) and delivery of p-EGFR to early endosomes. In order to keep the model as simple as possible, we described these mechanisms in a coarse grained model by an exponential delay with constant δτ. Since the dephosphorylation rate depends on the amount of p-EGFR per endosome, we expanded the mass flux equation usually applied in these cases with an equation that describes the number of endosomes carrying p-EGFR. Our experimental data suggest a significant redistribution of EGFR from the plasma membrane into endosomes even at very low doses of EGF (see Figure 1E, green curve. Compare 0.5 with 10 ng/ml). A simple mechanism to explain this is the internalization of ligand-unoccupied EGFR upon EGF stimulation, for example by formation of EGFR oligomers at the plasma membrane (Ariotti et al., 2010; Hofman et al., 2010). Another possible mechanism includes transient activation of p38 (Faust et al., 2012) by EGFR signalling that leads to acceleration of unoccupied receptor internalization (Zwang and Yarden, 2006; Faust et al., 2012). Therefore, we modelled the rate of EGFR-positive vesicle formation as Kv=kv0+kv1SpqQvq+Spq, where Sp is p-EGFR on plasma membrane, q is Hill coefficient, Qv is a characteristic constant. We considered that the ratio of EGF loaded/unloaded EGFR in the vesicles is equal to the weighted ratio p-EGFR/EGFR on plasma membrane with weight factor w. Importantly, the use of this term in the model gave the best description of the time course of total EGFR, but was not essential to explain the p-EGFR dynamics in individual endosomes (data not shown).dSmdt=−kin·Sm·cEGF(1−e−(tδτ)2)+kout·Smp−Kv·SmSm+w·Smp·Sm+kre_out·SredSmpdt=kin·Sm·cEGF(1−e−(tδτ)2)−kout·Smp−Kv·w·SmpSm+w·Smp·SmpdSpedt=Kv·w·SmpSm+w·Smp·Smp−(β1+Sper(Q·Npe)r+Sper(β2−β1))·SpedSedt=Kv·SmSm+w·Smp·Sm+(β1+Sper(Q·Npe)r+Sper(β2−β1))·Spe−kreSe−kleSedSredt=kreSe−kre_out·SredNpedt=Kvsvw·SmpSm+w·Smp·w·Smp−ρ·Npe2+f·Npe,where,
Sm is the amount of non-phosphorylated EGFR on the plasma membrane,
cEGF is the amount of EGF in the extracellular medium,
Smp is the amount of p-EGFR on plasma membrane,
Se is the non-phosphorylated EGFR on early endosomes,
Spe is the amount of non-phosphorylated EGFR on early endosomes,
Sre is the amount of EGFR on recycling endosomes,
Kv is the rate of EGF-stimulated EGFR-positive vesicle formation (see above),
Npe is the number early endosomes with p-EGFR,
kin is the rate of EGF binding to EGFR,
kout is the rate of EGF release from EGFR,
kre is the rate of sorting of EGFR from early to recycling endosomes,
kle is the rate of sorting of EGFR from early to late endosomes,
kre_out is the rate of delivery of EGFR from recycling endosomes to plasma membrane,
sv is the mean amount of p-EGFR per endocytic vesicle,
β1, β2 are minimum and maximum dephosphorylation rates,
r is a Hill coefficient of dephosphorylation rate,
Q is the characteristic amount of p-EGFR at which the dephosphorylation has ½ maximal rate,
ρ is the early endosome homotypic fusion rate (measured as number of events/minute/endosome),
f is the early endosome homotypic fission rate.
Equation 5 describes the total amount of non-phosphorylated EGFR on the plasma membrane (Sm). The first term describes the loss of non-phosphorylated EGFR that becomes phosphorylated upon EGF binding. The second term describes the increase in non-phosphorylated EGFR upon release of EGF from p-EGFR and concomitant dephosphorylation. The third term describes the amount of non-phosphorylated EGFR which is internalized upon EGF-stimulated endocytosis (see above). The last term describes the recycling of EGFR to the plasma membrane.
Equation 6 describes the total amount of p-EGFR on the plasma membrane (Smp). The first term describes the phosphorylation of EGFR upon EGF binding, the second the dephosphorylation following EGF release, and the third its endocytosis.
Equation 7 describes the amount of p-EGFR in EEA1-positive early endosomes (Spe). The first term describes the endocytosis of p-EGFR and the second its dephosphorylation. Note that the equation includes a sigmoidal function β1+Sper(Q·Npe)r+Sper(β2−β1) for the dephosphorylation rate.
Equation 8 describes the amount of non-phosphorylated EGFR on early endosomes (Se). The first term describes EGF-stimulated endocytosis of ligand-free EGFR. The second term describes the increase in the amount of non-phosphorylated EGFR through dephosphorylation of p-EGFR. The third and fourth terms describe the sorting of EGFR to recycling and late endosomes, respectively.
Equation 9 describes the amount of EGFR on recycling endosomes (Sre). The first term describes the delivery of EGFR from early endosomes and the second its recycling to the plasma membrane.
Equation 10 describes the number of EEA1-positive early endosomes containing p-EGFR (Npe). For simplicity, we considered that the p-EGFR is evenly distributed between endosomes. The first term describes the endocytosis of p-EGFR, the second the homotypic fusion of early endosomes, and the third their homotypic fission.
The model was fitted to the experimental data which included time courses of p-EGFR and EGFR colocalization to EEA1 (Kalaidzidis et al., 2015) upon stimulation with four different concentrations (0.5, 1.0, 5.0, and 10.0 ng/ml) of EGF (Figure 3A,B). Fitting was performed with FitModel software (Zeigerer, 2012) (http://pluk.mpi-cbg.de/projects/fitmodel). Since the amount p-EGFR was measured experimentally in arbitrary FRET intensity units, the modelled amount of p-EGFR was scaled before a comparison with the experimental data. The scaling factor was found by the least square formula scale=∑i=1Ndi·siσi2∑i=1Nsi2σi2, where di, σi; i = 1…N are experimental values and their SEMs; si are model predictions for the respective time points. The model prediction of p-EGFR modulation by reduction of the early endosome homotypic fusion rate is presented on Figure 3C. The model and fit parameters are provided in the text format and in the format of FitModel software in the Source code 1 (Model.zip).
Knock-down and phenotype characterization in Hela EGFR BAC cells
HeLa EGFR BAC cells were reverse transfected with 5 nM siRNA oligonucleotides per gene using the oligonucleotides given in Table 2.10.7554/eLife.06156.037
List of siRNAs used for down-regulation of endosomal proteins
DOI:http://dx.doi.org/10.7554/eLife.06156.037
Gene name
siRNA library
siRNA ID
EEA1
Ambion Silencer
139147
Rabenosyn5
Ambion Silencer
292470
Vps45
Ambion Silencer
136363
Hrs
Qiagen
SI00067305
Hrs
Qiagen
SI00288239
Hrs
Qiagen
SI02659650
Vps24
Invitrogen
148627
Vps24
Invitrogen
148628
Vps24
Qiagen
SI00760515
Snf8
Invitrogen
140086
Snf8
Qiagen
SI00375641
Snf8
Qiagen
SI00375648
Transfection was carried out using Interferin (Polyplus transfection) together with the selected oligonucleotides following the manufacturer's instructions or treated only with Interferin (mock). 72 hr after transfection total protein extracts were prepared to measure down-regulation of the targeted proteins by western blotting using antibodies previously described for EEA1 and Rabenosyn5 (Collinet et al., 2010). To measure the redistribution of EGFR in endosomes, cells were incubated with 10 ng/ml EGF (Invitrogen), fixed, and processed for quantitative microscopy. Image acquisition and analysis were done as described earlier. Measurement of p-EGFR was done using the FRET assay described above. To measure EGFR transport kinetics, cells were incubated in serum-free medium for 1 min with 10 ng/ml EGF (Invitrogen), washed with serum-free medium, and chased for different time points. Cells were then fixed and samples were processed for quantitative microscopy analysis as explained above.
To measure degradation of EGFR, cells were incubated for 1 hr with 10 μg/ml Cyclohexamide before stimulation with 10 ng/ml EGF for different time points. Total protein extracts were prepared and analysed by western blotting using rabbit monoclonal anti-EGFR (Cell Signaling, New England BioLabs, Massachusetts, USA) and mouse anti γ-tubulin (Antibody Facility, MPI-CBG, Germany) antibodies. To measure activated EGFR at the plasma membrane, cells were incubated with 100 ng/ml EGF-AlexaFluor 488 for 10 min on ice to prevent endocytosis, fixed with PFA, stained with a rabbit anti-AlexaFluor 488 antibody fraction (Invitrogen) to enhance the fluorescent signal and imaged as described above.
Phosphatase siRNA screen
HeLa EGFR BAC cells were reverse transfected with the protocol described above with 5 nM of the oligonucleotides in Table 3. After 72 hr, cells were stimulated with 10 ng/ml EGF for 30 min, fixed with PFA, and stained using a rabbit monoclonal anti-p-EGFR (Tyr 1068) antibody as described above. Images were acquired with an automated spinning-disk confocal microscope (OPERA, Evotec Technologies-PerkinElmer) with a 40×/0.9 NA water immersion objective. Settings were adjusted to minimize pixel-intensity saturation and maximize the dynamic range. Around 30 images for each siRNA oligonucleotide were collected. Images with less than three cells were excluded from analysis. Image analysis was performed with MotionTracking as described above.10.7554/eLife.06156.038
List of genes for PTP siRNA screen
DOI:http://dx.doi.org/10.7554/eLife.06156.038
Gene symbol
Gene ID
siRNa ID
Sequence 5′–3′
PTPN13
5783
5783-HSS108838
UCACAUUUCUGAACCAACUAGACAA
PTPN13
5783
5783-HSS184076
CAUCAGACUCUAAGCAACAUGGUAU
PTPN13
5783
5783-HSS184077
CCAUUGAGGGUAAUCUCCAGCUAUU
PTPN13
5783
5783-NM_080683.1_1459
GAAACACCCUUUGAAGGCAACUUAA
PTPRK
5796
5796-HSS108869
CCCAUCCAAGUGGAAUGUAUGUCUU
PTPRK
5796
5796-HSS108870
GGUCAUUCUUGAAACUGAUACUUCA
PTPRK
5796
5796-HSS184093
CCGCGCAAAGGAUACAACAUCUAUU
PTPRK
5796
5796-NM_002844.2_975
CCGCUUCCUUCAGAUUGCAAGAAGU
PTPRA
5786
5786-HSS108844
CCAGUUCACGGAUGCCAGAACAGAA
PTPRA
5786
5786-HSS108845
GCAUUCUCAGAUUAUGCCAACUUCA
PTPRA
5786
5786-HSS108846
GGCACCAACAUUCAGCCCAAAUAUA
PTPRA
5786
5786-NM_080841.2_1383
CGCCUCAUCACUCAGUUCCACUUUA
PTPRR
5801
5801-HSS108880
AGUUGAGGUUCUGGUUAUCAGUGUA
PTPN9
5780
5780-HSS108830
CCCUCAUUGACUUCUUGAGAGUGGU
PTPRR
5801
5801-HSS108882
GGUACACCUCAUGGCCUGAUCACAA
PTPN9
5780
5780-HSS108831
ACCUCAUGAGGAACCUCUUCGUUCU
PTPRR
5801
5801-HSS184097
CAAGAGAGAAGAGGGUCCAACGUAU
PTPN9
5780
5780-HSS184065
CGCUGUCUUGGAAUGUGGCUGUCAA
PTPRR
5801
5801-NM_130846.1_1022
CAGUGGCAAGGAGAAAGCCUUCAUU
PTPN9
5780
5780-NM_002833.2_1369
CAUCCAAGAGUUGGUGGACUAUGUU
PTPN2
5771
5771-HSS108817
GGAAGACUUAUCUCCUGCCUUUGAU
PTPN2
5771
5771-HSS108818
GAGCGGGAGUUCGAAGAGUUGGAUA
PTPN2
5771
5771-HSS184039
GAGAUUCUCAUACAUGGCUAUAAUA
PTPN2
5771
5771-NM_002828.2_1178
CCGAUGUACAGGACUUUCCUCUAAA
DUSP2
1844
1844-HSS140936
GCUCUGCCACCAUCUGUCUGGCAUA
PTPN3
5774
5774-HSS108820
GGCGUGGUACAGACCUUUAAAGUUA
PTPRE
5791
5791-HSS108853
UCUGGGAAUGGAAAUCCCACACUAU
DUSP2
1844
1844-HSS140937
GCUGCUGUCCCGAUCUGUGCUCUGA
PTPN3
5774
5774-HSS108821
GAGCUGUCCGCUCAUUUGCUGACUU
PTPRE
5791
5791-HSS108854
ACGAGACUUUCUGGUCACUCUCAAU
DUSP2
1844
1844-HSS140938
GGCAUCACAGCCGUCCUCAACGUGU
PTPN3
5774
5774-HSS108822
CCACCCGGGUAUUAUUGCAGGGAAA
PTPRE
5791
5791-HSS108855
GGAACAGUAUGAAUUCUGCUACAAA
DUSP2
1844
1844-NM_004418.3_925
UGGACGAGGCCUUUGACUUCGUUAA
PTPN3
5774
5774-NM_002829.2_621
CAAUCAGAAGCAGAAUCCUGCUAUA
PTPRE
5791
5791-NM_130435.2_1499
GAGCAGGAUAAAUGCUACCAGUAUU
PTPRF
5792
5792-HSS108856
CCCAUCAUCCAAGACGUCAUGCUAG
PTPRF
5792
5792-HSS108858
GGACAGCAGUUCACGUGGGAGAAUU
PTPRF
5792
5792-HSS184088
CAGCUGUGCCCUUUAAGAUUCUGUA
PTPRF
5792
5792-NM_130440.2_6013
CAGCUUUGACCACUAUGCAACGUAA
PTP4A2
8073
8073-HSS140957
GAUAACUCACAACCCUACCAAUGCU
DUSP6
1848
1848-HSS176270
GAGAGCAGCAGCGACUGGAACGAGA
PTP4A2
8073
8073-HSS140958
GCGUUCAAUUCCAAACAGCUGCUUU
DUSP6
1848
1848-HSS176271
UGGCAUUAGCCGCUCAGUCACUGUG
PTP4A2
8073
8073-HSS188476
GGUUCGAGUUUGUGAUGCUACAUAU
DUSP6
1848
1848-HSS176272
UGGCUUACCUUAUGCAGAAGCUCAA
PTP4A2
8073
8073-NM_080392.2_1123
UCGAGUUUGUGAUGCUACAUAUGAU
DUSP6
1848
1848-NM_022652.2_1097
CAUGUGACAACAGGGUUCCAGCACA
PTPRM
5797
5797-HSS108871
CCGAGUGAGGCUGCAGACAAUAGAA
PTP4A3
11156
11156-NM_007079.2_423
UCAGCACCUUCAUUGAGGACCUGAA
PTPN18
26469
26469-HSS120076
GCUGCCUUAUGAUCAGACGCGAGUA
PTPN14
5784
5784-HSS108841
UCAUGGGAAUGAAGAAGCCUUGUAU
PTPRM
5797
5797-HSS108872
CAGGCUCUGGUUACAGGGCAUUGAU
PTP4A3
11156
11156-NM_007079.2_460
UACCACUGUGGUGCGUGUGUGUGAA
PTPN18
26469
26469-HSS120077
UCGAGAGAUAGAGAAUGGGCGGAAA
PTPN14
5784
5784-HSS108843
GCCGCUGAUGUUGGCAGCAUUGAAU
PTPRM
5797
5797-HSS108873
CCCGACGCUUCAUUGCUUCAUUUAA
PTP4A3
11156
11156-NM_007079.2_473
CGUGUGUGUGAAGUGACCUAUGACA
PTPN18
26469
26469-HSS120078
CCCACCUGACUUCAGUCUCUUUGAU
PTPN14
5784
5784-HSS184078
GAUAUCAGUAUUACCUGCAAGUCAA
PTPRM
5797
5797-NM_002845.3_1217
CCGACGCUUCAUUGCUUCAUUUAAU
PTP4A3
11156
11156-NM_007079.2_678
CCAUCAACAGCAAGCAGCUCACCUA
PTPN18
26469
26469-NM_014369.2_835
UCAGUCUCUUUGAUGUGGUCCUUAA
PTPN14
5784
5784-NM_005401.3_3394
CACGAAGUUUCGAACGGAUUCUGUU
PTPN1
5770
5770-HSS108816
GAGUGAUGGAGAAAGGUUCGUUAAA
PTPN1
5770
5770-HSS184025
CAUGAAGCCAGUGACUUCCCAUGUA
PTPN1
5770
5770-HSS184026
CGAGAGAUCUUACAUUUCCACUAUA
PTPN1
5770
5770-NM_002827.2_507
CAGAGUGAUGGAGAAAGGUUCGUUA
PTPRJ
5795
5795-HSS108867
GCGACUUCAUAUGUAUUCUCCAUCA
PTPRJ
5795
5795-HSS184091
CGGGUUCUUCUUGAAAGCAUUGGAA
PTPRJ
5795
5795-HSS184092
GAGCAGCCAUGAUGCAGAAUCAUUU
PTPRJ
5795
5795-NM_002843.3_1838
CGGGUAGAAAUAACCACCAACCAAA
PTPN12
5782
5782-HSS108835
GCCACAGGAAUUAAGUUCAGAUCUA
PTP4A1
7803
7803-HSS111748
GCAACUUCUGUAUUUGGAGAAGUAU
PTPN12
5782
5782-HSS108836
GCCUCUUGAUGAGAAAGGACAUGUA
PTP4A1
7803
7803-HSS111749
UCAAAGAUUCCAACGGUCAUAGAAA
PTPN12
5782
5782-HSS108837
UCUGAUGGUGCUGUGACCCAGAAUA
PTP4A1
7803
7803-HSS111750
CCAACCAAUGCGACCUUAAACAAAU
PTPN12
5782
5782-NM_002835.2_554
CAGGACACUCUUACUUGAAUUUCAA
PTP4A1
7803
7803-NM_003463.3_1382
AACCAGAUUGUUGAUGACUGGUUAA
PTPN11
5781
5781-HSS108834
ACAUGGAACAUCACGGGCAAUUAAA
PTPN11
5781
5781-HSS184068
CAGACAGAAGCACAGUACCGAUUUA
PTPN11
5781
5781-HSS184069
GAAAGGGCACGAAUAUACAAAUAUU
PTPN11
5781
5781-NM_002834.3_5519
CAGGAUGCCUUUGUUAGGAUCUGUA
MAPK signalling measurements
72 hr after transfection, HeLa EGFR BAC cells were stimulated with 10 ng/ml EGF for different time points. Then, total protein extracts were prepared and analysed by western blotting using rabbit monoclonal anti-phospho-Erk1/2 (Thr202/Tyr204) (Cell Signaling, New England BioLabs) and mouse monoclonal anti-Erk1/2 (Cell Signaling, New England BioLabs) antibodies. For quantification, phospho-Erk1/2 intensity values were first normalized by the total Erk1/2 signal to control for differences in lane loading. For every blot, these values were normalized by the mean intensity amplitude per blot and then scaled by the mean difference between knock-down and mock-treated samples per experiment to account for experimental variability. To measure c-Fos activation, cells were stimulated with 10 ng/ml EGF for 30 min, fixed with PFA, and permeabilized with 0.5% Triton in PBS and 5% BSA as a blocking reagent. Cells were stained with a rabbit monoclonal anti-phospho-c-Fos (Ser32) antibody (Cell Signaling, New England BioLabs) and processed for image analysis. To measure Erk1/2 activation in PC12 cells, cells were stimulated with 100 ng/ml EGF or 50 ng/ml NGF and stained with a rabbit monoclonal anti-phospho-Erk1/2 (Thr202/Tyr204) (Cell Signaling, New England BioLabs) using the same protocol as above. 10 images per condition were acquired using a laser-scanning confocal microscope (Duoscan, Zeiss) with a 40×/1.3 oil objective. Image analysis was carried out as described above.
Animals
All animal studies were conducted in accordance with German animal welfare legislation and in strict pathogen-free conditions in the animal facility of the Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany. Protocols were approved by the Institutional Animal Welfare Officer (Tierschutzbeauftragter), and necessary licenses were obtained from the regional Ethical Commission for Animal Experimentation of Dresden, Germany (Tierversuchskommission, Landesdirektion Dresden).
Hepatoblast isolation and culture
Foetal hepatic cells were isolated from C57BL/6JOlaHsd mice, maintained in the animal facility of the MPI-CBG. Pregnancies were dated by the presence of a vaginal plug (embryonic day (E) 0.5). Hepatoblasts were prepared from E14.5 liver as described previously (Kamiya et al., 1999). Delta-like 1 (Dlk1) + hepatoblasts were isolated from the E14.5 hepatic cells as described previously with minor modifications (Tanimizu et al., 2003). Briefly, cells were blocked with an anti-mouse CD16/32 (BD Biosciences) and stained with a FITC-conjugated anti-Dlk1 antibody (MBL International, Massachusetts, USA) followed by anti-FITC Microbeads (Miltenyi Biotec GmbH, Germany). The labelled cells were separated using a MACS Cell Separation Column (Miltenyi Biotec). Dlk1+ cells were resuspended in DMEM (PAA Laboratories GmbH, Austria) containing 5% FBS, 2 mM L-glutamine (PAA Laboratories GmbH), 100 μM MEM Non-Essential Amino Acids (PAA Laboratories GmbH), 0.1 μM dexamethasone (Sigma–Aldrich), 100 Units/ml penicillin (PAA Laboratories GmbH), 100 μg/ml streptomycin (PAA Laboratories GmbH), and 4% BD Matrigel Basement Membrane Matrix (BD Biosciences), and seeded on a μ-slide 8-well (Ibidi GmbH, Germany) coated with fibronectin (Sigma–Aldrich, Germany). To measure Erk1/2 activation, cells were starved for 24 hr before stimulation with either 10 ng/ml EGF or HGF (R&D systems, Minnesota, USA). Total cell lysates were prepared and analysed using the same protocol and antibodies described above.
EEA1 staining after growth factor stimulation in PC12 cells or hepatoblasts
We used a clone of PC12 cells, PC12 Nsc-1 (Cellomics Inc., Maryland, USA) cells, due to their increased growth rate and decreased cell clumping, which facilitate imaging experiments (Hahn et al., 2009). Cells were grown following the manufacturer's instructions. PC12 cells were starved for 36 hr before stimulation either with 100 ng/ml EGF (Invitrogen) or 50 ng/ml NGF (R&D Systems) for 30 min. E14.5 Dlk1+ hepatoblasts were starved for 24 hr before stimulation with either 10 ng/ml EGF or HGF (R&D systems). Then, cells were fixed with 3% para-formaldehyde and stained with a mouse monoclonal anti-EEA1 (BD Biosciences Pharmingen). A fluorescently conjugated goat anti-mouse-AlexaFluor 555 secondary antibody (Molecular Probes, Invitrogen) revealed the antigen signal. Image acquisition and image analysis were performed as described above.
Triple knock-down and phenotype characterization in PC12 Nsc-1 cells
PC12 Nsc-1 cells were electroporated with 100 nM Stealth Select siRNA oligonucleotides (Invitrogen) (EEA1: 5′—GAA AGC AGC UCA ACU UGC UAC UGA A—3′, 3′—UUC AGU AGC AAG UUG AGC UGC UUU C—5′; Rabenosyn5: 5′—GGG CCU CAC ACU GAU CUU GCC UAU U—3′, 3′—AAU AGG CAA GAU CAG UGU GAG GCC C—5′; Vps45: 5′—GAC CCG GCA UGA AGG UAC UUC UCA U—3′, 3′—AUG AGA AGU ACC UUC AUG CCG GGU C—5′; Syntaxin-13: 5′—CCA AGG UGA UCU GAU UGA UAG CAU A—3′, 3′—UAU GCU AUC AAU CAG AUC ACC UUG G—5′; Syntaxin-6: 5′—GGA UGC UGG AGU GAC GGA UCG AUA U—3′, 3′—AUA UCG AUC CGU CAC UCC AGC AUC C—5′) or electroporated without siRNAs (Mock) using the Amaxa Cell line Nucleofector Kit V (Lonza, Switzerland) following the manufacturer's instructions. 36 hr after electroporation, cells were placed in serum-free medium. 72 hr after electroporation total protein extracts were prepared to measure down-regulation of the targeted proteins by western blotting with a mouse monoclonal anti-Syntaxin-13 (Synaptic Sytems, Germany) or a mouse monoclonal anti-Syntaxin-6 (Transduction Laboratories, BD Biosciences) antibody. To measure EGF transport, cells were stimulated with 100 ng/ml EGF-Alexafluor 555 (Molecular Probes, Invitrogen) for 1 min, washed with serum-free medium, and chased for different times. Then, cells were fixed and processed for microscopy as described above.
PC12 Nsc-1 differentiation-proliferation assay
Cells were starved for 36 hr and then stimulated in serum-free medium with 100 ng/ml EGF (Invitrogen) or 50 ng/ml NGF (R&D Systems) for 24 hr at 37°C and 5% CO2. During the last 3 hr, 5-ethynyl-2′ –deoxyuridine (EdU) was added at a final concentration of 10 μM. Then, cells were fixed and stained with Click-iT AlexaFluor 647 Azide (Molecular Probes, Invitrogen) following the manufacturer's instructions. Afterwards, cells were stained with a mouse monoclonal anti-β-III tubulin antibody (Chemicon International, Millipore) and a fluorescently conjugated goat anti-mouse-AlexaFluor 555 (Molecular Probes, Invitrogen) to reveal the antigen signal. Nuclei were stained with DAPI. 20 images per condition were acquired using a laser-scanning confocal microscope (Duoscan, Zeiss) with a 20×/0.8 objective. Image processing was carried out as described above. Images were inspected manually for process formation; cells with processes were defined as those having thin β-III tubulin-positive processes longer than 1 μm. The β-III tubulin expression was measured by total immunofluorescence intensity normalized by the frame area covered by cells to account for frame-to-frame variability in cell number.
Acknowledgements
We acknowledge T Galvez, G O'Sullivan, MP McShane, S Eaton, J Rink, J Howard, and P Bastiaens for discussions and comments on the manuscript. We thank Anja Zeigerer and Sarah Seifert for help in the experiments with primary mouse hepatocytes. We acknowledge the MPI-CBG services and facilities, in particular J Peychl for the management of the Light Microscopy Facility, C Möbius (HT-TDS) for assistance in automated image acquisition, and I Poser for generation of the stable HeLa cell lines. This work was financially supported by the Virtual Liver initiative (www.virtual-liver.de) funded by the German Federal Ministry of Research and Education (BMBF), the Max Planck Society (MPG), and the German Research Foundation (DFG). RV was supported by a grant from the Gottlieb Daimler und Karl Benz Stiftung.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
RV, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
HN, Established procedures for the isolation, culture and staining conditions of the embryonic hepatoblasts, Contributed unpublished essential data or reagents
PDC-Z, Developed the mathematical model and simulations, Analysis and interpretation of data
YK, Developed the FRET correction algorithm, Developed method to estimate the number of fluorescent molecules in light microscopy images, Supervised the image analysis, Developed the mathematical model and simulations, Conception and design, Analysis and interpretation of data, Drafting or revising the article
MZ, Directed Study, Conception and design, Drafting or revising the article
Ethics
Animal experimentation: All animal studies were conducted in accordance with German animal welfare legislation and in strict pathogen-free conditions in the animal facility of the Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany. Protocols were approved by the Institutional Animal Welfare Officer (Tierschutzbeauftragter) under the license Anzeige der Tötung von Tieren zu wissenschaftlichen Zwecken AZ: 24-9168.24-9/2009-1 (valid from 2009 until 31.12.2012) and AZ: 24-9168.24-9/2012-1 (valid from 30.4.2012 through 30.4.2015), obtained from the regional Ethical Commission for Animal Experimentation of Dresden, Germany (Tierversuchskommission, Landesdirektion Dresden).
Additional files10.7554/eLife.06156.039
System of differential equations for the mathematical model of p-EGFR endocytosis. The ZIP file contains the system of differential equations of the model with non-linear dephosporylation term and experimental data which were used to fit the model (see Figure 3 of main text). The data are provided in the FitModel format (http://pluk.mpi-cbg.de/projects/fitmodel) and as a simple text.
DOI:http://dx.doi.org/10.7554/eLife.06156.039
Major dataset
The following previously published dataset was used:
LuC, MiLZ, GreyMJ, ZhuJ, GraefE, YokoyamaS, SpringerTA, 2010, The Extracellular and Transmembrane Domain Interfaces in Epidermal Growth Factor Receptor Signaling, 3NJP; Publicly available at RCSB Protein Data Bank (http://www.rcsb.org).
ReferencesAgazieYMHaymanMJ2003Molecular mechanism for a role of SHP2 in epidermal growth factor receptor signalingMolecular and Cellular Biology237875788610.1128/MCB.23.21.7875-7886.2003AriottiNLiangHXuYZhangYYonekuboYInderKDuGPartonRGHancockJFPlowmanSJ2010Epidermal growth factor receptor activation remodels the plasma membrane lipid environment to induce nanocluster formationMolecular and Cellular Biology303795380410.1128/MCB.01615-09BacheKGRaiborgCMehlumAMadshusIHStenmarkH2002Phosphorylation of Hrs downstream of the epidermal growth factor receptorEuropean Journal of Biochemistry2693881388710.1046/j.1432-1033.2002.03046.xBrandhorstDZwillingDRizzoliSOLippertULangTJahnR2006Homotypic fusion of early endosomes: SNAREs do not determine fusion specificityProceedings of the National Academy of Sciences of USA1032701270610.1073/pnas.0511138103BrankatschkBWichertSPJohnsonSDSchaadORossnerMJGruenbergJ2012Regulation of the EGF transcriptional response by endocytic sortingScience Signaling5ra2110.1126/scisignal.2002351BurkePSchoolerKWileyHS2001Regulation of epidermal growth factor receptor signaling by endocytosis and intracellular traffickingMolecular Biology of the Cell121897191010.1091/mbc.12.6.1897CavalliVVilboisFCortiMMarcoteMJTamuraKKarinMArkinstallSGruenbergJ2001The stress-induced MAP kinase p38 regulates endocytic trafficking via the GDI: Rab5 complexMolecular Cell742143210.1016/S1097-2765(01)00189-7ChristoforidisSMcBrideHMBuroyneRDZerialM1999The Rab5 effector EEA1 is a core component of endosome dockingNature39762162510.1038/17618ChungIAkitaRVandlenRToomreDSchlessingerJMellmanI2010Spatial control of EGF receptor activation by reversible dimerization on living cellsNature46478378710.1038/nature08827CollinetCStöterMBradshawCRSamusikNRinkJCKenskiDHabermannBBuchholzFHenschelRMuellerMSNagelWEFavaEKalaidzidisYZerialM2010Systems survey of endocytosis by multiparametric image analysisNature46424324910.1038/nature08779CoumailleauFFürthauerMKnoblichJAGonzález-GaitánM2009Directional Delta and Notch trafficking in Sara endosomes during asymmetric cell divisionNature4581051105510.1038/nature07854DamkeHBabaTWarnockDESchmidSL1994Induction of mutant dynamin specifically blocks endocytic coated vesicle formationThe Journal of Cell Biology12791593410.1083/jcb.127.4.915DeribeYLWildPChandrashakerACurakJSchmidtMHKalaidzidisYMilutinovicNKratchmarovaIBuerkleLFetchkoMJSchmidtPKittanakomSBrownKRJurisicaIBlagoevBZerialMStagljarIDikicI2009Regulation of epidermal growth factor receptor trafficking by lysine deacetylase HDAC6Science Signaling2ra8410.1126/scisignal.2000576Di GuglielmoGMBaassPCOuWPosnerBIBergeronJJ1994Compartmentalization of SHC, GRB2 and mSOS and hyperphosphorylation of Raf-1 by EGF but not insulin in liver parenchymaThe EMBO Journal1342694277DunnWAHubbardAL1984Receptor-mediated endocytosis of epidermal growth factor by hepatocytes in the Perfused Rat liver: ligand and receptor dynamicsThe Journal of Cell Biology982148215910.1083/jcb.98.6.2148EdwardsRH2007The neurotransmitter cycle and quantal sizeNeuron5583585810.1016/j.neuron.2007.09.001FaustDSchmittCOeschFOesch-BartlomowiczBSchreckIWeissCDietrichC2012Differential p38-dependent signalling in response to cellular stress and mitogenic stimulation in fibroblastsCell Communication and Signaling10610.1186/1478-811X-10-6FelderSLaVinJUllrichASchlessingerJ1992Kinetics of binding, endocytosis, and recycling of EGF receptor mutantsThe Journal of Cell Biology11720321210.1083/jcb.117.1.203ForetLDawsonJEVillaseñorRCollinetCDeutschABruschLZerialMKalaidzidisYJülicherF2012A general theoretical framework to infer endosomal network dynamics from quantitative image analysisCurrent Biology221381139010.1016/j.cub.2012.06.021FrenchARSudlowGPWileyHSLauffenburgerDA1994Postendocytic trafficking of epidermal growth factor-receptor complexes is mediated through saturable and specific endosomal interactionsThe Journal of Biological Chemistry2691574915755GalperinESorkinA2008Endosomal targeting of MEK2 requires RAF, MEK kinase activity and clathrin-mediated endocytosisTraffic91776179010.1111/j.1600-0854.2008.00788.xGordonGWBerryGLiangXHLevineBHermanB1998Quantitative fluorescence Resonance Energy transfer measurements using fluorescence microscopyBiophysical Journal742702271310.1016/S0006-3495(98)77976-7HahnATJonesJTMeyerT2009Quantitative analysis of cell cycle phase durations and PC12 differentiation using fluorescent biosensorsCell Cycle81044105210.4161/cc.8.7.8042HofmanEGBaderANVoortmanJvan den HeuvelDJSigismundSVerkleijAJGerritsenHCvan Bergen en HenegouwenPM2010Ligand-induced EGF receptor oligomerization is kinase-dependent and enhances internalizationThe Journal of Biological Chemistry285394813948910.1074/jbc.M110.164731IrannejadRTomshineJCTomshineJRChevalierMMahoneyJPSteyaertJRasmussenSGSunaharaRKEl-SamadHHuangBvon ZastrowM2013Conformational biosensors reveal GPCR signalling from endosomesNature49553453810.1038/nature12000KalaidzidisYKalaidzidisIZerialM2015A probabilistic method to quantify the colocalization of markers on intracellular vesicular structures visualized by light microscopyAIP Conference Proceedings164158010.1063/1.4906025KamiyaAKinoshitaTItoYMatsuiTMorikawaYSenbaENakashimaKTagaTYoshidaKKishimotoTMiyajimaA1999Fetal liver development requires a paracrine action of oncostatin M through the gp130 signal transducerThe EMBO Journal182127213610.1093/emboj/18.8.2127KatzMAmitIYardenY2007Regulation of MAPKs by growth factors and receptor tyrosine kinasesBiochimica Et Biophysica Acta17731161117610.1016/j.bbamcr.2007.01.002KholodenkoBN2002MAP kinase cascade signaling and endocytic trafficking: a marriage of convenience?Trends in Cell Biology1217317710.1016/S0962-8924(02)02251-1KholodenkoBN2007Untangling the signalling wiresNature Cell Biology924724910.1038/ncb0307-247KholodenkoBNDeminOVMoehrenGHoekJB1999Quantification of short term signaling by the epidermal growth factor receptorThe Journal of Biological Chemistry274301693018110.1074/jbc.274.42.30169KholodenkoBNHancockJFKolchW2006Signalling ballet in space and timeNature Reviews Molecular Cell Biology1141442610.1038/nrm2901KleimanLBMaiwaldTConzelmannHLauffenburgerDASorgerPK2011Rapid phospho-turnover by receptor tyrosine kinase impacts downstream signaling and drug bindingMolecular Cell4372373710.1016/j.molcel.2011.07.014KondoSMiuraT2010Reaction-diffusion model as a framework for understanding biological pattern formationScience3291616162010.1126/science.1179047LadburyJEAroldST2012Noise in cellular signaling pathways: causes and effectsTrends Biochem. Sci3717317810.1016/j.tibs.2012.01.001LampeAHauckeVSigristSJHeilemannMSchmoranzerJ2012Multi-colour direct STORM with red emitting carbocyanianesBiology of the Cell10422923710.1111/boc.201100011LampugnaniMGOrsenigoFGaglianiMCTacchettiCDejanaE2006Vascular endothelial cadherin conrols VEGFR-2 internalization and signaling from intracellular compartmentsThe Journal of Cell Biology17459360410.1083/jcb.200602080LanzettiLRybinVMalabarbaMGChristoforidisSScitaGZerialMDi FiorePP2000The Eps8 protein coordinates EGF receptor signalling through Rac and trafficking through Rab5Nature40837437710.1038/35042605MacéGMiaczynskaMZerialMNebredaAR2005Phosphorylation of EEA1 by p38 MAP kinase regulates mu opioid receptor endocytosisThe EMBO Journal243235324610.1038/sj.emboj.7600799MalerødLStuffersJSBrechAStenmarkH2007Vps22/EAP30 in ESCRT-II mediates endosomal sorting of growth factor and chemokine receptors destined for lysosomal degradationTraffic81617162910.1111/j.1600-0854.2007.00630.xMarounCRJaujokasMAHolgado-MadrugadaMWongAJParkM2000The tyrosine phosphatase SHP-2 is required for sustained activation of extracelular signal-regulated kinase and epithelial morphogenesis downstream from the met receptor tyrosine kinaseMolecular and Cellular Biology208513852510.1128/MCB.20.22.8513-8525.2000MarshallCJ1995Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activationCell8017918510.1016/0092-8674(95)90401-8MiaczynskaMChristoforidisSGinerAShevchenkoAUttenweiler-JosephSHabermannBWilmMPartonRGZerialM2004APPL proteins link Rab5 to nuclear signal transduction via an endosomal compartmentCell11644545610.1016/S0092-8674(04)00117-5MorandellSStasykTSkvortsovSAscherSHuberLA2008Quantitative proteomics and phosphoproteomics reveal novel insights into complexity and dynamics of the EGFR signalling networkProteomics84383440110.1002/pmic.200800204NagashimaTShimodairaHIdeKNakakukiTTaniYTakahashiKYumotoNHatakeyamaM2007Quantitative transcriptional control of ErbB receptor signaling undergoes graded to biphasic response for cell differentiationThe Journal of Biological Chemistry2824045405610.1074/jbc.M608653200NakakukiTBirtwistleMRSaekiYYumotoNIdeKNagashimaTBruschLOgunnaikeBAOkada-HatakeyamaMKholodenkoBN2010Ligand-specific c-Fos expression emerges from the spatiotemporal control of ErbB network dynamicsCell14188489610.1016/j.cell.2010.03.054OhuchiTMaruokaSSakudoAAraiT2002Assay-based quantitative analysis of PC12 cell differentiationJournal of Neuroscience Methods1181810.1016/S0165-0270(02)00116-4OhyaTMiaczynskaMCoskunULommerBRungeADrechselDKalaidzidisYZerialM2010Reconstitution of Rab- and SNARE-dependent membrane fusion by synthetic endosomesNature4591091109710.1038/nature08107PelkmansLFavaEGrabnerHHannusMHabermannBKrauszEZerialM2005Genome-wide analysis of human kinases in clathrin- and caveola/raft-mediated endocytosisNature436788610.1038/nature03571PiperRCKatzmannDJ2007Biogenesis and function of multivesicular bodiesAnnual Review of Cell and Developmental Biology2351954710.1146/annurev.cellbio.23.090506.123319PoserISarovMHutchinsJRHérichéJKToyodaYPozniakovskyAWeiglDNitzscheAHegemannBBirdAWPelletierLKittlerRHuaSNaumannRAugsburgMSykoraMMHofemeisterHZhangYNasmythKWhiteKPDietzelSMechtlerKDurbinRStewartAFPetersJMBuchholzFHymanAA2008BAC TransgeneOmics: a high-throughput method for exploration of protein function in mammalsNature Methods540941510.1038/nmeth.1199PressWHTeukolskySAVetterlingWTFlanneryBP1992Numerical recipes in c. the art of scientific computingNew YorkCambridge University Press412425ResatHEwaldJADixonDAWileyHS2003An integrated model of epidermal growth factor receptor trafficking and signal transductionBiophysical Journal8573074310.1016/S0006-3495(03)74516-0RinkJGhigoEKalaidzidisYZerialM2005Rab conversion as a mechanism of progression from early to late endosomesCell12273574910.1016/j.cell.2005.06.043RowPEClagueMJUrbéS2005Growth factors induce differential phosphorylation profiles of the Hrs-STAM complex: a common node in signalling networks with signal-specific propertiesThe Biochemical Journal38962963610.1042/BJ20050067SantosSDVerveerPJBastiaensPI2007Growth factor-induced MAPK network topology shapes Erk responses determining PC-12 cell fateNature Cell Biology932433010.1038/ncb1543SasagawaSOzakiYFujitaKKurodaS2005Prediction and validation of the distinct dynamics of transient and sustained ERK activationNature Cell Biology736537310.1038/ncb1233SchenckAGoto-SilvaLCollinetCRhinnMGinerAHabermannBBrandMZerialM2008The endosomal protein Appl1 mediates Akt substrate specificity and cell survival in vertebrate developmentCell13348649710.1016/j.cell.2008.02.044SigismundSArgenzioETosoniDCavallaroEPoloSDi FiorePP2008Clathrin-mediated internalization is essential for sustained EGFR signaling but dispensable for degradationDevelopmental Cell1520921910.1016/j.devcel.2008.06.012SigismundSAlgisiVNappoGConteAPascoluttiRCuomoABonaldiTArgenzioEVerhoefLGMasperoEBianchiFCapuaniFCilibertoAPoloSDi FiorePP2013Threshold-controled ubiquitination of the EGFR directs receptor fateThe EMBO Journal322140215710.1038/emboj.2013.149SorkinA2001Internalization of the epidermal growth factor receptor: role in signalingBiochemical Society Transactions2948048410.1042/BST0290480SorkinAGohLK2009Endocytosis and intracellular trafficking of ErbBsExperimental Cell Research31568369610.1016/j.yexcr.2008.07.029SousaLPLaxIShenHFergusonSMDe CamilliPSchlessingerJ2012Suppression of EGFR endocytosis by dynamin depletion reveals that EGFR signalling occurs primarily at the plasma membraneProceedings of the National Academy of Sciences of USA1094419442410.1073/pnas.1200164109StasykTSchiefermeierNSkvortsovSZwierzinaHPeränenJBonnGKHuberLA2007Identification of endosomal epidermal growth factor receptor signaling targets by functional organelle proteomicsMolecular & Cellular Proteomics690892210.1074/mcp.M600463-MCP200TallGGBarbieriMAStahlPDHorazdovskyBF2001Ras-activated endocytosis is mediated by the Rab5 guanine nucleotide exchange activity of RIN1Developmental Cell1738210.1016/S1534-5807(01)00008-9TanimizuNNishikawaMSaitoHTsujimuraTMiyajimaA2003Isolation of hepatoblasts based on the expression of Dlk/Pref-1Journal of Cell Science11617751778610.1242/jcs.00388TarcicGBoguslavskySKWakimJKiuchiTLiuAReinitzFNathansonDTakahashiTMischelPSNgTYardenY2009An unbiased screen identifies DEP-1 tumor suppressor as a phosphatase controlling EGFR endocytosisCurrent Biology191788179810.1016/j.cub.2009.09.048TeisDWunderlichWHuberLA2002Localization of the MP1-MAPK scaffold complex to endosomes is mediated by p14 and required for signal transductionDevelopmental Cell380381410.1016/S1534-5807(02)00364-7TianTHardingAInderKPlowmanSPartonRGHancockJF2007Plasma membrane nanoswitches generate high-fidelity Ras signal transductionNature Cell Biology990591410.1038/ncb1615TuringAM1952The chemical basis of morphogenesisPhilosophical Transactions of the Royal Society of London Series B, Biological Sciences237377210.1098/rstb.1952.0012UmebayashiKStenmarkHYoshimoriT2008Ubc4/5 and c-Cbl continue to ubiquitinate EGF receptor after internalization to facilitate polyubiquitination and degradationMolecular Biology of the Cell193454346210.1091/mbc.E07-10-0988UrbéSMillsIGStenmarkHKitamuraNClagueMJ2000Endosomal localization and receptor dynamics determine tyrosine phosphorylation of hepatocyte growth factor-regulated tyrosine kinase substrateMolecular and Cellular Biology207685769210.1128/MCB.20.20.7685-7692.2000van de LindeSLöschbergerAKleinTHeidbrederMWolterSHeilemannMSauerM2011Direct stochastic optical reconstruction microscopy with standard fluorescent probesNatur Protocols6991100910.1038/nprot.2011.336VerveerPJWoutersFSReynoldsARBastiaensPIH2000Quantitative imaging of Lateral ErbB1 receptor signal propagation in the plasma membraneScience2901567157010.1126/science.290.5496.1567VieiraAVLamazeCSchmidSL1996Control of EGF receptor signaling by clathrin-mediated endocytosisScience2742086208910.1126/science.274.5295.2086WileyHS2003Trafficking of the ErbB receptors and its influence on signalingExperimental Cell Research284788810.1016/S0014-4827(03)00002-8WileyHSHerbstJJWalshBJLauffenburgerDARosenfeldMGGillGN1991The role of tyrosine kinase activity in endocytosis, compartmentation, and down-regulation of the epidermal growth factor receptorThe Journal of Biological Chemistry2661108311094WoutersFSBastiaensPI1999Fluorescence lifetime imaging of receptor tyrosine kinase activity in cellsCurrent Biology91127113010.1016/S0960-9822(99)80484-9YorkRDYaoHDillonTElligCLEckertSPMcCleskeyEWStorkPJ1998Rap1 mediates sustained MAP kinase activation induced by nerve growth factorNature39262262610.1038/33451ZeigererAGilleronJBogoradRLMarsicoGNonakaHSeifertSEpstein-BarashHKuchimanchiSPengCGRudaVMDel Conte-ZerialPHengstlerJGKalaidzidisYKotelianskyVZerialM2012Rab5 is necessary for the biogenesis of the endolysosomal system in vivoNature48546547010.1038/nature11133ZwangYYardenY2006p38 MAP kinase mediates stress-induced internalization of EGFR: implications for cancer chemotherapyThe EMBO Journal254195420610.1038/sj.emboj.760129710.7554/eLife.06156.040Decision letterPfefferSuzanne RReviewing editorStanford University, United States
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Regulation of EGFR signal transduction by analogue-to-digital conversion in endosomes”for consideration at eLife. Your article has been favorably evaluated by Randy Schekman (Senior editor) and 3 reviewers, one of whom (Suzanne Pfeffer) is a member of our Board of Reviewing Editors.
The reviewers thought that the manuscript would be improved by clarification of the text in three areas (no additional experiments are needed).
1) Please report EGFR molecule number per endosomal volume and clarify that if there are 100 EGFR molecules per endosome, this is the result of roughly how many coated vesicles fusing to bring early endosomes to that number?
2) Please clarify your logic that EGFR kinase activity is regulating phosphatase activity that, in turn, is regulating endosome pEGFR levels. It is difficult to envision how regulating something like local enzymatic activities could give rise to such precise levels of p-EGFR in endosomes. Instead, it seems more compatible with a process whereby there is a single scaffold (or similar protein assembly) per endosome that binds p-EGFR and protects it from dephosphorylation. This is supported by your observation that at low concentrations of EGF, blocking kinase activity does not change the total levels of p-EGFR, which would be expected if it controlled phosphatase activity, but does reduce the number of endosomes in which p-EGFR is found. This argues for a role of kinase activity in the segregation process by which a set number of p-EGFR molecules are associated with an endosome, not phosphatase activity.
Thus, increasing the levels of p-EGFR by using high concentrations of EGF would reduce the fraction associated with the scaffolds, resulting in an increase in fractional dephosphorylation. This hypothesis does correlate higher total EGFR kinase activity to higher p-EGFR dephosphorylation, but only indirectly. As a consequence of this correlation, one could develop a good descriptive model that functionally links the two processes and, for example, could give rise to the prediction that the total amount of pEGFR is dependent on the fusion/fission rate of endosomes. If p-EGFR flowed into endosomes too rapidly such that there was insufficient time to assemble the protective scaffold, their net rate of dephosphorylation would be higher, thus give rise to a positive correlation between EGFR kinase activity and dephosphorylation, and the predicted negative relationship between fusion rates and p-EGFR content per endosome.
(Please discuss this in the text to clarify your thinking for the reader. It would be best to provide a few possible models and state that more work is needed to distinguish between them.)
3) The paper would benefit from a one page summary discussing overall molecular mechanisms for what they observe for non-EGFR experts. This would include the mechanisms by which different growth factors influence fission and fusion of endosomes, how Hrs and Shp2 relate to p-EGFR in different cell types stimulated with different growth factors (or large versus small endosomes).
10.7554/eLife.06156.041Author response
The reviewers thought that the manuscript would be improved by clarification of the text in three areas (no additional experiments are needed).
1) Please report EGFR molecule number per endosomal volume and clarify that if there are 100 EGFR molecules per endosome, this is the result of roughly how many coated vesicles fusing to bring early endosomes to that number?
We estimated the number of receptors per µm3 of endosomal volume (“we estimated an average of 102±38 and 76±29 (Mean±SEM) molecules of EGFR and p-EGFR per endosome 30 minutes after EGF (10 ng/ml) internalization (Figure 2 – figure supplement 3), corresponding to 707±265 and 527±202 molecules per μm3 of endosomal volume (apparent, assessed by light microscopy), respectively”). To estimate the number of CCVs required to bring 100 EGFR per endosome, we used both geometrical calculations and the intensity based method described in the Methods.
2) Please clarify your logic that EGFR kinase activity is regulating phosphatase activity that, in turn, is regulating endosome pEGFR levels. It is difficult to envision how regulating something like local enzymatic activities could give rise to such precise levels of p-EGFR in endosomes. Instead, it seems more compatible with a process whereby there is a single scaffold (or similar protein assembly) per endosome that binds p-EGFR and protects it from dephosphorylation. This is supported by your observation that at low concentrations of EGF, blocking kinase activity does not change the total levels of p-EGFR, which would be expected if it controlled phosphatase activity, but does reduce the number of endosomes in which p-EGFR is found. This argues for a role of kinase activity in the segregation process by which a set number of p-EGFR molecules are associated with an endosome, not phosphatase activity.
Thus, increasing the levels of p-EGFR by using high concentrations of EGF would reduce the fraction associated with the scaffolds, resulting in an increase in fractional dephosphorylation. This hypothesis does correlate higher total EGFR kinase activity to higher p-EGFR dephosphorylation, but only indirectly. As a consequence of this correlation, one could develop a good descriptive model that functionally links the two processes and, for example, could give rise to the prediction that the total amount of pEGFR is dependent on the fusion/fission rate of endosomes. If p-EGFR flowed into endosomes too rapidly such that there was insufficient time to assemble the protective scaffold, their net rate of dephosphorylation would be higher, thus give rise to a positive correlation between EGFR kinase activity and dephosphorylation, and the predicted negative relationship between fusion rates and p-EGFR content per endosome.
(Please discuss this in the text to clarify your thinking for the reader. It would be best to provide a few possible models and state that more work is needed to distinguish between them.)
We took the reviewers’ suggestion and added a scholarly discussion on the molecular mechanisms that could be responsible for the formation of p-EGFR quanta at the end of the Discussion, considering which experimental data are consistent or inconsistent with each mechanism. We thank the reviewers for their proposal as we believe this discussion is now clearer and very stimulating. We acknowledge that this is only possible with eLife given that the printed journals normally demand to trim the text.
3) The paper would benefit from a one page summary discussing overall molecular mechanisms for what they observe for non-EGFR experts. This would include the mechanisms by which different growth factors influence fission and fusion of endosomes, how Hrs and Shp2 relate to p-EGFR in different cell types stimulated with different growth factors (or large versus small endosomes).
The explanation from point 2 is written as a summary that details how Hrs and SHP2 could also regulate the formation of quanta for other RTKs. We also report the evidence for general molecular mechanisms whereby different RTKs could regulate fusion/fission.
diff --git a/elife06303.xml b/elife06303.xml
new file mode 100644
index 0000000..90e088d
--- /dev/null
+++ b/elife06303.xml
@@ -0,0 +1 @@
+elifeeLifeeLifeeLife2050-084XeLife Sciences Publications, Ltd0630310.7554/eLife.06303Research advanceCell biologyDevelopmental biology and stem cellsRegions within a single epidermal cell of Drosophila can be planar polarised independentlyRoviraMiguel1†SaavedraPedro1†‡CasalJoséhttp://orcid.org/0000-0002-5149-13351*LawrencePeter A1*Department of Zoology, University of Cambridge, Cambridge, United KingdomVijayRaghavanKReviewing editorNational Centre for Biological Sciences, Tata Institute for Fundamental Research, IndiaFor correspondence: jec85@cam.ac.uk (JC);pal38@cam.ac.uk (PAL)
These authors contributed equally to this work
Institute of Molecular and Cell Biology, Proteos, Singapore
Planar cell polarity (PCP), the coordinated and consistent orientation of cells in the plane of epithelial sheets, is a fundamental and conserved property of animals and plants. Up to now, the smallest unit expressing PCP has been considered to be an entire single cell. We report that, in the larval epidermis of Drosophila, different subdomains of one cell can have opposite polarities. In larvae, PCP is driven by the Dachsous/Fat system; we show that the polarity of a subdomain within one cell is its response to levels of Dachsous/Fat in the membranes of contacting cells. During larval development, cells rearrange (Saavedra et al., 2014) and when two subdomains of a single cell have different types of neighbouring cells, then these subdomains can become polarised in opposite directions. We conclude that polarisation depends on a local comparison of the amounts of Dachsous and Fat within opposing regions of a cell's membrane.
DOI:http://dx.doi.org/10.7554/eLife.06303.001
Author keywordsmorphogensplanar cell polaritydachsousfatfour-jointedgradientsResearch organismD. melanogasterhttp://dx.doi.org/10.13039/100004440Wellcome TrustWT096645MALawrencePeter AThe funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementBuilding on previous work (Saaverda et al., 2014), we show that the Dachsous/Fat system can polarise different parts of a single cell in opposite ways in the Drosophila larva.Introduction
Epithelial cells are polarised within the plane of the epithelium and can display consistent orientation across extensive tracts of cells. This property, known as planar cell polarity (PCP) is revealed by polarised structures such as hairs on insect wings or in the skin of mammals. At least two genetic systems are involved in PCP and these are conserved between Drosophila and the mouse (Casal et al., 2006; Lawrence et al., 2007; Goodrich and Strutt, 2011). PCP is envisaged as a cellular property, the smallest unit of manifest polarity being an entire single cell. However, we now show in Drosophila that different domains within a single cell can have mutually opposing polarities. These multipolar cells occur in the normal larval epidermis and are detectable because the cells are large and some are decorated with several pointed denticles. The Dachsous/Fat (Ds/Ft) system acts at intercellular contacts (Strutt and Strutt, 2002; Ma et al., 2003; Casal et al., 2006); we provide evidence that the polarity of a domain within one cell is its response to the levels of Ds/Ft in neighbouring cells. When another domain of that same responding cell has different neighbours, it can acquire the opposite polarity. We conclude that polarisation of a domain results from a comparison of the amounts of Ds and Ft in different regions of the cell membrane. This comparison is made between limited regions of membranes on opposite sides of the same cell that face each other along the anterior to posterior axis. We conjecture that ‘conduits’ span across the cell and mediate this comparison. In each region of the cell, the orientation of the conduits, a consequence of the comparison, cues the polarity of denticles.
The later larval stages of <italic>Drosophila</italic>
As we have shown recently (Saavedra et al., 2014), the epidermis of the later larval stages of Drosophila, the second and third stages, offers considerable advantages for the analysis of PCP. The development of single individuals can be followed and genetic mosaics can be made and studied in vivo (Saavedra et al., 2014). The epidermal cells are large (ca 30 µm across in the third stage) and show their polarity by forming, first, actin-rich predenticles near the cell membrane and then, oriented denticles in the cuticle that they secrete. There are about seven rows of denticulate cells in each segment; denticle rows 0, 1, and 4 point forwards and rows 2, 3, 5, and 6 point backwards (Figure 1A,B). There are two rows of muscle attachments called tendon cells, T1 and T2, that lie between rows 1 and 2, and 4 and 5, respectively (Saavedra et al., 2014) (Figure 1). As the first stage develops into the second, there are neither cell divisions nor cell deaths; nevertheless, the epidermal cells rearrange and some change their identities (Saavedra et al., 2014)—for example, the tendon cells form denticles in the embryo and early larva (Dilks and DiNardo, 2010) but not in the second and third stages; some cells even change their polarities (Saavedra et al., 2014).10.7554/eLife.06303.002
Part of the wild type ventral epidermis.
(A) Ventral abdominal epidermis, including the central midline, showing rows of predenticles 1–6. (B) The denticles made later by the same larva. (C) A model: ds expression combined with expression of fj, gives the presumed pattern of Ds activity and explains the orientation of the rows 0–6. Each row points towards the neighbouring cell with the most Ds. The line within a cell (usually sloped) indicates that each cell has different amounts of Ds at its anterior and posterior limits (see Figure 4). In all the figures, one individual is imaged first at mid second stage and later after moulting to third stage. Note almost exact correspondence between predenticles and denticles in all cases. Cell boundaries (Ecad) are shown in red and actin highlighted in green (see ‘Materials and methods’). Cells shaded in blue belong to the posterior compartment (Lawrence and Struhl, 1996). T1 and T2 indicate the two rows of tendon cells, shaded in grey in C. Although tendon cells do not show actin predenticles, they show characteristic actin palisade-like structures (Saavedra et al., 2014). The blue dotted lines are transects to locate the diagrammatic cross sections shown at the right of each figure. Anterior is to the left.
DOI:http://dx.doi.org/10.7554/eLife.06303.002
Results and discussionDistribution of Ds activity in the segment
Protein interactions between neighbouring cells are at the core of known PCP systems (Goodrich and Strutt, 2011). In one of these systems, the protocadherins Ds and Ft form heterodimeric bridges between cells (reviewed in Lawrence et al., 2007; Thomas and Strutt, 2012). The deployment and orientation of Ds-Ft bridges within different parts of a cell membrane depend on the amounts of Ds and Ft activity in neighbouring cells. For example, a cell with a low level of Ds presents more Ft (than Ds) on its membrane and this draws more Ds (than Ft) to the abutting membrane of the neighbour; thereby affecting the distribution of dimers within the next cell. In this way, the relative numbers and orientations of heterodimers allow a comparison between a cell's anterior and posterior neighbours so that it orients its denticles towards the neighbour with the higher Ds, and/or lower Ft, activity (Ma et al., 2003; Casal et al., 2006; Matakatsu and Blair, 2006).
In Figure 1C, we show a hypothetical model of the segmental landscape of Ds activity in the epidermis of later larval stages. This model derives from mutant phenotypes of the third stage larvae (Casal et al., 2006; Donoughe and DiNardo, 2011) and experiments in which the distribution of Ds was manipulated, also at later stages (Repiso et al., 2010; Donoughe and DiNardo, 2011). The model also depends on the pattern of expression of four-jointed (Fj), a kinase that activates Ft and deactivates Ds (Brittle et al., 2010; Simon et al., 2010). fj is much more strongly expressed in the tendon cells than elsewhere—it should lower the activity of Ds in these cells—and graded in cells from rows 2 (high) to 4 (low) (Saavedra et al., in preparation). These pieces of evidence taken together argue for, but do not prove, the segmental landscape of Ds activity shown in Figure 1C. The hypothetical landscape can explain the orientation of all the denticle rows.
Atypical cells and multipolarity
If the relevant cells of the larva (cells from row 0 to row 6 and including the two rows of tendon cells) were stacked in 10 parallel rows like the bricks in a wall (as in Figure 1A), our model would be a sufficient explanation for the polarity of all the cells. But in reality, the arrangement of the cells is less orderly. Consider the cells of row 4. A few of these cells are tilted from the mediolateral axis; they take up ‘atypical’ positions, contributing to two different rows of cells in the normal stack (one is shown in Figure 2A,B, shaded magenta and Figure 2—figure supplement 1). In such a cell, one portion occupies territory between a row 3 cell (in which Ds activity is medium) and a T2 cell (in which Ds activity is low). Thus, this portion of the atypical cell has neighbours exactly like an ideal row 4 cell and its denticles point forwards towards the neighbouring row 3 cell (Figure 2A–D and Figure 2—figure supplement 1).10.7554/eLife.06303.003
Atypical cells.
(A–D) One atypical and multipolar cell, largely in row 4, is shown, in B–D (shaded in magenta). The transects shown as dotted lines in C and G are illustrated in D and H with the presumed amounts of Ds and Fj as well as the presumed activity of Ds. (E–H) One atypical cell of row 2 is shown; labelling as in other figures. See also Figure 2—figure supplement 1.
DOI:http://dx.doi.org/10.7554/eLife.06303.003
10.7554/eLife.06303.004
Atypical cells: more examples.
An example (A-D) showing two atypical cells, one in row 2, one in row 4. Even though much of the row 2 cell abuts, not T1 as is ‘typical’, but in other row 2 cells, the polarities of all denticles are always normal (Table 1). The row 4 atypical cell is of interest because it has only a small promontory that abuts another row 4 cell, and yet this small promontory has one posteriorly oriented denticle. Presented as the other figures. Related to Figure 2.
DOI:http://dx.doi.org/10.7554/eLife.06303.004
The neighbouring row 3 cell is presumed to have more Ds activity than the T2 cell (Figure 2D and Figure 2—figure supplement 1). However, the other portion of the same atypical cell intervenes between a row 3 and a normal row 4 cell and the denticles in that portion point backwards; again towards the neighbouring cell with higher Ds activity (in this case, a row 4 cell). Note that the backwards-pointing polarity adopted by this domain of the atypical cell does not, and is not expected to, affect the polarity of neighbouring cells. Its anterior neighbour, a row 3 cell, lies between a row 2 and a row 4, as does any normal row 3 cell, whereas its posterior neighbour, a row 4 cell, abuts a T2 cell that has a low Ds activity (a lower Ds activity than this portion of the atypical cell finds at its anterior interface). Therefore, under our hypothesis, cells touching this domain of the atypical row 4 cell do not differ, with respect to the Ds/Ft activities of their neighbours, from normal row 3 and 4 cells and consequently show normal polarity: thus, the row 3 cell points its denticles posteriorly, and the row 4 cell points its denticles anteriorly.
To quantitate, we selected atypical cells for study and then ask does the orientation of denticles in one part of a cell correlate with the anterior and posterior neighbours of that part? The answer is very clearly yes (Table 1). We explain below that these multipolar cells tell us that a portion of the membrane of one cell can compare itself with that in a facing portion of the same cell and this comparison polarises that particular domain of the cell. By this means a cell reads the Ds activities of its anterior and posterior neighbours and responds accordingly. In the case of the atypical row 4 cells, even though all their anterior neighbours are of the same type (row 3 cells), the neighbours at the posterior membrane are of two different types (T2 and row 4); accordingly, the two different regions of the cell manifest opposing polarities.10.7554/eLife.06303.005
Atypical cells: quantitation of denticle polarities in relation to neighbouring cells showing the effect of the Ds/Ft system
DOI:http://dx.doi.org/10.7554/eLife.06303.005
Wild type
ds−ft−
Anterior neighbour
Denticle polarity of atypical Row 2 cells†
Posterior neighbour
Anterior neighbour
Denticle polarity of atypical Row 2 cells§
Posterior neighbour
Anteriorly
Posteriorly
Anteriorly
Posteriorly
T1 cell
0
52
Row 3 cell
T1 cell
16
21
Row 3 cell
Row 2 cell
0
35
Row 3 cell
Row 2 cell
14
13
Row 3 cell
Anterior neighbour
Denticle polarity of atypical Row 4 cells‡
Posterior neighbour
Anterior neighbour
Denticle polarity of atypical Row 4 cells¶
Posterior neighbour
Anteriorly
Posteriorly
Anteriorly
Posteriorly
Row 3 cell
110
8*
T2 cell
Row 3 cell
54*
37
T2 cell
Row 3 cell
8**
41
Row 4 cell
Row 3 cell
24
20**
Row 4 cell
Denticles of 29 atypical cells.
Denticles of 27 atypical cells. 8 of 8* (and 6 of 8**) were predenticles located in ambiguous positions and their denticles were arbitrarily allocated to those classes favouring the null hypothesis. Fisher exact test p-value: <2.2−16.
Denticles of 23 atypical cells. Fisher exact test p-value: 0.6135.
Denticles of 24 atypical cells. 1 of 54* (and 1 of 20**) were predenticles located in ambiguous positions and their denticles were arbitrarily allocated to those classes disfavouring the null hypothesis. Fisher exact test p-value: 0.7104.
Numbers in bold emphasise the main result of the table that is, the effect of neighbours on denticle polarities in the wildtype. This effect does not exist in the mutant larvae.
Atypical cells occur in other regions of the segment, for example in row 2. These cells have a mix of neighbours also, and anatomically are equivalent to the atypical cells near row 4; however, all their predenticles and denticles point backwards as they do in the wild type (Figure 2E–H and Figure 2—figure supplement 1). This fits exactly with the model (Figure 2H and Figure 2—figure supplement 1), because a cell, or part of a cell, that intervenes between T1 and row 3 (i.e., like a normal row 2 cell) makes backward-pointing denticles; they point away from the tendon cells (where Ds is low) and towards cells where the Ds activity is presumed to be higher. The situation is the same for a cell or part of a cell that inserts itself between another row 2 cell and a row 3 cell (because, according to the model of the segmental landscape of Ds activity, row 3 has a higher Ds activity than row 2 [Figure 1C and Figure 2H; Saavedra et al., in preparation]) and they therefore make backwardly oriented denticles. If we quantitate as before, again we find complete agreement between the orientation of the denticles and the presumed Ds level of the neighbours, although in this case all the denticles point backwards (Table 1).
Atypical cells also occur in row 3; these cells have portions inserted between a different row 3 and a row 4 cell and their denticles point backwards toward the neighbour cell with higher Ds activity, that is row 4 (not shown).
Testing the role of the Ds/Ft system
Up to now, we have hypothesised that the orientations of all denticles both in the normal and the multipolar cells are primarily, or only, determined by the Ds/Ft system. To test this hypothesis, we first looked at larvae in which the Ds/Ft system is broken. That system depends on heterodimeric bridges made up of Ds and Ft molecules; these bridges cannot form without either or both of these proteins. Using ds−ft− larvae, we found that atypical cells do occur both in rows 2 and 4 and are anatomically similar to those in wild type. However, the orientation of denticles in these cells is equally indeterminate in both rows and is independent of neighbours (Figure 3A–D, Table 1). This is presumably and simply because, without the Ds-Ft intercellular bridges, a cell cannot compare neighbours to ascertain where to point its denticles. It also argues that there is no other PCP agent, apart from the Ds/Ft system, that is responsible for these multipolar cells in the larva.10.7554/eLife.06303.006
Testing the model.
(A–D) An epidermis lacking both the ds and ft genes. The cell shaded in magenta is a row 4 cell which has an atypical disposition. The predenticle and denticle orientation is variable and awry. (E–H) A clone of two cells over-expressing ds (marked in yellow, E and shaded in yellow, F–H) imposes an orientation on all or parts of the neighbouring wild-type cells. Note that one reoriented cell (*) appears to have no direct contact with clonal cells expressing ds but only with neighbours of such cells. We have seen this in other cases. In the adult abdomen such propagation extends to several cell diameters (Casal et al., 2002; Casal et al., 2006) and has also been reported in larvae (Repiso et al., 2010; Donoughe and DiNardo, 2011). See also Figure 3—figure supplement 1.
DOI:http://dx.doi.org/10.7554/eLife.06303.006
10.7554/eLife.06303.007
Breaches in rows of tendon cells.
(A–D) An example of the effect on cells of an interruption in the T1 row. This is associated with an unpolarised cell shaded in magenta. (E–G) shows the similar effect of a breach in the T2 row of cells. These results are interpreted in D and H and fit our model of Ds activity very well. Presented as the other figures. Related to Figure 3.
DOI:http://dx.doi.org/10.7554/eLife.06303.007
In a further test, we made small clones of marked cells that over-express ds and studied them in the larva in vivo. These clones were initiated early in embryogenesis, around the blastoderm stage and are small, ranging in size from 1 to 5 cells (Saavedra et al., 2014). They probably begin to make excess Ds in the embryo following the first one or two divisions that occur in the postblastoderm epidermis (Foe and Alberts, 1983; Hartenstein and Campos-Ortega, 1985) Consistent with observations on the adult of clones expressing high levels of ds (Casal et al., 2002), or on larvae when groups of cells over-express ds (Repiso et al., 2010; Donoughe and DiNardo, 2011), neighbouring wild type cells become repolarised so that their denticles point towards cells with excess Ds. In larvae, we now find that portions of cells are affected and the orientation of denticles in that part of the cell in contact with the ds-expressing cell can point inwards whilst other parts of the same cell, whose neighbours are not over-expressing ds, point in various directions. Figure 3E–H illustrates a nice example, several multipolar cells are induced by the clone and one entire cell is reoriented. In this and other examples, multipolar cells also appear occasionally to affect the polarity of their neighbouring cells (see legend, Figure 3). However, such a propagation of polarity seems to be limited to those portions of neighbouring cells that contact regions of the multipolar cells that show an abnormal polarity. Again, this finding supports our conclusion that polarity is the result of a local and independent comparison between facing membranes of cell domains. Note however that this local propagation of polarity does not affect row 4 cells. We believe that this observation can be explained if the low Ds activity of the T2 cells were to strongly influence the polarity of row 4 cells, strongly enough to counteract any effect that might originate from any multipolar or repolarised row 3 neighbours. Note that the marked ds-expressing cells themselves show no preferred polarity, their denticles pointing in diverse directions. This might be explained by the amount of Ds activity in such cells being much higher than all their neighbours, giving insufficient directional bias to fix their own polarities.
Gaps in the tendon cells
A different atypical situation occurs occasionally when there is an opening in the usually continuous rows of tendon cells. Such gaps often affect the orientation and pattern of denticles, as well as forming some multipolar cells. Examples of these are shown (Figure 3—figure supplement 1), and they find the same general explanation. Consider first a cell that is located next to a gap in the T1 tendon row and thus adjacent to a row 1 and a row 3 cell. The rows 1 and 3 have similar amounts of Ds activity and so the cell in question develops no defined polarity (magenta in Figure 3—figure supplement 1B–D). In another case, we see the T2 row of tendon cells is interrupted and in the breach there is a denticulate cell that comes to lie between a row 4 and a row 6 cell (Figure 3—figure supplement 1E–H). This creates an ambiguous situation as both these neighbours have similar amounts of Ds and give no directional cue. Consequently, the cell in question has predenticles and denticles that are placed in the middle of the cell (Figure 3—figure supplement 1E), giving rise to denticles that are spread over the apical surface of the cell and point in mixed directions (shown in magenta in Figure 2—figure supplement 1F–H). Again, both these situations fit with the model that cells or parts of cells point their denticles towards that neighbour cell that has the most Ds activity.
Refining the model
These results offer evidence that the earlier models of the Ds/Ft system (Strutt and Strutt, 2002; Casal et al., 2006) apply to these large larval cells, just as they do to the eye (Strutt and Strutt, 2002; Simon, 2004), the wing (Ma et al., 2003), and the abdomen, (Casal et al., 2006). For example, in all these organs and stages, the cell membrane of a clone containing excessive amounts of Ds will attract Ft to the abutting membranes of neighbouring cells and lead to a redistribution of Ds-Ft heterodimers in those cells, and even beyond (reviewed in Thomas and Strutt, 2012).
We proposed (Casal et al., 2006) that the polarity of a cell is the outcome of a comparison between its anterior and posterior membranes; that the amount and orientations of Ds-Ft heterodimers are compared and the denticles point towards that region of the cell membrane that has most Ft (i.e., abutting a neighbouring cell that has higher Ds levels). The multipolar cells provide new evidence for this hypothesis. First, note that the level of Ds in a row 3 cell should be higher than the level of Ds in a T2 cell but lower than the level of Ds in a row 4 cell (Figure 1C). Then, consider the following arguments: one that there is a comparison between the anterior and posterior membranes of the row 4 multipolar cell filled in magenta in Figure 2B–C. Two, that this comparison is local to different regions of that cell. Across the whole stretch of its anterior membrane this row 4 multipolar cell contacts two cells of row 3, each presenting equal amounts of Ds. However, within the posterior membrane of this same multipolar cell there are two separate regions, each abutting cells with different levels of Ds: a region abutting the T2 neighbour, and a region abutting the normal row 4 cell. The region abutting the T2 has a lower amount of Ft (as a result of the low amount of Ds in the T2 cell), and a higher amount of Ds (as a result of the high amount of Ft in the T2 cell) than the facing anterior membrane of the same cell, so its denticles point forwards. The region abutting the normal row 4 cell has a higher amount of Ft (as a result of the high amount of Ds in the normal row 4 cell) and a lower amount of Ds (as a result of the lower amount of Ft in the normal row 4 cell) than in the facing anterior membrane of the same cell, so its denticles point backwards.
Multipolarity tells us something new and unexpected: that the comparison is local to different regions of the cell. The two domains described above are comparing their facing anterior and posterior membranes, independent of each other. Therefore, the comparison cannot be a signal that pervades the whole cell, and instead multipolarity suggests the existence of oriented ‘conduits’ that link facing regions within the anterior and posterior membranes of a single cell. If such oriented conduits exist they could allow the directional transport and/or unequal stabilisation of components of the polarity machinery (such as Ds and Ft themselves). Our conception of these conduits is yet incomplete; but if their orientation were determined by the distribution of Ds and Ft, and if they also helped convey Ds and Ft across the cell, then together both these properties could constitute a feedback mechanism. Such a mechanism would make the polarity of cells more robust and also would affect the propagation of polarity from cell to cell. Understanding propagation is important because experiments suggest that the distribution of Ds and Ft within the membrane of one cell is partly determined by the distribution of these molecules in that cell's neighbours and also in cells beyond (reviewed in Thomas and Strutt, 2012).
The disposition of Ds-Ft heterodimers, as indicated in Figure 4, will be the result of these processes. In Figure 4, we imagine the distribution of Ft-Ds heterodimeric bridges in the epidermis and speculate in more detail how local comparisons might determine denticle polarity of the cells or parts of cells. Note that if Ds and Ft are transported across the cell between limited domains, there is no need to invoke free diffusion of these molecules (as in Mani et al., 2013; Abley et al., 2013). Similar conduits would be present in all cells, but normally, since they signal consistently in all parts of the cell, they would not be detected as separable elements.10.7554/eLife.06303.008
Polarised conduits in the cells.
A hypothetical view of how Ds and Ft polarise cells or parts of cells. All membranes contain both kinds of dimers: these are Ds in the cell x (Dsx), linked to Ft in the neighbouring cell y (Fty), or Ft in the cell x (Ftx) linked to Ds in the neighbouring cell y (Dsy). We proposed (Casal et al., 2006) that intercellular bridges consisting of heterodimers of Ds and Ft are asymmetrically distributed in a polarised cell and determine the polarity of that cell. In the diagram, we indicate a majority of Ds by yellow, and a majority of Ft by rufous. Some membranes contain similar numbers of Ds and Ft and these are shown with alternating blotches of the two colours. Arrows indicate the sign and the paths of the oriented conduits that span between facing and limited areas of membrane. Such conduits can give small parts of cells an individual polarity, as in the atypical cell in row 4 (red arrows). The tendon cells, T1 and T2, largely drive the segmental pattern of Ds activity—they have low Ds activity and therefore the majority of heterodimers formed between a T2 and a row 4 cell are Ds4–FtT2. Similarly at the boundary between a row 2 and a row 3 cell, the majority of the heterodimers are Ft2–Ds3; partly because at the opposite boundary between a T1 and a row 2 cell, the heterodimers are largely FtT1–Ds2. Where row 3 and row 4 cells meet in the wild type, they are imagined to have similar levels of Ds3–Ft4 and Ft3–Ds4 because, at that cell junction similar, but opposite, effects from very low Ds levels in both T1 and T2 tendon cells converge. However, in the red-arrowed region, at the anterior limit of the atypical cell, the heterodimers are mostly Ft3–Ds4. We propose that in this red-arrowed region, the deployment of heterodimers is the outcome of a different comparison made between facing subregions of the anterior and posterior limits of this atypical cell. What is different about this comparison? In this red-arrowed region, the cell's anterior neighbours (row 3) have less Ds than the posterior neighbour (a normal row 4). As a result, at the boundary between row 3 and atypical row 4, the heterodimers will be mostly Ft3–Ds4. However, at the facing boundary, between atypical and normal row 4, there are balancing influences from anterior and posterior directions, as in a boundary between a normal row 3 and a normal row 4, and as a consequence the amounts of Ds4–Ft4 and Ft4–Ds4 heterodimers should be similar.
DOI:http://dx.doi.org/10.7554/eLife.06303.008
There are hints at what these conduits might be: several authors described oriented microtubules in planar polarised cells (Fristrom and Fristrom, 1975; Eaton et al., 1996; Turner and Adler, 1998) and studied microtubule growth and polarity in the developing wing (Harumoto et al., 2010; Olofsson et al., 2014). In a particular part of the wing, and for a relatively short time, these authors noted that there are microtubules that are oriented proximodistally. In this part of the wing, they showed a slight preponderance (ca 5%) of oriented microtubules with the minus ends proximal and their plus ends distal and proposed that Ds and Ft are instrumental in this net orientation. Changing the distribution of Ds in a test part of the wing changed the net orientation of the microtubules (Harumoto et al., 2010). Our hypothesis of conduits could relate to these findings; microtubules could form all, part or none of these conduits. But, if they do it is not clear why, in most parts of the wing and for most of the time, the microtubules and the hairs are not co-oriented. In any case, a net orientation of microtubules might be read out as a net transport of vesicles from proximal to distal (Shimada et al., 2006; Harumoto et al., 2010; Gault et al., 2012). A similar process in the larval epidermal cells could lead to the subsequent orientation of predenticles in the membrane, but how they would do this is unknown. Microtubules have been observed in cells of the embryonic epidermis at the time that the denticles of the first larval stage are formed, although their actual polarity is not known (Price et al., 2006; Marcinkevicius and Zallen, 2013).
Multipolarity in other cell types
Motile cells such as fibroblasts or Dictyostelium tend to extend lamellipodia in different directions at once, net movement resulting if one direction is favoured over others (Shi et al., 2013). It is not clear if this kind of multipolarity relates to PCP: a defining feature of PCP is that the orientation of a cell is fixed by cell interaction and this is not usually the case with isolated motile cells. However, in plants, pavement cells are genuinely multipolar. They show that even between two identical neighbours, local interactions can be of different sign and can organise the cytoskeleton in different ways to build cells that, at least near the periphery, have regions of opposing polarities (Xu et al., 2010).
It has been thought that PCP in animals involves the whole cell and includes organelles and the cytoplasm, structures that form in membranes such as cilia or stereocilia as well as pervasive outputs such as mechanical tension (Wallingford, 2010; Deans, 2013; Guillot and Lecuit, 2013). Multipolarity in the larval cells of Drosophila that we report makes it clear that PCP is, or can be, subcellular; consequently, some current models of PCP may need to be adapted. For instance, our observations raise the possibility that all cells are fundamentally ‘multipolar’ but, usually, all subregions of a cell are subject to consistent polarising influences and are co-oriented.
Materials and methodsMutations and transgenes
Flies were reared at 25°C on standard food. The Flybase (St Pierre et al., 2014) entries of the relevant constructs used in this work are the following: DE-cad::tomato: shgKI.T:Disc\RFP-tdTomato; sqh.utrp::GFP: Hsap\UTRNsqh.T:Avic\GFP-EGFP; UAS.cherry::moesin: MoeScer\UAS.P\T.T:Disc\RFP-mCherry; UAS.stinger::GFP: Avic\GFPStinger.Scer\UAS.T:nls-tra; UAS.ectoDs: dsecto.Scer\UAS; UAS.cd8::GFP: Mmus\Cd8aScer\UAS.T:Avic\GFP; tub>stop>Gal4: P{GAL4-αTub84B(FRT.CD2).P}; sry.FLP: Scer\FLP1sry-alpha; ds−: dsUA071; ft−: ft15.
Second stage larvae at the pre-third stage were mounted in a drop of Voltalef 10S oil on a microscope slide and imaged using a Leica SP5 confocal microscope. The larvae were carefully removed, kept at 25°C on an agar plate with fresh yeast paste until they moulted into the third stage; cuticles of third stage were prepared using a standard protocol (Saavedra et al., 2014).
For Table 1, predenticles of cells of rows 2 and 4 with atypical dispositions were classified as follows: predenticles localised in a domain of the cell that abutted a tendon cell (i.e., T1 or T2), predenticles localised in a domain that abutted a non-tendon cell (i.e., another row 2 or 4 cell), or predenticles localised in an intermediate domain. The orientations of the denticles formed by these predenticles were scored, one by one, in the third stage cuticles. Late second stage larvae with small clones of marked cells in the epidermis were obtained as previously described (Saavedra et al., 2014).
Acknowledgements
We thank the Bloomington Stock Center, Marco Antunes, Marcus Bischoff, Paola Cognini, Thomas Lecuit, and Eurico Morais de Sá for flies and members of the lab for discussions. MR was partially supported by an Erasmus Placement scholarship and PS was supported by studentships from Fundação para a Ciência e a Tecnologia and the Cambridge Philosophical Society. This work was kindly supported by the Wellcome Trust Investigator Award to PAL, WT096645MA
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
MR, Conception and design, Acquisition of data, Analysis and interpretation of data
PS, Conception and design, Acquisition of data, Analysis and interpretation of data
JC, Conception and design, Analysis and interpretation of data, Drafting or revising the article
PAL, Conception and design, Analysis and interpretation of data, Drafting or revising the article
ReferencesAbleyKDe ReuillePBStruttDBanghamAPrusinkiewiczPMareeAFGrieneisenVACoenE2013An intracellular partitioning-based framework for tissue cell polarity in plants and animalsDevelopment1402061207410.1242/dev.062984BrittleALRepisoACasalJLawrencePAStruttD2010Four-jointed modulates growth and planar polarity by reducing the affinity of dachsous for fatCurrent Biology2080381010.1016/j.cub.2010.03.056CasalJLawrencePAStruhlG2006Two separate molecular systems, Dachsous/Fat and Starry night/Frizzled, act independently to confer planar cell polarityDevelopment1334561457210.1242/dev.02641CasalJStruhlGLawrencePA2002Developmental compartments and planar polarity in DrosophilaCurrent Biology121189119810.1016/S0960-9822(02)00974-0DeansMR2013A balance of form and function: planar polarity and development of the vestibular maculaeSeminars in Cell & Developmental Biology2449049810.1016/j.semcdb.2013.03.001DilksSADiNardoS2010Non-cell-autonomous control of denticle diversity in the Drosophila embryoDevelopment1371395140410.1242/dev.045450DonougheSDiNardoS2011dachsous and frizzled contribute separately to planar polarity in the Drosophila ventral epidermisDevelopment1382751275910.1242/dev.063024EatonSWepfRSimonsK1996Roles for Rac1 and Cdc42 in planar polarization and hair outgrowth in the wing of DrosophilaThe Journal of Cell Biology1351277128910.1083/jcb.135.5.1277FoeVEAlbertsBM1983Studies of nuclear and cytoplasmic behavior during the 5 mitotic-cycles that precede gastrulation in Drosophila embryogenesisJournal of Cell Science613170FristromDFristromJW1975The mechanism of evagination of imaginal discs of Drosophila melanogaster: I. General considerationsDevelopmental Biology4312310.1016/0012-1606(75)90127-XGaultWJOlguinPWeberUMlodzikM2012Drosophila CK1-γ, gilgamesh, controls PCP-mediated morphogenesis through regulation of vesicle traffickingThe Journal of Cell Biology19660562110.1083/jcb.201107137GoodrichLVStruttD2011Principles of planar polarity in animal developmentDevelopment1381877189210.1242/dev.054080GuillotCLecuitT2013Mechanics of epithelial tissue homeostasis and morphogenesisScience3401185118910.1126/science.1235249HartensteinVCampos-OrtegaJA1985Fate-mapping in wild-type Drosophila melanogasterWilhelm Roux's Archives of Developmental Biology19418119510.1007/BF00848246HarumotoTItoMShimadaYKobayashiTJUedaHRLuBUemuraT2010Atypical cadherins Dachsous and Fat control dynamics of noncentrosomal microtubules in planar cell polarityDevelopmental Cell1938940110.1016/j.devcel.2010.08.004LawrencePAStruhlG1996Morphogens, compartments, and pattern: lessons from Drosophila?Cell8595196110.1016/S0092-8674(00)81297-0LawrencePAStruhlGCasalJ2007Planar cell polarity: one or two pathways?Nature Reviews Genetics855556310.1038/nrg2125MaDYangCHMcNeillHSimonMAAxelrodJD2003Fidelity in planar cell polarity signallingNature42154354710.1038/nature01366ManiMGoyalSIrvineKDShraimanBI2013Collective polarization model for gradient sensing via Dachsous-Fat intercellular signalingProceedings of the National Academy of Sciences of USA110204202042510.1073/pnas.1307459110MarcinkeviciusEZallenJA2013Regulation of cytoskeletal organization and junctional remodeling by the atypical cadherin FatDevelopment14043344310.1242/dev.083949MatakatsuHBlairSS2006Separating the adhesive and signaling functions of the Fat and Dachsous protocadherinsDevelopment1332315232410.1242/dev.02401OlofssonJSharpKAMatisMChoBAxelrodJD2014Prickle/spiny-legs isoforms control the polarity of the apical microtubule network in planar cell polarityDevelopment1412866287410.1242/dev.105932PriceMHRobertsDMMcCartneyBMJezuitEPeiferM2006Cytoskeletal dynamics and cell signaling during planar polarity establishment in the Drosophila embryonic denticleJournal of Cell Science11940341510.1242/jcs.02761RepisoASaavedraPCasalJLawrencePA2010Planar cell polarity: the orientation of larval denticles in Drosophila appears to depend on gradients of Dachsous and FatDevelopment1373411341510.1242/dev.047126SaavedraPVincentJPPalaciosIMLawrencePACasalJ2014Plasticity of both planar cell polarity and cell identity during the development of DrosophilaeLife3e0156910.7554/eLife.01569ShiCHuangCHDevreotesPNIglesiasPA2013Interaction of motility, directional sensing, and polarity modules recreates the behaviors of chemotaxing cellsPLOS Computational Biology9e100312210.1371/journal.pcbi.1003122ShimadaYYonemuraSOhkuraHStruttDUemuraT2006Polarized transport of Frizzled along the planar microtubule arrays in Drosophila wing epitheliumDevelopmental Cell1020922210.1016/j.devcel.2005.11.016SimonMA2004Planar cell polarity in the Drosophila eye is directed by graded Four-jointed and Dachsous expressionDevelopment1316175618410.1242/dev.01550SimonMAXuAIshikawaHOIrvineKD2010Modulation of fat:dachsous binding by the cadherin domain kinase four-jointedCurrent Biology2081181710.1016/j.cub.2010.04.016St PierreSEPontingLStefancsikRMcQuiltonPFlyBaseC2014FlyBase 102–advanced approaches to interrogating FlyBaseNucleic Acids Research42D780D78810.1093/nar/gkt1092StruttHStruttD2002Nonautonomous planar polarity patterning in Drosophila: dishevelled-independent functions of frizzledDevelopmental Cell385186310.1016/S1534-5807(02)00363-5ThomasCStruttD2012The roles of the cadherins Fat and Dachsous in planar polarity specification in DrosophilaDevelopmental Dynamics241273910.1002/dvdy.22736TurnerCMAdlerPN1998Distinct roles for the actin and microtubule cytoskeletons in the morphogenesis of epidermal hairs during wing development in DrosophilaMechanisms of Development7018119210.1016/S0925-4773(97)00194-9WallingfordJB2010Planar cell polarity signaling, cilia and polarized ciliary beatingCurrent Opinion in Cell Biology2259760410.1016/j.ceb.2010.07.011XuTWenMNagawaSFuYChenJGWuMJPerrot-RechenmannCFrimlJJonesAMYangZ2010Cell surface-and Rho GTPase-based Auxin signaling controls cellular Interdigitation in ArabidopsisCell1439911010.1016/j.cell.2010.09.00310.7554/eLife.06303.009Decision letterVijayRaghavanKReviewing editorNational Centre for Biological Sciences, Tata Institute for Fundamental Research, India
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Thank you for sending your work entitled “Planar cell polarity: regions within a single cell can be polarised independently” for consideration at eLife. Your article has been favourably evaluated by K VijayRaghavan (Senior editor and Reviewing editor) and 2 reviewers. As you can see the substantial comments are quite readily addressable.
The Reviewing editor and the reviewers discussed their comments before we reached this decision, and the Reviewing editor has assembled the following comments to help you prepare a revised submission.
Lawrence and coworkers have shown in previous work that the Drosophila larval epidermis serves as excellent model system to study the Ft/Ds pathway, in particular, as in larvae emerging denticles can be followed relative to individual cell borders over time. Depending on the row of cells in a segment, denticles either point anteriorly or posterior, respectively. In addition, in late stages (L3), there are two rows of tendon cells that lack denticles and serve as borders of Ds activity (as they apparently express high levels of Fj repressing Ds activity based on a paper under preparation and not available to the reviewer). Here, Rovira et al. propose a model that each cell compares Ds activity levels of its anterior and posterior neighbours and grows the denticle towards the neighbour with higher Ds activity. To test, they make use of the fact that the larval epithelium is not a perfect wall of bricks, but that some of the very large cells contact 2 cells of different rows on one of their sides (with anticipated distinct activities of Ds).
Rovira et al. show that such cells can point denticles in both orientations, of course not randomly as in Ft/Ds mutants, but according to the (hypothetical) Ds activity difference between the neighbours in contact at that position. This is highly intriguing. On the wings of Drosophila PCP mutants, cells frequently produce two hairs with divergent polarity, with the individual hairs matching the polarity of different neighbouring cells.
However, the embryonic multipolar cells described in the manuscript are wild-type and therefore more amenable to meaningful analysis. In retrospect, the striking results are perhaps not completely surprising as they fit older models in which Ds and Ft across a membrane can recruit each other to the boundary. Importantly, over-expression of Ds in a cell can repolarise cells (known), but do so for parts of a cell only that it is in contact with (leaving the rest of the cell polarised as instructed by neighbours; but see below), which is intriguing.
The Irvine (Ambegaonkar, 2012) and Strutt (Brittle, 2013) labs have made tagged Ds/Ft constructs that may be expressed in mosaics in the larva (the authors state that the epithelial cells divide after they can induce flip-pout clones; it thus may be possible to use their FLP to make mitotic clones). We do not ask that this be done for acceptance of the paper as a 'Research Advance' for eLife, but only suggest this as possible direction the authors may like to consider.
While the work presented finds astonishing differences of polarity within a single cell, a lot in the paper is based on the proposed model (without looking at Ft or Ds levels).
One concern is that the model shown in Figure 4 does not obviously match the mechanism described in the text (paragraph two of subsection headed “Refining the model” in the Results and Discussion section). While the model is just that and not merely a summary of the realists, it will serve all well if the authors re-craft the text to bring clarity and avoid concerns in about what the authors mean to convey.
Specifically, the Ds landscape shown in Figure 1C can produce the observed denticle pattern if each cell points its denticles towards the neighbouring cell with the highest Ds activity. A cell can sense the levels of Ds in neighbouring cells by monitoring the orientation of Ft-Ds dimers that are shared with each neighbour. In effect, a cell points its denticles towards the region of its own membrane that has the greatest proportion of Ft (rather than Ds) within Ft-Ds dimers.
However, in Figure 4, both the boundary between row 3 and 4 cells, and the atypical boundary between row 4 and 4 cells, are represented as having approximately equal amounts of Ft-Ds dimers in each orientation. If this were the case, how would the red arrow (oriented conduit) form in the atypical cell, since initially both 3/4 and 4/4 boundaries would have similar Ft-Ds orientations? For the model to work, the whole 3/4 boundary should have a majority (but not all) of Ft-Ds dimers with Ds in the row 4 cell.
Some further points about the model, which can be elaborated in the revised manuscript. It is not clear why the 3/4 boundary to the left of the red arrow has Ft-Ds dimers that are all oriented with Ds in the row 4 cell. Why does this differ from the rest of the 3/4 boundary? The figure legend states that “this deployment of heterodimers is the result of a local comparison made between facing subregions of the anterior and posterior limits of this atypical cell”. There is some (apparent?) circularity between cause and effect here. Our understanding is that the local comparison is between Ft and Ds distribution in the facing subregions, not that the comparison determines the distribution of Ft and Ds. Do clarify this in the revised manuscript.
Some more points:
Introduction section: “This comparison appears to involve limited regions of membranes that face each other along the anterior to posterior axis and depends on oriented conduits that span between the two membranes.” This sentence is confusing as it implies that ‘oriented conduits’ are required for the Ft/Ds comparison process, although my understanding is that the oriented conduits are proposed to result from the Ft/Ds comparison. In addition, the existence of ‘oriented conduits’ appears from this sentence to be established, whereas they are actually a proposal of the model. Please correct this.
10.7554/eLife.06303.010Author response
Lawrence and coworkers have shown in previous work that the Drosophila larval epidermis serves as excellent model system to study the Ft/Ds pathway, in particular, as in larvae emerging denticles can be followed relative to individual cell borders over time. Depending on the row of cells in a segment, denticles either point anteriorly or posterior, respectively. In addition, in late stages (L3), there are two rows of tendon cells that lack denticles and serve as borders of Ds activity (as they apparently express high levels of Fj repressing Ds activity based on a paper under preparation and not available to the reviewer). Here, Rovira et al. propose a model that each cell compares Ds activity levels of its anterior and posterior neighbours and grows the denticle towards the neighbour with higher Ds activity. To test, they make use of the fact that the larval epithelium is not a perfect wall of bricks, but that some of the very large cells contact 2 cells of different rows on one of their sides (with anticipated distinct activities of Ds).
Rovira et al. show that such cells can point denticles in both orientations, of course not randomly as in Ft/Ds mutants, but according to the (hypothetical) Ds activity difference between the neighbours in contact at that position. This is highly intriguing. On the wings of Drosophila PCP mutants, cells frequently produce two hairs with divergent polarity, with the individual hairs matching the polarity of different neighbouring cells.
We are interested in knowing where this point is published, as if it is true we ought to and would like to refer to it, can the reviewer oblige with a reference?
However, the embryonic multipolar cells described in the manuscript are wild-type and therefore more amenable to meaningful analysis. In retrospect, the striking results are perhaps not completely surprising as they fit older models in which Ds and Ft across a membrane can recruit each other to the boundary. Importantly, over-expression of Ds in a cell can repolarise cells (known), but do so for parts of a cell only that it is in contact with (leaving the rest of the cell polarised as instructed by neighbours; but see below), which is intriguing.
The Irvine (Ambegaonkar, 2012) and Strutt (Brittle, 2013) labs have made tagged Ds/Ft constructs that may be expressed in mosaics in the larva (the authors state that the epithelial cells divide after they can induce flip-pout clones; it thus may be possible to use their FLP to make mitotic clones). We do not ask that this be done for acceptance of the paper as a 'Research Advance' for eLife, but only suggest this as possible direction the authors may like to consider.
Yes, thank you. We are aware of these constructs and had already begun to explore in this direction and have been making new genetic stocks. The discovery in this paper of multipolar cells can, we think, be exploited to learn more but obviously that will take time.
While the work presented finds astonishing differences of polarity within a single cell, a lot in the paper is based on the proposed model (without looking at Ft or Ds levels).
Yes, we are sorry about this. The model is evidenced in another experimental paper (Saavedra et al., in preparation) that is not yet complete. We have been doing some difficult experiments to complete it and the hope is that it should be finished and submitted to a specialised journal within a couple of months. We didn't want to wait to publish our discovery of multipolar cells, nor to “bury” this discovery in a long and detailed paper about denticle patterns.
One concern is that the model shown inFigure 4does not obviously match the mechanism described in the text (paragraph two of subsection headed “Refining the model” in the Results and Discussion section). While the model is just that and not merely a summary of the realists, it will serve all well if the authors re-craft the text to bring clarity and avoid concerns in about what the authors mean to convey.
Okay, see below.
Specifically, the Ds landscape shown inFigure 1Ccan produce the observed denticle pattern if each cell points its denticles towards the neighbouring cell with the highest Ds activity. A cell can sense the levels of Ds in neighbouring cells by monitoring the orientation of Ft-Ds dimers that are shared with each neighbour. In effect, a cell points its denticles towards the region of its own membrane that has the greatest proportion of Ft (rather than Ds) within Ft-Ds dimers.
Agreed.
However, inFigure 4, both the boundary between row 3 and 4 cells, and the atypical boundary between row 4 and 4 cells, are represented as having approximately equal amounts of Ft-Ds dimers in each orientation. If this were the case, how would the red arrow (oriented conduit) form in the atypical cell, since initially both 3/4 and 4/4 boundaries would have similar Ft-Ds orientations? For the model to work, the whole 3/4 boundary should have a majority (but not all) of Ft-Ds dimers with Ds in the row 4 cell.
This description reveals that we have not explained the model in Figure 4 adequately. Therefore, we have expanded and hopefully improved the arguments that lead to Figure 4 both in the text and in Figure 4 legend.
Some further points about the model, which can be elaborated in the revised manuscript. It is not clear why the 3/4 boundary to the left of the red arrow has Ft-Ds dimers that are all oriented with Ds in the row 4 cell. Why does this differ from the rest of the 3/4 boundary? The figure legend states that “this deployment of heterodimers is the result of a local comparison made between facing subregions of the anterior and posterior limits of this atypical cell”. There is some (apparent?) circularity between cause and effect here. Our understanding is that the local comparison is between Ft and Ds distribution in the facing subregions, not that the comparison determines the distribution of Ft and Ds. Do clarify this in the revised manuscript.
Thank you, we have tried to do this.
Some more points:
Introduction section: “This comparison appears to involve limited regions of membranes that face each other along the anterior to posterior axis and depends on oriented conduits that span between the two membranes.” This sentence is confusing as it implies that ‘oriented conduits’ are required for the Ft/Ds comparison process, although my understanding is that the oriented conduits are proposed to result from the Ft/Ds comparison. In addition, the existence of 'oriented conduits' appears from this sentence to be established, whereas they are actually a proposal of the model. Please correct this.
We do hypothesise that the “oriented conduits” are polarised in consequence of a comparison of the facing membranes (across the cell) and also are the agents that compare facing membranes. And we have tried to be clearer about what we mean by causes and effects.
In eukaryotes, most integral membrane proteins are synthesized, integrated into the membrane, and folded properly in the endoplasmic reticulum (ER). We screened the mutants affecting rhabdomeric expression of rhodopsin 1 (Rh1) in the Drosophila photoreceptors and found that dPob/EMC3, EMC1, and EMC8/9, Drosophila homologs of subunits of ER membrane protein complex (EMC), are essential for stabilization of immature Rh1 in an earlier step than that at which another Rh1-specific chaperone (NinaA) acts. dPob/EMC3 localizes to the ER and associates with EMC1 and calnexin. Moreover, EMC is required for the stable expression of other multi-pass transmembrane proteins such as minor rhodopsins Rh3 and Rh4, transient receptor potential, and Na+K+-ATPase, but not for a secreted protein or type I single-pass transmembrane proteins. Furthermore, we found that dPob/EMC3 deficiency induces rhabdomere degeneration in a light-independent manner. These results collectively indicate that EMC is a key factor in the biogenesis of multi-pass transmembrane proteins, including Rh1, and its loss causes retinal degeneration.
DOI:http://dx.doi.org/10.7554/eLife.06306.001
10.7554/eLife.06306.002eLife digest
The membranes that surround cells contain many proteins, and those that span the entire width of the membrane are known as transmembrane proteins. Rhodopsin is one such transmembrane protein that is found in the light-sensitive ‘photoreceptor’ cells of the eye, where it plays an essential role in vision.
Transmembrane proteins are made inside the cell and are inserted into the membrane surrounding a compartment called the endoplasmic reticulum. Here, they mature and ‘fold’ into their correct three-dimensional shape with help from chaperone proteins. Once correctly folded, the transmembrane proteins can be transported to the cell membrane. Incorrect folding of proteins can have severe consequences; if rhodopsin is incorrectly folded the photoreceptor cells can die, leading to blindness in humans and other animals.
Experiments carried out in zebrafish have shown that the chaperone protein Pob is required for the survival of photoreceptor cells. Pob is part of a group or ‘complex’ of chaperone proteins in the endoplasmic reticulum called the EMC complex. This suggests that the EMC complex may be involved in folding rhodopsin, but the details remain unclear.
Here, Satoh et al. studied the role of the EMC complex in the folding of rhodopsin in fruit flies. This involved examining hundreds of flies that carried a variety of genetic mutations and that also had low levels of rhodopsin. The experiments show that dPob—the fly version of Pob—and two other proteins in the EMC complex are required for newly-made rhodopsin to be stabilized. If photoreceptor cells are missing proteins from the complex, the light-sensitive structures in the eye degenerate.
Rhodopsin is known as a ‘multi-pass’ membrane protein because it crosses the membrane multiple times. Satoh et al. found that the EMC complex is also required for the folding of other multi-pass membrane proteins in photoreceptor cells. The next challenge will be to reveal how the EMC complex is able to specifically target this type of transmembrane protein.
DOI:http://dx.doi.org/10.7554/eLife.06306.002
Author keywordsrhodopsinEMCchaperontransmembrane proteinResearch organismD. melanogasterhttp://dx.doi.org/10.13039/501100002241Japan Science and Technology Agency (JST)PRESTO (Precursory Research for Embryonic Science and Technology) 25-J-J4215SatohAkiko Khttp://dx.doi.org/10.13039/501100001691Japan Society for the Promotion of Science (JSPS)21687005 KAKENHISatohAkiko Khttp://dx.doi.org/10.13039/501100001691Japan Society for the Promotion of Science (JSPS)21113510 KAKENHISatohAkiko Khttp://dx.doi.org/10.13039/501100001691Japan Society for the Promotion of Science (JSPS)30529037 KAKENHISatohAkiko Khttp://dx.doi.org/10.13039/100007428The Naito Foundation25-040920SatohAkiko Khttp://dx.doi.org/10.13039/100008273The Novartis Foundation25-050421SatohAkiko Khttp://dx.doi.org/10.13039/501100006094The Hayashi Memorial Foundation for Female Natural Scientists25-051022SatohAkiko KThe funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.elife-xml-version2.0Author impact statementA membrane protein complex in the endoplasmic reticulum is a key factor for the biogenesis of multi-pass transmembrane proteins, including Rh1, and its loss causes retinal degeneration.Introduction
In eukaryotes, most integral membrane proteins are synthesized, integrated into the membrane, and folded properly in the endoplasmic reticulum (ER). Molecular chaperones and folding enzymes are required for the folding of the integral membrane proteins in the ER. A comprehensive approach in yeast to identify genes required for protein folding in the ER identified the ER membrane protein complex (EMC), which comprises six subunits (Jonikas et al., 2009). Another report studying the comprehensive interaction map of ER-associated degradation (ERAD) machinery revealed that EMC contains four and three additional subunits in mammals and Drosophila, respectively (Christianson et al., 2011). The deletions of each emc1–6 gene causes the unfolded protein response (UPR), presumably caused by the accumulation of misfolded proteins (Jonikas et al., 2009). Meanwhile, a recent study showed that EMC also facilitates lipid transfer from ER to mitochondria (Lahiri et al., 2014).
In photoreceptors, the massive biosynthesis of rhodopsin demands chaperones in the ER. In the vertebrate retina, rhodopsin interacts with the ER degradation enhancing α-mannosidase-like 1 (EDEM1) protein and a DnaJ/Hsp40 chaperone (HSJ1B) (Chapple and Cheetham, 2003; Kosmaoglou et al., 2009). Meanwhile, in Drosophila photoreceptors, rhodopsin 1 (Rh1) sequentially interacts with chaperones calnexin99A (Cnx), NinaA, and Xport before exiting from the ER (Colley et al., 1991; Rosenbaum et al., 2006, 2011). Defects in rhodopsin biosynthesis and trafficking cause retinal degeneration in both Drosophila and humans; more than 120 mutations in the rhodopsin gene are associated with human retinitis pigmentosa (Mendes et al., 2005; Xiong and Bellen, 2013). The overwhelming majority of these mutations lead to misfolded rhodopsin, which aggregates in the secretory pathway (Hartong et al., 2006). Thus, it is important to understand the mechanisms underlying the folding and trafficking of rhodopsin as well as retinal degeneration caused by misfolded rhodopsin.
In zebrafish the partial optokinetic response b (pob)a1 mutant exhibits red cone photoreceptor degeneration (Brockerhoff et al., 1997; Taylor et al., 2005). The localization of overexpressed zebrafish Pob protein in cultured cells in the early secretory pathway including the ER and Golgi body indicates that Pob is involved in red cone rhodopsin transport (Taylor et al., 2005). The zebrafish pob gene is the homolog of a subunit of EMC, EMC3. Here we report the function of dPob, Drosophila pob homolog, on Rh1 maturation, photoreceptor maintenance, and expression of other multi-pass membrane proteins.
ResultsdPob is essential for maturation and transport of Rh1
Retinal mosaic screening using the FLP/FRT method and two-color fluorescent live imaging was used to identify the genes essential for Rh1 maturation and transport (Satoh et al., 2013). For selected lines exhibiting defects in Rh1 accumulation in the live imaging screening, the immunocytochemical distribution of Rh1 was investigated to evaluate the phenotype with respect to transport and morphogenesis (Table 2, Satoh et al., 2013). Among them, CG6750e02662 (Kyoto stock number: 114504) exhibits severe Arrestin2::GFP and Rh1 reduction in rhabdomeres (Figure 1A,C) with normal ommatidial organization. CG6750e02662 has an insertion of a piggyBac transposon right downstream of the stop codon of CG6750 (Figure 1B). The phenotype was reverted by the precise excision of the piggyBac transposon or transgenically-expressed CG6750 (data not shown); this indicates Rh1 reduction is caused by reduced CG6750 gene function. CG6750 shares 65% identity and 82% similarity with zebrafish pob and 27% identity and 44% similarity with yeast EMC3. Because CG6750 is likely to be the homolog of zebrafish pob, we designated CG6750 as ‘dPob’ and analyzed its functions in Rh1 transport and retinal morphogenesis.10.7554/eLife.06306.003
Identification of CG6750 as an essential gene for rhodopsin 1 (Rh1) biosynthesis.
(A) Observation of fluorescent protein localizations in CG6750e02662 mosaic retinas by the water immersion technique. RFP (red) indicates wild-type photoreceptors (R1–R8). Arrestin2::GFP (green) shows endogenous Rh1 localization in R1–R6 peripheral photoreceptors. (B) Schematic drawing of CG6750 and insertion/deletion mutants. The dPob-null mutant allele, dPob∆4, was created by the recombination of two FRTs on dPobf07762 and dPobCB−0279−3 using an FRT/FLP-based deletion method. (C, D) Immunostaining of dPobe02662 (C) and dPob∆4 (D) retinas expressing RFP as a wild-type cell marker (magenta) by anti-Rh1 antibody (green). Asterisks show mutant cells. Scale bar: 5 μm (A, C, D).
DOI:http://dx.doi.org/10.7554/eLife.06306.003
To address the possibility that the severe reduction of Rh1 protein in dPobe02662 mutant is caused by the reduction of mRNA, Rh1 mRNA was quantified in whole-eye clones of the mutant. When compared with control FRT40A whole-eye clone, relative mRNA levels normalized to Act5C were, Rh1: 0.51 (n = 4, S.D. = 0.24); trp: 0.31 (n = 4, S.D. = 0.17); and Arr2: 0.49 (n = 4, S.D. = 0.24). Thus, the great reduction of the Rh1 protein level in dPobe02662 clones could not be interpreted by the reduction of mRNA.
As expected from the position of the insertion, dPob was still weakly expressed in dPobe02662 homozygous photoreceptors (Figure 2B,C), so it was classified as a hypomorphic allele. To further investigate the function of dPob, dPob∆4, a null mutant allele lacking the entire coding sequence of dPob, was created using an FRT/FLP-based deletion method (Figure 1B) (Parks et al., 2004). Unlike dPobe02662, which gives escapers up to the late pupal stage, dPob∆4 flies were lethal in the first instar larval stage. Immunostaining of dPob∆4 mosaic retinas shows a great reduction of Rh1 in dPob∆4 homozygous photoreceptors, similar to dPobe02662 homozygous photoreceptors (Figure 1D).10.7554/eLife.06306.004
Construction of antisera against dPob.
(A) Immunoblotting of wild-type (+/+) and dPobe02662 homozygous (−/−) extracts from whole larvae using antiserum against dPob N- and C-terminal polypeptides. (B) Immunostaining of a dPobe02662 mosaic retina expressing RFP (red) as a wild-type cell marker (not shown) by rat anti-dPob-C1 antiserum (blue) and phalloidin (green). Asterisks show dPobe02662 homozygous photoreceptors. (C, D) Immunostaining of wild-type retinas by anti-dPob (green) and anti-NinaA (C) or anti-HDEL (D) antisera. Scale bar: 5 μm (B–D).
DOI:http://dx.doi.org/10.7554/eLife.06306.004
Next, antisera against dPob (Figure 2) were created to investigate dPob localization in fly photoreceptors. Four antisera (three against the N-terminal and one against the C-terminal) recognized a single ∼27 kD band in wild-type head homogenates by immunoblotting (Figure 2A). This band was greatly reduced in dPobe02662 homozygous head homogenates, indicating that these four antisera recognized dPob and that the molecular weight of dPob is ∼27 kD. In immunostaining dPobe02662 mosaic retinas, two of the C-terminal antisera (dPob-C1 and dPob-C3) produced similar staining patterns in the cytoplasm of wild-type cells which were reduced in dPobe02662 homozygous photoreceptors (Figure 2B and Figure 3B), indicating that these two antisera recognized dPob in tissue. Because dPob-C3 antiserum had the highest reactivity, we used it in further experiments. Anti-dPob reactivity co-localized with ER markers NinaA and HDEL (Figure 2C,D), indicating ER localization of dPob in fly photoreceptors.10.7554/eLife.06306.005
dPob stabilizes rhodopsin 1 (Rh1) apoprotein.
(A) Immunostaining of a dPob∆4 mosaic retina from a fly reared in vitamin A (VA)-deficient medium by anti-Rh1 antibody. Asterisks show dPob∆4 homozygous photoreceptors. (B–D) Immunostaining of a wild-type (B), ninaAp263(C), or dPob∆4 (D) ommatidium of flies reared in normal vitamin A-containing medium. (E) Immunostaining of a dPobe02662 mosaic retina in ninaAp263 homozygous mutant background from a fly reared in normal medium. Asterisks show dPob∆4 homozygous photoreceptors. Scale bar: 5 μm (A–E).
DOI:http://dx.doi.org/10.7554/eLife.06306.005
dPob is essential for the biosynthesis of Rh1 apoprotein
Rh1 comprises opsin (an apoprotein) and 11-cis retinal (a chromophore). Without the chromophore, newly synthesized Rh1 apoprotein accumulates in the ER as an N-glycosylated immature form (Ozaki et al., 1993). To investigate whether dPob is essential for the accumulation of Rh1 apoprotein in the ER, dPob∆4 mosaic retinas were observed in flies reared in medium lacking vitamin A, the source of the chromophore (Figure 3A). Rh1 apoprotein was greatly reduced in dPob∆4 photoreceptor cells, indicating that dPob is essential for the early stage of Rh1 biosynthesis before chromophore binding in the ER.
NinaA, the rhodopsin-specific peptidyl-prolyl-cis-trans-isomerase, is a known Rh1 chaperone. In contrast to dPob deficiency, which lacks both Rh1 apoprotein and mature Rh1 (Figure 3D), loss of NinaA causes accumulation of Rh1 apoprotein in the ER similar to that observed in the chromophore-depleted condition (Colley et al., 1991) (Figure 3C). To investigate the epistatic interaction between dPob and NinaA for Rh1 synthesis, Rh1 apoprotein was observed in the dPob∆4/ninaAp263 double mutant. Rh1 apoprotein was greatly reduced in dPob∆4/ninaAp263 double-mutant photoreceptors, similar to that in the dPob∆4 single mutant (Figure 3E). This indicates that dPob is epistatic to NinaA. Cnx is also an Rh1 chaperone and is known to be epistatic to NinaA. Rh1 apoprotein is greatly reduced in both the cnx1 mutant and cnx1/ninaAp269 double mutant (Rosenbaum et al., 2006), suggesting that dPob functions in the same stage or a stage close to that in which Cnx functions.
Other mutants with dPob-like phenotype
The null mutant of dPob shows a characteristic phenotype with no detectable protein expression of Rh1 and very weakened expression of other multiple-transmembrane domain proteins such as Na+K+-ATPase in the mosaic retina (see below). We did not find any other mutant lines with such a phenotype in the course of mosaic screening among 546 insertional mutants described previously (Satoh et al., 2013). To explore other mutants showing phenotypes similar to the dPob null mutant, we examined a collection of 233 mutant lines deficient in Rh1 accumulation in photoreceptor rhabdomeres obtained in an ongoing ethyl methanesulfonate (EMS) mutagenesis screening. The detail of the screening will be published elsewhere; at present the Rh1 accumulation mutant collection covers three chromosome arms, approximately 60% of the Drosophila melanogaster genome. Under the assumption of a Poisson distribution of the mutants on genes, the collection stochastically covers more than 80% of genes in those arms. The distribution of Rh1 and Na+K+-ATPase was examined for 55 lines of mutants on the right arm of the third chromosome, 93 lines of mutants on the right arm of the second chromosome, and 85 mutants on the left arm of the second chromosome. Among them, only two lines—665G on the right arm of the third chromosome and 008J on the right arm of the second chromosome—showed a dPob null-like phenotype in the mean distribution of Rh1 and Na+K+-ATPase in the mosaic retina (Figure 4A,C).10.7554/eLife.06306.006
Loss of rhodopsin 1 (Rh1) apoprotein in EMC1 and EMC8/9 deficiency.
Immunostaining of a EMC1655G mosaic retina (A, B) or a EMC8/9008J mosaic retina (C, D) reared in normal (A, C) and vitamin A-deficient media (B, D). Asterisks show EMC1655G or EMC8/9008J homozygous photoreceptors. RFP (red) indicates wild-type photoreceptors (R1–R8). (A, C) Na+K+-ATPase, green; Rh1, blue; RFP, red. (B, D) Rh1, green; RFP, magenta. Scale bar: 5 μm (A–D).
DOI:http://dx.doi.org/10.7554/eLife.06306.006
Meiotic recombination mapping and RFLP analysis (Berger et al., 2001) were used to map the mutations responsible for the dPob-like phenotype of 008J and 655G. Close linkage of the mutation responsible for the dPob-like phenotype of 655G indicated that the responsible gene is located close to the proximal FRT. Since CG2943 gene, the potential Drosophila homolog of EMC1, is also close to the proximal FRT, CG2943 was recognized as a candidate of the responsible gene of 655G. As expected, Df(3R)BSC747, which is lacking the CG2943 gene, failed to complement the lethality of 655G. Targeted re-sequencing in the vicinity of CG2943 revealed that 655G has a two-base deletion at 3R:3729838-3729839 which causes a frame-shift mutation of CG2943, causing185aa deletion from I730 to C-terminus adding polypeptide of RTVRGQESGKQQCLEFLASSANAPRGAPVLYTAHNS. The only membrane-spanning helix of CG2943 is lost in this frame-shift mutation.
RFLP analysis narrowed down the cytology of the responsible gene of 008J to 58D2−59D11. Whole genome re-sequencing revealed that the 008J chromosome obtained three unique mutations in the mapped region compared with the starter stock: one silent mutation on CG30274 at 2R:18714026, a missense mutation on MED23 (E329K) at 2R:18777637, and one nonsense mutation on CG3501 at 2R:18770005 which turns Q40 to a stop codon. Complementation with the deficiencies over the MED23 (BSC783, BSC784) excluded the missense mutation on MED23 from the candidate mutation responsible for the dPob-like phenotype. The amino acid sequence of CG3501 shows 38% and 39% identity to the human EMC8 and EMC9, respectively, and no other gene similar to EMC8/9 was found in the Drosophila genome. Based on these results, we identified 655G and 008J as a loss of functional mutation of EMC1 and EMC8/9 of Drosophila and named these alleles EMC1655G and EMC8/9008J.
We investigated whether EMC1 and EMC8/9 are necessary for the accumulation of Rh1 apoprotein in the ER using EMC1655G and EMC8/9008J mosaic retinas reared in medium lacking vitamin A (Figure 4B,D). Rh1 apoprotein was greatly reduced in both EMC1655G and EMC8/9008J photoreceptor cells, indicating that EMC1 and EMC8/9 are also essential for the early stage of Rh1 biosynthesis, like dPob.
EMC1 binds to dPob and Cnx
To investigate if EMCs form a complex and bind to Rh1 apoprotein, we performed a co-immunoprecipitation assay (Figure 5). Since C-terminally tagged dPob protein did not predominantly localize to the ER in vivo (data not shown), GFP-tagged EMC1 protein (EMC1::GFP) was used as the bait. A protein-trap line expressing GFP-tagged sec61alpha protein (sec61::GFP) which localizes in the ER membrane was used as a negative control. Since the overall expression level of EMC1::GFP was strong, hs-Gal4 driver was used to activate UAS:EMC1::GFP for most of the experiments. To analyze the interaction between EMC1 and Rh1 apoprotein, Rh1-Gal4 driver was also used because the expression of EMC1::GFP was stronger in the photoreceptors (data not shown). For the Rh1-Gal4 experiment, flies were reared in a medium lacking vitamin A to accumulate Rh1 apoprotein in the ER. Membrane fraction was recovered from the adult heads, the membrane proteins extracted by CHAPS from the adult head membrane fraction were bound to anti-GFP magnetic beads, and the elutions were analyzed by immunoblotting with antibodies against GFP, Rh1, dPob, and Cnx.10.7554/eLife.06306.007
Co-immunoprecipitation of EMC1::GFP with dPob and calnexin (Cnx).
Immunoblotting of precipitates with anti-GFP antibody from the head extract was prepared from Rh1-Gal4/UAS-EMC1::GFP or sec61::GFP flies reared in a vitamin A (VA)-deficient medium (left) or heat shock (hs)-Gal4/UAS-EMC1::GFP or sec61::GFP flies reared in a vitamin A-containing normal medium (right). The mature form of rhodopsin 1 (Rh1) is accumulated in the rhabdomeres in normal medium but not in vitamin A-deficient medium. Instead of the mature form, an N-glycosylated immature form of Rh1 with a larger molecular weight accumulated in the endoplasmic reticulum of flies reared in the vitamin A-deficient medium. In both input extracts prepared from Rh1-Gal4/UAS-EMC1::GFP or sec61::GFP flies there is a band with the same position as EMC1GFP; this band will be the protein cross-reacting to anti-GFP antibody.
DOI:http://dx.doi.org/10.7554/eLife.06306.007
EMC1::GFP and sec61::GFP were concentrated in the immunoprecipitated extract from flies expressing either in the photoreceptor or in the whole head. dPob was co-immunoprecipitated with EMC1::GFP much more strongly than with sec61::GFP. Cnx was also well co-immunoprecipitated with EMC1::GFP but was barely detectable with sec61::GFP. However, Rh1 was not co-immunoprecipitated with EMC1::GFP from vitamin A-deficient photoreceptors accumulating immature Rh1 apoprotein in the ER. These results indicate that dPob and EMC1 are in a complex in vivo, as shown in yeast, and Cnx can also be associated with the complex, which is consistent with the result of epistatic analysis; the stage at which dPob works on the expression of Rh1 apoprotein is close to that of Cnx. Despite the requirement for the expression of Rh1 and co-localization with immature Rh1 apoprotein in the ER, EMC1 does not stably bind to Rh1, indicating that the EMC complex is only temporarily associated with Rh1 apoprotein.
EMC/dPob is required for the expression of multi-pass membrane proteins
To investigate the substrate specificity of EMC/dPob, we investigated the expressions of secreted or transmembrane proteins other than Rh1 in dPob∆4 mosaic retinas. In dPob∆4 photoreceptors, multi-pass membrane proteins, the alpha subunit of Na+K+-ATPase (Figure 6A) and transient receptor potential (TRP) (Figure 6B), were greatly reduced and neither anti-Rh3 nor anti-Rh4 staining was detected (Figure 6C,D). On the other hand, the type I single-pass membrane proteins Crb (Figure 6B) and DE-Cad (Figure 6E) were localized normally in the stalks and adherence junctions in dPob∆4 photoreceptors. Similarly, a type II single-pass membrane protein Nrt (Figure 6G) and a type VI single-pass membrane protein Syx1A (Figure 6H) were localized normally in Golgi units and on the plasma membrane in Pob∆4 photoreceptors. Eys, a secreted protein that expands the inter-rhabdomeric space (IRS) (Husain et al., 2006; Zelhof et al., 2006), was also secreted normally in dPob∆4 ommatidia, as expected from the near-normal size of the IRS (Figure 6I). Two other type I single-pass membrane proteins expressed in retinal cone cells, Neuroglian (Nrg) and Fasiclin III (FasIII), exhibited normal localization in contact sites between cone cells and cone cell feet (Figure 6J,K). Only one type II single-pass membrane protein, the beta subunit of Na+K+-ATPase (Nrv), showed deficient expression in Pob∆4 photoreceptors (Figure 6F). As alpha and beta subunits of Na+K+-ATPase are assembled into a heterodimer within the ER and then transported to the plasma membrane, the absence of Nrv in Pob∆4 photoreceptors can be interpreted as a consequence of the lack of the multi-pass alpha subunit. These results indicate that dPob is essential for the normal biosynthesis of multi-pass membrane proteins but not for single-pass membrane proteins or secreted proteins.10.7554/eLife.06306.008
Essential role of dPob in the biosynthesis of multi-pass transmembrane proteins.
Immunostaining of a dPob∆4 mosaic retina (A–H) or a dPobe02662 mosaic retina (I). Asterisks show dPob homozygous photoreceptors. (A) Na+K+-ATPase, green; Rh1, magenta. (B) Crb, green; TRP1, magenta. (C, D) Rh3 (C) and Rh4 (D), green; RFP (wild-type cell marker), magenta. Although the boundary between dPob∆4 and wild-type cells is unclear, all green signals are attached to RFP-expressing cell bodies, indicating that mutant R7 cells do not express Rh3 (C) or Rh4 (D). (E) DE-Cad staining. (F) Nrv, the beta subunit of Na+K+-ATPase, green; dMPPE, magenta. (G) Nrt staining. (H) Syx1A staining. (I) Eys staining. (J) Nrg, blue; F-actin, red; GFP-nls (wild-type cell marker), green. (K) FasIII staining. (L) Na+K+-ATPase, green; Rh1, magenta. Scale bar: 2 μm (A, B), 10 μm (C, D), 2 μm (E–I).
DOI:http://dx.doi.org/10.7554/eLife.06306.008
EMC1655G- and EMC8/9008J-deficient photoreceptors show similar substrate specificity to dPob∆4-deficient photoreceptors (Figure 6 and Figure 7). In both mutants, accumulation of the membrane proteins with multiple transmembrane domains, Na+K+-ATPase (Figure 4A,C), Rh3, Rh4 and TRP (Figure 7A,C), on the plasma membrane are greatly reduced in the photoreceptors. However, a type I single-pass transmembrane protein, Crb, is localized intensively in the stalks in EMC1655G or EMC8/9008J mutant photoreceptors (Figure 7B,D). A type II single-pass membrane protein, Nrt, and a type VI single-pass membrane protein, Syx1A, is localized normally in Golgi units and on the plasma membrane in EMC1655G and EMC8/9008J photoreceptors, respectively (Figure 7C,F). Eys was also secreted normally and formed a near-normal size of IRS in EMC1655G or EMC8/9008J mutant ommatidia (Figure 7B,D). Similar to Pob∆4 photoreceptors, a type II single-pass membrane protein, the beta subunit of Na+K+-ATPase (Nrv) was not detected in the plasma membrane of EMC1655G or EMC8/9008J photoreceptors (data not shown).10.7554/eLife.06306.009
Essential role of EMC1 and EMC8/9 in the biosynthesis of multi-pass transmembrane proteins.
Immunostaining of a EMC1655G mosaic retina (A, B, C) or a EMC8/9008J mosaic retina (D, E, F). (A, D) Left: Rh3, middle: Rh4, right: TRP in green, RFP in magenda. (B, E) Eys in green, Crb in blue, and RFP, wild-type cell marker in red. (C, F) Left: dMPPE, middle: Nrt, right: Syx1A in green, RFP in magenda. Scale bar: 10 μm (left and middle in A, D), 5 μm (right in A, D), 5 μm (B, C, E, F).
DOI:http://dx.doi.org/10.7554/eLife.06306.009
We observed the expression of dMPPE (Cao et al., 2011), a Golgi luminal metallophosphoesterase, anchored by a type II transmembrane helix in the N-terminal region and another transmembrane helix in the C-terminal region. dMPPE was expressed normally in Pob∆4, EMC1655G, and EMC8/9008J mutant photoreceptors (Figures 6F, 7C,F). As two transmembrane helices of dMPPE are separated from each other by the enzymatic domain, these two helices might not associate but behave as two separate transmembrane helices. The EMC requirement for proteins with two transmembrane helices therefore remains unclear.
ER membrane amplification in dPob-deficient photoreceptors
Electron microscopic observation of thin sections of late pupal flies showed massive amplification of the ER membrane in both dPobe02662 and dPob∆4 photoreceptors (Figure 8A–C) despite the substantial reduction in immature Rh1 apoprotein. In dPobe02662 photoreceptors the ER maintains its sheet structures: the number and length of the sheets was greatly increased but their lumens were almost normal with slight swelling and the sheets were aligned at a regular distance. Meanwhile, in dPob∆4 photoreceptors the ER sheet structures were no longer maintained and the cytoplasmic space was filled with ER membrane with a larger luminal space. Golgi bodies were also swollen and dilated, and sometimes vesiculated (Figure 8A–C, insets). Moreover, concordant with the reduction in Rh1, the rhabdomeres in dPob mutant photoreceptors were quite small and thin but the adherence junctions and basolateral membrane exhibited normal morphology. ER membrane amplification and rhabdomere membrane reduction therefore represent the most prominent phenotype in dPob-deficient photoreceptors.10.7554/eLife.06306.010
Endoplasmic reticulum membrane amplification and unfolded protein response (UPR) induced in <italic>dPob</italic><sup><italic>∆4</italic></sup> photoreceptor.
(A–C) Electron microscopy of late pupal photoreceptors: wild-type (A), dPobe02662 (B), and dPob∆4 photoreceptors (C). Arrow indicate adherens junctions. Insets show Golgi bodies. (D, E) Immunostaining of a dPobe02662 mosaic retina. dPob is shown in green and KDEL (D) or HDEL (E) are shown in magenta. Asterisks show dPob∆4 homozygous photoreceptors. Scale bar: 1 μm (A–C), 5 μm (D, E).
DOI:http://dx.doi.org/10.7554/eLife.06306.010
The massive amplification of the ER membrane in both dPobe02662 and dPob∆4 photoreceptors prompted us to quantify the amounts of residual ER proteins using anti-KDEL and HDEL antibodies. KDEL and HDEL sequences are signals for ER retention, and Drosophila ER resident chaperones including Hsp70–3 and PDI contain these sequences (Bobinnec et al., 2003; Ryoo et al., 2007). Corresponding to ER membrane amplification, anti-HDEL and anti-KDEL staining were greatly increased in dPob-deficient photoreceptors (Figure 8D,E).
Upregulated unfolded protein responses in dPob-deficient photoreceptors
Accumulation of unfolded proteins in the ER invokes the UPR, which includes activation of the transcription of chaperones and related genes, suppression of translation and enhanced degradation of unfolded protein. The UPR is regulated by some unique intracellular signal transduction pathways. Therefore, mutants lacking the function of a gene essential for folding or degradation of unfolded protein probably exhibit UPR. In fact, the yeast Pob homolog, EMC3, was identified by screening of mutants exhibiting upregulated UPR. ER amplification and chaperone induction, which we observed in dPob-deficient photoreceptors, are also common outcomes of UPR. We therefore examined whether UPR is induced in dPob-deficient photoreceptors. First we used the Xbp1:GFP sensor, which is an established method for detecting UPRs in flies (Ryoo et al., 2007). During UPR, Ire1 catalyzes an unconventional splicing of a small intron from the xbp1 mRNA, enabling translation into an active transcription factor (Yoshida et al., 2001). Using this mechanism, Xbp1:GFP sensor, a fused transcript of Drosophila Xbp1 and GFP translated only after the unconventional splicing by Ire1, can be used as a reporter of one of the UPR transduction pathways (Ryoo et al., 2007). In both dPob∆4 and dPobe02662 mutant mosaic retinas expressed Xbp1:GFP sensor in all R1−6 photoreceptors, and Xbp1:GFP fusion proteins were detected in the dPob mutant photoreceptors but not in the wild-type (Figure 9A and data not shown). Next, we examined the level of eukaryotic translation Initiation Factor 2α (eIF2α) phosphorylation because UPR is well known to induce eIF2α phosphorylation to attenuate protein translation on the ER membrane in a transduction pathway independent from IreI/Xbp1 (Ron and Walter, 2007; Cao and Kaufman, 2012). Anti-phospho-eIF2α signals were stronger in both dPob∆4 and dPobe02662 photoreceptors than in wild-type photoreceptors (Figure 9B and data not shown). These results indicate that UPR is induced in the dPob-deficient photoreceptors, similar to EMC mutant.10.7554/eLife.06306.011
Unfolded protein response (UPR) induced in dPob<sup>∆4</sup> photoreceptor.
(A) Projection image from the Z-series section with a 1 μm interval of dPob∆4 mosaic retina expressing RFP (magenta) as a wild-type cell marker and Xbp1:GFP as a UPR sensor. The Xbp1:GFP signal (green) is enhanced by immunostaining using anti-GFP antibody. Asterisks show dPob∆4 homozygous photoreceptors. (B) Immunostaining of a dPob∆4 mosaic retina expressing RFP (magenta) as a wild-type cell marker. Phosphorylated eukaryotic translation Initiation Factor 2α is shown in green. Asterisks show dPob∆4 homozygous photoreceptors.
DOI:http://dx.doi.org/10.7554/eLife.06306.011
Rhabdomere development and degeneration in <italic>dPob</italic> null mutant
Because the synthesis of many membrane proteins was affected in dPob mutant cells, we observed the phenotype of dPob mutant throughout the developmental processes of photoreceptors. Despite the lack of many membrane proteins, ommatidial formation was not affected in dPob∆4 photoreceptors in mosaic retina; adherence junctions formed normally (Figure 6E) and the apical membrane was well differentiated into stalks and rhabdomeres (identified with Crb and phosphorylated moesin, respectively) (Figure 6B and data not shown) (Karagiosis and Ready, 2004). The IRS was formed normally and rhabdomeres were still separated by IRSs (Figure 8A–C). We observed dPob∆4 mosaic retinas at 58% and 75% pupal development (pd) by electron microscopy (Figure 10A,B). The wild-type photoreceptors at 58% pd had already begun to amplify the rhabdomere membranes. The rhabdomeres were shorter in dPob∆4 photoreceptors than in wild-type photoreceptors, but the difference in their appearance was subtle at this stage. Until 75% pd, the microvilli of wild-type rhabdomeres were ∼0.5 μm long and packed tightly. However, the microvilli of dPob∆4 rhabdomeres at 73% pd retained almost the same length and appearance as those at 58% pd, which is the same as the dPob∆4 rhabdomeres of the late pupal retina (Figures 10A,B and 8C). ER membrane expansion and dilation were already apparent at 58% pd. These results indicate that dPob does not inhibit overall photoreceptor development and morphogenesis but does affect microvilli elongation and rhabdomere formation.10.7554/eLife.06306.012
Development and degeneration of <italic>dPob</italic><sup><italic>∆4</italic></sup> photoreceptor rhabdomeres.
Electron microscopy of pupal and adult dPob∆4 mosaic retinas. Asterisks show dPob∆4 homozygous photoreceptors. Scale bar: 1 μm. (A, B) dPob∆4 mosaic ommatidia from 58% pupal development (A) and 73% pupal development (B) under constant light (L) condition. (C–F) dPob∆4 mosaic ommatidia from flies reared in complete darkness (D) (C, E) or under 12 hr light/12 hr dark conditions (D, F). Ommatidia from 3-day-old (C, D) and 17-day-old (E, F) flies. (D, inset) dPob∆4 R5 photoreceptor rhabdomere at higher magnification.
DOI:http://dx.doi.org/10.7554/eLife.06306.012
Because zebrafish pob was identified as the responsible gene of poba1 mutant which exhibits red cone photoreceptor degeneration (Brockerhoff et al., 1997; Taylor et al., 2005), we investigated photoreceptor degeneration of the dPob null mutant. Three-day-old dPob∆4 mosaic retinas from flies reared under dark or 12 hr light/12 hr dark cycles were observed by electron microscopy (Figure 10C,D). In both conditions the rhabdomeres of dPob∆4 photoreceptors invaginated into the cytoplasm, indicating that dPob-deficient rhabdomeres undergo retinal degeneration in a light-independent manner, like Rh1 null mutants (Kumar and Ready, 1995). No microvilli or invaginations were observed in 17-day-old dPob∆4 mosaic retinas, suggesting most invaginated microvilli had degraded before day 17 (Figure 10E,F). Such rhabdomere degeneration was observed not only in R1–6 peripheral photoreceptors but also in R7 central photoreceptors. Therefore, dPob is an essential protein for maintenance of retinal structure, similar to the zebrafish pob gene.
Discussion
The present study shows that dPob, the Drosophila homolog of a subunit of EMC, EMC3, localizes in the ER and is essential for Rh1 accumulation of the rhabdomeres. The deficiency of each of two other EMC subunits, EMC1 and EMC8/9, also shows absence of Rh1 on the rhabdomeres. Mammalian EMC8 and EMC9 were identified together with EMC7 and EMC10 by high-content proteomics strategy (Christianson et al., 2011). Unlike EMC1−6 subunits, EMC8 and EMC9 do not have a transmembrane helix or signal peptide and no experimental data have been reported to show the functions of these subunits. We observed that Drosophila EMC8/9-deficient cells lack accumulation of Rh1 apoprotein in the ER and impaired biosynthesis of the multi-pass transmembrane proteins. These phenotypes in EMC8/9 deficiency are indistinguishable from those in dPob and EMC1 mutant cells, suggesting that EMC8/9 work together with EMC1 and dPob. This is the first functional study of the additional subunits of EMC, which are lacking in yeast.
We found that null mutants of EMC subunits are defective in expressing the multi-pass transmembrane proteins rhodopsins, TRP, and the alpha subunit of Na+K+-ATPase, which have seven, six, and eight transmembrane helices, respectively. In contrast, the EMC null mutants adequately express type I, type II, or type IV single-pass membrane proteins. Our observation on the substrate specificity of EMC is mostly consistent with previous reports. Jonikas et al. (2009) found that EMC mutants and a strain overexpressing a misfolded transmembrane protein, sec61-2p or KWS, had a similar genetic interaction pattern and suggested that EMC works as a chaperone for transmembrane proteins. A recent study in Caenorhabditis elegans using a hypomorphic EMC6 allele and RNAi knock-down of emc1–6 genes showed results partially consistent with our study; at least two pentameric Cys-loop receptors, AcR and GABAA, consisting of subunits with four transmembrane helices, were significantly decreased in the hypomorphic EMC6 mutants but GLR-1, a tetrameric AMPA-like glutamate receptor with four transmembrane helices and a type I single-pass transmembrane EGF receptor, was not affected (Richard et al., 2013). Despite its four transmembrane helices, GLR-1 was normally expressed in the hypomorphic emc6 mutant of the nematode; however, these results may indicate that the residual activity of EMC was sufficient for the expression of GLR-1. The degree of requirement of EMC activity can vary for each membrane protein. In fact, in a dPob hypomorphic allele, dPobe02662, near-normal expression of Na+K+-ATPase was detected (Figure 6I) despite a severe reduction in a dPob null allele, dPob∆4. Overall, the results observed in the dPob null mutant does not conflict with previous studies but rather clarifies the role of EMC in the biosynthesis of multi-pass transmembrane proteins. Because of the limited availability of antibodies, we could not show a clear threshold for the number of transmembrane helices in the substrates for EMC activity. In total, the data presented to date indicate that EMC affects the expression of membrane proteins with four or more transmembrane helices.
Co-immunoprecipitation of dPob/EMC3 and Cnx by EMC1 indicates that EMC components and Cnx can form a complex. The photoreceptors of an amorphic mutant of Cnx show complete loss of Rh1 apoprotein (Rosenbaum et al., 2006), just as shown in dPob, EMC1 or EMC8/9 mutants. Moreover, both Cnx and EMC3 are epistatic to the mutant of the rhodopsin-specific chaperone, NinaA, which accumulates Rh1 apoprotein in the ER. These results indicate that EMC and Cnx can work together in the Rh1 biosynthetic cascade prior to NinaA. Cnx, the most studied chaperone of N-glycosylated membrane proteins, recognizes improperly folded proteins and facilitates folding and quality control of glycoproteins through the calnexin cycle, which prevents ER export of misfolded proteins (Williams, 2006). One possible explanation for our result is that the EMC-Cnx complex is required for multi-pass membrane proteins to be incorporated into the calnexin cycle. If the EMC-Cnx complex is a chaperone of Rh1, physical interaction is expected between ER-accumulated Rh1 apoprotein and the EMC-Cnx complex. Indeed, it is reported that Cnx is co-immunoprecipitated with Drosophila Rh1 (Rosenbaum et al., 2006). However, in this study, Rh1 apoprotein accumulated in the chromophore-depleted photoreceptor cells was not co-immunoprecipitated with EMC1. Thus, even if EMC is a Rh1 chaperone, our result indicates that EMC is unlikely to be working in the calnexin cycle or acting as a buffer of properly folded Rh1 apoprotein ready to bind the chromophore 11-cis retinal.
In addition to preventing the export of immature protein by the calnexin cycle, Cnx is also known to recognize the nascent polypeptides co-translationally (Chen et al., 1995). The dual role of Cnx might explain the observations that both dPob/EMC3 and Cnx are epistatic to another ER resident chaperone, NinaA, whereas Cnx but not the EMC-Cnx complex binds to Rh1. These results imply that the EMC-Cnx complex is more likely to be involved in the earlier processes such as membrane integration or co-translational folding than in the folding of fully translated membrane-integrated Rh1 apoprotein.
In spite of the absence of Rh1 apoprotein, UPR is much more upregulated in the EMC3 null mutant than in the NinaA null mutant which accumulates Rh1 apoprotein in the ER. The elevated UPR without accumulation of Rh1 apoprotein in the dPob mutant photoreceptor can be explained either by the quick degradation of Rh1 apoprotein or by accumulation of the single-pass membrane proteins abandoned by the multi-pass binding partner.
Newly synthesized secreted proteins co-translationally translocate across the membrane through the translocons Sec61 in eukaryotic ER or SecYEG in the plasma membrane of bacteria. The translocons also mediate integration of the transmembrane helix of the integral membrane protein into the lipid bilayer (Park and Rapoport, 2012). In bacteria, mitochondria and chloroplasts, YidC/Oxa1/Alb3 proteins specifically facilitate insertion, folding, and assembly of many transmembrane proteins (Wang and Dalbey, 2011). In the ER membrane of eukaryotes, in addition to the translocon, other components such as translocon-associated protein/signal sequence receptor (TRAP/SSR) complex and translocating chain-associating membrane protein (TRAM) complex are required for the membrane insertion of the transmembrane helix. Most of the newly synthesized multi-pass membrane proteins are co-translationally integrated into the ER membrane through the translocon complex. Although the mechanism of this process is yet to be fully understood, it is assumed that only one or two transmembrane helices can be stored in the translocon channel and the lateral gate and that the next set of newly synthesized transmembrane helices displace them (Rapoport et al., 2004; Cymer et al., 2014). In the case of nascent chain of bovine rhodopsin, translocon associates with transmembrane helices sequentially, and TRAM temporarily associates with the second transmembrane helix (Ismail et al., 2008). EMC may be involved in these co-translational membrane integration or co-translational folding processes.
Zebrafish pob was identified as the responsible gene of poba1 mutant, which exhibits red cone photoreceptor degeneration (Brockerhoff et al., 1997; Taylor et al., 2005). Because only red cone photoreceptors degenerated in zebrafish poba1 mutant, pob is postulated as a gene with a red cone-specific function. However, the identification of the poba1 mutation as hypomorphic together with pob expression in all photoreceptors, as well as its localization in the early secretory pathway, suggests that Pob has a general function rather than being red cone-specific (Taylor et al., 2005). We found that dPob-deficient rhabdomeres undergo retinal degeneration in a light-independent manner, like Rh1 null mutants (Kumar and Ready, 1995). Rhabdomere degeneration was observed not only in R1–6 peripheral photoreceptors but also in R7 central photoreceptors. Our results indicate that dPob is an essential protein for the maintenance of retinal structure, similar to the zebrafish pob gene.
Materials and methods<italic>Drosophila</italic> stocks and genetics
Flies were reared at 20–25°C in 12 hr light/12 hr dark cycles and fed standard cornmeal/glucose/agar/yeast food unless noted otherwise. Vitamin A-deficient food contained 1% agar, 10% dry yeast, 10% sucrose, 0.02% cholesterol, 0.5% propionate, and 0.05% methyl 4-hydroxybenzoate.
UAS-Xbp1::GFP was a gift from H Ryoo at New York University and other Drosophila stocks obtained from Bloomington Stock Center (BL) or the Kyoto Drosophila Genetic Resource Center (KY) are referred to with their respective sources and stock numbers.
dPob deletion mutants were made using a standard induced FLP/FRT recombination method (Parks et al., 2004). Trans-heterozygous PBac(WH)f07762 (BL19109) and P (RS3)CB−0279−3 (KY123106) males carrying hs-FLP (BL6876) were heat treated three times at 37°C for 1 hr at larval stages. SM6a-balanced offspring were genotyped using PCR to select the recombinant carrying both the proximal side of PBac(WH)f07762 and the distal side of P (RS3)CB−0279−3 with the following primers: 5′-CTCCTTGCCAGCTTCTGC-3′ and 5′-TCGCTGTCTCACTCAGACTCA-3′ for P (RS3)CB−0279−3, and 5′–CCACCGAAGAGGCCTACTATT-3′ and 5′-TCCAAGCGGCGACTGAGATG-3′ for PBac(WH)f07762.
Transgenic flies for UAS-dPob, UAS-EMC1::GFP
The entire coding region of the dPob gene was amplified from a cDNA clone LD37839 (DGRC: Drosophila Genomics Resource Center, Bloomington, IN, USA) and cloned into pTW (DGRC) to construct pP{UAST-dPob}. To construct pP{UAST-EMC1::GFP}, the entire coding region of CG2943 except the stop codon was amplified from a cDNA clone LD19064 (DGRC) and cloned into pTWG (DGRC). Plasmids were injected into embryos by BestGene Inc. (Chino Hills, CA, USA) to generate transgenic lines.
Live imaging of fluorescent proteins expressed in photoreceptors
Fluorescent proteins expressed in photoreceptors were imaged by water-immersion technique.
y w ey-FLP;CG6750e02662FRT40A/ CyO y+ (KY114504) was mated with w;P3RFP FRT40A/SM1;Rh1-Arrestin2::GFP eye-FLP/TM6B (Satoh et al., 2013). Late pupae of the siblings with GFP-positive RFP mosaic retina were attached to the slide glass using double-sided sticky tape and the pupal cases around the heads were removed. The pupae were chilled on ice, embedded in 0.5% agarose, and observed using an FV1000 confocal microscope equipped with a LUMPlanFI water-immersion 40× objective (Olympus, Tokyo, Japan). Arrestin2::GFP specifically binds to activated rhodopsin (Satoh et al., 2010). Rh1 was activated by a 477 nm solid-state laser to bind Arr2:GFP and GFP. The wild-type marker P3RFP is DsRed gene under the control of three Pax3 binding sites and labels photoreceptors (Bischof et al., 2007).
EMS mutagenesis and screening
The precise method of screening, whole genome re-sequencing, will be described elsewhere. Briefly, second or third chromosomes carrying P-element vector with FRT on 40A, 42D, or 82B (Berger et al., 2001) were isogenized and used as the starter strains. EMS was fed to males in a basic protocol (Bökel, 2008) and mosaic retinas were generated on F1 or F2. The estimated number of lethal mutations introduced per chromosome arm was 0.8–1.8. The mutants were screened based on the distribution of Arr2-GFP by confocal live imaging under water-immersion lens using 3xP3-RFP as the wild-type marker, as previously described for the screening of insertional mutants (Satoh et al., 2013).
Mapping and determination of mutations
Meiotic recombination mapping was carried out by the standard method (Bökel, 2008). Briefly, to allow meiotic recombination between the proximal FRT, the phenotype-responsible mutation and a distal miniature w+ marker, flies carrying isogenized chromosome of 008J and 655G were crossed with flies with isogenized P{EP755} and P{EP381} which carry miniature-w+ marker, respectively. Female offspring carrying the mutated chromosome and the miniature-w+-marked chromosome were crossed with males carrying FRT42D, P3RFP, and Rh1Arr2GFP. The resulting adult offspring with w+ mosaic, which means maternally inherited both FRT and w+, were observed using live imaging to judge whether the mutation responsible for the dPob-like phenotype had been inherited. The recovered flies were individually digested in 50 µl of 200 ng/µl Proteinase K in 10 mM Tris-Cl (pH 8.2), 1 mM EDTA, and 25 mM NaCl at 55°C for 1 hr and heat inactivated at 85°C for 30 min and at 95°C for 5 min. 0.5 µl of the digested solution were used as the template of PCR amplification for RFLP analysis according to the method described in the FlySNP database (Chen et al., 2008; http://flysnp.imp.ac.at/index.php). The mutation responsible for the dPob-like phenotype of 008J was mapped between SNP markers 1417 and 1518 defined in the FlySNP database.
Whole-genome and targeted re-sequence of EMS-generated mutants
For the whole genome re-sequencing of the 008J mutant, the second chromosome was balanced over a balancer, CyO, P{Dfd-GMR-nvYFP}(Bloomington stock number 23230) to facilitate the isolation of homozygous embryo. Using REPLI-G single cell kit (QIAGEN, Hilden, Germany), the genomic DNA was amplified from two 008J homozygous embryos independently. A sequencing library was prepared using Nextera DNA sample preparation kit (Illumina, San Diego, CA, USA) for each embryo and 2 × 250 bp reads were obtained using MiSeq v2 kit (Illumina). Reads were mapped to release five of the Drosophila melanogaster genome using BWA 0.7.5a. The RFLP-mapped region of 008J was covered by reads with an average depth of 23.2× and width of 99.5%. Mapped reads were processed using picard-tools 1.99 and Genome Analysis Tool Kit 2.7-2 (GATK, Broad Institute, Cambridge, MA, USA). SNVs and Indels were called using Haplotypecaller in GATK. SNVs and Indels were subtracted by the ones of the isogenized starter stock to extract the unique variants in 008J and annotated using SnpSift (Cingolani, 2012). The point mutation on 2R:18770005 was verified by capillary sequencing of PCR-amplified fragment using 5′ GTCGCGGTCACACTTTCTAG 3′ and 5′ CTGCAGCGTCATCAGTTTGT 3′ as primers.
For targeted re-sequencing of 655G, a region including CG2943 was amplified from a heterozygous fly of the 655G mutant chromosome and the starter chromosome using KOD FX Neo DNA polymerase and 5′ TTTTGTTCTTGTTGGGCGACTCCTTTTCCGTCTC 3′ and 5′ AGGCTGTGTCTTTGTTGTTTTGGCGTTGTCGTC 3′ as primers. Reads covering the CG2943 gene region at a depth of 2213–6436 were obtained using MiSeq and mapped, as described above. The sequence was confirmed by capillary sequencing and PCR using 5′ GCAAGAATCCCATCGAGCAT 3′ and 5′ CCTTCTTCACGTCCCTGAGT 3′ as primers.
Antisera against dPob and CNX99a
Fragments of cDNA encoding V28-D104 (dPob-N) or G173-S247 (dPob-C1) of dPob were amplified from a cDNA clone, LD37839 (Drosophila Genomics Resource Center, Bloomington, IN, USA) and cloned into pDONR-211 using Gateway BP Clonase II and then into pET-161 expression vector using Gateway LR Clonase II (Life Technologies, Carlsbad, CA, USA). The fusion proteins with 6xHis-tag were expressed in BL21-Star (DE3) (Life Technologies) and purified using Ni-NTA Agarose (QIAGEN). To obtain antisera, rabbits were immunized six times with 300 µg dPob-N fusion protein (Operon, Tokyo, Japan) and three rats were immunized six times with 125 µg dPob-C1 fusion protein (Biogate, Gifu, Japan). Antisera against Drosophila Cnx were raised by immunizing a rabbit four times with 400 to 200 µg of synthetic peptide corresponding to C-terminal 24 amino acids of Cnx99a protein conjugated to KLH (Sigma Aldrich Japan, Tokyo, Japan).
Immunoblotting
Immunoblotting was performed as described previously (Satoh et al., 1997). The antibodies used were as follows: rabbit anti-dPob–N-terminal (dPob-N) (1:2000 concentrated supernatant) (made by the authors of this paper), three rat anti-dPob–C-terminal antibodies (dPob-C1-3) (1:2000 concentrated supernatant) (made by the authors of this paper) as primary antibodies. HRP-conjugated anti-rat or anti-rabbit IgG antibody (1:20,000, Life Technologies) was used as a secondary antibody. For co-immunoprecipitation, 1:2000 rabbit anti-dPob-N, 1:2000 rabbit anti-Cnx99A, 1:2000 rabbit anti-GFP (Life Technologies), mouse anti-Rh1 monoclonal antibody 4C5, and detected by biotinylated secondary antibodies followed by HRP-conjugated avidin. Signals were visualized using enhanced chemiluminescence (Clality Western blotting ECL Substrate; BioRad, Hercules, CA, USA) and imaged using ChemiDoc XRS+ (BioRad).
Immunohistochemistry
Fixation and staining were performed as described previously (Satoh and Ready, 2005). The primary antisera were as follows: rabbit anti-Rh1 (1:1000) (Satoh et al., 2005), chicken anti-Rh1 (1:1000) (Satoh et al., 2013), mouse monoclonal anti-HDEL (1:100) (Santa Cruz Biotechnology, Dallas, TX, USA), mouse monoclonal anti-KDEL (1:100) (Assay Designs, Ann Arbor, MI, USA), rabbit anti-NinaA (1:300) (gift from Dr Zuker, Colombia University), mouse monoclonal anti-Na+K+-ATPase α subunit (1:500 ascite) (DSHB, Iowa City, IA, USA), rat monoclonal anti-DE-cad (1:20 supernatant) (DSHB), mouse monoclonal anti-Syx1A (1:20 supernatant) (DSHB), mouse monoclonal anti-Nrt (1:20 supernatant) (DSHB), mouse monoclonal anti-Nrv (1:20 supernatant) (DSHB), mouse monoclonal anti-FasIII (1:20 supernatant) (DSHB), mouse monoclonal anti-Nrg (1:20 supernatant) (DSHB), mouse monoclonal anti-Chp (24B10) (1:20 supernatant) (DSHB), rat anti-Crb (gift from Dr Tepass, University of Toronto), rabbit anti-TRP (gift from Dr Montell, Johns Hopkins University), rabbit anti-dMPPE (1:50) (gift from Dr Han, Southeast University), and rabbit anti-phosphorylated eIF2α (1:300) (Cell Signaling Technologies, Danvers, MA, USA). The secondary antibodies used were anti-mouse, rabbit, rat, and chicken IgG labeled with Alexa Fluor 488, 568, and 647 (1:300) (Life Technologies) and Cy2 (1: 300) (GE Healthcare Life Sciences, Pittsburgh, PA, USA). Samples were examined and images recorded using a FV1000 confocal microscope (60×, 1.42-NA lens; Olympus, Tokyo, Japan). To minimize bleed-through, each signal in double- or triple-stained samples was imaged sequentially. Images were processed in accordance with the guidelines for proper digital image handling using ImageJ and/or Adobe Photoshop CS3.
Co-immunoprecipitation analysis of EMC complex
The EMC1 gene was cloned into a P-element vector pTWG using the Gateway System (Life Technologies) to express EMC1 protein-tagged GFP on the C-terminus under control of upstream activation sequence (UAS). Transgenic lines were generated by the BestGene Inc. (Chino Hills, CA, USA). UAST-EMC1-GFP(1M), a line carrying the transgene on the second chromosome, was crossed to Rh1-Gal4 line to express EMC1-GFP in the photoreceptor or to hs-Gal4 line to express EMC1-GFP in the whole body. A protein-trap line, Sec61alpha [ZCL0488] which constitutively expresses GFP-tagged Sec61alpha protein, was used as a control. To accumulate rhodopsin in the ER, flies were reared in the vitamin A-deficient medium in a Rh1-driven experiment. For heat-shock driven expression, newly eclosed adult fly flies were incubated at 37°C for 45 min a day before preparation. Within 0–1 days after eclosion, flies were frozen with liquid nitrogen and stored at −80°C. The heads were collected by sieving in liquid nitrogen, ground to powder and homogenized in buffer (50 mM Tris-Cl, 500 mM NaCl, pH 7.5) containing 1:200 Protein inhibitor cocktail VI (Calbiochem, San Diego, CA, UAS) using BioMasher II (Wako Pure Chemical, Osaka, Japan) with motor drive. Debris was removed by centrifugation at 950×g for 5 min and the membrane was precipitated by centrifugation at 21,500×g for 15 min. Approximately 30 µl of membrane pellet were solubilized by 130 µl of 1% CHAPS and placed on ice for 1 hr, and the insoluble membrane was removed by centrifugation at 21,500×g for 30 min. The extract was diluted fivefold by the buffer and 50 µl of Anti-GFP-Magnetic beads (MBL, Nagoya, Japan) were added and mixed by mild rotation for 18 hr. The magnetic beads were rinsed with 2× 100 μl of 0.1% CHAPS in buffer and the bound protein was extracted by incubation in 20 µl SDS-PAGE Sampling Buffer (BioRad) for 5 min at room temperature and an equal amount of Sampling Buffer with 2-mercaptoethanol was then added. The extracts were heat denatured for 5 min at 37°C. SDS-PAGE and immunoblotting was performed as described above.
Electron microscopy
Electron microscopy was performed as described previously (Satoh et al., 1997). Samples were observed on a JEM1200 or JEM1400 electron microscope (JEOL, Tokyo, Japan).
Quantification of relative expression of mRNA of Rh1, TRP, and Arr2 normalized by Act5C
Whole-eye mutant clones were generated using the FRT/GMR-hid method (Stowers and Schwarz, 1999). Both eyes were dissected from two adult flies per sample and cDNA was reverse-transcribed using SuperPrep Cell Lysis and RT Kit for qPCR (Toyobo, Osaka, Japan) according to the manufacturer’s instructions. Eyes with whole-eye clones of FRT40A were used as a control to obtain the relative standard curves. qPCR reactions were performed using the StepOne real-time PCR system (Life Technologies) and KOD SYBR qPCR Mix (Toyobo, Osaka, Japan), according to the manufacturers’ instructions. PCR condition was 98°C for 2 min, followed by 40 cycles at 98°C for 15 s, 55°C for 15 s, and 68°C for 45 s, and a melt curve stage of 95°C for 30 s, 60°C for 1 min, and 0.3°C/s increments to 98°C, with primers of Rh1: (ninaE-qF1:5′-GTGGACACCATACCTGGTC-3′ and ninaE-qR1:5′-GCGATATTTCGGATGGCTG-3′), Arr2: (Arr2-qF1:5′-AAGGATCGCCATGGTATCG-3′ and Arr2-qR1:5′-TACGAGATGACAATACCACAGG-3′), TRP: (Trp-qF2:5′-GAATACACGGAGATGCGTC-3′ and Trp-qF2:5′-CTCGAGTTCCATGGATGTG-3′), Act5C: (5′-GCTTGTCTGGGCAAGAGGAT-3′ and 5′-CTGGAACCACACAACATGCG-3′). The relative expression levels were normalized by Act5C.
Acknowledgements
We thank Drs U Tepass, C Montell, C Zuker, H Ryoo, and J Han who kindly provided fly stocks and reagents. We also thank the Bloomington Stock Center and the Drosophila Genetic Resource Center of the Kyoto Institute of Technology for fly stocks. This study was supported by grants from the Naito Foundation (25-040920), the Novartis Foundation (25-050421), the Hayashi Memorial Foundation for Female Natural Scientists (25-051022), PRESTO (25-J-J4215), and KAKENHI (21687005, 21113510, and 23113712) to ASK. This study was also supported by grants from the Global Centers of Excellence Program ‘Advanced Systems-Biology: Designing The Biological Function’ from the Japanese Ministry of Education, Culture, Sports, Science, and Technology. Whole genome and targeted re-sequencing was carried out at the Analysis Center of Life Science, Natural Science Center for Basic Research and Development, Hiroshima University.
Additional informationCompeting interests
The authors declare that no competing interests exist.
Author contributions
TS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
AKS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article
AO, Acquisition of data, Drafting or revising the article
ZL, Acquisition of data, Drafting or revising the article
TI, Acquisition of data, Drafting or revising the article
ReferencesBergerJSuzukiTSentiKAStubbsJSchaffnerGDicksonBJ2001Genetic mapping with SNP markers in DrosophilaNature Genetics2947548110.1038/ng773BischofJMaedaRKHedigerMKarchFBaslerK2007An optimized transgenesis system for Drosophila using germ-line-specific phiC31 integrasesProceedings of the National Academy of Sciences of USA1043312331710.1073/pnas.0611511104BobinnecYMarcaillouCMorinXDebecA2003Dynamics of the endoplasmic reticulum during early development of Drosophila melanogasterCell Motility and the Cytoskeleton5421722510.1002/cm.10094BrockerhoffSEHurleyJBNiemiGADowlingJE1997A new form of inherited red-blindness identified in zebrafishThe Journal of Neuroscience1742364242BökelC2008EMS screens: from mutagenesis to screening and mappingMethods in Molecular Biology42011913810.1007/978-1-59745-583-1_7CaoJLiYXiaWReddigKHuWXieWLiHSHanJ2011A Drosophila metallophosphoesterase mediates deglycosylation of rhodopsinThe EMBO Journal303701371310.1038/emboj.2011.254CaoSSKaufmanRJ2012Unfolded protein responseCurrent Biology22R622R62610.1016/j.cub.2012.07.004ChappleJPCheethamME2003The chaperone environment at the cytoplasmic face of the endoplasmic reticulum can modulate rhodopsin processing and inclusion formationThe Journal of Biological Chemistry278190871909410.1074/jbc.M212349200ChenDAhlfordASchnorrerFKalchhauserIFellnerMViràghEKissISyvänenACDicksonBJ2008High-resolution, high-throughput SNP mapping in Drosophila melanogasterNature Methods532332910.1038/nmeth.1191ChenWHeleniusJBraakmanIHeleniusA1995Cotranslational folding and calnexin binding during glycoprotein synthesisProceedings of the National Academy of Sciences of USA926229623310.1073/pnas.92.14.6229ChristiansonJCOlzmannJAShalerTASowaMEBennettEJRichterCMTylerREGreenblattEJHarperJWKopitoRR2011Defining human ERAD networks through an integrative mapping strategyNature Cell Biology149310510.1038/ncb2383CingolaniPPatelVMCoonMNguyenTLandSJRudenDMLuX2012Using Drosophila melanogaster as a model for genotoxic chemical mutational studies with a new program, SnpSiftFrontiers in Genetics33510.3389/fgene.2012.00035ColleyNJBakerEKStamnesMAZukerCS1991The cyclophilin homolog ninaA is required in the secretory pathwayCell6725526310.1016/0092-8674(91)90177-ZCymerFvon HeijneGWhiteSH2014Mechanisms of integral membrane protein insertion and foldingJournal of Molecular Biology10.1016/j.jmb.2014.09.014HartongDTBersonELDryjaTP2006Retinitis pigmentosaLancet3681795180910.1016/S0140-6736(06)69740-7HusainNPellikkaMHongHKlimentovaTChoeKMClandininTRTepassU2006The agrin/perlecan-related protein eyes shut is essential for epithelial lumen formation in the Drosophila retinaDevelopmental Cell1148349310.1016/j.devcel.2006.08.012IsmailNCrawshawSGCrossBCHaagsmaACHighS2008Specific transmembrane segments are selectively delayed at the ER translocon during opsin biogenesisThe Biochemical Journal41149550610.1042/BJ20071597JonikasMCCollinsSRDenicVOhEQuanEMSchmidVWeibezahnJSchwappachBWalterPWeissmanJSSchuldinerM2009Comprehensive characterization of genes required for protein folding in the endoplasmic reticulumScience3231693169710.1126/science.1167983KaragiosisSAReadyDF2004Moesin contributes an essential structural role in Drosophila photoreceptor morphogenesisDevelopment13172573210.1242/dev.00976KosmaoglouMKanugaNAguilàMGarrigaPCheethamME2009A dual role for EDEM1 in the processing of rod opsinJournal of Cell Science1224465447210.1242/jcs.055228KumarJPReadyDF1995Rhodopsin plays an essential structural role in Drosophila photoreceptor developmentDevelopment12143594370LahiriSChaoJTTavassoliSWongAKChoudharyVYoungBPLoewenCJPrinzWA2014A conserved endoplasmic reticulum membrane protein complex (EMC) facilitates phospholipid transfer from the ER to mitochondriaPLOS Biology12e100196910.1371/journal.pbio.1001969MendesHFvan der SpuyJChappleJPCheethamME2005Mechanisms of cell death in rhodopsin retinitis pigmentosa: implications for therapyTrends in Molecular Medicine1117718510.1016/j.molmed.2005.02.007OzakiKNagataniHOzakiMTokunagaF1993Maturation of major Drosophila rhodopsin, ninaE, requires chromophore 3-hydroxyretinalNeuron101113111910.1016/0896-6273(93)90059-ZParkERapoportTA2012Mechanisms of Sec61/SecY-mediated protein translocation across membranesAnnual Review of Biophysics41214010.1146/annurev-biophys-050511-102312ParksALCookKRBelvinMDompeNAFawcettRHuppertKTanLRWinterCGBogartKPDealJEDeal-HerrMEGrantDMarcinkoMMiyazakiWYRobertsonSShawKJTabiosMVysotskaiaVZhaoLAndradeRSEdgarKAHowieEKillpackKMilashBNortonAThaoDWhittakerKWinnerMAFriedmanLMargolisJSingerMAKopczynskiCCurtisDKaufmanTCPlowmanGDDuykGFrancis-LangHL2004Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster genomeNature Genetics3628829210.1038/ng1312RapoportTAGoderVHeinrichSUMatlackKE2004Membrane-protein integration and the role of the translocation channelTrends in Cell Biology1456857510.1016/j.tcb.2004.09.002RichardMBoulinTRobertVJRichmondJEBessereauJL2013Biosynthesis of ionotropic acetylcholine receptors requires the evolutionarily conserved ER membrane complexProceedings of the National Academy of Sciences of USA110E1055E106310.1073/pnas.1216154110RonDWalterP2007Signal integration in the endoplasmic reticulum unfolded protein responseNature Reviews Molecular Cell Biology851952910.1038/nrm2199RosenbaumEEBrehmKSVasiljevicELiuCHHardieRCColleyNJ2011XPORT-dependent transport of TRP and rhodopsinNeuron7260261510.1016/j.neuron.2011.09.016RosenbaumEEHardieRCColleyNJ2006Calnexin is essential for rhodopsin maturation, Ca2+ regulation, and photoreceptor cell survivalNeuron4922924110.1016/j.neuron.2005.12.011RyooHDDomingosPMKangMJStellerH2007Unfolded protein response in a Drosophila model for retinal degenerationThe EMBO Journal2624225210.1038/sj.emboj.7601477SatohATokunagaFKawamuraSOzakiK1997In situ inhibition of vesicle transport and protein processing in the dominant negative Rab1 mutant of DrosophilaJournal of Cell Science11029432953SatohAKO'TousaJEOzakiKReadyDF2005Rab11 mediates post-Golgi trafficking of rhodopsin to the photosensitive apical membrane of Drosophila photoreceptorsDevelopment1321487149710.1242/dev.01704SatohAKReadyDF2005Arrestin1 mediates light-dependent rhodopsin endocytosis and cell survivalCurrent Biology151722173310.1016/j.cub.2005.08.064SatohAKXiaHYanLLiuCHHardieRCReadyDF2010Arrestin translocation is stoichiometric to rhodopsin isomerization and accelerated by phototransduction in Drosophila photoreceptorsNeuron67997100810.1016/j.neuron.2010.08.024SatohTInagakiTLiuZWatanabeRSatohAK2013GPI biosynthesis is essential for rhodopsin sorting at the trans-Golgi network in Drosophila photoreceptorsDevelopment14038539410.1242/dev.083683StowersRSSchwarzTL1999A genetic method for generating Drosophila eyes composed exclusively of mitotic clones of a single genotypeGenetics15216311639TaylorMRKikkawaSDiez-JuanARamamurthyVKawakamiKCarmelietPBrockerhoffSE2005The zebrafish pob gene encodes a novel protein required for survival of red cone photoreceptor cellsGenetics17026327310.1534/genetics.104.036434WangPDalbeyRE2011Inserting membrane proteins: the YidC/Oxa1/Alb3 machinery in bacteria, mitochondria, and chloroplastsBiochimica et Biophysica Acta180886687510.1016/j.bbamem.2010.08.014WilliamsDB2006Beyond lectins: the calnexin/calreticulin chaperone system of the endoplasmic reticulumJournal of Cell Science11961562310.1242/jcs.02856XiongBBellenHJ2013Rhodopsin homeostasis and retinal degeneration: lessons from the flyTrends in Neurosciences3665266010.1016/j.tins.2013.08.003YoshidaHMatsuiTYamamotoAOkadaTMoriK2001XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factorCell10788189110.1016/S0092-8674(01)00611-0ZelhofACHardyRWBeckerAZukerCS2006Transforming the architecture of compound eyesNature44369669910.1038/nature0512810.7554/eLife.06306.013Decision letterRonDavidReviewing editorUniversity of Cambridge, United Kingdom
eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see review process). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.
Congratulations: we are very pleased to inform you that your article, “dPob/EMC is essential for biosynthesis of rhodopsin and other multi-pass membrane proteins in Drosophila photoreceptors”, has been accepted for publication in eLife, subject to revisions of grammar and text. We also hope you would give due consideration to the minor comments of reviewers 1 and 3. The Reviewing Editor for your submission was David Ron.
Reviewer #1
In Figure 5, right panel it is not clear why the sec61::GFP input shows a band the size of the EMC1::GFP; what is the nature of this band?
Equally why is the +/- vitamin A experiment performed with different driver lines? wouldn't it be better to do both with the Rh1-Gal4 driver? Please give the rationale of using the two drivers.
Reviewer #3
1) The authors show that EMC1::GFP does not co-immuno-precipitate with Rh1. Does calnexin co-IP with Rh1? I ask because the authors conclude (in the Discussion) that “ EMC is unlikely to be working in the calnexin cycle”, and it is difficult for me to understand why the authors have come to that conclusion.
2) The authors report negative results with the ERAD component mutants, EDEM1, EDEM2 and VCP/+. I don't see the point of showing such negative data, as one cannot draw any conclusions. Perhaps the individual EDEM mutants do not show Rh1 phenotype as they are redundant (in mammals, EDEMs are known to be genetically redundant). Also, VCP's effect was assessed in a heterozygous condition, and a negative result here simply means that there is no dominant genetic effect. How about analyzing homozygous clones? Alternatively, since ERAD is not a major point of this work, the authors may want to take out this part, as it does not contribute to the overall story.
[Editors’ note: a previous version of this study was rejected after peer review, but the authors submitted for reconsideration. The previous decision letter after peer review is shown below.]
Thank you for choosing to send your work entitled “dPob is essential for rhodopsin maturation in Drosophila photoreceptors” for consideration at eLife. Your full submission has been evaluated by Randy Schekman (Senior editor) and 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors. Between them, the three reviewers had expertise in Drosophila genetics, the study of assembly and maturation of membrane transporters and ER protein folding homeostasis. Their decision was reached after discussions between the reviewers.
The reviewers noted the significance of your discovery of a role for a Drosophila member of the EMC complex in the maturation of rhodopsin. Your discovery that dPob/EMC3 acts upstream of ninaA in rhodopsin maturation was specifically credited with supporting a role for this member of the EMC complex in the early steps of the maturation of a multi-pass membrane protein. However, the expert reviewers were unanimous in their view that these finding do not add up to an advance of sufficient measure to merit publication in eLife.
It is possible that a different manuscript, for example one with further details on genetic lesions in other Drosophila EMC homologs or one that carried the biochemical analysis of the dPob/EMC3 lesion further would had received a warmer welcome at eLife. Unfortunately, the manuscript before us is too far off this mark to suggest specific experiments that would render it suitable for this journal.
Thus we are left with no choice but to return the manuscript to you in the hope that you will find the reviewers’ comments (appended below) of use.
Reviewer #1
A mosaic screen for mutations in (otherwise essential) genes required for rhodopsin trafficking and maturation in the fly eye led the authors to dPod, in whom homozygosity for a hypomorphic allele compromised Rh1 expression in the rhabdomeres.
dPod is a fly homolog of a zebra fish gene mutations in which compromise color vision and of EMC3 a yeast gene whose compromise activates the unfolded protein response and whose encoded protein forms a stable complex with several other proteins that collectively constitute the ER Membrane Complex (or EMC).
Since its identification in 2009, the EMC has been postulated to play a role in the biogenesis of transmembrane proteins. This paper provides several pieces of information that support that notion.
The most important findings pertain to:
1) The failure of Rh1 to accumulate in the ER of dPod mutant flies that are deprived of vitamin A.
2) The epistatic relationship between dPod and ninaA, wherein the lack of dPod preempts the accumulation of Rh1 normally obsereved in ninaA mutants.
3) The selectivity of dPod mutations in compromising the trafficking of multi-pass transmembrane proteins, whilst preserving that of secreted and single pass proteins.
Together these findings point to site of dPod action early during the biogenesis of multi-pass transmembrane proteins.
Other findings, such as the altered ER morphology in mutant tissue and the activation of the UPR are less informative and more anticipated by the yeast work.
Thus the crucial issue for the reviewers is to establish the strength of the experimental data and, importantly the degree to which it advances our understanding of the EMC's biological role. If the application of this Drosophila genetic system has provided strong evidence favoring a selective role for this EMC component in early biogenesis of multi-pass transmembrane proteins, then the paper would be of interest to a broad readership regardless of the degree to which these findings were anticipated by prior speculation. If, however, the additional experimental data derived from the Drosophila system were deemed merely incremental, the paper would be more suited for a specialist journal.
Reviewer #2
In this manuscript, Satoh and colleagues report a genetic analysis of dPob, a Drosophila homolog of zebrafish Pob and yeast EMC4. The data presented here supports the idea that this gene is involved in the maturation of rhodopsins and other multipass transmembrane domain proteins in the endoplasmic reticulum. Consistently, the loss of dPob leads to the activation of the Unfolded Protein Response, and a defect in rhabdomere development.
EMC genes have been identified through a large-scale yeast genetic interaction screen, but their precise physiological roles had remained unclear. A couple of reports, based on studies in zebrafish and C. elegans, indicate that they are involved in membrane protein maturation. Although the authors have done a solid job in their characterization of dPob function in Drosophila, it appears to be an extension of the C. elegans work (Richard et al., 2013), but with more analysis of potential substrates. Moreover, this study does not address many of the pressing questions regarding the EMC complex. Does dPob's role in Drosophila reflect the entire EMC complex function? What is the mechanistic basis of the specificity of dPob towards multispan membrane proteins? Based on these grounds, whether the overall novelty and scope of this study justifies the publication of eLife can be subject to debate. Below are a few specific comments along these lines.
1) The authors had identified dPob through a previously reported genetic screen (Satoh et al., 2013). Were there other EMC homologs identified in the screen? If it is the case, the authors should highlight those results, as it would indicate that the observed effect is not an isolated function of dPob.
2) EMC genes are grouped because of their genetic interactions with each other in yeast. This study is potentially significant because the function of EMCs still remain poorly understood. Naturally, one must ask whether other EMC gene homologs show similar effects in rhodopsin maturation or not. The authors can examine this through in vivo RNAi, or better, by use of available mutant alleles.
3) Does dPob physically interact with rhodopsin and other multipass membrane proteins? Immuno-precipitation experiments may be useful.
4) Is there any mechanistic insight as to how dPob would specifically recognize multipass membrane proteins? Although the authors show the dPob locus in Figure 1, they do not describe any domain structures in the gene. The fact that dPob specifically affects transmembrane domain is an intriguing phenomenon, but a mechanistic insight is lacking.
5) Does dPob physically interact with rhodopsins and other targets? A simple immune-precipitation experiment should provide answers.
Reviewer #3
The manuscript by Satoh et al. adds dPob, a subunit of the Drosophila EMC complex, to the list of factors required for rhodopsin biogenesis, an important process with biomedical relevance. The authors show that in the absence of dPob several rhodopsins do not mature and their steady-state levels are strongly decreased. Besides rhodopsins, two other polytopic membrane proteins, Na, K-ATPase alpha subunit and the TRP channel, were affected while four proteins with a single transmembrane segment and one secreted protein were not affected. These are interesting observations that do indeed suggest a role of dPob in the early biogenesis of polytopic membrane proteins.
However, I have major doubts whether the set of experiments adds much to our understanding of the specific function of dPob. Chaperone activity and protein folding are mentioned all over the manuscript but neither are ever addressed directly. Instead, steady-state levels of proteins and induction of the UPR are used as proxies to infer the folding status of the putative substrates. Both phenomena could be rather indirectly caused by lack of dPob function. In my opinion the manuscript does not achieve more than narrowing down which proteins are affected by loss of dPob function in a non-systematic fashion. The numbers of substrates in the affected and unaffected classes are low (furthermore, 'with a single membrane-spanning domain' is not an accurate topological classification for a membrane protein). Potentially, the distinctions between affected and unaffected proteins could fall into completely different categories: half-life of the precursor, activity of the precursor, trafficking machinery used by the precursor to leave the ER, activity of the precursor at the ER to name just a few speculative ideas. Therefore, the last sentences of the Conclusion section are strongly over-stated.
This work is solid and interesting. It presents a good set of figures. I don't perceive the manuscript as outstanding because the actual function of dPob is inferred from correlations with highly complex steady-state effects on selected individual proteins or on global ER homeostasis. From an outstanding paper I would expect either a systematic and unbiased approach to identifying the substrates of the putative chaperone complex or a delineation of the features that it recognizes in its substrates or some direct evidence for the actual chaperone activity proposed.
10.7554/eLife.06306.014Author response
We made some changes regarding the reviewers’ comments.
Reviewer #1
InFigure 5, right panel it is not clear why the sec61::GFP input shows a band the size of the EMC1::GFP; what is the nature of this band?
We think this band indicated a protein cross-reacting to anti-GFP antibody. We added the following sentence to the figure legend.
In both input extracts prepared from Rh1-Gal4/UAS-EMC1::GFP or sec61::GFP flies, there is the band with the same position to EMC1::GFP: this band will be the protein cross-reacting to anti GFP antibody
Equally why is the +/- vitamin A experiment performed with different driver lines? wouldn't it be better to do both with the Rh1-Gal4 driver? Please give the rationale of using the two drivers.
Since the overall expression level of EMC1::GFP was strong, hs-Gal4 driver was used to activate UAS:EMC1::GFP for the most of experiment. To analyze the interaction between EMC1 and Rh1-apoprotein, Rh1-Gal4 driver was also used because the expression of EMC1::GFP was stronger in the photoreceptors. We added this sentence in the result section.
Reviewer #3
1) The authors show that EMC1::GFP does not co-immuno-precipitate with Rh1. Does calnexin co-IP with Rh1? I ask because the authors conclude (in the Discussion) that “ EMC is unlikely to be working in the calnexin cycle”, and it is difficult for me to understand why the authors have come to that conclusion.
Indeed, it is reported that Cnx is co-immunoprecipitated with Drosophila Rh1 (Rosenbaum, 2006).
Our result indicated that EMC1-GFP co-IPs with Cnx. If EMC-Cnx complex work for Rh1 in calnexin-cycle, physical interaction is expected between ER-accumulated Rh1 apoprotein and EMC-Cnx complex. However, EMC1::GFP does not co-IP with Rh1, suggesting EMC-Cnx complex does not function in the calnexin-cycle but Cnx does.
In addition to the calnexin cycle, Cnx is also know to recognizes nascent polypeptides co-translationally in the ER lumen (Chen, 1995). We think it is more likely EMC-Cnx complex functions in this process.
2) The authors report negative results with the ERAD component mutants, EDEM1, EDEM2 and VCP/+. I don't see the point of showing such negative data, as one cannot draw any conclusions. Perhaps the individual EDEM mutants do not show Rh1 phenotype as they are redundant (in mammals, EDEMs are known to be genetically redundant). Also, VCP's effect was assessed in a heterozygous condition, and a negative result here simply means that there is no dominant genetic effect. How about analyzing homozygous clones? Alternatively, since ERAD is not a major point of this work, the authors may want to take out this part, as it does not contribute to the overall story.
Regarding the redundancy for EDEM, it is shown that RNAi knockdown of EDEM1 increases aggregation of P23H folding-mutant of human rhodopsin (Kosmaoglou, 2009). Drosophila has only two EDEM orthologs, EDEM1 and EDEM2 in the genome. We used mutant clones lacking dPob, EDEM1 and EDEM2, but did not see Rh1 apoprotein accumulation in the ER.
Regarding the heterozygous usage for VCP, It is shown that The degradation of Rh1 intermediate in Rh1[P37H] folding-mutant is restored by the heterozygous mutation of TER9426-8 (Griciuc, 2010).
However, we also agree with the reviewer’s opinion: these negative data does not draw clear conclusions. Therefore, we will take out this part.
[Editors’ note: the author responses to the previous round of peer review follow.]
The reviewers noted the significance of your discovery of a role for a Drosophila member of the EMC complex in the maturation of rhodopsin. Your discovery that dPob/EMC3 acts upstream of ninaA in rhodopsin maturation was specifically credited with supporting a role for this member of the EMC complex in the early steps of the maturation of a multi-pass membrane protein. However, the expert reviewers were unanimous in their view that these finding do not add up to an advance of sufficient measure to merit publication in eLife.
It is possible that a different manuscript, for example one with further details on genetic lesions in other Drosophila EMC homologs or one that carried the biochemical analysis of the dPob/EMC3 lesion further would had received a warmer welcome at eLife. Unfortunately, the manuscript before us is too far off this mark to suggest specific experiments that would render it suitable for this journal.
Thus we are left with no choice but to return the manuscript to you in the hope that you will find the reviewers’ comments (appended below) of use as you prepare it for submission elsewhere.
We are pleased to read reviewer’s constructive criticism and suggestion, and tried to answer them with sincere. Our revised manuscript includes three sets of new data, and three minor changes in figures, based on reviewer’s suggestion. Based on our new data, we slightly shifted our conclusion from the original manuscript, but mostly the same. We believe now our conclusion, “EMC is essential for biosynthesis of rhodopsin and other multi-pass membrane proteins in Drosophila photoreceptors”, becomes more solid and reliable.
Three sets of new data are following:
1) In our original manuscript, we analyzed only a subunit of EMS, dPob/EMC3. However, in this revised manuscript, in a large scale screening of EMS induced mutants deficient in Rh1 expression, we identified only two mutants with dPob-like phenotype. These two mutants were carrying loss of function mutations of EMC subunits: one is on EMC1, and the other is on EMC8/9, which yeast lacks. We analyzed the accumulation of immature Rh1 and substrate specificities in these mutants. In both loss of function mutants for EMC1 and EMC8/9, immature Rh1 fails to accumulate in ER (Figure 4), and the expressions of multi-pass transmembrane proteins, but neither a secreted nor type-I, II and IV single-pass transmembrane proteins are greatly reduced (Figure 7). The substrate specificities shown in the deficiencies of EMC1 and EMC8/9 are exactly same as that in dPob deficiency. This is the first functional study of the additional subunits of EMC, which yeast lacks.
2) The absence of Rh1 apoprotein can be explained by the degradation of misfolded Rh1 in the EMC null mutants through the accelerated ERAD pathway. Thus, we investigated if mutations on ERAD components restore immature Rh1 in ER. However, unlike in the two folding mutants of Rh1 (Kang and Ryoo, 2009; Kosmaoglou et al., 2009; Griciuc et al., 2010), disturbance of ERAD activity did not restore expression of Rh1 in dPob/EMC3-deficient photoreceptors (Figure 9C, D). Together with the epistasis over ninaA, the higher susceptibility of the Rh1 to the ERAD pathway than that of the two folding mutants of Rh1, imply that the EMC complex is more likely to be involved in the earlier processes such as membrane integration or co-translational folding than in the folding of fully translated, membrane integrated Rh1-apoprotein.
3) We performed co-IP experiment using EMC1::GFP as a bait, because dPob::GFP is not functional. We could confirm EMC1::GFP interacts with dPob, but we failed to show significant interaction between EMC1::GFP and immature Rh1 apoprotein in the ER (Figure 5). This result also supports the EMC complex is likely involved in the early biogenesis for multi-pass membrane proteins than in the folding of fully translated, membrane integrated Rh1-apoprotein.
By this co-IP experiment, we also found EMC1::GFP interacts with calnexin (Cnx). It has been reported that the photoreceptors of an amorphic mutant of Cnx show complete loss of Rh1 apoprotein (Rosenbaum et al., 2006) just as we showed in dPob, EMC1 or EMC8/9 mutant. Moreover, both Cnx and EMC are epistatic to the mutant of the rhodopsin-specific chaperon NinaA, which accumulates Rh1 apoprotein in the ER. These results indicate that EMC and Cnx can work together in Rh1 biosynthetic cascade prior to NinaA works.
Three minor changes in figures are following:
1) We add Rh1 staining of WT, ninaAp263 mutant, dPobdelta4 mutant ommatidia (Figure 3B-D).
2) This revised manuscript includes the analysis of three single-pass transmembrane proteins and one double-pass transmembrane protein, which we did not work for the original manuscript (Figure 6F-H).
3) We removed Figure 9 in the original manuscript. We just described the results in the text.
Reviewer #1
[…] Other findings, such as the altered ER morphology in mutant tissue and the activation of the UPR are less informative and more anticipated by the yeast work.
Thus the crucial issue for the reviewers is to establish the strength of the experimental data and, importantly the degree to which it advances our understanding of the EMC's biological role. If the application of this Drosophila genetic system has provided strong evidence favoring a selective role for this EMC component in early biogenesis of multi-pass transmembrane proteins, then the paper would be of interest to a broad readership regardless of the degree to which these findings were anticipated by prior speculation. If, however, the additional experimental data derived from the Drosophila system were deemed merely incremental, the paper would be more suited for a specialist journal.
In this revised version of our paper, we investigated not only dPob/EMC3 deficiency, but also the deficiencies for other two subunits of EMC (see the answer for Reviewer#2 in detail). These two mutants were identified in a large scale screening of EMS-induced mutant deficient in Rh1 expression. Among 233 lines of Rh1-expression mutants, only two of them showed dPob-like phenotype (loss of Rh1 and NaK-ATPase but normal Eys expression), and these two were loss of function mutant of EMC subunits. Importantly, the deficiencies for three subunits of EMC gave the exactly same substrate specificity: they are essential for all of multi-pass transmembrane proteins, but not for a secreted protein or type-I, II and IV single-pass transmembrane proteins. Coincidence of the phenotypes in the deficiencies for three subunits of EMC provides the strong evidence favoring a selective role for this EMC component in early biogenesis of multi-pass transmembrane proteins.
Reviewer #2
[…] Although the authors have done a solid job in their characterization of dPob function in Drosophila, it appears to be an extension of the C. elegans work (Richard, 2013), but with more analysis of potential substrates. Moreover, this study does not address many of the pressing questions regarding the EMC complex. Does dPob's role in Drosophila reflect the entire EMC complex function? What is the mechanistic basis of the specificity of dPob towards multispan membrane proteins? Based on these grounds, whether the overall novelty and scope of this study justifies the publication of eLife can be subject to debate. Below are a few specific comments along these lines.
1) The authors had identified dPob through a previously reported genetic screen (Satoh, 2013). Were there other EMC homologs identified in the screen? If it is the case, the authors should highlight those results, as it would indicate that the observed effect is not an isolated function of dPob.
The previous screening among FRT-combined transposon insertion lines included only dPob/EMC3 among EMC subunits. The null mutant of dPob shows quite characteristic phenotype; no detectable protein expression of Rh1, and very weakened expression of other multiple-transmembrane domain proteins such as Na+K+-ATPase or TRP in the mosaic retina. We did not find any other mutant lines with such phenotype in the course of the mosaic screening among 546 insertional mutants described previously (Satoh et al., 2013). To explorer other mutants showing phenotypes similar to dPob null mutant, we examined a collection of 233 mutant lines deficient of Rh1 accumulation in photoreceptor rhabdomeres, obtained in an ongoing ethyl methanesulfonate (EMS) mutagenesis screening. Among them, only two lines, 665G and 008J showed dPob-like phenotype in the mean of distribution of Rh1 and Na+K+-ATPase in the mosaic retina. 665G and 008J turned out to be frame-shift and nonsense mutations of EMC1 and EMC8/9, respectively. Thus, we included the results of analysis for these mutations to our revised manuscript.
2) EMC genes are grouped because of their genetic interactions with each other in yeast. This study is potentially significant because the function of EMCs still remain poorly understood. Naturally, one must ask whether other EMC gene homologs show similar effects in rhodopsin maturation or not. The authors can examine this through in vivo RNAi, or better, by use of available mutant alleles.
We found the null mutants of EMC1 and EMC7/8 show the phenotype which is characteristic to the dPob null mutant. Mutants of other subunits of EMC were not available. We did not perform RNAi experiment of other EMCs because the phenotype expected in RNAi experiment will be hypomorphic, and we knew the hypomorphic mutant of dPob shows just reduced Rh1 expression. Previously, we have tried RNAi screening of genes required for the Rh1 expression, and found out that too many genes show reduced Rh1 expression when derived by GMR-Gal4, probably because of the off-target effect, and none of them were phenocopied by null allele of the genes. Contrary, when RNAi was derived by Rh1-Gal4, the genes known to be required for Rh1 transport, showed no phenotype. Altogether, in Drosophila photoreceptor, we think RNAi experiments will not provide phenotype specific enough to conclude something.
3) Does dPob physically interact with rhodopsin and other multipass membrane proteins? Immuno-precipitation experiments may be useful.
We performed co-IP experiment using EMC1::GFP as a bait, because dPob::GFP did not localize to the ER. We could confirm EMC1::GFP interacts with dPob, but we failed to show the stable interaction between EMC1::GFP and Rh1 apoprotein accumulated in the ER in the VA-condition.
In addition, we found the mutations of ERAD components fail to restore Rh1 immature in dPob mutant cells. From these two results with epistasis over ninaA, we now think EMC complex is more likely to be involved in the earlier processes such as membrane integration or co-translational folding than in the folding of fully-translated, membrane-integrated Rh1-apoprotein.
Interestingly, we found Calnexin (Cnx) binds to EMC1::GFP in the co-IP experiment. The photoreceptors of an amorphic mutant of Cnx show complete loss of Rh1 apoprotein (Rosenbaum, 2006) just as shown in dPob, EMC1 or EMC8/9 mutant. Moreover, both Cnx and EMC are epistatic to the mutant of the rhodopsin-specific chaperon NinaA, which accumulates Rh1 apoprotein in the ER. These results indicate that EMC and Cnx can work together in Rh1 biosynthetic cascade prior to NinaA works.
4) Is there any mechanistic insight as to how dPob would specifically recognize multipass membrane proteins? Although the authors show the dPob locus inFigure 1, they do not describe any domain structures in the gene. The fact that dPob specifically affects transmembrane domain is an intriguing phenomenon, but a mechanistic insight is lacking.
We now think EMC complex is more likely to be involved in the earlier processes such as membrane integration or co-translational folding, than in the folding of fully-translated, membrane-integrated Rh1-apoprotein (see above for the reason of this).
Most of the newly synthesized multi-pass membrane proteins are co-translationally integrated into the ER membrane through the translocon complex, and only one or two transmembrane helices can be stored in the translocon channel and the lateral gate. The helices associated with the translocon are displaced by the next set of newly synthesized transmembrane helices in the membrane-integration of multi-pass membrane proteins (Cymer, 2014; Rapoport, 2004). We assume that EMC may bind and stabilize the displaced helices, facilities the displacement, or involves to the integration, however, only sophisticated biochemistry with in vitro translation system can address the issue. We are applying for funding to start it, though we do not have enough resource to do it now.
5) Does dPob physically interact with rhodopsins and other targets? A simple immune-precipitation experiment should provide answers.
We answered to this question in point 3 above.
Reviewer #3
[…] This work is solid and interesting. It presents a good set of figures. I don't perceive the manuscript as outstanding because the actual function of dPob is inferred from correlations with highly complex steady-state effects on selected individual proteins or on global ER homeostasis. From an outstanding paper I would expect either a systematic and unbiased approach to identifying the substrates of the putative chaperone complex or a delineation of the features that it recognizes in its substrates or some direct evidence for the actual chaperone activity proposed.
Because of the limitation of our model system, we could not perform a systematic approach to identifying the substrates. Instead, we looked more substrates and we used more accurate topological classification for membrane proteins.
Now we do not think chaperon is the only possible function of EMC. We showed that weakening ERAD pathway does not suppress the loss of Rh1 by slowing down ERAD-dependent degradation. We also performed Co-IP with EMC1::GFP but did not detect stable binding with ER-accumulated Rh1 apoprotein. Our new results indicate it is more possible that EMC has some function on earlier steps than folding of fully translated protein, such as stabilizing translation, co-translational folding or membrane integration.