This repository has been archived by the owner on Mar 2, 2018. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 7
/
elife01008.xml
1 lines (1 loc) · 226 KB
/
elife01008.xml
1
<?xml version="1.0" encoding="UTF-8"?><!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.1d1 20130915//EN" "JATS-archivearticle1.dtd"><article article-type="research-article" dtd-version="1.1d1" xmlns:xlink="http://www.w3.org/1999/xlink"><front><journal-meta><journal-id journal-id-type="nlm-ta">elife</journal-id><journal-id journal-id-type="hwp">eLife</journal-id><journal-id journal-id-type="publisher-id">eLife</journal-id><journal-title-group><journal-title>eLife</journal-title></journal-title-group><issn publication-format="electronic">2050-084X</issn><publisher><publisher-name>eLife Sciences Publications, Ltd</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">01008</article-id><article-id pub-id-type="doi">10.7554/eLife.01008</article-id><article-categories><subj-group subj-group-type="display-channel"><subject>Research article</subject></subj-group><subj-group subj-group-type="heading"><subject>Biochemistry</subject></subj-group><subj-group subj-group-type="heading"><subject>Biophysics and structural biology</subject></subj-group></article-categories><title-group><article-title>Three-color single molecule imaging shows WASP detachment from Arp2/3 complex triggers actin filament branch formation</article-title></title-group><contrib-group><contrib contrib-type="author" equal-contrib="yes" id="author-5567"><name><surname>Smith</surname><given-names>Benjamin A</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="fn" rid="equal-contrib">†</xref><xref ref-type="fn" rid="pa1">‡</xref><xref ref-type="fn" rid="con1"/><xref ref-type="fn" rid="conf4"/></contrib><contrib contrib-type="author" equal-contrib="yes" id="author-5694"><name><surname>Padrick</surname><given-names>Shae B</given-names></name><xref ref-type="aff" rid="aff2"/><xref ref-type="fn" rid="equal-contrib">†</xref><xref ref-type="fn" rid="con2"/><xref ref-type="fn" rid="conf4"/></contrib><contrib contrib-type="author" id="author-5695"><name><surname>Doolittle</surname><given-names>Lynda K</given-names></name><xref ref-type="aff" rid="aff2"/><xref ref-type="fn" rid="con3"/><xref ref-type="fn" rid="conf4"/></contrib><contrib contrib-type="author" id="author-5696"><name><surname>Daugherty-Clarke</surname><given-names>Karen</given-names></name><xref ref-type="aff" rid="aff3"/><xref ref-type="aff" rid="aff4"/><xref ref-type="other" rid="par-4"/><xref ref-type="fn" rid="con4"/><xref ref-type="fn" rid="conf4"/></contrib><contrib contrib-type="author" id="author-5697"><name><surname>Corrêa</surname><given-names>Ivan R</given-names><suffix>Jr</suffix></name><xref ref-type="aff" rid="aff5"/><xref ref-type="fn" rid="con6"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5698"><name><surname>Xu</surname><given-names>Ming-Qun</given-names></name><xref ref-type="aff" rid="aff5"/><xref ref-type="fn" rid="con5"/><xref ref-type="fn" rid="conf2"/></contrib><contrib contrib-type="author" id="author-1093"><name><surname>Goode</surname><given-names>Bruce L</given-names></name><xref ref-type="aff" rid="aff3"/><xref ref-type="aff" rid="aff4"/><xref ref-type="other" rid="par-4"/><xref ref-type="other" rid="par-6"/><xref ref-type="fn" rid="con7"/><xref ref-type="fn" rid="conf3"/></contrib><contrib contrib-type="author" corresp="yes" id="author-4358"><name><surname>Rosen</surname><given-names>Michael K</given-names></name><xref ref-type="aff" rid="aff2"/><xref ref-type="corresp" rid="cor1">*</xref><xref ref-type="other" rid="par-2"/><xref ref-type="other" rid="par-3"/><xref ref-type="other" rid="par-5"/><xref ref-type="fn" rid="con8"/><xref ref-type="fn" rid="conf4"/></contrib><contrib contrib-type="author" corresp="yes" id="author-5568"><name><surname>Gelles</surname><given-names>Jeff</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="corresp" rid="cor2">*</xref><xref ref-type="other" rid="par-1"/><xref ref-type="other" rid="par-7"/><xref ref-type="fn" rid="con9"/><xref ref-type="fn" rid="conf4"/></contrib><aff id="aff1"><institution content-type="dept">Department of Biochemistry</institution>, <institution>Brandeis University</institution>, <addr-line><named-content content-type="city">Waltham</named-content></addr-line>, <country>United States</country></aff><aff id="aff2"><institution content-type="dept">Department of Biophysics</institution>, <institution>Howard Hughes Medical Institute, University of Texas Southwestern Medical Center</institution>, <addr-line><named-content content-type="city">Dallas</named-content></addr-line>, <country>United States</country></aff><aff id="aff3"><institution content-type="dept">Rosenstiel Basic Medical Sciences Research Center</institution>, <institution>Brandeis University</institution>, <addr-line><named-content content-type="city">Waltham</named-content></addr-line>, <country>United States</country></aff><aff id="aff4"><institution content-type="dept">Department of Biology</institution>, <institution>Brandeis University</institution>, <addr-line><named-content content-type="city">Waltham</named-content></addr-line>, <country>United States</country></aff><aff id="aff5"><institution>New England Biolabs</institution>, <addr-line><named-content content-type="city">Ipswich</named-content></addr-line>, <country>United States</country></aff></contrib-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Sundquist</surname><given-names>Wesley</given-names></name><role>Reviewing editor</role><aff><institution>University of Utah</institution>, <country>United States</country></aff></contrib></contrib-group><author-notes><corresp id="cor1"><label>*</label>For correspondence: <email>michael.rosen@utsouthwestern.edu</email> (MKR);</corresp><corresp id="cor2"><label>*</label>For correspondence: <email>gelles@brandeis.edu</email> (JG)</corresp><fn fn-type="con" id="equal-contrib"><label>†</label><p>These authors contributed equally to this work</p></fn><fn fn-type="present-address" id="pa1"><label>‡</label><p>Biogen Idec, Cambridge, United States</p></fn></author-notes><pub-date date-type="pub" publication-format="electronic"><day>03</day><month>09</month><year>2013</year></pub-date><pub-date pub-type="collection"><year>2013</year></pub-date><volume>2</volume><elocation-id>e01008</elocation-id><history><date date-type="received"><day>27</day><month>05</month><year>2013</year></date><date date-type="accepted"><day>29</day><month>07</month><year>2013</year></date></history><permissions><copyright-statement>© 2013, Smith et al</copyright-statement><copyright-year>2013</copyright-year><copyright-holder>Smith et al</copyright-holder><license xlink:href="http://creativecommons.org/licenses/by/3.0/"><license-p>This article is distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/3.0/">Creative Commons Attribution License</ext-link>, which permits unrestricted use and redistribution provided that the original author and source are credited.</license-p></license></permissions><self-uri content-type="pdf" xlink:href="elife01008.pdf"/><related-article ext-link-type="doi" id="ra1" related-article-type="article-reference" xlink:href="10.7554/eLife.00884"/><abstract><object-id pub-id-type="doi">10.7554/eLife.01008.001</object-id><p>During cell locomotion and endocytosis, membrane-tethered WASP proteins stimulate actin filament nucleation by the Arp2/3 complex. This process generates highly branched arrays of filaments that grow toward the membrane to which they are tethered, a conflict that seemingly would restrict filament growth. Using three-color single-molecule imaging in vitro we revealed how the dynamic associations of Arp2/3 complex with mother filament and WASP are temporally coordinated with initiation of daughter filament growth. We found that WASP proteins dissociated from filament-bound Arp2/3 complex prior to new filament growth. Further, mutations that accelerated release of WASP from filament-bound Arp2/3 complex proportionally accelerated branch formation. These data suggest that while WASP promotes formation of pre-nucleation complexes, filament growth cannot occur until it is triggered by WASP release. This provides a mechanism by which membrane-bound WASP proteins can stimulate network growth without restraining it.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.001">http://dx.doi.org/10.7554/eLife.01008.001</ext-link></p></abstract><abstract abstract-type="executive-summary"><object-id pub-id-type="doi">10.7554/eLife.01008.002</object-id><title>eLife digest</title><p>Most cells are neither perfect spheres nor amorphous blobs, but instead have characteristic shapes that enable them to carry out specific roles within tissues or organs. These shapes are established by a type of scaffolding, called the cytoskeleton, that gives structure to the cell, and also forms networks over which other proteins, and even organelles, can travel.</p><p>The filaments that make up the cytoskeleton are composed of various proteins, one of which is called actin. Cellular actin filaments can grow by adding new actin molecules, and actin filaments can also have ‘branches’ that fork out from the mother filament. Branches grow out of an assembly of seven proteins known as the Arp2/3 complex, which attaches to the side of the mother filament. Branch growth is triggered by binding to the Arp2/3 complex of an additional protein, WASP, but the sequence of events required to initiate a new branch is not well understood. In particular, WASP is bound to cell membranes; at some point it must detach from the Arp2/3 complex so that the nearness of the membrane does not interfere with the growth of branches. Now, Smith et al. uncover how branch formation is triggered, and define a new role played by WASP in this process.</p><p>It is known that a specific region of the WASP protein called the VCA domain binds to both the Arp2/3 complex and actin. Smith et al. studied how this domain could initiate branch formation, and showed that a pair of VCA domains linked to each other, along with an Arp2/3 complex, could interact jointly with an existing actin filament before a new branch formed. However, new branches did not form unless the VCA-domain pair detached from the actin filament, leaving the Arp2/3 complex behind. Additionally, Smith et al. found that mutant VCA-domain pairs detached from the actin filament at different rates, which then determined the chance that a new branch formed.</p><p>These findings—and those of Helgeson and Nolen published concurrently in <italic>eLife</italic>—suggest that, in cells, two WASP proteins first recruit the Arp2/3 complex to the membrane, and that together they interact with an existing actin filament. The WASP proteins then release the filament, and only then does the Arp2/3 complex initiate the formation of an actin branch. Since the Arp2/3 complex is no longer attached to WASP, subsequent growth of the branch is not physically limited by linkage to the membrane.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.002">http://dx.doi.org/10.7554/eLife.01008.002</ext-link></p></abstract><kwd-group kwd-group-type="author-keywords"><title>Author keywords</title><kwd>TIRF</kwd><kwd>WH2</kwd><kwd>nucleation</kwd><kwd>Wiskott-Aldrich syndrome protein</kwd><kwd>verprolin homology</kwd><kwd>activation</kwd></kwd-group><kwd-group kwd-group-type="research-organism"><title>Research organism</title><kwd>Human</kwd><kwd><italic>S. cerevisiae</italic></kwd></kwd-group><funding-group><award-group id="par-1"><funding-source><institution-wrap><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>GM43369</award-id><principal-award-recipient><name><surname>Gelles</surname><given-names>Jeff</given-names></name></principal-award-recipient></award-group><award-group id="par-2"><funding-source><institution-wrap><institution>Howard Hughes Medical Institute</institution></institution-wrap></funding-source><principal-award-recipient><name><surname>Rosen</surname><given-names>Michael K</given-names></name></principal-award-recipient></award-group><award-group id="par-3"><funding-source><institution-wrap><institution>Welch Foundation</institution></institution-wrap></funding-source><award-id>I-1544</award-id><principal-award-recipient><name><surname>Rosen</surname><given-names>Michael K</given-names></name></principal-award-recipient></award-group><award-group id="par-4"><funding-source><institution-wrap><institution>National Science Foundation</institution></institution-wrap></funding-source><award-id>MRSEC 0820492</award-id><principal-award-recipient><name><surname>Daugherty-Clarke</surname><given-names>Karen</given-names></name><name><surname>Goode</surname><given-names>Bruce L</given-names></name></principal-award-recipient></award-group><award-group id="par-5"><funding-source><institution-wrap><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>GM56322</award-id><principal-award-recipient><name><surname>Rosen</surname><given-names>Michael K</given-names></name></principal-award-recipient></award-group><award-group id="par-6"><funding-source><institution-wrap><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>GM63691</award-id><principal-award-recipient><name><surname>Goode</surname><given-names>Bruce L</given-names></name></principal-award-recipient></award-group><award-group id="par-7"><funding-source><institution-wrap><institution>National Institutes of Health</institution></institution-wrap></funding-source><award-id>GM098143</award-id><principal-award-recipient><name><surname>Gelles</surname><given-names>Jeff</given-names></name></principal-award-recipient></award-group><funding-statement>The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.</funding-statement></funding-group><custom-meta-group><custom-meta><meta-name>elife-xml-version</meta-name><meta-value>2</meta-value></custom-meta><custom-meta specific-use="meta-only"><meta-name>Author impact statement</meta-name><meta-value>New insights into the mechanism responsible for the nucleation of branches on networks of actin filaments shed light on a process that is essential for eukaryotic cell motility and endocytosis.</meta-value></custom-meta></custom-meta-group></article-meta></front><body><sec id="s1" sec-type="intro"><title>Introduction</title><p>Control of actin dynamics is essential to many cellular processes, including motility, vesicle trafficking, and cell division (<xref ref-type="bibr" rid="bib51">Pollard and Cooper, 2009</xref>). The Actin related protein 2/Actin related protein 3 (Arp2/3) complex nucleates new (<italic>daughter</italic>) filaments from the sides of existing (<italic>mother</italic>) filaments in response to activating stimulus from the Wiskott-Aldrich Syndrome Protein (WASP) family (<xref ref-type="bibr" rid="bib49">Pollard, 2007</xref>; <xref ref-type="bibr" rid="bib47">Padrick and Rosen, 2010</xref>; <xref ref-type="bibr" rid="bib7">Campellone and Welch, 2010</xref>). Membrane-associated WASP proteins integrate upstream signals and activate Arp2/3 complex in the correct place and time to produce actin structures that perform a variety of cellular functions including motility. The verprolin homology-central-acidic (VCA) domain of WASP family proteins binds to monomeric actin and the Arp2/3 complex, and is both necessary and sufficient for the WASP proteins to stimulate nucleation (<xref ref-type="bibr" rid="bib40">Machesky and Insall, 1998</xref>; <xref ref-type="bibr" rid="bib43">Miki and Takenawa, 1998</xref>; <xref ref-type="bibr" rid="bib39">Machesky et al., 1999</xref>; <xref ref-type="bibr" rid="bib54">Rohatgi et al., 1999</xref>; <xref ref-type="bibr" rid="bib49">Pollard, 2007</xref>).</p><p>VCA acts to promote daughter nucleation by Arp2/3 complex in several ways. VCA engagement promotes a conformational change in Arp2/3 complex that repositions Arp2 and Arp3 (<xref ref-type="bibr" rid="bib52">Robinson et al., 2001</xref>; <xref ref-type="bibr" rid="bib24">Goley et al., 2004</xref>; <xref ref-type="bibr" rid="bib53">Rodal et al., 2005</xref>; <xref ref-type="bibr" rid="bib55">Rouiller et al., 2008</xref>; <xref ref-type="bibr" rid="bib68">Xu et al., 2012</xref>; <xref ref-type="bibr" rid="bib27">Hetrick et al., 2013</xref>). This conformational change is thought to be required to initiate daughter filament growth. Further, the initial monomers of the nucleated filament are delivered by the V region, or WASP homology 2 (WH2) domain (<xref ref-type="bibr" rid="bib54">Rohatgi et al., 1999</xref>; <xref ref-type="bibr" rid="bib26">Hertzog et al., 2004</xref>; <xref ref-type="bibr" rid="bib31">Irobi et al., 2004</xref>; <xref ref-type="bibr" rid="bib8">Chereau et al., 2005</xref>; <xref ref-type="bibr" rid="bib45">Padrick et al., 2011</xref>). However, the Arp2/3-VCA-actin complex is not active on its own; an additional stimulus must be provided by the mother filament (<xref ref-type="bibr" rid="bib44">Mullins et al., 1998</xref>; <xref ref-type="bibr" rid="bib1">Achard et al., 2010</xref>). Thus, when stimulated by VCA the Arp2/3 complex only nucleates filaments from the sides of existing filaments. This produces branched filament arrays in vitro (<xref ref-type="bibr" rid="bib44">Mullins et al., 1998</xref>; <xref ref-type="bibr" rid="bib39">Machesky et al., 1999</xref>; <xref ref-type="bibr" rid="bib5">Blanchoin et al., 2000</xref>; <xref ref-type="bibr" rid="bib2">Amann and Pollard, 2001</xref>; <xref ref-type="bibr" rid="bib1">Achard et al., 2010</xref>), which resemble the branched networks found in cells (<xref ref-type="bibr" rid="bib61">Svitkina and Borisy, 1999</xref>; <xref ref-type="bibr" rid="bib64">Vinzenz et al., 2012</xref>). An additional feature of the system is that simultaneous binding of two VCA peptides to Arp2/3 complex greatly potentiates daughter nucleation (<xref ref-type="bibr" rid="bib45">Padrick et al., 2011</xref>). It is likely that WASP oligomerization is a broadly used mechanism of activation (<xref ref-type="bibr" rid="bib46">Padrick et al., 2008</xref>), and a number of cellular factors that dimerize or multimerize WASP have been identified (<xref ref-type="bibr" rid="bib47">Padrick and Rosen, 2010</xref>).</p><p>The molecular interactions and structural rearrangements outlined above contribute to VCA acting at more than one step in the nucleation pathway, although the pathway is not fully defined. It is established that VCA stimulates branch formation by accelerating the association of Arp2/3 complex with the mother filament (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). In addition, there is evidence from kinetic analyses that a VCA-dependent ‘activation step’ follows filament binding during nucleation (<xref ref-type="bibr" rid="bib41">Marchand et al., 2001</xref>; <xref ref-type="bibr" rid="bib70">Zalevsky et al., 2001</xref>; <xref ref-type="bibr" rid="bib4">Beltzner and Pollard, 2008</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). The nature of this step remains unknown but has been hypothesized to arise from conformational changes in the Arp2/3 complex (<xref ref-type="bibr" rid="bib41">Marchand et al., 2001</xref>; <xref ref-type="bibr" rid="bib4">Beltzner and Pollard, 2008</xref>). The activation step substantially limits the efficiency of nucleation (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>).</p><p>An interesting feature of WASP activation of daughter nucleation is that Arp2/3 complex must associate with membrane bound activators at an early stage in the process and yet be separated from those activators at a subsequent stage. In cells, branched filament networks have their barbed ends directed toward membranes (<xref ref-type="bibr" rid="bib58">Small et al., 1978</xref>; <xref ref-type="bibr" rid="bib61">Svitkina and Borisy, 1999</xref>; <xref ref-type="bibr" rid="bib50">Pollard and Borisy, 2003</xref>; <xref ref-type="bibr" rid="bib64">Vinzenz et al., 2012</xref>). The characteristic geometry of the branches nucleated by the Arp2/3 complex dictates that both mother and daughter filaments grow toward the membrane. However, WASP proteins are linked with activators on the membrane (<xref ref-type="bibr" rid="bib47">Padrick and Rosen, 2010</xref>), so that VCA-bound Arp2/3 complex should be tethered to the membrane. This tethering creates a steric problem, in that the growing ends of the filaments are held against, and possibly have their growth limited by, the membrane. However, this problem is eventually resolved (<xref ref-type="fig" rid="fig1">Figure 1</xref>, right). In lamellipodia, VCA-containing WAVE proteins stay associated with the leading edge, while the Arp2/3 complex is distributed throughout the actin mesh (<xref ref-type="bibr" rid="bib36">Lai et al., 2008</xref>). In both budding and fission yeasts, Arp2/3 complex is separated from membrane-bound activators during endocytosis (<xref ref-type="bibr" rid="bib33">Kaksonen et al., 2003</xref>; <xref ref-type="bibr" rid="bib57">Sirotkin et al., 2005</xref>). Analogously, in propulsive actin ‘comet tails’ Arp2/3 complex is found throughout the tail while its activators stay (largely) associated with the motile bacterium, virus, or vesicle (<xref ref-type="bibr" rid="bib66">Welch et al., 1997</xref>; <xref ref-type="bibr" rid="bib19">Egile et al., 1999</xref>; <xref ref-type="bibr" rid="bib38">Loisel et al., 1999</xref>; <xref ref-type="bibr" rid="bib62">Taunton et al., 2000</xref>; <xref ref-type="bibr" rid="bib65">Weisswange et al., 2009</xref>). In vitro, branches are released from the budding yeast WASP family member Las17 (<xref ref-type="bibr" rid="bib42">Martin et al., 2006</xref>). In all of these systems Arp2/3 complex can disengage from the surface attached activator within a short time of the onset of daughter filament growth.<fig id="fig1" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.003</object-id><label>Figure 1.</label><caption><title>Pathway of Arp2/3 complex mediated actin branch formation activated by WASP protein dimers on the inside surface of a cell membrane, as deduced from previous studies.</title><p>Within the white arrow Arp2/3 complex is activated by VCA, detaches from the membrane and initiates daughter filament elongation. The order of these steps and how they are coordinated remains unclear.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.003">http://dx.doi.org/10.7554/eLife.01008.003</ext-link></p></caption><graphic xlink:href="elife01008f001"/></fig></p><p>These observations raise a fundamental question: how is the binding of Arp2/3 complex to filament sides coordinated with binding and release of VCA and the initiation and growth of the daughter filament (<xref ref-type="fig" rid="fig1">Figure 1</xref>)? Previous studies have proposed that VCA may dissociate from Arp2/3 complex prior to initiation of daughter filament growth. The observation that ATP hydrolysis by Arp2 decreases affinity for VCA suggested a possible trigger for dissociation (<xref ref-type="bibr" rid="bib13">Dayel et al., 2001</xref>; <xref ref-type="bibr" rid="bib14">Dayel and Mullins, 2004</xref>), although hydrolysis was later shown to be dispensable for filament nucleation and to instead control disassembly of the branch (<xref ref-type="bibr" rid="bib42">Martin et al., 2006</xref>; <xref ref-type="bibr" rid="bib30">Ingerman et al., 2013</xref>). Consideration of biochemical and structural data on WH2-actin interactions (<xref ref-type="bibr" rid="bib19">Egile et al., 1999</xref>; <xref ref-type="bibr" rid="bib28">Higgs et al., 1999</xref>; <xref ref-type="bibr" rid="bib26">Hertzog et al., 2004</xref>; <xref ref-type="bibr" rid="bib31">Irobi et al., 2004</xref>; <xref ref-type="bibr" rid="bib8">Chereau et al., 2005</xref>; <xref ref-type="bibr" rid="bib6">Boczkowska et al., 2008</xref>) led to speculation that during Arp2/3-mediated nucleation, the WH2 region of VCA may initially block the barbed end of actin delivered to the Arp2 or Arp3 subunit and thus must move or dissociate prior to daughter filament elongation (<xref ref-type="bibr" rid="bib45">Padrick et al., 2011</xref>; <xref ref-type="bibr" rid="bib63">Ti et al., 2011</xref>; <xref ref-type="bibr" rid="bib27">Hetrick et al., 2013</xref>). However, direct observation of the sequence of molecular events associated with the initiation of daughter filament growth and the release of VCA has been lacking.</p><p>Multi-wavelength single-molecule fluorescence colocalization methods (<xref ref-type="bibr" rid="bib29">Hoskins et al., 2011</xref>; <xref ref-type="bibr" rid="bib23">Friedman and Gelles, 2012</xref>) are a powerful approach to elucidating the reaction pathways and identifying key regulated steps in processes that involve multiple macromolecular components. Here we used three-color single molecule fluorescence experiments to directly visualize the sequence and rates of the key steps in the pathway through which VCA dimers, Arp2/3 complexes, and actin filaments associate with one another and generate a new actin branch. The work reveals that the activation step in nucleation is likely the release of VCA dimers from the nascent branch, such that VCA dissociation is the trigger for daughter filament growth. The problem of filament growth against a membrane to which it is tethered is therefore solved by a mechanism in which release from the membrane tether is required for filament initiation.</p></sec><sec id="s2" sec-type="results"><title>Results</title><sec id="s2-1"><title>VCA is released from filament-bound Arp2/3 complex before initiation of daughter filament growth</title><p>To follow the coordination of VCA association with Arp2/3 complex during actin branch formation we labeled each protein with a fluorescent probe and visualized their colocalization dynamics using colocalization single molecule spectroscopy (CoSMoS) (<xref ref-type="bibr" rid="bib22">Friedman et al., 2006</xref>; <xref ref-type="bibr" rid="bib29">Hoskins et al., 2011</xref>; <xref ref-type="bibr" rid="bib23">Friedman and Gelles, 2012</xref>). Actin was labeled with a blue-excited dye (on 10% of monomers) and tagged with biotin (on 1% of monomers) to enable tethering to microscope slides. <italic>Saccharomyces cerevisiae</italic> Arp2/3 complex was labeled with a red-excited dye targeted to a SNAP tag fused to the C-terminus of the Arc18 (ArpC3) subunit (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). We used a green-excited, Cy3 dye bis-maleimide derivative to label and covalently dimerize the VCA from N-WASP, hereafter called diVCA, which includes the second WH2 motif (V) through the C-terminus of the protein. Like other dimeric VCAs (<xref ref-type="bibr" rid="bib46">Padrick et al., 2008</xref>) (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>), this diVCA construct was able to stimulate the activity of the Arp2/3 complex (<xref ref-type="fig" rid="fig2s2">Figure 2—figure supplement 2</xref>; ‘Materials and methods’) at low-nanomolar concentrations suitable for single molecule imaging, whereas labeled monomeric VCA constructs did not.</p><p>Using this combination of tagged proteins (<xref ref-type="fig" rid="fig2">Figure 2A</xref>) we directly observed individual Arp2/3 complex and diVCA molecules binding to immobilized actin filaments and nucleating new branches. During this process, single molecules of Arp2/3 complex and diVCA were observed to bind together to locations on filament sides (e.g., <xref ref-type="fig" rid="fig2">Figure 2B–C</xref> at <italic>t</italic> = 0; <xref ref-type="fig" rid="fig2s3">Figure 2—figure supplement 3</xref>). In nearly all cases (83 ± 9% S.E., based on 877 Arp2/3 complex observations), Arp2/3 complex and diVCA arrived simultaneously within the experimental time resolution (0.15 s), indicating that diVCA was bound to Arp2/3 complex prior to filament engagement, and that both proteins bound to filaments as a unit (‘Materials and methods’). The sparse labeling of actin and high fluorescence intensity of mother filaments did not permit us to detect the arrival of actin monomers with VCA and Arp2/3 complex. However, under the reaction conditions (5 nM diVCA and 1 μM actin), with a <italic>K</italic><sub>D</sub> of diVCA for actin of ∼300 nM (which is only slightly altered by the presence of Arp2/3 complex, <xref ref-type="fig" rid="fig2s4">Figure 2—figure supplement 4</xref>), we expect ∼77% of diVCAs to have at least one actin bound. Thus, most of the diVCA-Arp2/3 complexes observed to bind mother filament should contain actin and thus have all the molecular factors needed for nucleation. We hereafter refer to this filament-bound complex as the ‘nascent branch’, an intermediate in the pathway to daughter filament assembly.<fig-group><fig id="fig2" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.004</object-id><label>Figure 2.</label><caption><title>Rapid release of dimeric VCA from the nascent branch precedes nucleation.</title><p>(<bold>A</bold>) Design of an experiment to observe diVCA-activated branch nucleation by Arp2/3 complex on the sides of surface-immobilized actin filaments. Blue, green and red stars denote fluorescent dye labels AlexaFluor 488 (AF488), Cyanine 3 (Cy3), and Dy649 that are excited with blue, green, and red lasers, respectively. (<bold>B</bold>) Image sequence of the same microscope field of view taken at each of the three dye wavelengths (rows) at five selected time points (<italic>t</italic>; columns). Images record the colocalization of an individual Arp2/3 complex and diVCA molecule at <italic>t</italic> = 0 (yellow arrowhead) followed by nucleation and growth of a daughter filament at that location (red arrowhead). Solution contained 5 nM Cy3-diVCA, 5 nM SNAP-tagged Arp2/3 complex labeled with Dy649 (Arp2/3-SNAP649), and 1 µM actin, 10% AF488-labeled. Bar: 1 μm. See <xref ref-type="other" rid="video1">Video 1</xref>. (<bold>C</bold>) Recordings of daughter filament length and branch site fluorescence intensities from the nucleation event in <bold>B</bold>. Arrow marks the time of daughter filament nucleation estimated by extrapolating the daughter length fit line to zero length (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). Plot at bottom is a magnified view showing that Arp2/3 complex and diVCA labels appear simultaneously (<italic>t</italic> = 0) followed by rapid release of diVCA (<italic>t</italic> = 0.2 s). (<bold>D</bold>) Cumulative lifetime distributions of Arp2/3 complex and diVCA on filament sides after binding of an Arp2/3-diVCA complex to the filament (<italic>N</italic> = 752). Smooth lines indicate two- (diVCA) or three-exponential (Arp2/3 complex) fits yielding the indicated fit parameters (‘Materials and methods’). Main plot shows the data for time <10 s; inset shows the full distribution with the exception of one outlier. (<bold>E</bold>) Comparison of the time (±S.E.) of daughter filament initiation with the time of diVCA release from the nascent branch in individual branch nucleation events by diVCA-Arp2/3 complexes.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.004">http://dx.doi.org/10.7554/eLife.01008.004</ext-link></p></caption><graphic xlink:href="elife01008f002"/></fig><fig id="fig2s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.005</object-id><label>Figure 2—figure supplement 1.</label><caption><title>N-WASP VCA dimers bind tightly to Arp2/3 complex and stimulate its actin nucleation activity.</title><p>(<bold>A</bold>) Schematic of Glutathione-S-transferase (GST) fusions with N-WASP VCA-containing domains. N-terminal GST-fusions, which are naturally tight dimers, were made to VVCA (N-WASP residues 393–505), VCA (residues 432–505), and NL-VCA (NL [gray] is a sequence, residues 420-–430, found N-terminal to the human WASP VCA). (<bold>B</bold>) GST-VCA stimulates actin polymerization by Arp2/3 complex. Records indicate pyrene actin fluorescence increase due to filament polymerization 2 µM actin, 10% pyrene-labeled and bovine Arp2/3 complex, with or without GST-VCA. (<bold>C</bold>) GST-VVCA is more active at saturation than is GST-VCA. Bars indicate the number (±S.E.) of actin filament barbed ends present at the midpoint of polymerization assays like those shown in <bold>B</bold>. The concentration series show that stimulation by all constructs saturates at or below 25 nM and that dimer with two V domains per subunit (GST-VVCA) is more active that the single V domain construct (GST-VCA) even when an extended linker is incorporated into the latter (GST-NL-VCA). (<bold>D</bold>) A dye-labeled, single cysteine mutant N-WASP VVCA-A462C-AF594 binds to bovine Arp2/3 complex, as determined by fluorescence anisotropy (points). Fit to a direct binding isotherm yields <italic>K</italic><sub>D</sub> = 150 ± 5 nM (68% C.I.). (<bold>E</bold>) Binding of GST-VVCA and GST-VCA to bovine Arp2/3 complex (200 nM). Points indicate interference with binding of VVCA-A462C-AF594 in a fluorescence anisotropy assay like that in D caused by addition of the indicated amount of unlabeled competitor GST-VCA or GST-VVCA. Fits to a competition-binding isotherm (lines) yield the dissociation equilibrium constants (±S.E.) shown in the inset.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.005">http://dx.doi.org/10.7554/eLife.01008.005</ext-link></p></caption><graphic xlink:href="elife01008fs001"/></fig><fig id="fig2s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.006</object-id><label>Figure 2—figure supplement 2.</label><caption><title>VCA dimers crosslinked through Cy3 stimulate Arp2/3 to a similar extent as GST-VCA dimers.</title><p>(<bold>A</bold>) Schematics of N-WASP VVCA and VCA dimers made via fusion to GST (as in <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>) or by covalent crosslinking of single-cysteine mutant subunits with a bis-maleimide Cy3. The diXVCA material is similar to diVVCA but has the first V domain mutated at several sites to reduce binding to actin. (<bold>B</bold>) Sequence alignment of the diXVCA construct with wild-type N-WASP VVCA (residues 393–505 of human N-WASP). White rectangle, C(GGS)<sub>4</sub> linker; X, V, C, and A domains are indicated in the same colors as in <bold>A</bold>. Cys residues are highlighted; the C431A mutation was made so that the XVCA peptide has only a single cys to react with the Cy3 bis-maleimide crosslinker. (<bold>C</bold> and <bold>D</bold>) Pyrene-actin fluorescence records (as in <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1B</xref>) indicating rates of actin filament polymerization in the presence of 10 nM yeast Arp2/3 complex and 2 μM rabbit muscle actin supplemented with 25 nM (<bold>C</bold>) or 200 nM (<bold>D</bold>) of the indicated VCA dimers. The records show that diXVCA is of somewhat higher nucleation activity (maximum slope) than diVCA, but the insertion of the X domain sequence does not increase the activity to the levels seen for diVVCA. Thus, the length and nature of the linkage between the two VCA segments in the dimer may have a subtle effect on the activity, but is not the origin of large difference in activity between diVCA and diVVCA. This reduced activity of the diVCA construct, together with the significant basal branch nucleation activity of <italic>S. cerevisiae</italic> Arp2/3 complex in the absence of WASP proteins (<xref ref-type="bibr" rid="bib67">Wen and Rubenstein, 2005</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>) likely contribute to the modest (∼twofold; see text) activation of branch formation by di-VCA we observe.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.006">http://dx.doi.org/10.7554/eLife.01008.006</ext-link></p></caption><graphic xlink:href="elife01008fs002"/></fig><fig id="fig2s3" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.007</object-id><label>Figure 2—figure supplement 3.</label><caption><title>Arp2/3 complex and diVCA usually bind to and release from filaments as a unit when no daughter filament is formed.</title><p>Left: Merged images (<xref ref-type="fig" rid="fig2">Figure 2B</xref>) and fluorescence intensity records (<xref ref-type="fig" rid="fig2">Figure 2C</xref>) showing an example of simultaneous binding (<italic>t</italic> = 0) and release (dashed vertical line) of diVCA and Arp2/3 complex on an actin filament. Right: Three more examples. Experimental conditions were the same as in <xref ref-type="fig" rid="fig2">Figure 2</xref>. Bar: 1 µm.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.007">http://dx.doi.org/10.7554/eLife.01008.007</ext-link></p></caption><graphic xlink:href="elife01008fs003"/></fig><fig id="fig2s4" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.008</object-id><label>Figure 2—figure supplement 4.</label><caption><title>Association with Arp2/3 complex does not affect binding of VCA to actin.</title><p>(<bold>A</bold>) Binding, measured using fluorescence anisotropy (points), during titration of rabbit muscle actin into 20 nM VCA-AF488. Fit to a binding isotherm (line) yielded <italic>K</italic><sub>D</sub> 220 ± 10 (68% C.I.) nM. (<bold>B</bold>) Competition binding experiment to determine the <italic>K</italic><sub>D</sub> of diVCA for actin. Fluorescence anisotropy of 20 nM VCA-AF488 was monitored in the presence of 150 nM rabbit muscle actin, and the indicated concentration of diVCA. Fit to a competition binding isotherm (line) yielded <italic>K</italic><sub>D</sub> 340 ± 60 nM (see text). (<bold>C</bold>) Binding affinity of non-polymerizable actin (‘Materials and methods’) for VCA is minimally perturbed by the presence of yeast Arp2/3 complex. Fluorescence anisotropy of 10 nM of VCA-AF594 was measured in the presence of increasing concentrations of non-polymerizable actin (red circles), yeast Arp2/3 complex (blue squares), or non-polymerizable actin in the presence of 300 nM yeast Arp2/3 complex (green inverted triangles). Fits to single site binding isotherms (lines) yielded <italic>K</italic><sub>D</sub> 70 ± 10 nM, 65 ± 4 nM, and 105 nM ± 15 nM, respectively. Changes in anisotropy are expected to be dominated by the binding of actin, as the fluorophore location is proximal to the N-terminus of the V domain, and away from the Arp2/3 complex binding motifs. Increased fluorescence anisotropy in the presence of both Arp2/3 complex and actin indicates that both species can bind simultaneously.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.008">http://dx.doi.org/10.7554/eLife.01008.008</ext-link></p></caption><graphic xlink:href="elife01008fs004"/></fig></fig-group><media content-type="glencoe play-in-place height-250 width-310" id="video1" mime-subtype="mp4" mimetype="video" xlink:href="elife01008v001.mp4"><object-id pub-id-type="doi">10.7554/eLife.01008.009</object-id><label>Video 1.</label><caption><title>The diVCA-stimulated actin branch formation event shown in <xref ref-type="fig" rid="fig2">Figure 2B–C</xref>.</title><p>Red: Arp2/3-SNAP649 (5 nM in solution); green: Cy3-diVCA (5 nM); blue: actin-AF488 (1 µM, 10% labeled). Arp2/3 complex and diVCA images were recorded every 0.05 s; actin images were recorded every ∼12 s. Playback rate: real time. Bar: 1 μm. Many diVCA-Arp2/3 complexes are observed to transiently associate with actin filaments. One such nascent branch complex (yellow arrowhead) releases diVCA shortly after it appears, leaving Arp2/3 complex stably associated with the mother filament (red arrowhead), where it subsequently initiates daughter filament elongation.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.009">http://dx.doi.org/10.7554/eLife.01008.009</ext-link></p></caption></media></p><p>Most nascent branches, although containing the necessary components for nucleation, dissociated quickly (typically in <1 s) without producing a daughter filament. This is consistent with our previous observations that the vast majority of Arp2/3 complex binding events are non-productive (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). In nearly all cases (97.3 ± 0.4%, based on 758 nascent branch observations), diVCA and Arp2/3 complex dissociated simultaneously as a unit (<xref ref-type="fig" rid="fig2s3">Figure 2—figure supplement 3</xref>). The on-filament lifetime distributions of diVCA and Arp2/3 complex have identical short (τ<sub>1</sub> and τ<sub>2</sub>) components (<xref ref-type="fig" rid="fig2">Figure 2D</xref> at time <1 s), consistent with the conclusion that the two molecules are released from the filament as an Arp2/3-diVCA complex.</p><p>In contrast to the large majority of non-productive filament encounters, a small fraction of Arp2/3-diVCA filament binding events led to formation of a daughter filament. A merit of the single-molecule approach is that we could characterize these rare productive events (e.g., <xref ref-type="fig" rid="fig2">Figure 2B–C</xref>) independently of the excess of non-productive events. In the productive events, Arp2/3 complex and diVCA release were not simultaneous. There was no evidence that the Arp2/3 complexes which formed branches ever dissociated; the value of τ<sub>3</sub> is set by the photobleaching lifetime of the dye-labeled Arp2/3 complex (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). In contrast, diVCA dissociated rapidly (typically in <1 s) from the productive Arp2/3-diVCA-filament complexes. Consistent with this observation, the on-filament lifetime distribution of diVCA lacks a long component (τ<sub>3</sub>) that is present in the on-filament lifetime distribution of Arp2/3 complex (<xref ref-type="fig" rid="fig2">Figure 2D</xref> at time > approximately 4 s). Thus, the data demonstrate that daughter nucleation is essentially always accompanied by Arp2/3 complex retention and diVCA release.</p><p>To determine whether diVCA release occurs before or after the onset of daughter filament growth, we measured the time at which each daughter filament initiated elongation by extrapolating daughter length records (as in <xref ref-type="fig" rid="fig2">Figure 2C</xref>, top). These filament initiation times were then compared to the times of diVCA release from the same nascent branch. Filament initiation time measurements were imprecise because of the uncertainties inherent in measuring daughter filaments of sub-micrometer lengths. Nevertheless, within this experimental uncertainty we observed that the initiation of daughter filament growth always occurred at or after the time of diVCA release (41 observations; <xref ref-type="fig" rid="fig2">Figure 2E</xref>). This was true even in the comparatively rare cases in which diVCA persisted on the nascent branch for times >1 s before dissociating. Taken together, these data suggest that the daughter filament cannot initiate unless and until VCA is released from the nascent branch. Thus, diVCA release may serve as the trigger for daughter growth.</p></sec><sec id="s2-2"><title>VCA does not bind Arp2/3 complex after branch formation</title><p>Next we asked whether diVCA can bind to Arp2/3 complex after branches have formed, in order to better understand how WASP recruits and activates free Arp2/3 complex yet does not stay bound to Arp2/3 complex in branch junctions and restrict network growth (‘Introduction’). To address this question, we tethered individual dye- and biotin-labeled Arp2/3 complexes to the microscope slide and visualized the binding of freely diffusing diVCA and (non-biotinylated) actin filaments (<xref ref-type="fig" rid="fig3">Figure 3A</xref>). Most of the tethered Arp2/3 complexes (>80%) were observed to bind diVCA. Binding lasted for tens or hundreds of seconds when no filaments were nearby (e.g., <xref ref-type="fig" rid="fig3">Figure 3B</xref> at time <0). With 2 nM diVCA in solution, individual tethered Arp2/3 complexes were nearly continuously occupied, suggesting a dissociation equilibrium constant <italic>K</italic><sub>D</sub> < 2 nM similar to bulk affinity measurements on other diVCA and Arp2/3 complex species (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1E</xref>). Further, we never observed the nucleation of a new actin filament from an isolated surface-tethered diVCA-Arp2/3 complex, consistent with previous conclusions that Arp2/3 complex cannot nucleate a daughter unless it is bound to a pre-existing mother filament (<xref ref-type="bibr" rid="bib39">Machesky et al., 1999</xref>; <xref ref-type="bibr" rid="bib5">Blanchoin et al., 2000</xref>; <xref ref-type="bibr" rid="bib1">Achard et al., 2010</xref>).<fig id="fig3" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.010</object-id><label>Figure 3.</label><caption><title>VCA dimers form long-lived complexes with Arp2/3 complex before filament binding but not after branch formation.</title><p>(<bold>A</bold>) Experimental design to observe diVCA binding and nucleation of actin filaments on immobilized Arp2/3 complexes. Arp2/3-SNAP was tethered to the slide surface via a bi-functional SNAP substrate that incorporated both a Dy649 dye and a biotin-terminated PEG chain; we monitored binding of fluorescently labeled diVCA and actin filaments from solution. (<bold>B</bold>) Example record showing the length of a nucleated daughter filament and the fluorescence intensity from actin (blue) and diVCA (green) at an individual tethered Arp2/3 complex molecule. The solution contained 1 μM actin (10% AF488 labeled) and Cy3-diVCA (2 nM). Fluorescence from the tethered Arp2/3 complex (red trace) remained steady and above background (red dash) throughout. Schematics show the inferred complexes present at the indicated times. Time zero is the time of diVCA release.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.010">http://dx.doi.org/10.7554/eLife.01008.010</ext-link></p></caption><graphic xlink:href="elife01008f003"/></fig></p><p>In some cases, we did observe association of a mother filament (formed in solution) with some tethered Arp2/3-diVCA complexes, followed by growth of a daughter filament. This yielded a branch junction that remained stably colocalized with the tethered Arp2/3 complex. In those events (e.g., <xref ref-type="fig" rid="fig3">Figure 3B</xref> at time >0), branch formation essentially abolished binding of diVCA; any appearance of diVCA on branch junctions was transient (<0.1 s) and occurred at a low frequency (∼0.7 × 10<sup>6</sup> M<sup>−1</sup> s<sup>−1</sup>; <italic>N</italic> = 2) comparable to non-specific binding at randomly chosen points on the microscope slide (1.0 ± 0.6 × 10<sup>6</sup> M<sup>−1</sup> s<sup>−1</sup>). The data demonstrate that the affinity of diVCA for isolated Arp2/3 complex is high, whereas the diVCA affinity for Arp2/3 complex in the branch junction is comparatively low. The low affinity of diVCA for the branch is consistent with data from previous studies (<xref ref-type="bibr" rid="bib20">Egile et al., 2005</xref>; <xref ref-type="bibr" rid="bib42">Martin et al., 2006</xref>). In addition, our measurements suggest that in the cell, once WASP proteins dissociate and the branch forms the Arp2/3 complex incorporated in the branch junction is unlikely to reassociate with membrane-linked WASP proteins and thus will not restrict filament network growth.</p></sec><sec id="s2-3"><title>Targeted mutations alter the kinetic stability of ternary complexes of diVCA, actin monomers, and Arp2/3 complex</title><p>To challenge the model that diVCA release from the nascent branch is required to initiate daughter filament growth, we next engineered a series of mutations in diVCA (<xref ref-type="fig" rid="fig4">Figure 4A</xref>; <xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1</xref>). The goal was to modestly perturb VCA interactions with Arp2/3 complex or actin without altering the reaction pathway by which diVCA stimulates branch formation. Guided by previous biochemical data (<xref ref-type="bibr" rid="bib70">Zalevsky et al., 2001</xref>; <xref ref-type="bibr" rid="bib48">Panchal et al., 2003</xref>; <xref ref-type="bibr" rid="bib8">Chereau et al., 2005</xref>), we mutated each of the three regions of N-WASP VCA. The D435S/A436D mutation in the V-region (diVCA-V*) was designed to perturb actin affinity (<xref ref-type="bibr" rid="bib8">Chereau et al., 2005</xref>), whereas mutations in the C-region (I467A, diVCA-C*) and A-region (Δ486–488, diVCA-A*) were designed to perturb interactions with Arp2/3 complex (<xref ref-type="bibr" rid="bib70">Zalevsky et al., 2001</xref>; <xref ref-type="bibr" rid="bib48">Panchal et al., 2003</xref>).<fig-group><fig id="fig4" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.011</object-id><label>Figure 4.</label><caption><title>diVCA mutations alter the stability of Arp2/3 complex-diVCA-actin monomer assemblies.</title><p>(<bold>A</bold>) Arrangement of V, C, and A domains in native N-WASP and in the diVCA constructs used in this study (w.t. is wild-type). Asterisks mark the domains bearing targeted mutations (substitution of one or two residues, or a three-residue deletion; <xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1A</xref>) in the three mutant constructs. (<bold>B</bold>) Fluorescence anisotropy detected binding of AF488-labeled N-WASP VCA with rabbit muscle actin, in the presence of competitor wild-type (same data as in <xref ref-type="fig" rid="fig2s4">Figure 2—figure supplement 4B</xref>) or mutant Cy3-diVCA constructs (symbols). Data were fit (lines) with competition binding isotherms incorporating the coupled equilibria (‘Materials and methods’) yielding <italic>K</italic><sub>D</sub> values 340 ± 60 (S.E.) nM for wild-type diVCA, 660 ± 80 nM for diVCA-V*, 260 ± 40 nM for diVCA-C*, and 250 ± 40 nM for diVCA-A*. (<bold>C</bold>) Example Cy3-diVCA fluorescence intensity records recorded on individual tethered Arp2/3 complexes (<xref ref-type="fig" rid="fig3">Figure 3</xref>): Cy3-diVCA wild-type or C* mutant (0.5 nM) molecules binding and dissociating in the presence of 1 μM actin monomers but no filament. (<bold>D</bold>) Cumulative lifetime distributions of diVCA-Arp2/3 complexes in the presence of monomeric actin observed in records like those in <bold>B</bold>. Smooth lines are biexponential fits (<xref ref-type="table" rid="tbl1">Table 1</xref>). Inset is a magnified view of the indicated data range.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.011">http://dx.doi.org/10.7554/eLife.01008.011</ext-link></p></caption><graphic xlink:href="elife01008f004"/></fig><fig id="fig4s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.012</object-id><label>Figure 4—figure supplement 1.</label><caption><title>Design of the diVCA mutant constructs and characterization of Arp2/3 complex binding by the diVCA-C* mutant.</title><p>(<bold>A</bold>) Sequence alignments showing the substitution (colored residues) and deletion (dash) mutations in the diVCA mutant constructs. Lines mark the regions of VCA that interact with G-actin and with Arp2/3 complex. The C(GGS)<sub>4</sub> linker and the cys residue used to react with Cy3 bis-maleimide are marked. (<bold>B</bold>) Binding of GST-VCA and GST-VCA-C* to bovine Arp2/3 complex assessed by a competition assay. Conditions and analysis as in <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1E</xref>; blue curve is repeated from that panel. The fits reveal <italic>K</italic><sub>D</sub> values (68% C.I.) indicating that GST-VCA-C* has a ∼fourfold weaker affinity for Arp2/3 complex than does GST-VCA (inset).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.012">http://dx.doi.org/10.7554/eLife.01008.012</ext-link></p></caption><graphic xlink:href="elife01008fs005"/></fig><fig id="fig4s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.013</object-id><label>Figure 4—figure supplement 2.</label><caption><title>Single molecule analysis of wild-type and mutant diVCA binding to and dissociating from tethered Arp2/3 complex.</title><p>(<bold>A</bold>)–(<bold>C</bold>) Binding rate measurements. (<bold>A</bold>) Example curves from single experiments showing the cumulative distribution of the lifetimes of Arp2/3 complexes unoccupied by diVCA constructs. diVCA constructs were present at 0.1 nM; other conditions as in <xref ref-type="fig" rid="fig4">Figure 4D</xref>. Fitting to a single exponential function (smooth curves) yielded binding rates. (<bold>B</bold>) Measured binding rates (±S.E.) were proportional to diVCA-C* concentration. (<bold>C</bold>) Second order binding rate constants, <italic>k</italic><sub>V+</sub> (±S.E.), calculated from experiments like those in A (<xref ref-type="table" rid="tbl1">Table 1</xref>) at diVCA construct concentrations in the range 0.1–1.0 nM. (<bold>D</bold>) The effect of photobleaching on the slow component (τ<sub>V,2</sub>; <xref ref-type="table" rid="tbl1">Table 1</xref>) of the lifetime distributions for association of diVCA constructs with tethered Arp2/3 complex (<xref ref-type="fig" rid="fig4">Figure 4D</xref>). Experiments were performed over a range of excitation green laser powers and the dependence of the observed dissociation rate (1/τ<sub>V,2</sub>; ±S.E.) on power was globally fit to quantify the photobleaching rate (slope) and the photobleaching-corrected dissociation rates for each diVCA construct (intercepts). The τ<sub>V,1</sub> and τ<sub>V,2</sub> values reported in <xref ref-type="table" rid="tbl1">Table 1</xref> were taken from experiments at the weakest laser powers and were equal within experimental error to the values obtained after photobleaching correction.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.013">http://dx.doi.org/10.7554/eLife.01008.013</ext-link></p></caption><graphic xlink:href="elife01008fs006"/></fig></fig-group></p><p>We measured the affinity of the mutants for monomeric actin using a fluorescence anisotropy competition assay. Here, labeled VCA reported on binding to monomeric actin by an increase in fluorescence anisotropy (as in <xref ref-type="fig" rid="fig2s4">Figure 2—figure supplement 4A</xref>). As expected from the design, wild-type diVCA, diVCA-C*, and diVCA-A* had similar affinities for actin, while diVCA-V* bound more weakly (∼650 nM vs ∼300 nM for wild-type; <xref ref-type="fig" rid="fig4">Figure 4B</xref>).</p><p>We next compared the rates of wild-type and mutant diVCAs binding to and dissociating from Arp2/3 complex using surface-tethered Arp2/3 complexes in the presence of actin monomers but not bound to actin filament. For wild-type diVCA and all three mutants, we observed repeated association and dissociation of the diVCA molecules with surface-tethered Arp2/3 complex (e.g., <xref ref-type="fig" rid="fig4">Figure 4C</xref>). These events allowed measurement of the lifetime distributions for the diVCA bound and dissociated states of Arp2/3 complex, and thus determination of the binding and dissociation rate constants. The mutations had only modest effects, at most 2.7-fold, on the rate of binding of diVCA to Arp2/3 complex (<xref ref-type="fig" rid="fig4s2">Figure 4—figure supplement 2A–C</xref>; <xref ref-type="table" rid="tbl1">Table 1</xref>). The dissociation rates varied over a wide range, with the A* mutant dissociating from Arp2/3 complex more slowly than wild-type, and the V* and C* mutants dissociating more rapidly (<xref ref-type="fig" rid="fig4">Figure 4D</xref>). Both wild-type and mutant complexes displayed lifetime distributions that were fit well with two exponential components (<xref ref-type="fig" rid="fig4">Figure 4D</xref>, <xref ref-type="fig" rid="fig4s2">Figure 4—figure supplement 2D</xref>, <xref ref-type="table" rid="tbl1">Table 1</xref>), indicating that at least two distinct diVCA-Arp2/3 complex assemblies or conformations were present. In a more detailed analysis with the V* mutant, we saw no evidence that the short and long lifetime components segregated into different subpopulations of individual Arp2/3 complexes. These observations suggest even individual diVCA-Arp2/3 complexes participated in multiple states. The presence of multiple different complexes is consistent with previous observations including that Arp2/3 complex has two (or more) binding sites for VCA (<xref ref-type="bibr" rid="bib45">Padrick et al., 2011</xref>; <xref ref-type="bibr" rid="bib63">Ti et al., 2011</xref>; <xref ref-type="bibr" rid="bib68">Xu et al., 2012</xref>), and that alternative conformations of Arp2/3 complex (<xref ref-type="bibr" rid="bib24">Goley et al., 2004</xref>; <xref ref-type="bibr" rid="bib53">Rodal et al., 2005</xref>) may have different affinities for VCA.<table-wrap id="tbl1" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.014</object-id><label>Table 1.</label><caption><p>Colocalization kinetics and activities of diVCA and Arp2/3 complex</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.014">http://dx.doi.org/10.7554/eLife.01008.014</ext-link></p></caption><table frame="hsides" rules="groups"><thead><tr><th/><th>no VCA</th><th>diVCA w.t.</th><th>diVCA V*</th><th>diVCA C*</th><th>diVCA A*</th></tr></thead><tbody><tr><td colspan="6">Arp2/3 complex off filament (<xref ref-type="fig" rid="fig4">Figure 4D</xref>, <xref ref-type="fig" rid="fig4s2">Figure 4—figure supplement 2</xref>)</td></tr><tr><td> <italic>N</italic> (groups)</td><td/><td>5</td><td>3</td><td>3</td><td>1</td></tr><tr><td> <italic>k</italic><sub>V+</sub> (10<sup>7</sup> M<sup>−1</sup> s<sup>−1</sup>)</td><td/><td align="char" char="plusmn">16±5 (SEM)</td><td align="char" char="plusmn">7±3 (SEM)</td><td align="char" char="plusmn">6±1 (SEM)</td><td align="char" char="plusmn">16±1 (SE)</td></tr><tr><td> <italic>N</italic><sub>V</sub></td><td/><td>1528</td><td>957</td><td>1339</td><td>1262</td></tr><tr><td> τ<sub>V1</sub> (s)</td><td/><td align="char" char="plusmn">8±1</td><td align="char" char="plusmn">3.6±0.4</td><td align="char" char="plusmn">2.4±0.1</td><td align="char" char="plusmn">40±10</td></tr><tr><td> A<sub>V1</sub></td><td/><td align="char" char="plusmn">24±3%</td><td align="char" char="plusmn">42±2%</td><td align="char" char="plusmn">98±2%</td><td align="char" char="plusmn">28±9%</td></tr><tr><td> τ<sub>V2</sub> (s)</td><td/><td align="char" char="plusmn">61±3</td><td align="char" char="plusmn">56±3</td><td align="char" char="plusmn">14±6</td><td align="char" char="plusmn">250±30</td></tr><tr><td colspan="6">Arp2/3 complex on filament (<xref ref-type="fig" rid="fig6">Figure 6B</xref>)</td></tr><tr><td> <italic>N</italic><sub>A</sub></td><td>715</td><td>877</td><td>407</td><td>1089</td><td>597</td></tr><tr><td> <italic>k</italic><sub>A+</sub> (10<sup>4</sup> M<sup>−1</sup> s<sup>−1</sup>)</td><td align="char" char="plusmn">2.0±0.3</td><td align="char" char="plusmn">2.1±0.3</td><td align="char" char="plusmn">1.3±0.2</td><td align="char" char="plusmn">2±1</td><td align="char" char="plusmn">0.8±0.2</td></tr><tr><td> <italic>f</italic><sub>AV</sub></td><td/><td align="char" char="plusmn">0.83±0.09</td><td align="char" char="plusmn">0.7±0.1</td><td align="char" char="plusmn">0.59±0.08</td><td align="char" char="plusmn">0.7±0.1</td></tr><tr><td> <italic>f</italic><sub>V−</sub></td><td/><td align="char" char="plusmn">0.026±0.004</td><td align="char" char="plusmn">0.041±0.009</td><td align="char" char="plusmn">0.018±0.003</td><td align="char" char="plusmn">0.015±0.004</td></tr><tr><td> <italic>f</italic><sub>B</sub></td><td align="char" char="plusmn">0.006±0.002</td><td align="char" char="plusmn">0.008±0.002</td><td align="char" char="plusmn">0.029±0.008</td><td align="char" char="plusmn">0.015±0.003</td><td align="char" char="plusmn">0.013±0.003</td></tr><tr><td colspan="6">Arp2/3 complex at branch sites (<xref ref-type="fig" rid="fig5">Figure 5A</xref>, <xref ref-type="fig" rid="fig6">Figure 6C–E</xref>)</td></tr><tr><td> <italic>N</italic><sub>B</sub></td><td/><td>57</td><td>69</td><td>69</td><td>40</td></tr><tr><td> τ<sub>V</sub><sup>*</sup> (s)</td><td/><td align="char" char="plusmn">0.7±0.1</td><td align="char" char="plusmn">0.54±0.08</td><td align="char" char="plusmn">0.37±0.04</td><td align="char" char="plusmn">0.7±0.2</td></tr><tr><td> <italic>k</italic><sub>V</sub><sup>*</sup> (s<sup>−1</sup>)</td><td/><td align="char" char="plusmn">0.04±0.01</td><td align="char" char="plusmn">0.08±0.02</td><td align="char" char="plusmn">0.05±0.01</td><td align="char" char="plusmn">0.022±0.008</td></tr><tr><td> <italic>k</italic><sub>B</sub> (M<sup>−1</sup> s<sup>−1</sup>)</td><td align="char" char="plusmn">120±40</td><td align="char" char="plusmn">160±50</td><td align="char" char="plusmn">320±90</td><td align="char" char="plusmn">200±100</td><td align="char" char="plusmn">100±30</td></tr></tbody></table><table-wrap-foot><fn><p>Parameter descriptions:</p></fn><fn><p><italic>N</italic> = number of groups of observations used to calculate binding rate of diVCA to isolated Arp2/3 complexes.</p></fn><fn><p><italic>k</italic><sub>V+</sub> = second order rate constant for diVCA binding to Arp2/3 complexes.</p></fn><fn><p><italic>N</italic><sub>V</sub> = number of observations of diVCA on isolated Arp2/3 complexes.</p></fn><fn><p>τ<sub>V1</sub> = first characteristic lifetime of diVCA on isolated Arp2/3 complexes.</p></fn><fn><p>A<sub>V1</sub> = percent of diVCA that dissociate from Arp2/3 complexes with time constant τ<sub>V1</sub>.</p></fn><fn><p>τ<sub>V2</sub> = second characteristic lifetime of diVCA on isolated Arp2/3 complexes.</p></fn><fn><p><italic>N</italic><sub>A</sub> = number of observations of Arp2/3 complexes on the sides of select filaments.</p></fn><fn><p><italic>k</italic><sub>A+</sub> = second order rate constant for Arp2/3 complex binding filament sides (per filament subunit).</p></fn><fn><p><italic>f</italic><sub>AV</sub> = fraction of Arp2/3 complexes that bind filament sides coincident with diVCA.</p></fn><fn><p><italic>f</italic><sub>V-</sub> = fraction of diVCA-Arp2/3-filament complexes that release diVCA.</p></fn><fn><p><italic>f</italic><sub>B</sub> = fraction of diVCA-Arp2/3-filament complexes that nucleate a daughter filament.</p></fn><fn><p><italic>N</italic><sub>B</sub> = number of observations of branch formation from diVCA-Arp2/3-filament complexes.</p></fn><fn><p>τ<sub>V</sub><sup>*</sup> = mean lifetime of diVCA on nascent branches.</p></fn><fn><p><italic>k</italic><sub>V</sub><sup>*</sup> = rate constant for diVCA release from the nascent branch.</p></fn><fn><p><italic>k</italic><sub>B</sub> = second order rate constant for branch formation (per mother filament subunit).</p></fn></table-wrap-foot></table-wrap></p><p>From these experiments, we conclude that we have created a panel of mutants that modestly alter the association of diVCA with its binding partners. The V* and C* mutants, by disrupting interactions with Arp2/3 complex and monomeric actin, respectively, produce ternary diVCA-actin-Arp2/3 complexes that are less kinetically stable than wild-type. Conversely, the A* mutant produces a more stable ternary complex than wild-type.</p></sec><sec id="s2-4"><title>Rate of diVCA release from the nascent branch limits rate of daughter filament nucleation</title><p>The diVCA mutants have distinct activities in stimulating branch nucleation by Arp2/3 complex. In bulk solution (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1</xref>) wild-type diVCA boosted Arp2/3 complex-dependent nucleation by 2.0 ± 0.5-fold (p=0.017; measured by concentration of filament barbed ends at the midpoint of the reaction, <xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1B</xref>). The V* and C* mutants were more potent than wild-type (p=0.018 and 0.053, respectively), stimulating nucleation 10 ± 3-fold and 6 ± 3-fold. Conversely, the A* mutant was less potent than wild-type (p=0.040), stimulating only 1.4 ± 0.4 fold. In all of these experiments, the diVCA construct was at 25 nM, a concentration that produces near-maximal stimulation (<xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2A</xref>).</p><p>Similar effects of the mutant diVCA constructs were seen in real-time observations of individual filaments being nucleated (<xref ref-type="fig" rid="fig5">Figure 5A</xref>). As expected, all three diVCAs were able to promote nucleation from the sides of existing filaments. More importantly, the mutant diVCA constructs shared with wild-type the key molecular behaviors discussed previously. Like wild-type, mutant diVCA constructs released from the nascent branch prior to initiation of the daughter filament (<xref ref-type="fig" rid="fig5">Figure 5B</xref>; <xref ref-type="fig" rid="fig5s3">Figure 5—figure supplement 3</xref>), and did not bind to Arp2/3 complex after the branch formed (<xref ref-type="fig" rid="fig5">Figure 5C,D</xref>; this later behavior could not be verified for the A* mutant because of its low activity in the experiments using tethered Arp2/3 complexes). Overall, these results suggest that the mutant constructs stimulate Arp2/3 complex by the same mechanism as wild-type. Moreover, the rank order of the diVCA construct nucleation activities was identical in the single-molecule measurements of the rate of branch formation observed on existing filaments (<xref ref-type="fig" rid="fig5">Figure 5A</xref>) and in bulk measurements of concentration of filaments generated in solution (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1B</xref>). Thus, the mutants provide a range of activities both above and below wild-type that can be seen in both experimental modes, and the mechanism by which the mutants stimulate branch formation appears to be identical to wild-type.<fig-group><fig id="fig5" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.015</object-id><label>Figure 5.</label><caption><title>diVCA constructs differ in the rate but not the pathway of activity in stimulating branch formation.</title><p>(<bold>A</bold>) Rate (±S.E.) of initiation of daughter filament growth by Arp2/3 complex in the absence or presence of diVCA wild-type and mutant constructs. <italic>k</italic><sub>B</sub>, the second order rate constant for the appearance of branches on existing filaments, per subunit, was calculated from observations of branch formation on existing filaments, as in <xref ref-type="fig" rid="fig2">Figure 2</xref> (‘Materials and methods’). (<bold>B</bold>) Comparison of the time (±S.E.) of daughter filament initiation with the time of diVCA release from the nascent branch for wild-type (data replotted from <xref ref-type="fig" rid="fig2">Figure 2E</xref>) and mutant constructs (<xref ref-type="fig" rid="fig5s3">Figure 5—figure supplement 3</xref>). (<bold>C</bold> and <bold>D</bold>) Example records showing the length of a nucleated daughter filament and the fluorescence intensity from actin, diVCA, and individual tethered Arp2/3 complex molecules, as in <xref ref-type="fig" rid="fig3">Figure 3</xref>. Mutant Cy3-diVCA was 0.5 nM V* in C, or 1.0 nM C* in <bold>D</bold>. The merged fluorescence images in <bold>C</bold> were recorded at the indicated times and the white squares mark the area from which the fluorescence was integrated to produce the intensity records. Scale bar, 1 µm. Both mutants bound readily to tethered Arp2/3 prior to but not after branch formation.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.015">http://dx.doi.org/10.7554/eLife.01008.015</ext-link></p></caption><graphic xlink:href="elife01008f005"/></fig><fig id="fig5s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.016</object-id><label>Figure 5—figure supplement 1.</label><caption><title>Analysis of diVCA mutant activities in bulk actin polymerization assays.</title><p>(<bold>A</bold>) Stimulation of actin filament assembly by 10 nM Arp2/3-SNAP complex activated by 25 nM wild-type or mutant diVCA constructs. Plots show records of pyrene fluorescence in assays containing 2 µM actin, 5% pyrene-labeled. (<bold>B</bold>) Actin nucleation activities from the data in <bold>A</bold> and a replicate experiment (mean ± S.D.; ‘Materials and methods’).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.016">http://dx.doi.org/10.7554/eLife.01008.016</ext-link></p></caption><graphic xlink:href="elife01008fs007"/></fig><fig id="fig5s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.017</object-id><label>Figure 5—figure supplement 2.</label><caption><title>Saturation of stimulation of Arp2/3 complex actin nucleation activity by diVCA constructs.</title><p>(<bold>A</bold>) Actin nucleation activity measured as in <xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1</xref> with 10 nM yeast Arp2/3 complex and specified diVCA construct concentrations. Activity saturates above 50 nM for three constructs; for diVCA-A* activity is too low to detect saturation in this assay. (<bold>B</bold>) Saturation of diVCA and diVCA-A* in competing with stimulation of actin nucleation by VVCA. Mean values from 3–4 replicate measurements. Error bars indicate S.E.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.017">http://dx.doi.org/10.7554/eLife.01008.017</ext-link></p></caption><graphic xlink:href="elife01008fs008"/></fig><fig id="fig5s3" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.018</object-id><label>Figure 5—figure supplement 3.</label><caption><title>Relationship relation between diVCA release from the nascent branch and daughter filament initiation for each of the three mutant diVCA constructs.</title><p>These panels show the data presented in <xref ref-type="fig" rid="fig5">Figure 5B</xref> replotted separately to make the error bars more visible. V*, C*, and A* data sets contain 41, 49, and 27 observations, respectively, where the measured daughter initiation time was >0.1 s. Asterisks mark two observations in which daughter filament initiation occurred significantly before disappearance of the diVCA fluorescent spot. We speculate that in these rare events the diVCA molecule may have become irreversibly crosslinked at or near the branch junction and that spot disappearance is caused by dye photobleaching, not by diVCA release. Consistent with this explanation, both of these outliers show uncharacteristically long diVCA release times.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.018">http://dx.doi.org/10.7554/eLife.01008.018</ext-link></p></caption><graphic xlink:href="elife01008fs009"/></fig></fig-group></p><p>Since high diVCA activity appears to correspond with weak binding to Arp2/3 complex, we next tested the hypotheses that the rate of diVCA release from the nascent branch quantitatively explains the nucleation activities of different mutants. While we already measured the dissociation rates of the diVCA constructs from Arp2/3 complex when the latter is not bound to the side of a mother filament (<xref ref-type="fig" rid="fig4">Figure 4D</xref>; <xref ref-type="table" rid="tbl1">Table 1</xref>), the rates of dissociation from the filament-bound Arp2/3 complex (the nascent branch) might be different.</p><p>Our observations (<xref ref-type="fig" rid="fig2">Figure 2</xref>) showed that once a nascent branch forms, it can have multiple fates (<xref ref-type="fig" rid="fig6">Figure 6A</xref>). Most often, the Arp2/3-diVCA complex simply dissociates from the filament (<xref ref-type="fig" rid="fig6">Figure 6A</xref>, thick red arrow). In a small fraction of the nascent branches, <italic>f</italic><sub>V−</sub>, diVCA departs first, leaving behind Arp2/3 complex bound to mother filament (<xref ref-type="fig" rid="fig6">Figure 6A</xref>, activated complex) where it may subsequently initiate a daughter filament. Furthermore, only a fraction of activated complexes subsequently formed a branch, so that the fraction of nascent branches that successfully produced a daughter filament, <italic>f</italic><sub>B</sub>, is less than <italic>f</italic><sub>V−</sub>. For some of the mutant constructs <italic>f</italic><sub>V−</sub> or <italic>f</italic><sub>B</sub> values (<xref ref-type="fig" rid="fig6">Figure 6B</xref>) could not be unambiguously distinguished from wild type in pairwise comparisons (p=0.05–0.15); others differed significantly (p=0.005–0.05) from wild-type (asterisks in <xref ref-type="fig" rid="fig6">Figure 6B</xref>).<fig-group><fig id="fig6" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.019</object-id><label>Figure 6.</label><caption><title>Release of diVCA from nascent branches is rare and limits the rate of daughter nucleation.</title><p>(<bold>A</bold>) Schematic mechanism of diVCA stimulated branch formation (see text). The key activation step, release of diVCA from the nascent branch, is highlighted. (<bold>B</bold>) Classification of nascent branch fates observed in single molecule experiments (e.g., <xref ref-type="fig" rid="fig2">Figure 2B and C</xref>; <xref ref-type="fig" rid="fig2s3">Figure 2—figure supplement 3</xref>). Overall bar height indicates the fraction (±S.E.) of nascent branches that release diVCA leaving behind a filament-bound Arp2/3 complex. Filled bar height shows the fraction (±S.E.) of nascent branches that nucleate a daughter filament. (<bold>C</bold>) Cumulative lifetime distributions of diVCA molecules on the subset of filament-bound Arp2/3 complexes observed to produce branches in single-molecule experiments. Inset: mean lifetimes (±S.E.). (<bold>D</bold>) Rate constants (±S.E.) for diVCA dissociation from the nascent branch, calculated from the mean lifetimes in <bold>C</bold> and release efficiencies in <bold>B</bold>. (<bold>E</bold>) Correlation between the rate constant of diVCA-stimulated Arp2/3 complex branch nucleation (from <xref ref-type="fig" rid="fig5">Figure 5A</xref>) and the rate constant of diVCA release from the nascent branch (from <bold>D</bold>). Correlation coefficient <italic>r</italic> = 0.9928 is unlikely to arise by coincidence (p=0.0045). Dotted line is a linear fit constrained to pass through the origin.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.019">http://dx.doi.org/10.7554/eLife.01008.019</ext-link></p></caption><graphic xlink:href="elife01008f006"/></fig><fig id="fig6s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.01008.020</object-id><label>Figure 6—figure supplement 1.</label><caption><title>Correlation between the actin nucleation activity of Arp2/3 activated by wild-type and mutant diVCA constructs (from <xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1B</xref>) and the rate constant of diVCA release from the nascent branch (from <xref ref-type="fig" rid="fig4">Figure 4F</xref>).</title><p>Dotted line is a linear fit constrained to pass through the origin. The correlation coefficient r = 0.9598 is unlikely to be coincidental (p=0.027; ‘Materials and methods’). Error bars indicate S.E.s.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.020">http://dx.doi.org/10.7554/eLife.01008.020</ext-link></p></caption><graphic xlink:href="elife01008fs010"/></fig></fig-group></p><p>We also measured the lifetimes of diVCA constructs on the nascent branches that ultimately produced daughter filaments. The mean lifetime (τ<sub>V</sub><sup>*</sup>, equal to the reciprocal of the sum of the rate constants for diVCA departure and Arp2/3 complex dissociation; ‘Materials and methods’) differed little between the different diVCA constructs (<xref ref-type="fig" rid="fig6">Figure 6C</xref>). Only the C* mutant had a significantly shorter mean lifetime on the nascent branch than wild-type diVCA (2 ± 1-fold, p=0.015), whereas the V* and A* mutants did not (p=0.17 and 0.48, respectively). This observation suggests (‘Materials and methods’) that the mutations do not substantially alter the rate of the main pathway of nascent complex breakdown, which is the dissociation of the intact Arp2/3-diVCA complex from filament sides (<xref ref-type="fig" rid="fig6">Figure 6A</xref>, thick red arrow). Instead, the mutations principally affect the fraction of nascent complexes that release diVCA, and are productive in forming branches.</p><p>Based on the measurements of <xref ref-type="fig" rid="fig6">Figure 6B,C</xref>, we can calculate the rate constant for dissociation of wild-type and mutant diVCAs from the nascent branch (<xref ref-type="fig" rid="fig6">Figure 6A</xref>, <italic>k</italic><sub>V−</sub><sup>*</sup>) as <italic>f</italic><sub>V−</sub>/τ<sub>V</sub><sup>*</sup> (<xref ref-type="fig" rid="fig6">Figure 6D</xref>). Strikingly, this rate of release of diVCA from the nascent branch is proportional, within experimental uncertainty (<xref ref-type="fig" rid="fig6">Figure 6E</xref>), to the branch formation rate constant <italic>k</italic><sub>B</sub> measured above (<xref ref-type="fig" rid="fig5">Figure 5A</xref>; <xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1</xref>). Despite the relatively subtle effects of the mutants on the measured rate constants, the high correlation coefficient (<italic>r</italic> = 0.993) is unlikely to be coincidental (p=0.0045; ‘Materials and methods’). This relationship provides strong support for the hypotheses that diVCA release is a prerequisite for daughter filament formation, and that the rate of release of diVCA limits the rate at which nascent branches initiate daughter filament growth.</p></sec></sec><sec id="s3" sec-type="discussion"><title>Discussion</title><p>By placing three different colors of fluorescent labels on Arp2/3 complex, diVCA, and actin, we directly observed the sequence and kinetics of key steps in the branch nucleation pathway. The observations confirm our previous results (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>) that branch nucleation is inefficient even at near-saturating VCA protein concentrations, with only a small fraction of Arp2/3 complex-mother filament associations yielding branches. In contrast to the tight binding of diVCA to Arp2/3 complex in solution, we show that diVCA binding is undetectable when Arp2/3 complex is incorporated into a branch junction.</p><p>Significantly, we observed that the branch formation process is strictly dependent on release of diVCA from the filament-bound Arp2/3-diVCA complex. Taken together with earlier work, the data support a specific mechanism for diVCA-stimulated actin nucleation by Arp2/3 complex in which diVCA plays a dual role (<xref ref-type="fig" rid="fig6">Figure 6A</xref>). Initially, diVCA stimulates assembly of the nascent branch by associating tightly with Arp2/3 complex and actin monomers, which promotes the binding of Arp2/3 complex to the sides of filaments. However, once the nascent branch forms, diVCA plays an inhibitory role: diVCA must dissociate before daughter filament can grow. In mutants that alter the interactions of diVCA with Arp2/3 complex or monomeric actin, alterations of the rate at which diVCA leaves the nascent branch exactly parallel changes in the efficiency of branch formation. These observations strongly suggest that diVCA dissociation is the key, rate-limiting step in daughter filament nucleation.</p><p>The key mechanistic conclusions reached here with dimerized VCA constructs in vitro are likely to apply to the mechanism of activation of branch nucleation by native WASP oligomers in vivo. Data on dimeric VCA showing a minimum crosslinker length for high activity (<xref ref-type="bibr" rid="bib45">Padrick et al., 2011</xref>) strongly suggests that the system passes through a form with two VCAs bound at some point during nucleation. The crosslinkers and spacer sequences used here allow a maximum VCA separation of 122 Å (<xref ref-type="bibr" rid="bib45">Padrick et al., 2011</xref>). During Arp2/3 complex activation in diverse contexts in living cells, WASP family proteins are separated by similar or even smaller distances. For example, when dimerized by activators such as EspFu, the WASP protein VCA regions would be ∼80 Å apart if the proline rich region is considered as a random coil polymer (<xref ref-type="bibr" rid="bib46">Padrick et al., 2008</xref>). Similarly, ActA density on the surface of <italic>Listeria</italic> separates its VCA-like sequences by 19 nm, close enough to function as VCA dimers (<xref ref-type="bibr" rid="bib21">Footer et al., 2008</xref>). N-WASP proteins recruited to moving PIP2-rich vesicles have a high local density, such that their average spacing is <50 Å (<xref ref-type="bibr" rid="bib9">Co et al., 2007</xref>), and in rocketing vesicles, N-WASP proteins are recruited by Nck onto the vesicle surface such that their average separation is ∼100 Å (<xref ref-type="bibr" rid="bib15">Ditlev et al., 2012</xref>). Bzz1 and Cdc15 are the activators of the fission yeast WASP family protein Wsp1 during endocytosis, and they array their SH3 domains (which engage Wsp1) at a density sufficient to bring VCA regions from adjacent Wsp1 proteins within 50–80 Å (<xref ref-type="bibr" rid="bib3">Arasada and Pollard, 2011</xref>).</p><p>The reaction scheme in <xref ref-type="fig" rid="fig6">Figure 6A</xref> describes the essential features of the Arp2/3 complex-dependent filament nucleation process. This mechanism, derived from direct observation of single molecules as opposed to fitting of bulk data, is broadly consistent with proposed kinetic schemes for Arp2/3 complex nucleation of filaments (<xref ref-type="bibr" rid="bib70">Zalevsky et al., 2001</xref>; <xref ref-type="bibr" rid="bib4">Beltzner and Pollard, 2008</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). The dominant pathway is that VCA and actin monomer associate with Arp2/3 complex in solution, and this complex then binds to an existing filament, after which an activation step occurs, which allows the daughter filament to elongate. Here we add two informative points. First, we ascribe a distinct mechanism to the activation step, the release of diVCA. Previously this step had been ascribed to structural rearrangements within the Arp2/3 complex (<xref ref-type="bibr" rid="bib13">Dayel et al., 2001</xref>; <xref ref-type="bibr" rid="bib70">Zalevsky et al., 2001</xref>; <xref ref-type="bibr" rid="bib4">Beltzner and Pollard, 2008</xref>). Second, our measurements indicate that most engagements of diVCA-Arp2/3 complexes with the mother filament are resolved by dissociation of the complex from filament without releasing VCA, an idea distinct from previous models (<xref ref-type="bibr" rid="bib4">Beltzner and Pollard, 2008</xref>). It is possible that the observed inefficiency in branch formation is to allow for positive regulation of branch formation by factors not present in our experiment (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). Consistent with this idea, cortactin has been recently demonstrated to accelerate release of WASP proteins from Arp2/3 complex (<xref ref-type="bibr" rid="bib25">Helgeson and Nolen, 2013</xref>). Since WASP proteins are tethered to the cell membrane, it is also possible that mechanical tension between the filament network and the membrane plays a similar role in promoting WASP release and consequent daughter growth, which might allow alteration of cell motility in response to mechanical stimuli.</p><p>While our scheme encompasses key features of the mechanism of diVCA stimulation of Arp2/3 complex-mediated branch formation, it should be noted that the scheme shown in <xref ref-type="fig" rid="fig6">Figure 6A</xref> is not complete. The clearest indication of this is that the diVCA lifetime distribution observed in <xref ref-type="fig" rid="fig2">Figure 2D</xref> is multi-exponential, whereas the scheme of <xref ref-type="fig" rid="fig6">Figure 6A</xref> predicts only a simple single-exponential distribution. The data can be explained if there are two or more conformations of filament-bound diVCA-Arp2/3 complexes that differ in kinetic stability. This proposal is consistent with our previous kinetic analysis with monomeric VCA (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>) and with demonstrations that Arp2/3 complex exists in multiple conformations in solution (<xref ref-type="bibr" rid="bib24">Goley et al., 2004</xref>; <xref ref-type="bibr" rid="bib53">Rodal et al., 2005</xref>; <xref ref-type="bibr" rid="bib68">Xu et al., 2012</xref>). Formulating a more complete kinetic mechanism of diVCA stimulation that accounts for the multiple diVCA-Arp2/3 complex conformations will require additional data.</p><p>Our model in which association of VCA with the nascent branch inhibits initiation of daughter filament growth is consistent with the known interaction of the WH2 motif (V-region) with actin in a conserved cleft involved in longitudinal filament contacts (<xref ref-type="bibr" rid="bib26">Hertzog et al., 2004</xref>; <xref ref-type="bibr" rid="bib31">Irobi et al., 2004</xref>; <xref ref-type="bibr" rid="bib8">Chereau et al., 2005</xref>). We suspect that inhibition of daughter filament elongation from nascent branches results from the WH2 motif staying engaged with the cleft, occluding addition of the next actin subunit in the daughter filament (<xref ref-type="bibr" rid="bib19">Egile et al., 1999</xref>; <xref ref-type="bibr" rid="bib28">Higgs et al., 1999</xref>). Further support for this model is realized by recent reports showing that VCA peptides covalently crosslinked to actin monomers are inactive in stimulating Arp2/3 complex-mediated actin nucleation (<xref ref-type="bibr" rid="bib6">Boczkowska et al., 2008</xref>; <xref ref-type="bibr" rid="bib63">Ti et al., 2011</xref>). Moreover, the C-region of VCA likely occupies a similar location on Arp2 and Arp3 that the V-region occupies on actin, preventing insertion of subdomain 2 of the actin recruited by the V-region into the cleft on the bound Arp subunit (<xref ref-type="bibr" rid="bib26">Hertzog et al., 2004</xref>; <xref ref-type="bibr" rid="bib31">Irobi et al., 2004</xref>; <xref ref-type="bibr" rid="bib8">Chereau et al., 2005</xref>; <xref ref-type="bibr" rid="bib6">Boczkowska et al., 2008</xref>; <xref ref-type="bibr" rid="bib45">Padrick et al., 2011</xref>; <xref ref-type="bibr" rid="bib63">Ti et al., 2011</xref>). Thus, the molecular structures are consistent with the observed dual function of VCA in both stimulating daughter nucleation (by recruiting actin monomers to Arp2/3 complex) and in suppressing daughter nucleation (by blocking assembly of the daughter filament). These structural models are also consistent with the low affinity of diVCA for branch junctions. If neither the V-region nor the C-region can bind the branch (when the actin and Arp clefts are occupied by D-loops of daughter filament actins), then the affinity of VCA for the branch may be of the order of that observed for isolated A-region. This affinity has been reported to be approximately 9 μM (<xref ref-type="bibr" rid="bib41">Marchand et al., 2001</xref>), consistent with the lack of observed binding under the conditions necessary for single molecule observations.</p><p>In cells, active WASP proteins are predominantly tethered to membranes. Taken together, our data suggests a straightforward mechanism by which WASP regulation of Arp2/3 complex can cause branches to form preferentially at membranes without having membrane attachment restrict network growth (<xref ref-type="fig" rid="fig7">Figure 7</xref>). Binding of Arp2/3 complex to VCA dimers in the absence of mother filaments is of high affinity and of long lifetime, allowing dimerization of WASP by upstream activators to promote recruitment of Arp2/3 complex to the membrane surface (<xref ref-type="fig" rid="fig7">Figure 7</xref>, ‘WASP dimer association with Arp2/3 complex’). Association with VCA promotes filament binding by Arp2/3 complex (‘Arp2/3 complex engagement of filament’), but that process is readily reversible and the nascent branch most frequently is simply lost through dissociation (‘Filament release’). More rarely, WASP detaches (‘WASP release’) leaving Arp2/3 complex associated with the mother filament. This severs the direct linkage to the membrane, and only then allows the daughter filament to nucleate and grow (‘Initiation of daughter filament elongation’). Our data show that once the branch forms, WASP does not rebind. The dissociation of WASP prior to nucleation and the lack of rebinding provide an appealing explanation for how WASP dimers stimulate branched network formation at the membrane without interfering with network growth.<fig id="fig7" position="float"><object-id pub-id-type="doi">10.7554/eLife.01008.021</object-id><label>Figure 7.</label><caption><title>Model of WASP-Arp2/3 complex stimulated actin branch formation at cell membranes (see text).</title><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.01008.021">http://dx.doi.org/10.7554/eLife.01008.021</ext-link></p></caption><graphic xlink:href="elife01008f007"/></fig></p></sec><sec id="s4" sec-type="materials|methods"><title>Materials and methods</title><sec id="s4-1"><title>Protein design, labeling, and purification</title><p>The bis-maleimide crosslinked N-WASP VCA dimers were prepared from human N-WASP VCA (amino acid residues 432–505), with the sequence CGGSGGSGGSGGS appended at the N-terminus. This sequence was generated by PCR using overlapping primers encoding the N-terminal extension. The resulting PCR product was cloned into a plasmid derived from pGEX2T with a TEV protease site between the sequence for GST and the multiple cloning site (which was modified to include a 5′ NdeI site and 3′ BamHI site). Mutations were introduced by designing overlapping primers containing the desired mutations, which were used in PCR amplification of the region 5′ to the modification and 3′ to the modification. These products were used as template in a second round of PCR that fused them into a single cassette containing the desired mutation. The resulting product was digested with NdeI and BamHI and ligated into the above described expression vector. Cloned products were verified by DNA sequencing.</p><p>Crosslinked VCA dimers were prepared using a common strategy. Proteins were expressed in <italic>Escherichia coli</italic> BL21(DE3) T1<sup>R</sup> cells using IPTG induction at 37°C for 3 hr. Cells were harvested, resuspended in 25 ml of buffer (20 mM Tris pH 8.0, 50 mM NaCl, 2 mM EDTA, 1 mM DTT and 1 mM PMSF) per l of culture, and frozen at −80°C until needed. Cells were lysed using a cell disruptor and clarified by centrifugation at 19,500 rpm in a JA25.50 rotor (Beckman Coulter Inc., Brea, CA). Clarified lysate was purified using DEAE Sepharose FF (GE Healthcare Biosciences, Pittsburgh, PA) ion exchange chromatography, followed by Glutathione Sepharose (GE Healthcare Biosciences) chromatography, followed by cleavage with TEV protease. Cleaved VCA peptides were purified by SOURCE 15Q ion exchange chromatography. Next, the pooled VCA was concentrated and the DTT removed, by dilution and passing over a 0.5 ml SOURCE 15Q column in buffer lacking reducing agent. BMCy3 (#C959070, Toronto Research Chemicals, Toronto, Canada) was prepared as a 20 mM stock in anhydrous DMSO. VCA materials were quantified by absorption at 280 nm, and BMCy3 was added to a final concentration of one equivalent (about 80 μM depending on the preparation), split over three additions separated by 10 min incubations at room temperature. After a final 20 min incubation at room temperature, the reaction was quenched with 2 mM DTT, and the reaction was purified using SOURCE 15 Q ion exchange chromatography, followed by Superdex 200 gel filtration chromatography (GE Healthcare Biosciences). Purification was tracked using SDS-PAGE analysis, and concentrations were determined from absorption at 552 nm, using an extinction coefficient of 150,000 M<sup>−1</sup> cm<sup>−1</sup>.</p><p>GST-VCA and GST-VVCA were produced from vectors lacking the CGGSGGSGGSGGS N-terminal extension. Expression and purification followed a similar protocol as for the crosslinked VCA dimers, with the modification that GST was not cleaved from the product following Glutathione Sepharose purification. Purification continued with SOURCE 15Q ion exchange chromatography and Superdex 200 gel filtration chromatography. Concentrations were measured using absorbance at 280 nm.</p><p>The fluorescence anisotropy probe for actin binding, VCA-AF488, was produced from a similar vector encoding N-WASP VCA amino acid residues 430–505 (with the introduced mutations S430C and C431A, and lacking the CGGSGGSGGSGGS extension). Expression and purification was the same as for the VCA dimers, through the SOURCE 15Q step in buffer without reducing agent. At this point, 1 ml of 40 μM VCA was labeled with 150 μM AlexaFluor488 maleimide (#A-10254, Molecular Probes/Life Technologies, Grand Island, NY, 40 mM stock in anhydrous DMSO). After 2 hr reaction at room temperature, the reaction was quenched with 2 mM DTT and VCA-AF488 was purified by SOURCE15Q and Superdex 75 chromatography. Labeling efficiency was judged to be nearly 100% from a shift in mobility by SDS-PAGE (small shift judged using a reference of flanking unlabeled material). Concentration was measured as the absorbance at 492 nm, using an extinction coefficient of 71,000 M<sup>−1</sup> cm<sup>−1</sup>.</p><p>The fluorescence anisotropy probe for Arp2/3 complex binding, VVCA-A462C-A594, was produced by using a similar method to VCA-AF488 with two modifications. First, the vector used encoded N-WASP VCA amino acid residues 393–505, with two mutations, C431A and A462C. Second, the VCA was labeled with AlexaFluor 594 maleimide (#A-10256, Molecular Probes/Life Technologies, 40 mM stock in anhydrous DMSO), instead of AlexFluor 488 maleimide. Quantification was performed using absorbance at 588 nm and an extinction coefficient of 96,000 M<sup>−1</sup> cm<sup>−1</sup>. The fluorescence anisotropy probe used for examining simultaneous binding of yeast Arp2/3 complex and <italic>Drosophila</italic> 5C actin, VCA-AF594, was produced by using a similar method to VCA-AF488, but substituting the AlexaFluor 594 maleimide (mentioned above) for AlexaFluor 488 maleimide.</p><p><italic>S. cerevisiae</italic> Arp2/3 complex (used in <xref ref-type="fig" rid="fig2s2 fig2s4">Figure 2—figure supplements 2 and 4</xref>, <xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2</xref>) was purified from commercial baker’s yeast (#05020, Red Star Yeast Company, Milwaukee, WI) using a method adapted from published protocols (<xref ref-type="bibr" rid="bib19">Egile et al., 1999</xref>; <xref ref-type="bibr" rid="bib37">Lechler et al., 2001</xref>) with added SOURCE 15Q and Superdex 200 chromatography steps (<xref ref-type="bibr" rid="bib17">Doolittle et al., 2013b</xref>). Endogenous bovine Arp2/3 complex (used in <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1</xref>, <xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1B</xref>) was purified from calf thymus using previously described methods (<xref ref-type="bibr" rid="bib28">Higgs et al., 1999</xref>; <xref ref-type="bibr" rid="bib18">Doolittle et al., 2013c</xref>). SNAP-tagged Arp2/3 complex was purified from recombinant <italic>S. cerevisiae</italic> and labeled as previously described (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>), except that labeling used BG-649 and BG-649-PEG-biotin (‘Synthesis of SNAP-tag substrates BG-649 and BG-649-PEG-biotin’ below) to yield Arp2/3-SNAP649 and Arp2/3-SNAP649-biotin.</p><p>Rabbit muscle actin, pyrene-labeled actin, AlexaFluor488-labeled actin, and biotinylated actin were purified as described (<xref ref-type="bibr" rid="bib60">Spudich and Watt, 1971</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). Non-polymerizable <italic>Drosophila melanogaster</italic> 5C actin was prepared according to established methods (<xref ref-type="bibr" rid="bib32">Joel et al., 2004</xref>), but with the mutation D287A/V288A/D289A instead of A204E/P243K. Characterization of this mutation will be described elsewhere (<xref ref-type="bibr" rid="bib69">Zahm et al., 2013</xref>).</p></sec><sec id="s4-2"><title>Synthesis of SNAP-tag substrates BG-649 and BG-649-PEG-biotin</title><p>In the syntheses, commercially available compounds were used without further purification and reaction yields are not optimized. Reversed-phase high-performance liquid chromatography (HPLC) was performed on Agilent LC/MS Single Quad System 1200 Series (analytical) and Agilent 1100 Preparative-scale Purification System (semi-preparative). Analytical HPLC was performed on Waters Atlantis T3 C18 column (2.1 × 150 mm, 5 µm particle size) at a flow rate of 0.5 ml/min with a binary gradient from Phase A (0.1 M triethyl ammonium bicarbonate [TEAB] or 0.1% trifluoroacetic acid [TFA] in water) to Phase B (acetonitrile) and monitored by absorbance at 280 nm. Semi-preparative HPLC was performed on VYDAC 218 TP series C18 polymeric reversed-phase column (22 × 250 mm, 10 µm particle size) at a flow rate of 20 ml/min. Mass spectra were recorded by electrospray ionization (ESI) on an Agilent 6210 Time-of-Flight (TOF) or 6120 Quadrupole LC/MS system.</p><p>BG-649 was prepared by reacting the building block BG-NH<sub>2</sub> (New England Biolabs, Ipswich, MA) with the dye <italic>N</italic>-hydroxysuccinimide ester DY-649 NHS (Dyomics GmbH, Jena, Germany) as described previously (<xref ref-type="bibr" rid="bib34">Keppler et al., 2004</xref>). BG-NH<sub>2</sub> (0.54 mg, 2.0 µmol) was dissolved in anhydrous DMF (0.5 ml). DY-649 NHS (2.0 mg, 2.0 µmol) and triethylamine (0.4 µl, 3.0 µmol) were added and the reaction mixture stirred overnight at room temperature. The solvent was removed under vacuum and the product purified by reversed-phase HPLC using 0.1 M TEAB/acetonitrile gradient. Yield: 74%. BG-649: ESI-MS <italic>m/z</italic> 1095.2 [M-H]<sup>−</sup> (calculated for C<sub>48</sub>H<sub>56</sub>N<sub>8</sub>O<sub>14</sub>S<sub>4</sub>, <italic>m/z</italic> 1095.3).<fig id="fig8" position="anchor"><graphic xlink:href="elife01008f008"/></fig></p><p>The bifunctional BG-649-PEG-Biotin (which includes both a DY-649 dye and a biotin moiety) was prepared by successive couplings of commercially available α-<italic>N</italic>-Fmoc-ε-<italic>N</italic>-Dde-lysine (Merck KGaA, Darmstadt, Germany) with BG-NH<sub>2</sub> (New England Biolabs, Ipswich, MA), <italic>N</italic>-(+)-biotin-6-aminocaproic acid <italic>N</italic>-succinimidyl ester (Sigma-Aldrich, St. Louis, MO) and DY-649 NHS (Dyomics GmbH, Jena, Germany) according to synthetic route described previously (<xref ref-type="bibr" rid="bib35">Kindermann et al., 2004</xref>). BG-649-PEG-Biotin was synthesized as follows: BG-NH<sub>2</sub> (250.0 mg, 0.92 mmol) was dissolved in anhydrous DMF (8 ml). HBTU (<italic>N,N,N′,N</italic>′-Tetramethyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate) (368.0 mg, 0.97 mmol), triethylamine (135 µl, 0.97 mmol), and Fmoc-Lys(Dde)-OH (515.5 mg, 0.97 mmol) were added and the reaction mixture stirred overnight at room temperature. The reaction mixture was poured onto water (80 ml). The white solid was collected by filtration, washed twice with water, and dried in dessicator under vacuum overnight. Yield: 91%. BG-Lys(Dde)-Fmoc (50 mg, 63.7 µmol) was dissolved in anhydrous in DMF (5 ml). Et<sub>2</sub>NH (19.8 µl, 191.1 µmol) was added and the reaction mixture stirred overnight at room temperature. The solvent was removed under vacuum for 6 hr and the residue dissolved in DMF (3 ml). Fmoc-12-amino-4,7,10-trioxadodecanoic acid (29.7 mg, 66.9 µmol), triethylamine (26.6 µl, 191.1 µmol) and HBTU (36.3 mg, 95.6 µmol) were added and the reaction mixture stirred for 1 hr at room temperature. The reaction completion was monitored by LC/MS. The solvent was removed under vacuum and the product purified by reversed-phase HPLC using 0.1 M TEAB/acetonitrile gradient. Yield: 50%. BG-Lys(Dde)-PEG-NHFmoc: ESI-MS <italic>m/z</italic> 988.4 [M+H]<sup>+</sup> (calculated for C<sub>53</sub>H<sub>65</sub>N<sub>9</sub>O<sub>10</sub>, <italic>m/z</italic> 988.5). BG-Lys(Dde)-PEG-NHFmoc (31.6 mg, 31.9 µmol) was dissolved in anhydrous DMF (2 ml). Et<sub>2</sub>NH (9.9 µl, 95.7 µmol) was added and the reaction mixture stirred overnight at room temperature. The solvent was removed under vacuum for 6 hr and the residue dissolved in DMF (2 ml). <italic>N</italic>-(+)-biotin-6-aminocaproic acid NHS (14.5 mg, 31.9 µmol) and triethylamine (13.3 µl, 95.7 mmol) were added and the reaction mixture stirred overnight at room temperature. The reaction completion was monitored by LC/MS. A 2% solution of hydrazine in DMF (0.5 ml) was added and the reaction mixture stirred for 1 hr at room temperature. The solvent was removed under vacuum and the product purified by reversed-phase HPLC using 0.1% TFA in water/acetonitrile gradient. Yield: 75%. BG-Lys(NH<sub>2</sub>)-PEG-Biotin: ESI-TOFMS <italic>m/z</italic> 939.4873 [M-H]<sup>−</sup> (calculated for C<sub>44</sub>H<sub>68</sub>N<sub>12</sub>O<sub>9</sub>S, <italic>m/z</italic> 939.4880). BG-Lys(NH<sub>2</sub>)-PEG-Biotin (2.3 mg, 2.13 µmol) was dissolved in anhydrous DMF (1 ml). DY-649 NHS (2.1 mg, 2.13 µmol) and triethylamine (0.45 µl, 3.2 µmol) were added and the reaction mixture stirred overnight at room temperature. The solvent was removed under vacuum and the product purified by reversed-phase HPLC using 0.1 M TEAB/acetonitrile gradient. Yield: 73%. BG-649-PEG-Biotin: ESI-TOFMS <italic>m/z</italic> 882.3166 [M-2H]<sup>2−</sup> (calculated for C<sub>79</sub>H<sub>110</sub>N<sub>14</sub>O<sub>22</sub>S<sub>5</sub>, <italic>m/z</italic> 882.3188).<fig id="fig9" position="anchor"><graphic xlink:href="elife01008f009"/></fig></p></sec><sec id="s4-3"><title>Actin assembly kinetics</title><p>Actin assembly kinetics measurements were performed in a fashion similar to previously published (<xref ref-type="bibr" rid="bib10">Cooper et al., 1983</xref>; <xref ref-type="bibr" rid="bib12">D’Agostino and Goode, 2005</xref>; <xref ref-type="bibr" rid="bib46">Padrick et al., 2008</xref>, <xref ref-type="bibr" rid="bib45">2011</xref>; <xref ref-type="bibr" rid="bib16">Doolittle et al., 2013a</xref>). Briefly, a rabbit muscle actin stock (5% pyrene labeled in <xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1</xref> or 10% pyrene labeled in <xref ref-type="fig" rid="fig2s1 fig2s2">Figure 2—figure supplement 1 and 2</xref>, and <xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2</xref>) in buffer G (2 mM Tris pH 8.0, 200 μM CaCl<sub>2</sub>, 1 mM NaN<sub>3</sub>, 100 μM ATP, 0.5 mM DTT) was combined with 1/10th volume 10 mM EGTA, 1 mM MgCl<sub>2</sub>, and then with enough buffer G-Mg (same as buffer G but substituting MgCl<sub>2</sub> for CaCl<sub>2</sub>) to dilute the overall actin concentration to 4 μM. After a 2 min incubation in this buffer, the actin solution was combined with an equal volume of Arp2/3 complex and VCA materials in double strength KMEI buffer (such that the final solution was 10 mM imidazole pH 7.0, 50 mM KCl, 1 mM EGTA, 1 mM MgCl<sub>2</sub>, 0.5 mM DTT, with one half concentration of buffer G carrying over from the actin solution). For experiments designed to enable quantitative comparison of actin assembly rates (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1</xref>) to results from single molecule analysis (<xref ref-type="fig" rid="fig6">Figure 6D</xref>, <xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1</xref>) the buffer was supplemented with additional components (final concentrations: 10 mM DTT, 0.2 mM ATP, 15 mM glucose, 0.02 mg/ml catalase, 0.1 mg/ml glucose oxidase, 0.1% bovine serum albumin, 1 mM 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), 1 mM 4-nitrobenzyl alcohol, and 0.5 mM propyl gallate) so as to maintain similar buffer conditions in the two experiments. Reactions were then immediately placed in a cuvette in a PTI Quantamaster spectrofluorometer. Pyrene-actin fluorescence was observed over time by exciting at 365 nm and observing at 407 nm. For some assays (<xref ref-type="fig" rid="fig2s1 fig2s2">Figure 2—figure supplement 1 and 2</xref>, <xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2</xref>), 10% pyrene labeling was used, and the reactions were placed into 96-well plates and followed using a plate reader (VarioSkan Flash, Thermo Scientific, Hudson, NH).</p><p>Actin filament barbed end concentrations were evaluated at 50% polymerization (<italic>t</italic><sub>50</sub>) by first scaling the pyrene fluorescence intensity over the full range of filament concentrations (0–1.9 µM), then fitting the slope (actin assembly rate) between 42% and 58% polymerization. These rates were then divided by the approximate filament elongation rate (10 subunits per second) to obtain the barbed end concentrations (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1B</xref>). Fold stimulation of branch formation by diVCA was calculated by subtracting the barbed ends created by actin alone from the total barbed ends formed in the presence of Arp2/3 complex, then dividing the concentrations of these excess barbed ends formed in the presence diVCA by those formed in the absence of diVCA.</p></sec><sec id="s4-4"><title>Fluorescence anisotropy</title><p>Binding of VCA to actin was monitored by fluorescence anisotropy (<xref ref-type="fig" rid="fig2s4">Figure 2—figure supplement 4</xref>, <xref ref-type="fig" rid="fig4">Figure 4B</xref>). 20 nM N-WASP VCA-AF488 was mixed with the indicated concentrations of actin and Cy3-diVCA competitor with buffer additions to bring the final mixture to 10 mM imidazole pH 7.0, 50 mM KCl, 1 mM EGTA, 1 mM MgCl<sub>2</sub>, 0.5 mM DTT, 0.1 mM ATP, and 1/10th residual concentration of buffer G. The mixtures were incubated for 3 min at room temperature prior to placing in a 3 mm by 3 mm cuvette in a T-form PTI Quantamaster Spectrafluorometer, equipped with Glan-Thompson polarizers. Emission intensity was averaged for 3 min and converted to anisotropy values, after correcting for background signal intensity and G-factor. Competition binding experiments (including N-WASP diVCA competition) were performed with 20 nM N-WASP VCA-AF488, 200 nM rabbit muscle actin, and the indicated concentrations of VCA dimers. Binding isotherms (both direct and competition binding) were fit to a complete competition binding solution for a single site receptor, using Levenberg-Marquardt nonlinear least squares methods, with bound and free fluorescence anisotropy as fit parameters. Direct binding isotherms were fit with the concentration of competitor ligand set to zero. Fitting of the competition-binding isotherm used the direct binding <italic>K</italic><sub>D</sub> that was obtained from fitting the direct binding isotherm. Fit values for free fluorescence anisotropy values were similar to the actual free anisotropy, and the values of free and bound fluorescence anisotropy determined from competition binding experiments was similar to that of the direct binding system.</p><p>Binding of VCA to bovine Arp2/3 complex in solution (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1D–E</xref>, <xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1B</xref>) was monitored by fluorescence anisotropy using the same basic protocol as for actin binding, with a few modifications. First, the reporter was 20 nM VVCA-A462C-AF594. Next, as there was no actin present, there was no residual buffer G in the mixture. Finally, the mixture was incubated for 10 min prior to measurement, and the fluorescence intensity data averaged for 5 min. Solution binding of VCA to non-polymerizable <italic>Drosophila</italic> 5C actin, and to yeast Arp2/3 complex, was followed by fluorescence anisotropy. Acquisition and processing was similar to the actin binding assay described above, except 10 nM VCA-A594 was used, and either <italic>Drosophila</italic> 5C actin was added or endogenous budding yeast Arp2/3 complex was added. From fitting the Arp2/3 complex titration curve, 300 nM was judged to have essentially complete binding, and this was added to a separate titration of actin with Arp2/3 complex present. In fitting all three data sets, a single binding site was assumed on VCA.</p></sec><sec id="s4-5"><title>Colocalization single molecule spectroscopy (CoSMoS)</title><p>Single molecule imaging was performed on a custom built multi-wavelength total internal reflection fluorescence (TIRF) microscope, described previously (<xref ref-type="bibr" rid="bib22">Friedman et al., 2006</xref>; <xref ref-type="bibr" rid="bib29">Hoskins et al., 2011</xref>; <xref ref-type="bibr" rid="bib23">Friedman and Gelles, 2012</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). Briefly, the microscope design permitted selective fluorescence excitation of molecules immobilized on the surface of a glass flow chamber (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>) using three lasers at wavelengths 488 nm, 532 nm, and 633 nm. Emissions were split into short (<635 nm) and long (>635 nm) wavelengths and focused on different locations on the camera, to allow for simultaneous acquisition of two-color fluorescence (<xref ref-type="bibr" rid="bib22">Friedman et al., 2006</xref>). Experiments were performed using two different modes of operation described below, one for recording Arp2/3 complex and diVCA interacting with tethered filaments, and another for recording diVCA and filaments interacting with tethered Arp2/3 complexes.</p><p>All experiments were performed in TIRF buffer: 50 mM KCl, 1 mM MgCl<sub>2</sub>, 1 mM EGTA, 10 mM imidazole pH 7.2, 10 mM DTT, 0.2 mM ATP, 15 mM glucose, 0.02 mg/ml catalase, 0.1 mg/ml glucose oxidase, 0.1% bovine serum albumin (BSA), and 2% dextran. To suppress blinking of the fluorophores, TIRF buffer was supplemented with a mixture of triplet state quenchers: 1 mM 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), 1 mM 4-nitrobenzyl alcohol, and 0.5 mM propyl gallate, which were dissolved in DMSO at a 200× stock concentration. Prior to flow chamber assembly (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>), surfaces were first cleaned by sonication in detergent (2% Micro-90, 1 hr), ethanol (1 hr), KOH (0.1 M, 30 min), and deionized water (10 min), then coated with a mixture of methoxy-poly (ethylene glycol)-silane and biotin-poly (ethylene glycol)-silane (mPEG-sil-2000 and biotin-PEG-sil-3400; Laysan Bio Inc., Arab, AL) in 80% ethanol, pH 2 HCl, baked overnight at 70°C. Immediately before each experiment, the chambers were incubated in 0.5 mg/ml BSA and 0.03 mg/ml streptavidin in successive washes.</p><p>The oxygen scavenging activity of the glucose/glucose oxidase/catalase system in our buffers resulted in a gradual decrease in pH over time through the course of our experiments (<xref ref-type="bibr" rid="bib56">Shi et al., 2010</xref>). We measured this pH change in our microscope flow chambers using a pH-sensitive fluorophore (SNARF-4F, pKa ∼6.4; Molecular Probes/Life Technologies), by recording the ratio of fluorescence emission intensities in the two color channels of our microscope, exciting at fixed wavelength (533 nm). The ratiometric response from 0.1 µM SNARF-4F was first calibrated using strong buffers (0.1 M phosphate buffer) over a pH range of 5.8–8.0 (increment 0.2). The time dependent pH of our TIRF buffer was then measured and found to decrease from 7.0 to 6.5 over the typical time course of our experiments (∼30 min). The effect of pH on Arp2/3 complex mediated actin filament assembly was also tested and showed that decreasing the pH from 7 to 6 increased nucleation twofold without VCA, while stimulation by diVCA and mutants differed by <1.5-fold. Thus, we expect that the buffer conditions used for the single molecule experiments had only a minor influence on diVCA stimulation of branch formation by Arp2/3 complex.</p><p>For tethered-filament experiments, a mixture of 10% AF488 labeled actin, 1% biotinated actin, and unlabeled actin was allowed to polymerize for 2–8 hr at 3 μM in TIRF buffer without dextran. Preassembled filaments were then diluted 40-fold, flowed into the microscope observation chamber coated with a 1:100 mixture of biotin-PEG-silane:PEG-silane, and allowed to adhere to the surface. The chamber was then rinsed with TIRF buffer and the reaction mixture was introduced: 1 μM actin (10% AF488 labeled), 5 nM Arp2/3-SNAP649, and 5 nM Cy3-diVCA. Data was recorded for 15–30 min in cycles of a single 50 ms frame of fluorescence emission using 488 nm laser excitation (to image actin filaments) followed by continuous acquisition of 200 frames (50 ms per frame) of emissions using dual 532 nm and 633 nm excitation (to image Arp2/3 complex and diVCA). Each cycle was followed by a ∼1 s delay during which the focus was automatically adjusted (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>), bringing the total cycle interval to ∼12 s.</p><p>For tethered-Arp2/3 experiments, 1 nM Arp2/3-SNAP-649-biotin was introduced into a microscope observation chamber coated with a 1:2000 mixture of biotin-PEG-silane:PEG-silane, and allowed to adhere to the surface. The chamber was then rinsed with TIRF buffer and the reaction mixture was introduced: 1 μM actin (10% AF488 labeled) and 0.1–1 nM Cy3-diVCA. Data was recorded for 30–50 min in cycles of a single frame of fluorescence emission using dual 488 nm and 633 nm excitation (to image actin filaments and tethered-Arp2/3) followed by continuous acquisition of emissions using 532 nm excitation (to image VCA). For these experiments the frame duration was varied in the range 0.05–1 s and the number of frames per cycle was adjusted such that each cycle interval was 14–28 s. The frame duration was increased to allow for the use of lower power excitation, so as to characterize and correct for the effect of photobleaching on the observed lifetimes of Cy3-diVCA bound to tethered Arp2/3 complexes (<xref ref-type="fig" rid="fig4s2">Figure 4—figure supplement 2D</xref>).</p><p>For each experiment, we calibrated the transformation matrix for superimposing images from the split emission recordings by imaging fluorescent beads that emitted in both the short- and long-wavelength color channels of the microscope. This transformation corrected for the translocation of the two color images on the camera, as well as differences in the rotation and magnification in the two channels. For records of colocalization on tethered Arp2/3 complexes we also corrected for drift in the microscope stage using automated tracking of the centers of 2D-Gaussian-fit fluorescence emissions from individual Arp2/3-SNAP649-biotin molecules stably tethered to the microscope slide.</p></sec><sec id="s4-6"><title>CoSMoS data analysis</title><p>Image processing and kinetic analysis was performed in ImageJ (National Institutes of Health, Bethesda, MD), Origin (OriginLab Corp., Northampton, MA) and with custom programs developed in Matlab (Mathworks, Natick, MA).</p><sec id="s4-6-1"><title>Colocalization dynamics before, during, and after branch formation</title><p>To monitor Arp2/3 complex and diVCA colocalization dynamics on the sides of tethered filaments during branch formation (<xref ref-type="fig" rid="fig2">Figure 2</xref>), we first identified sites of branch formation in image sequences and integrated the fluorescence intensities in a 4 × 4 pixel (0.54 × 0.54 µm) area centered on the branch junctions. Fluorescence intensity traces were created from data in the long- and short-wavelength channels of the microscope, corresponding to Arp2/3-SNAP649 and Cy3-diVCA emissions, respectively. Daughter filament lengths were measured from AF488-actin images by manual tracing the filament contours, and nucleation times were calculated by extrapolating from linear fits of elongation data (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). Errors in measurement of filament lengths, possibly due to non-uniform incorporation of labeled subunits or fluctuations of filament ends away from the microscope slide (and therefore out of the TIRF excitation field), sometimes led to nucleation times that preceded arrival of Arp2/3 complex on the mother filament (typically ∼20% of nucleation events). To ensure that nucleation times were accurate, we restricted our analysis in <xref ref-type="fig" rid="fig2">Figure 2E</xref> to nucleation events that occurred >0.1 s after Arp2/3 complex appeared on the mother filament side.</p><p>To monitor colocalization dynamics of diVCA and filaments on tethered Arp2/3 complexes, we first identified the locations where individual Arp2/3 complexes were stably tethered and integrated the fluorescence intensities at those locations in both emission channels. To specifically analyze diVCA-Arp2/3 interactions when Arp2/3 complex is not bound to filament sides, the Arp2/3-SNAP649-biotin and AF488-actin emission were used to identify periods where each Arp2/3 complex was tethered and no filament colocalized. diVCA binding and dissociation kinetics were then analyzed during those periods at each tethered-Arp2/3 location, as described below. In the rare cases when a mother filament colocalized with tethered-Arp2/3 and nucleation occurred, daughter filament lengths were measured (as above) and the interactions of diVCA with branch junctions were separately analyzed, in periods following nucleation during which tethered-Arp2/3 was continuously observed.</p></sec><sec id="s4-6-2"><title>Binding and dissociation kinetics</title><p>To detect times at which diVCA and Arp2/3 complex associated and detached from tethered filament sides, as well as times of association and dissociation of diVCA from tethered Arp2/3 complexes, a combination of automated colocalization detection algorithms (<xref ref-type="bibr" rid="bib11">Crocker and Grier, 1996</xref>; <xref ref-type="bibr" rid="bib23">Friedman and Gelles, 2012</xref>; <xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>) and manual inspection of images were used. First, the spatial and temporal coordinates of single molecule fluorescence were detected using custom particle tracking software (<xref ref-type="bibr" rid="bib11">Crocker and Grier, 1996</xref>), applied to image sequences that had been filtered by averaging of sequential frames (5-frame sliding average for Arp2/3 complex, 2-frame sliding average for diVCA). These coordinates were then compared to manually selected target sites: either a mask that circumscribed an observed tethered filament or the location of a tethered Arp2/3 complex. Colocalization was scored if the coordinates agreed within ∼0.8 µm from the center line of a tethered filament mask or within ∼0.3 µm of a tethered-Arp2/3 location. These colocalization records were used in combination with integrated fluorescence intensity traces and image sequences (not subjected to adjacent frame averaging) to guide the ultimate discrimination of bound and dissociated intervals.</p><p>Quantification of the kinetics of diVCA-Arp2/3 and Arp2/3-filament interactions was achieved by fitting distributions of bound and dissociated intervals to one-, two-, or three-exponential functions, as described previously (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>). Fits were performed with maximum likelihood algorithms and errors were determined by bootstrapping. The times until first appearance of an Arp2/3 complex on resolvable (0.4 µm) segments of tethered filaments were fit with a single-exponential function to find the second order rate constants for filament side binding (<italic>k</italic><sub>A+</sub>; <xref ref-type="table" rid="tbl1">Table 1</xref>). The distribution of lifetimes of coincidently bound Arp2/3 complex and diVCA on filament sides were fit with three- and two-exponential functions, respectively (<xref ref-type="fig" rid="fig2">Figure 2D</xref>; see below for how coincidence was determined). Given the limited number of observations of branch formation events (<70 for each diVCA construct) the dwell times of diVCA on filament-bound Arp2/3 complexes prior to nucleation were not fit to an exponential decay, rather, they were averaged to determine the mean lifetime of diVCA on the nascent branch (τ<sub>V</sub><sup>*</sup>; <xref ref-type="fig" rid="fig6">Figure 6C</xref>). For tethered-Arp2/3 experiments, the second order rate constant for diVCA binding was obtained from single-exponential fits to the intervals where tethered-Arp2/3 was unoccupied by diVCA (<italic>k</italic><sub>V+</sub>; <xref ref-type="fig" rid="fig4s2">Figure 4—figure supplement 2A–C</xref>), whereas the characteristic lifetimes were obtained from double-exponential fits to the distributions of intervals where tethered-Arp2/3 was occupied by diVCA (τ<sub>V1</sub>, τ<sub>V2</sub>; <xref ref-type="fig" rid="fig4">Figure 4D</xref>). There was no evidence that the biexponential distribution of diVCA-Arp2/3 complexes reflects two distinct forms of Arp2/3 complex that do not interconvert on the timescale of our experiment (∼30 min). This was assessed for the V* mutant by calculating the fraction of diVCA-V*-Arp2/3 complexes that lasted <10 s for each tethered Arp2/3 complex observed, which showed a single distribution centered at ∼50%, consistent with the ensemble distribution (<xref ref-type="fig" rid="fig4">Figure 4D</xref>). The effect of photobleaching on the apparent dissociation rate of Cy3-diVCA constructs was assessed and corrected by analyzing the dependence of τ<sub>V2</sub> on the excitation laser power (<xref ref-type="fig" rid="fig4s2">Figure 4—figure supplement 2D</xref>). Laser power was measured by intercepting the beam path between dichroic mirror D3 and mirror M5 (Figure 1 in <xref ref-type="bibr" rid="bib22">Friedman et al., 2006</xref>). All binding rates and lifetime distributions were corrected for non-specific binding of Arp2/3 complex or diVCA to the microscope slide, as described previously (<xref ref-type="bibr" rid="bib59">Smith et al., 2013</xref>).</p></sec><sec id="s4-6-3"><title>Arp2/3-diVCA coincidence, co-release, and branching efficiencies on filament sides</title><p>For each Arp2/3 complex detected on the side of a tethered filament, we evaluated what fraction arrived complexed with diVCA (<italic>f</italic><sub>AV</sub>). Coincident arrival of Arp2/3 complex and diVCA was determined if a diVCA appeared within 0.27 µm (2 pixels) from the location and within 0.15 s (3 frames) of the arrival time of the Arp2/3 complex, given conservative estimates of spatial and temporal resolutions of the frame-averaged dual-channel particle tracking methods. Although we cannot distinguish between coincident arrival and sequential arrival of Arp2/3 complex and diVCA within 0.15 s, we would not expect that diVCA could bind so rapidly after formation of Arp2/3-filament complexes, given the measured rate of binding to Arp2/3 complexes off-filament and the diVCA concentration in solution (∼10<sup>8</sup> M<sup>−1</sup>s<sup>−1</sup> × 5 nM = ∼0.5 s<sup>−1</sup>). We also corrected the total number of Arp2/3 complexes that appeared on filament sides with and without diVCA by the expected number of Arp2/3-filament encounters missed because they were too transient (lifetime <0.1 s) to be detected. This correction factor (1/p<sub>0</sub>) was calculated using the multi-exponential fits to the Arp2/3-filament lifetime distributions (survival probability p[<italic>t</italic>]), with and without coincident diVCA, evaluated at the minimum detectible event duration: p<sub>0</sub> = p(0.1 s). This detection efficiency was typically 0.5–0.6. The same correction factor was applied to the second order rate constants of filament side binding, <italic>k</italic><sub>A+</sub>.</p><p>Similarly, co-release of an Arp2/3 complex and diVCA was determined if the disappearance of diVCA was within 0.15 s from the disappearance of the Arp2/3 complex with which it coincidently bound the filament side. The subset of diVCA-Arp2/3-filament complexes from which diVCA was released leaving Arp2/3 complex bound to the filament, for at least 0.15 s after diVCA release, was counted toward the diVCA release efficiency (<italic>f</italic><sub>V−</sub>). Also, the fraction of diVCA-Arp2/3-filament complexes or Arp2/3-filament complexes that nucleated a daughter filament was counted as the branch formation efficiency (<italic>f</italic><sub>B</sub>) with or without diVCA, respectively. The number of binding and release events in each category (Arp2/3 complex alone, coincident with diVCA, Arp2/3-diVCA co-released, and diVCA release before Arp2/3 complex) were corrected by the number of observations in each category on regions of the field of view that did not contain a filament, so as to correct for non-specific interactions with the microscope slide.</p><p>Finally, the rate of release of diVCA from the nascent branch was calculated as: <italic>k</italic><sub>V−</sub><sup>*</sup> = <italic>f</italic><sub>V−</sub>/τ<sub>V</sub><sup>*</sup> (<xref ref-type="fig" rid="fig6">Figure 6D</xref>). This relation results from the fact that the observed lifetime of the nascent branch intermediate is limited by the combination of the dissociation of diVCA from the filament-bound Arp2/3 complex (proceeds with rate <italic>k</italic><sub>V−</sub><sup>*</sup>) and the dissociation of diVCA-Arp2/3 complex from the filament (proceeds with rate <italic>k</italic><sub>A−</sub>), such that τ<sub>V</sub><sup>*</sup> = 1/(<italic>k</italic><sub>A−</sub> + <italic>k</italic><sub>V−</sub><sup>*</sup>). The lifetimes (<xref ref-type="fig" rid="fig6">Figure 6C</xref>) are very similar for all diVCA constructs because the dissociation of diVCA-Arp2/3 complexes from filament sides is dominant (<italic>k</italic><sub>A−</sub> >> <italic>k</italic><sub>V−</sub><sup>*</sup>) and <italic>k</italic><sub>A−</sub> is not significantly affected by VCA mutations. Note that the proposed mechanism (<xref ref-type="fig" rid="fig6">Figure 6A</xref>) predicts that the nominal lifetime of nascent branches is the same whether we select all nascent branches or only those that terminate in diVCA release and subsequently form branches. The fraction of nascent branches that release diVCA is determined by kinetic competition between the two pathways of nascent branch disassembly: diVCA release leaving an Arp2/3-filament complex (with rate <italic>k</italic><sub>V−</sub><sup>*</sup>; labeled green arrow in <xref ref-type="fig" rid="fig6">Figure 6A</xref>) and filament release leaving a diVCA-Arp2/3 complex (with rate <italic>k</italic><sub>A−</sub>; thick red arrow in <xref ref-type="fig" rid="fig6">Figure 6A</xref>). This yields <italic>f</italic><sub>V−</sub> = <italic>k</italic><sub>V−</sub><sup>*</sup>/(<italic>k</italic><sub>A−</sub> + <italic>k</italic><sub>V−</sub><sup>*</sup>), and thus we obtain <italic>k</italic><sub>V−</sub><sup>*</sup> = <italic>f</italic><sub>V−</sub>/τ<sub>V</sub><sup>*</sup>.</p><p>We compared the diVCA release rate to the second order rate of overall branch formation calculated from the single molecule data as: <italic>k</italic><sub>B</sub> = <italic>k</italic><sub>A+</sub>[<italic>f</italic><sub>AV</sub> <italic>f</italic><sub>B</sub>(<italic>+diVCA</italic>) + (1 − <italic>f</italic><sub>AV</sub>) <italic>f</italic><sub>B</sub>(<italic>−diVCA</italic>)] (<xref ref-type="fig" rid="fig5">Figure 5A</xref>, <xref ref-type="fig" rid="fig6">Figure 6E</xref>). The values of all parameters used in these calculations, for all diVCA constructs tested, are indicated in <xref ref-type="table" rid="tbl1">Table 1</xref>. To calculate the probability that the measured correlation coefficients between diVCA activity towards Arp2/3 complex and the rate of its release from the nascent branch (<xref ref-type="fig" rid="fig6">Figure 6E</xref> and <xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1</xref>) arose by chance (i.e., that the two quantities are not in fact correlated), the data for each diVCA construct was resampled (100,000 times) based on the anisotropic 2D-Gaussian distribution defined by the error bars on each measurement. In each sample, activity and release rate values for the four diVCA constructs were randomly permuted, and the correlation coefficient was calculated. The probability p was then calculated as the definite integral of the normalized distribution of the sample correlation coefficients from the correlation coefficient of the experimental data to one. To the extent that replicate measurements may have correlated sources of error, we may overestimate the magnitude of the errors and therefore underestimate p.</p><p>Standard errors on parameters determined from fits to distributions of binding times or lifetimes were calculated by bootstrapping. Binding rate constants were reported as means ± S.E. from multiple replicate experiments. Errors on the fractions of Arp2/3 complex observations that appeared coincident with diVCA (<italic>f</italic><sub>AV</sub>), those that released diVCA (<italic>f</italic><sub>V−</sub>), and those that formed branches (<italic>f</italic><sub>B</sub>), were calculated using counting statistics: δ<italic>f</italic> = √[<italic>f</italic> (1 − <italic>f</italic>)/<italic>N</italic>]. Errors on quantities calculated from multiple measured parameters (such as <italic>k</italic><sub>B</sub>), were calculated using propagation of errors.</p></sec></sec></sec></body><back><ack id="ack"><title>Acknowledgements</title><p>We thank L Friedman and J Chung for assistance with TIRF microscopy and analysis, CA Ydenberg for help with bulk actin assembly kinetics, and members of the Goode lab for help with actin purification and labeling, A Okonechnikov for single-molecule analysis software development, and L Helgeson and B Nolen for sharing the results of unpublished work.</p></ack><sec sec-type="additional-information"><title>Additional information</title><fn-group content-type="competing-interest"><title>Competing interests</title><fn fn-type="conflict" id="conf1"><p>IRC,New England Biolabs manufactures and sells the SNAP-tag system, components of which are used in this study.</p></fn><fn fn-type="conflict" id="conf2"><p>M-QX: New England Biolabs manufactures and sells the SNAP-tag system, components of which are used in this study.</p></fn><fn fn-type="conflict" id="conf3"><p>BLG: Reviewing editor, <italic>eLife</italic>.</p></fn><fn fn-type="conflict" id="conf4"><p>The other authors declare that no competing interests exist.</p></fn></fn-group><fn-group content-type="author-contribution"><title>Author contributions</title><fn fn-type="con" id="con1"><p>BAS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con2"><p>SBP, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con3"><p>LKD, Acquisition of data, Analysis and interpretation of data</p></fn><fn fn-type="con" id="con4"><p>KD-C, Acquisition of data, Analysis and interpretation of data</p></fn><fn fn-type="con" id="con5"><p>M-QX, Acquisition of data, Analysis and interpretation of data</p></fn><fn fn-type="con" id="con6"><p>IRC, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con7"><p>BLG, Conception and design, Drafting or revising the article</p></fn><fn fn-type="con" id="con8"><p>MKR, Conception and design, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con9"><p>JG, Conception and design, Analysis and interpretation of data, Drafting or revising the article</p></fn></fn-group></sec><ref-list><title>References</title><ref id="bib1"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Achard</surname><given-names>V</given-names></name><name><surname>Martiel</surname><given-names>J-L</given-names></name><name><surname>Michelot</surname><given-names>A</given-names></name><name><surname>Guérin</surname><given-names>C</given-names></name><name><surname>Reymann</surname><given-names>A-C</given-names></name><name><surname>Blanchoin</surname><given-names>L</given-names></name><etal/></person-group><year>2010</year><article-title>A ‘primer’-based mechanism underlies branched actin filament network formation and motility</article-title><source>Curr Biol</source><volume>20</volume><fpage>423</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.1016/j.cub.2009.12.056</pub-id></element-citation></ref><ref id="bib2"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Amann</surname><given-names>KJ</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>2001</year><article-title>Direct real-time observation of actin filament branching mediated by Arp2/3 complex using total internal reflection fluorescence microscopy</article-title><source>Proc Natl Acad Sci USA</source><volume>98</volume><fpage>15009</fpage><lpage>13</lpage><pub-id pub-id-type="doi">10.1073/pnas.211556398</pub-id></element-citation></ref><ref id="bib3"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Arasada</surname><given-names>R</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>2011</year><article-title>Distinct roles for f-bar proteins Cdc15p and Bzz1p in actin polymerization at sites of endocytosis in fission yeast</article-title><source>Curr Biol</source><volume>21</volume><fpage>1450</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1016/j.cub.2011.07.046</pub-id></element-citation></ref><ref id="bib4"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Beltzner</surname><given-names>CC</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>2008</year><article-title>Pathway of actin filament branch formation by Arp2/3 complex</article-title><source>J Biol Chem</source><volume>283</volume><fpage>7135</fpage><lpage>44</lpage><pub-id pub-id-type="doi">10.1074/jbc.M705894200</pub-id></element-citation></ref><ref id="bib5"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Blanchoin</surname><given-names>L</given-names></name><name><surname>Amann</surname><given-names>KJ</given-names></name><name><surname>Higgs</surname><given-names>HN</given-names></name><name><surname>Marchand</surname><given-names>JB</given-names></name><name><surname>Kaiser</surname><given-names>DA</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>2000</year><article-title>Direct observation of dendritic actin filament networks nucleated by Arp2/3 complex and WASP/Scar proteins</article-title><source>Nature</source><volume>404</volume><fpage>1007</fpage><lpage>11</lpage><pub-id pub-id-type="doi">10.1038/35010008</pub-id></element-citation></ref><ref id="bib6"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Boczkowska</surname><given-names>M</given-names></name><name><surname>Rebowski</surname><given-names>G</given-names></name><name><surname>Petoukhov</surname><given-names>MV</given-names></name><name><surname>Hayes</surname><given-names>DB</given-names></name><name><surname>Svergun</surname><given-names>DI</given-names></name><name><surname>Dominguez</surname><given-names>R</given-names></name></person-group><year>2008</year><article-title>X-ray scattering study of activated Arp2/3 complex with bound actin-WCA</article-title><source>Structure</source><volume>16</volume><fpage>695</fpage><lpage>704</lpage><pub-id pub-id-type="doi">10.1016/j.str.2008.02.013</pub-id></element-citation></ref><ref id="bib7"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Campellone</surname><given-names>KG</given-names></name><name><surname>Welch</surname><given-names>MD</given-names></name></person-group><year>2010</year><article-title>A nucleator arms race: cellular control of actin assembly</article-title><source>Nat Rev Mol Cell Biol</source><volume>11</volume><fpage>237</fpage><lpage>51</lpage><pub-id pub-id-type="doi">10.1038/nrm2867</pub-id></element-citation></ref><ref id="bib8"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chereau</surname><given-names>D</given-names></name><name><surname>Kerff</surname><given-names>F</given-names></name><name><surname>Graceffa</surname><given-names>P</given-names></name><name><surname>Grabarek</surname><given-names>Z</given-names></name><name><surname>Langsetmo</surname><given-names>K</given-names></name><name><surname>Dominguez</surname><given-names>R</given-names></name></person-group><year>2005</year><article-title>Actin-bound structures of Wiskott-Aldrich syndrome protein (WASP)-homology domain 2 and the implications for filament assembly</article-title><source>Proc Natl Acad Sci USA</source><volume>102</volume><fpage>16644</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1073/pnas.0507021102</pub-id></element-citation></ref><ref id="bib9"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Co</surname><given-names>C</given-names></name><name><surname>Wong</surname><given-names>DT</given-names></name><name><surname>Gierke</surname><given-names>S</given-names></name><name><surname>Chang</surname><given-names>V</given-names></name><name><surname>Taunton</surname><given-names>J</given-names></name></person-group><year>2007</year><article-title>Mechanism of actin network attachment to moving membranes: barbed end capture by N-WASP WH2 domains</article-title><source>Cell</source><volume>128</volume><fpage>901</fpage><lpage>13</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2006.12.049</pub-id></element-citation></ref><ref id="bib10"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cooper</surname><given-names>JA</given-names></name><name><surname>Walker</surname><given-names>SB</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>1983</year><article-title>Pyrene actin: documentation of the validity of a sensitive assay for actin polymerization</article-title><source>J Muscle Res Cell Motil</source><volume>4</volume><fpage>253</fpage><lpage>62</lpage></element-citation></ref><ref id="bib11"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Crocker</surname><given-names>JC</given-names></name><name><surname>Grier</surname><given-names>DG</given-names></name></person-group><year>1996</year><article-title>Methods of digital video microscopy for colloidal studies</article-title><source>J Colloid Interface Sci</source><volume>179</volume><fpage>298</fpage><lpage>310</lpage></element-citation></ref><ref id="bib12"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>D’Agostino</surname><given-names>JL</given-names></name><name><surname>Goode</surname><given-names>BL</given-names></name></person-group><year>2005</year><article-title>Dissection of Arp2/3 complex actin nucleation mechanism and distinct roles for its nucleation-promoting factors in <italic>Saccharomyces cerevisiae</italic></article-title><source>Genetics</source><volume>171</volume><fpage>35</fpage><lpage>47</lpage><pub-id pub-id-type="doi">10.1534/genetics.105.040634</pub-id></element-citation></ref><ref id="bib13"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dayel</surname><given-names>MJ</given-names></name><name><surname>Holleran</surname><given-names>EA</given-names></name><name><surname>Mullins</surname><given-names>RD</given-names></name></person-group><year>2001</year><article-title>Arp2/3 complex requires hydrolyzable ATP for nucleation of new actin filaments</article-title><source>Proc Natl Acad Sci USA</source><volume>98</volume><fpage>14871</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.1073/pnas.261419298</pub-id></element-citation></ref><ref id="bib14"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Dayel</surname><given-names>MJ</given-names></name><name><surname>Mullins</surname><given-names>RD</given-names></name></person-group><year>2004</year><article-title>Activation of Arp2/3 complex: addition of the first subunit of the new filament by a WASP protein triggers rapid ATP hydrolysis on Arp2</article-title><source>PLOS Biol</source><volume>2</volume><fpage>E91</fpage><pub-id pub-id-type="doi">10.1371/journal.pbio.0020091</pub-id></element-citation></ref><ref id="bib15"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ditlev</surname><given-names>JA</given-names></name><name><surname>Michalski</surname><given-names>PJ</given-names></name><name><surname>Huber</surname><given-names>G</given-names></name><name><surname>Rivera</surname><given-names>GM</given-names></name><name><surname>Mohler</surname><given-names>WA</given-names></name><name><surname>Loew</surname><given-names>LM</given-names></name><etal/></person-group><year>2012</year><article-title>Stoichiometry of Nck-dependent actin polymerization in living cells</article-title><source>J Cell biol</source><volume>197</volume><fpage>643</fpage><lpage>58</lpage><pub-id pub-id-type="doi">10.1083/jcb.201111113</pub-id></element-citation></ref><ref id="bib16"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Doolittle</surname><given-names>LK</given-names></name><name><surname>Rosen</surname><given-names>MK</given-names></name><name><surname>Padrick</surname><given-names>SB</given-names></name></person-group><year>2013a</year><article-title>Measurement and analysis of in vitro actin polymerization</article-title><source>Methods Mol Biol</source><volume>1046</volume><fpage>273</fpage><lpage>93</lpage><pub-id pub-id-type="doi">10.1007/978-1-62703-538-5_16</pub-id></element-citation></ref><ref id="bib17"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Doolittle</surname><given-names>LK</given-names></name><name><surname>Rosen</surname><given-names>MK</given-names></name><name><surname>Padrick</surname><given-names>SB</given-names></name></person-group><year>2013b</year><article-title>Purification of Arp2/3 complex from <italic>Saccharomyces cerevisiae</italic></article-title><source>Methods Mol Biol</source><volume>1046</volume><fpage>251</fpage><lpage>71</lpage><pub-id pub-id-type="doi">10.1007/978-1-62703-538-5_15</pub-id></element-citation></ref><ref id="bib18"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Doolittle</surname><given-names>LK</given-names></name><name><surname>Rosen</surname><given-names>MK</given-names></name><name><surname>Padrick</surname><given-names>SB</given-names></name></person-group><year>2013c</year><article-title>Purification of native Arp2/3 complex from bovine thymus</article-title><source>Methods Mol Biol</source><volume>1046</volume><fpage>231</fpage><lpage>50</lpage><pub-id pub-id-type="doi">10.1007/978-1-62703-538-5_14</pub-id></element-citation></ref><ref id="bib19"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Egile</surname><given-names>C</given-names></name><name><surname>Loisel</surname><given-names>TP</given-names></name><name><surname>Laurent</surname><given-names>V</given-names></name><name><surname>Li</surname><given-names>R</given-names></name><name><surname>Pantaloni</surname><given-names>D</given-names></name><name><surname>Sansonetti</surname><given-names>PJ</given-names></name><etal/></person-group><year>1999</year><article-title>Activation of the CDC42 effector N-WASP by the <italic>Shigella flexneri</italic> IcsA protein promotes actin nucleation by Arp2/3 complex and bacterial actin-based motility</article-title><source>J Cell Biol</source><volume>146</volume><fpage>1319</fpage><lpage>32</lpage><pub-id pub-id-type="doi">10.1083/jcb.146.6.1319</pub-id></element-citation></ref><ref id="bib20"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Egile</surname><given-names>C</given-names></name><name><surname>Rouiller</surname><given-names>I</given-names></name><name><surname>Xu</surname><given-names>X-P</given-names></name><name><surname>Volkmann</surname><given-names>N</given-names></name><name><surname>Li</surname><given-names>R</given-names></name><name><surname>Hanein</surname><given-names>D</given-names></name></person-group><year>2005</year><article-title>Mechanism of filament nucleation and branch stability revealed by the structure of the Arp2/3 complex at actin branch junctions</article-title><source>PLOS Biol</source><volume>3</volume><fpage>e383</fpage><pub-id pub-id-type="doi">10.1371/journal.pbio.0030383</pub-id></element-citation></ref><ref id="bib21"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Footer</surname><given-names>MJ</given-names></name><name><surname>Lyo</surname><given-names>JK</given-names></name><name><surname>Theriot</surname><given-names>JA</given-names></name></person-group><year>2008</year><article-title>Close packing of <italic>Listeria monocytogenes</italic> ActA, a natively unfolded protein, enhances F-actin assembly without dimerization</article-title><source>J Biol Chem</source><volume>283</volume><fpage>23852</fpage><lpage>62</lpage><pub-id pub-id-type="doi">10.1074/jbc.M803448200</pub-id></element-citation></ref><ref id="bib22"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Friedman</surname><given-names>LJ</given-names></name><name><surname>Chung</surname><given-names>J</given-names></name><name><surname>Gelles</surname><given-names>J</given-names></name></person-group><year>2006</year><article-title>Viewing dynamic assembly of molecular complexes by multi-wavelength single-molecule fluorescence</article-title><source>Biophys J</source><volume>91</volume><fpage>1023</fpage><lpage>31</lpage><pub-id pub-id-type="doi">10.1529/biophysj.106.084004</pub-id></element-citation></ref><ref id="bib23"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Friedman</surname><given-names>LJ</given-names></name><name><surname>Gelles</surname><given-names>J</given-names></name></person-group><year>2012</year><article-title>Mechanism of transcription initiation at an activator-dependent promoter defined by single-molecule observation</article-title><source>Cell</source><volume>148</volume><fpage>679</fpage><lpage>89</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2012.01.018</pub-id></element-citation></ref><ref id="bib24"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Goley</surname><given-names>ED</given-names></name><name><surname>Rodenbusch</surname><given-names>SE</given-names></name><name><surname>Martin</surname><given-names>AC</given-names></name><name><surname>Welch</surname><given-names>MD</given-names></name></person-group><year>2004</year><article-title>Critical conformational changes in the Arp2/3 complex are induced by nucleotide and nucleation promoting factor</article-title><source>Mol Cell</source><volume>16</volume><fpage>269</fpage><lpage>79</lpage><pub-id pub-id-type="doi">10.1016/j.molcel.2004.09.018</pub-id></element-citation></ref><ref id="bib25"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Helgeson</surname><given-names>LA</given-names></name><name><surname>Nolen</surname><given-names>BJ</given-names></name></person-group><year>2013</year><article-title>Mechanism of synergistic activation of Arp2/3 complex by cortactin and N-WASP</article-title><source>eLife</source><volume>2</volume><fpage>e00884</fpage><pub-id pub-id-type="doi">10.7554/eLife.00884</pub-id></element-citation></ref><ref id="bib26"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hertzog</surname><given-names>M</given-names></name><name><surname>van Heijenoort</surname><given-names>C</given-names></name><name><surname>Didry</surname><given-names>D</given-names></name><name><surname>Gaudier</surname><given-names>M</given-names></name><name><surname>Coutant</surname><given-names>J</given-names></name><name><surname>Gigant</surname><given-names>B</given-names></name><etal/></person-group><year>2004</year><article-title>The beta-thymosin/WH2 domain; structural basis for the switch from inhibition to promotion of actin assembly</article-title><source>Cell</source><volume>117</volume><fpage>611</fpage><lpage>23</lpage></element-citation></ref><ref id="bib27"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hetrick</surname><given-names>B</given-names></name><name><surname>Han</surname><given-names>MS</given-names></name><name><surname>Helgeson</surname><given-names>LA</given-names></name><name><surname>Nolen</surname><given-names>BJ</given-names></name></person-group><year>2013</year><article-title>Small molecules CK-666 and CK-869 inhibit actin-related protein 2/3 complex by blocking an activating conformational change</article-title><source>Chem Biol</source><volume>20</volume><fpage>701</fpage><lpage>12</lpage><pub-id pub-id-type="doi">10.1016/j.chembiol.2013.03.019</pub-id></element-citation></ref><ref id="bib28"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Higgs</surname><given-names>HN</given-names></name><name><surname>Blanchoin</surname><given-names>L</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>1999</year><article-title>Influence of the C terminus of Wiskott-Aldrich syndrome protein (WASp) and the Arp2/3 complex on actin polymerization</article-title><source>Biochemistry</source><volume>38</volume><fpage>15212</fpage><lpage>22</lpage></element-citation></ref><ref id="bib29"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hoskins</surname><given-names>AA</given-names></name><name><surname>Friedman</surname><given-names>LJ</given-names></name><name><surname>Gallagher</surname><given-names>SS</given-names></name><name><surname>Crawford</surname><given-names>DJ</given-names></name><name><surname>Anderson</surname><given-names>EG</given-names></name><name><surname>Wombacher</surname><given-names>R</given-names></name><etal/></person-group><year>2011</year><article-title>Ordered and dynamic assembly of single spliceosomes</article-title><source>Science</source><volume>331</volume><fpage>1289</fpage><lpage>95</lpage><pub-id pub-id-type="doi">10.1126/science.1198830</pub-id></element-citation></ref><ref id="bib30"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ingerman</surname><given-names>E</given-names></name><name><surname>Hsiao</surname><given-names>JY</given-names></name><name><surname>Mullins</surname><given-names>RD</given-names></name></person-group><year>2013</year><article-title>Arp2/3 complex ATP hydrolysis promotes lamellipodial actin network disassembly but is dispensable for assembly</article-title><source>J Cell Biol</source><volume>200</volume><fpage>619</fpage><lpage>33</lpage><pub-id pub-id-type="doi">10.1083/jcb.201211069</pub-id></element-citation></ref><ref id="bib31"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Irobi</surname><given-names>E</given-names></name><name><surname>Aguda</surname><given-names>AH</given-names></name><name><surname>Larsson</surname><given-names>M</given-names></name><name><surname>Guerin</surname><given-names>C</given-names></name><name><surname>Yin</surname><given-names>HL</given-names></name><name><surname>Burtnick</surname><given-names>LD</given-names></name><etal/></person-group><year>2004</year><article-title>Structural basis of actin sequestration by thymosin-ß4: implications for WH2 proteins</article-title><source>EMBO J</source><volume>23</volume><fpage>3599</fpage><lpage>608</lpage><pub-id pub-id-type="doi">10.1038/sj.emboj.7600372</pub-id></element-citation></ref><ref id="bib32"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Joel</surname><given-names>PB</given-names></name><name><surname>Fagnant</surname><given-names>PM</given-names></name><name><surname>Trybus</surname><given-names>KM</given-names></name></person-group><year>2004</year><article-title>Expression of a nonpolymerizable actin mutant in Sf9 cells</article-title><source>Biochemistry</source><volume>43</volume><fpage>11554</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1021/bi048899a</pub-id></element-citation></ref><ref id="bib33"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kaksonen</surname><given-names>M</given-names></name><name><surname>Sun</surname><given-names>Y</given-names></name><name><surname>Drubin</surname><given-names>DG</given-names></name></person-group><year>2003</year><article-title>A pathway for association of receptors, adaptors, and actin during endocytic internalization</article-title><source>Cell</source><volume>115</volume><fpage>475</fpage><lpage>87</lpage></element-citation></ref><ref id="bib34"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Keppler</surname><given-names>A</given-names></name><name><surname>Pick</surname><given-names>H</given-names></name><name><surname>Arrivoli</surname><given-names>C</given-names></name><name><surname>Vogel</surname><given-names>H</given-names></name><name><surname>Johnsson</surname><given-names>K</given-names></name></person-group><year>2004</year><article-title>Labeling of fusion proteins with synthetic fluorophores in live cells</article-title><source>Proc Natl Acad Sci USA</source><volume>101</volume><fpage>9955</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1073/pnas.0401923101</pub-id></element-citation></ref><ref id="bib35"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kindermann</surname><given-names>M</given-names></name><name><surname>Sielaff</surname><given-names>I</given-names></name><name><surname>Johnsson</surname><given-names>K</given-names></name></person-group><year>2004</year><article-title>Synthesis and characterization of bifunctional probes for the specific labeling of fusion proteins</article-title><source>Bioorg Med Chem Lett</source><volume>14</volume><fpage>2725</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.1016/j.bmcl.2004.03.078</pub-id></element-citation></ref><ref id="bib36"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lai</surname><given-names>FP</given-names></name><name><surname>Szczodrak</surname><given-names>M</given-names></name><name><surname>Block</surname><given-names>J</given-names></name><name><surname>Faix</surname><given-names>J</given-names></name><name><surname>Breitsprecher</surname><given-names>D</given-names></name><name><surname>Mannherz</surname><given-names>HG</given-names></name><etal/></person-group><year>2008</year><article-title>Arp2/3 complex interactions and actin network turnover in lamellipodia</article-title><source>EMBO J</source><volume>27</volume><fpage>982</fpage><lpage>92</lpage><pub-id pub-id-type="doi">10.1038/emboj.2008.34</pub-id></element-citation></ref><ref id="bib37"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lechler</surname><given-names>T</given-names></name><name><surname>Jonsdottir</surname><given-names>GA</given-names></name><name><surname>Klee</surname><given-names>SK</given-names></name><name><surname>Pellman</surname><given-names>D</given-names></name><name><surname>Li</surname><given-names>R</given-names></name></person-group><year>2001</year><article-title>A two-tiered mechanism by which Cdc42 controls the localization and activation of an Arp2/3-activating motor complex in yeast</article-title><source>J Cell Biol</source><volume>155</volume><fpage>261</fpage><lpage>70</lpage><pub-id pub-id-type="doi">10.1083/jcb.200104094</pub-id></element-citation></ref><ref id="bib38"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Loisel</surname><given-names>TP</given-names></name><name><surname>Boujemaa</surname><given-names>R</given-names></name><name><surname>Pantaloni</surname><given-names>D</given-names></name><name><surname>Carlier</surname><given-names>MF</given-names></name></person-group><year>1999</year><article-title>Reconstitution of actin-based motility of <italic>Listeria</italic> and <italic>Shigella</italic> using pure proteins</article-title><source>Nature</source><volume>401</volume><fpage>613</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.1038/44183</pub-id></element-citation></ref><ref id="bib39"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Machesky</surname><given-names>LM</given-names></name><name><surname>Mullins</surname><given-names>RD</given-names></name><name><surname>Higgs</surname><given-names>HN</given-names></name><name><surname>Kaiser</surname><given-names>DA</given-names></name><name><surname>Blanchoin</surname><given-names>L</given-names></name><name><surname>May</surname><given-names>RC</given-names></name><etal/></person-group><year>1999</year><article-title>Scar, a WASp-related protein, activates nucleation of actin filaments by the Arp2/3 complex</article-title><source>Proc Natl Acad Sci USA</source><volume>96</volume><fpage>3739</fpage><lpage>44</lpage><pub-id pub-id-type="doi">10.1073/pnas.96.7.3739</pub-id></element-citation></ref><ref id="bib40"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Machesky</surname><given-names>LM</given-names></name><name><surname>Insall</surname><given-names>RH</given-names></name></person-group><year>1998</year><article-title>Scar1 and the related Wiskott-Aldrich syndrome protein, WASP, regulate the actin cytoskeleton through the Arp2/3 complex</article-title><source>Curr Biol</source><volume>8</volume><fpage>1347</fpage><lpage>56</lpage><pub-id pub-id-type="doi">10.1016/S0960-9822(98)00015-3</pub-id></element-citation></ref><ref id="bib41"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Marchand</surname><given-names>JB</given-names></name><name><surname>Kaiser</surname><given-names>DA</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name><name><surname>Higgs</surname><given-names>HN</given-names></name></person-group><year>2001</year><article-title>Interaction of WASP/Scar proteins with actin and vertebrate Arp2/3 complex</article-title><source>Nat Cell Biol</source><volume>3</volume><fpage>76</fpage><lpage>82</lpage><pub-id pub-id-type="doi">10.1038/35050590</pub-id></element-citation></ref><ref id="bib42"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Martin</surname><given-names>AC</given-names></name><name><surname>Welch</surname><given-names>MD</given-names></name><name><surname>Drubin</surname><given-names>DG</given-names></name></person-group><year>2006</year><article-title>Arp2/3 ATP hydrolysis-catalysed branch dissociation is critical for endocytic force generation</article-title><source>Nat Cell Biol</source><volume>8</volume><fpage>826</fpage><lpage>33</lpage><pub-id pub-id-type="doi">10.1038/ncb1443</pub-id></element-citation></ref><ref id="bib43"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Miki</surname><given-names>H</given-names></name><name><surname>Takenawa</surname><given-names>T</given-names></name></person-group><year>1998</year><article-title>Direct binding of the verprolin-homology domain in N-WASP to actin is essential for cytoskeletal reorganization</article-title><source>Biochem Biophys Res Commun</source><volume>243</volume><fpage>73</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.1006/bbrc.1997.8064</pub-id></element-citation></ref><ref id="bib44"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Mullins</surname><given-names>RD</given-names></name><name><surname>Heuser</surname><given-names>JA</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>1998</year><article-title>The interaction of Arp2/3 complex with actin: nucleation, high affinity pointed end capping, and formation of branching networks of filaments</article-title><source>Proc Natl Acad Sci USA</source><volume>95</volume><fpage>6181</fpage><lpage>6</lpage></element-citation></ref><ref id="bib45"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Padrick</surname><given-names>SB</given-names></name><name><surname>Doolittle</surname><given-names>LK</given-names></name><name><surname>Brautigam</surname><given-names>CA</given-names></name><name><surname>King</surname><given-names>DS</given-names></name><name><surname>Rosen</surname><given-names>MK</given-names></name></person-group><year>2011</year><article-title>Arp2/3 complex is bound and activated by two WASP proteins</article-title><source>Proc Natl Acad Sci USA</source><volume>108</volume><fpage>E472</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1073/pnas.1100236108</pub-id></element-citation></ref><ref id="bib46"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Padrick</surname><given-names>SB</given-names></name><name><surname>Cheng</surname><given-names>H-C</given-names></name><name><surname>Ismail</surname><given-names>AM</given-names></name><name><surname>Panchal</surname><given-names>SC</given-names></name><name><surname>Doolittle</surname><given-names>LK</given-names></name><name><surname>Kim</surname><given-names>S</given-names></name><etal/></person-group><year>2008</year><article-title>Hierarchical regulation of WASP/WAVE proteins</article-title><source>Mol Cell</source><volume>32</volume><fpage>426</fpage><lpage>38</lpage><pub-id pub-id-type="doi">10.1016/j.molcel.2008.10.012</pub-id></element-citation></ref><ref id="bib47"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Padrick</surname><given-names>SB</given-names></name><name><surname>Rosen</surname><given-names>MK</given-names></name></person-group><year>2010</year><article-title>Physical mechanisms of signal integration by WASP family proteins</article-title><source>Annu Rev Biochem</source><volume>79</volume><fpage>707</fpage><lpage>35</lpage><pub-id pub-id-type="doi">10.1146/annurev.biochem.77.060407.135452</pub-id></element-citation></ref><ref id="bib48"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Panchal</surname><given-names>SC</given-names></name><name><surname>Kaiser</surname><given-names>DA</given-names></name><name><surname>Torres</surname><given-names>E</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name><name><surname>Rosen</surname><given-names>MK</given-names></name></person-group><year>2003</year><article-title>A conserved amphipathic helix in WASP/Scar proteins is essential for activation of Arp2/3 complex</article-title><source>Nat Struct Biol</source><volume>10</volume><fpage>591</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.1038/nsb952</pub-id></element-citation></ref><ref id="bib49"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>2007</year><article-title>Regulation of actin filament assembly by Arp2/3 complex and formins</article-title><source>Annu Rev Biophys Biomol Struct</source><volume>36</volume><fpage>451</fpage><lpage>77</lpage><pub-id pub-id-type="doi">10.1146/annurev.biophys.35.040405.101936</pub-id></element-citation></ref><ref id="bib50"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pollard</surname><given-names>TD</given-names></name><name><surname>Borisy</surname><given-names>GG</given-names></name></person-group><year>2003</year><article-title>Cellular motility driven by assembly and disassembly of actin filaments</article-title><source>Cell</source><volume>112</volume><fpage>453</fpage><lpage>65</lpage><pub-id pub-id-type="doi">10.1016/S0092-8674(03)00120-X</pub-id></element-citation></ref><ref id="bib51"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pollard</surname><given-names>TD</given-names></name><name><surname>Cooper</surname><given-names>JA</given-names></name></person-group><year>2009</year><article-title>Actin, a central player in cell shape and movement</article-title><source>Science</source><volume>326</volume><fpage>1208</fpage><lpage>12</lpage><pub-id pub-id-type="doi">10.1126/science.1175862</pub-id></element-citation></ref><ref id="bib52"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Robinson</surname><given-names>RC</given-names></name><name><surname>Turbedsky</surname><given-names>K</given-names></name><name><surname>Kaiser</surname><given-names>DA</given-names></name><name><surname>Marchand</surname><given-names>JB</given-names></name><name><surname>Higgs</surname><given-names>HN</given-names></name><name><surname>Choe</surname><given-names>S</given-names></name><etal/></person-group><year>2001</year><article-title>Crystal structure of Arp2/3 complex</article-title><source>Science</source><volume>294</volume><fpage>1679</fpage><lpage>84</lpage><pub-id pub-id-type="doi">10.1126/science.1066333</pub-id></element-citation></ref><ref id="bib53"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rodal</surname><given-names>AA</given-names></name><name><surname>Sokolova</surname><given-names>O</given-names></name><name><surname>Robins</surname><given-names>DB</given-names></name><name><surname>Daugherty</surname><given-names>KM</given-names></name><name><surname>Hippenmeyer</surname><given-names>S</given-names></name><name><surname>Riezman</surname><given-names>H</given-names></name><etal/></person-group><year>2005</year><article-title>Conformational changes in the Arp2/3 complex leading to actin nucleation</article-title><source>Nat Struct Mol Biol</source><volume>12</volume><fpage>26</fpage><lpage>31</lpage><pub-id pub-id-type="doi">10.1038/nsmb870</pub-id></element-citation></ref><ref id="bib54"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rohatgi</surname><given-names>R</given-names></name><name><surname>Ma</surname><given-names>L</given-names></name><name><surname>Miki</surname><given-names>H</given-names></name><name><surname>Lopez</surname><given-names>M</given-names></name><name><surname>Kirchhausen</surname><given-names>T</given-names></name><name><surname>Takenawa</surname><given-names>T</given-names></name><etal/></person-group><year>1999</year><article-title>The interaction between N-WASP and the Arp2/3 complex links Cdc42-dependent signals to actin assembly</article-title><source>Cell</source><volume>97</volume><fpage>221</fpage><lpage>31</lpage><pub-id pub-id-type="doi">10.1016/S0092-8674(00)80732-1</pub-id></element-citation></ref><ref id="bib55"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rouiller</surname><given-names>I</given-names></name><name><surname>Xu</surname><given-names>X-P</given-names></name><name><surname>Amann</surname><given-names>KJ</given-names></name><name><surname>Egile</surname><given-names>C</given-names></name><name><surname>Nickell</surname><given-names>S</given-names></name><name><surname>Nicastro</surname><given-names>D</given-names></name><etal/></person-group><year>2008</year><article-title>The structural basis of actin filament branching by the Arp2/3 complex</article-title><source>J Cell Biol</source><volume>180</volume><fpage>887</fpage><lpage>95</lpage><pub-id pub-id-type="doi">10.1083/jcb.200709092</pub-id></element-citation></ref><ref id="bib56"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Shi</surname><given-names>X</given-names></name><name><surname>Lim</surname><given-names>J</given-names></name><name><surname>Ha</surname><given-names>T</given-names></name></person-group><year>2010</year><article-title>Acidification of the oxygen scavenging system in single-molecule fluorescence studies: in situ sensing with a ratiometric dual-emission probe</article-title><source>Anal Chem</source><volume>82</volume><fpage>6132</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.1021/ac1008749</pub-id></element-citation></ref><ref id="bib57"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sirotkin</surname><given-names>V</given-names></name><name><surname>Beltzner</surname><given-names>CC</given-names></name><name><surname>Marchand</surname><given-names>J-B</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>2005</year><article-title>Interactions of WASp, myosin-I, and verprolin with Arp2/3 complex during actin patch assembly in fission yeast</article-title><source>J Cell Biol</source><volume>170</volume><fpage>637</fpage><lpage>48</lpage><pub-id pub-id-type="doi">10.1083/jcb.200502053</pub-id></element-citation></ref><ref id="bib58"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Small</surname><given-names>JV</given-names></name><name><surname>Isenberg</surname><given-names>G</given-names></name><name><surname>Celis</surname><given-names>JE</given-names></name></person-group><year>1978</year><article-title>Polarity of actin at the leading edge of cultured cells</article-title><source>Nature</source><volume>272</volume><fpage>638</fpage><lpage>9</lpage></element-citation></ref><ref id="bib59"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Smith</surname><given-names>BA</given-names></name><name><surname>Daugherty-Clarke</surname><given-names>K</given-names></name><name><surname>Goode</surname><given-names>BL</given-names></name><name><surname>Gelles</surname><given-names>J</given-names></name></person-group><year>2013</year><article-title>Pathway of actin filament branch formation by Arp2/3 complex revealed by single-molecule imaging</article-title><source>Proc Natl Acad Sci USA</source><volume>110</volume><fpage>1285</fpage><lpage>90</lpage><pub-id pub-id-type="doi">10.1073/pnas.1211164110</pub-id></element-citation></ref><ref id="bib60"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Spudich</surname><given-names>JA</given-names></name><name><surname>Watt</surname><given-names>S</given-names></name></person-group><year>1971</year><article-title>The regulation of rabbit skeletal muscle contraction. I. Biochemical studies of the interaction of the tropomyosin-troponin complex with actin and the proteolytic fragments of myosin</article-title><source>J Biol Chem</source><volume>246</volume><fpage>4866</fpage><lpage>71</lpage></element-citation></ref><ref id="bib61"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Svitkina</surname><given-names>TM</given-names></name><name><surname>Borisy</surname><given-names>GG</given-names></name></person-group><year>1999</year><article-title>Arp2/3 complex and actin depolymerizing factor/cofilin in dendritic organization and treadmilling of actin filament array in lamellipodia</article-title><source>J Cell Biol</source><volume>145</volume><fpage>1009</fpage><lpage>26</lpage><pub-id pub-id-type="doi">10.1083/jcb.145.5.1009</pub-id></element-citation></ref><ref id="bib62"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Taunton</surname><given-names>J</given-names></name><name><surname>Rowning</surname><given-names>BA</given-names></name><name><surname>Coughlin</surname><given-names>ML</given-names></name><name><surname>Wu</surname><given-names>M</given-names></name><name><surname>Moon</surname><given-names>RT</given-names></name><name><surname>Mitchison</surname><given-names>TJ</given-names></name><etal/></person-group><year>2000</year><article-title>Actin-dependent propulsion of endosomes and lysosomes by recruitment of N-WASP</article-title><source>J Cell Biol</source><volume>148</volume><fpage>519</fpage><lpage>30</lpage><pub-id pub-id-type="doi">10.1083/jcb.148.3.519</pub-id></element-citation></ref><ref id="bib63"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ti</surname><given-names>S-C</given-names></name><name><surname>Jurgenson</surname><given-names>CT</given-names></name><name><surname>Nolen</surname><given-names>BJ</given-names></name><name><surname>Pollard</surname><given-names>TD</given-names></name></person-group><year>2011</year><article-title>Structural and biochemical characterization of two binding sites for nucleation-promoting factor WASp-VCA on Arp2/3 complex</article-title><source>Proc Natl Acad Sci USA</source><volume>108</volume><fpage>E463</fpage><lpage>71</lpage><pub-id pub-id-type="doi">10.1073/pnas.1100125108</pub-id></element-citation></ref><ref id="bib64"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Vinzenz</surname><given-names>M</given-names></name><name><surname>Nemethova</surname><given-names>M</given-names></name><name><surname>Schur</surname><given-names>F</given-names></name><name><surname>Mueller</surname><given-names>J</given-names></name><name><surname>Narita</surname><given-names>A</given-names></name><name><surname>Urban</surname><given-names>E</given-names></name><etal/></person-group><year>2012</year><article-title>Actin branching in the initiation and maintenance of lamellipodia</article-title><source>J Cell Sci</source><volume>125</volume><fpage>2775</fpage><lpage>85</lpage><pub-id pub-id-type="doi">10.1242/jcs.107623</pub-id></element-citation></ref><ref id="bib65"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Weisswange</surname><given-names>I</given-names></name><name><surname>Newsome</surname><given-names>TP</given-names></name><name><surname>Schleich</surname><given-names>S</given-names></name><name><surname>Way</surname><given-names>M</given-names></name></person-group><year>2009</year><article-title>The rate of N-WASP exchange limits the extent of ARP2/3-complex-dependent actin-based motility</article-title><source>Nature</source><volume>458</volume><fpage>87</fpage><lpage>91</lpage><pub-id pub-id-type="doi">10.1038/nature07773</pub-id></element-citation></ref><ref id="bib66"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Welch</surname><given-names>MD</given-names></name><name><surname>Iwamatsu</surname><given-names>A</given-names></name><name><surname>Mitchison</surname><given-names>TJ</given-names></name></person-group><year>1997</year><article-title>Actin polymerization is induced by Arp2/3 protein complex at the surface of <italic>Listeria monocytogenes</italic></article-title><source>Nature</source><volume>385</volume><fpage>265</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1038/385265a0</pub-id></element-citation></ref><ref id="bib67"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wen</surname><given-names>K-K</given-names></name><name><surname>Rubenstein</surname><given-names>PA</given-names></name></person-group><year>2005</year><article-title>Acceleration of yeast actin polymerization by yeast Arp2/3 complex does not require an Arp2/3-activating protein</article-title><source>J Biol Chem</source><volume>280</volume><fpage>24168</fpage><lpage>74</lpage><pub-id pub-id-type="doi">10.1074/jbc.M502024200</pub-id></element-citation></ref><ref id="bib68"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xu</surname><given-names>X-P</given-names></name><name><surname>Rouiller</surname><given-names>I</given-names></name><name><surname>Slaughter</surname><given-names>BD</given-names></name><name><surname>Egile</surname><given-names>C</given-names></name><name><surname>Kim</surname><given-names>E</given-names></name><name><surname>Unruh</surname><given-names>JR</given-names></name><etal/></person-group><year>2012</year><article-title>Three-dimensional reconstructions of Arp2/3 complex with bound nucleation promoting factors</article-title><source>EMBO J</source><volume>31</volume><fpage>236</fpage><lpage>47</lpage><pub-id pub-id-type="doi">10.1038/emboj.2011.343</pub-id></element-citation></ref><ref id="bib69"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zahm</surname><given-names>JA</given-names></name><name><surname>Padrick</surname><given-names>SB</given-names></name><name><surname>Chen</surname><given-names>Z</given-names></name><name><surname>Pak</surname><given-names>CW</given-names></name><name><surname>Yunus</surname><given-names>AA</given-names></name><name><surname>Henry</surname><given-names>L</given-names></name><etal/></person-group><year>2013</year><article-title>Structure of a filament-like actin trimer bound to the bacterial effector VopL</article-title><source>Cell</source><comment>(In Press)</comment></element-citation></ref><ref id="bib70"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zalevsky</surname><given-names>J</given-names></name><name><surname>Lempert</surname><given-names>L</given-names></name><name><surname>Kranitz</surname><given-names>H</given-names></name><name><surname>Mullins</surname><given-names>RD</given-names></name></person-group><year>2001</year><article-title>Different WASP family proteins stimulate different Arp2/3 complex-dependent actin-nucleating activities</article-title><source>Curr Biol</source><volume>11</volume><fpage>1903</fpage><lpage>13</lpage><pub-id pub-id-type="doi">10.1016/S0960-9822(01)00603-0</pub-id></element-citation></ref></ref-list></back><sub-article article-type="article-commentary" id="SA1"><front-stub><article-id pub-id-type="doi">10.7554/eLife.01008.022</article-id><title-group><article-title>Decision letter</article-title></title-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Sundquist</surname><given-names>Wesley</given-names></name><role>Reviewing editor</role><aff><institution>University of Utah</institution>, <country>United States</country></aff></contrib></contrib-group></front-stub><body><boxed-text><p>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 <ext-link ext-link-type="uri" xlink:href="http://elife.elifesciences.org/review-process">review process</ext-link>). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.</p></boxed-text><p>Thank you for sending your work entitled “Tri-color single-molecule imaging shows WASP release from the nascent branch triggers actin nucleation by Arp2/3 complex” for consideration at <italic>eLife</italic>. Your article has been favorably evaluated by a Senior editor and 3 reviewers, one of whom, Wesley Sundquist, is a member of our Board of Reviewing Editors, and one of whom, Thomas Pollard, has agreed to reveal his identity.</p><p>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.</p><p>The authors address how WASP/NPF family members (as represented by the minimal dimerized VCA activation region) act to stimulate actin branching through the Arp2/3 complex. They conclude that: 1) diVCA stimulates initial assembly of the nascent branch by associating with Arp2/3 complex and actin and promoting association of this “nascent branch complex” to the side of the actin filament, and 2) diVCA (but not Arp2/3 complex) then dissociates to create a complex that is elongated into a branch. The latter point is the major new contribution of the manuscript.</p><p>In support of the idea that dissociation of the diVCA component triggers branch formation, the authors use three-color fluorescence experiments to demonstrate that: 1) Arp2/3 complex and diVCA (and presumably also actin monomers) bind filaments simultaneously. These nascent branch complexes usually associate non-productively and the complex typically dissociates as a unit. In a small subset of the cases, however, the diVCA component alone dissociates, and this almost invariably leads to branch formation. Furthermore, diVCA release and initiation of branch elongation occur at exactly the same time (within experimental error). The data in Figure 1 appear to be of high quality, and imply that release of diVCA is likely an obligatory step in the commitment to branch formation. 2) diVCA does not bind Arp2/3 complexes once branches have formed (Figure 2), again consistent with the idea that the presence of diVCA is incompatible with branch formation, and that diVCA departure induces a conformational change(s) that destroy the diVCA binding site and stimulates branch formation. 3) Mutations that alter the half life of diVCA within the nascent branch complexes also alter the rate of branching stimulation, and there is a strong positive correlation between the diVCA dissociation rate constant and the rate constant for branch formation (Figures 3 and 4). These observations support a model that explains how membrane-associated WASP complexes can stimulate actin branching, but then dissociate to create space for subsequent actin polymerization (Figure 5).</p><p>This work is interesting and important, being the first to document at the single molecule level the timing of the interactions of Arp2/3 complex with the partners required to make an actin filament branch – the nucleation promoting factor VCA, actin filaments and actin monomers. The data are generally of high quality, and enhance our understanding of the mechanism of Arp2/3 complex activation by revealing NPF release as a key step in the activation process. Moreover, the study is a nice example of the application of single-molecule imaging to address an important yet incompletely defined biochemical pathway. The manuscript should therefore be acceptable for publication once the following three important issues have been addressed: Additional points noted below are also important, and should be considered carefully and a response provided.</p><p>1) Figure 1: We would urge the authors to provide a more explanatory diagram at the beginning of the paper to draw in the outsider and help explain the points being made in the Introduction, and the key question that is being tested. Figure 1A does this to some extent, but it's small and could be much more explanatory. Figure 1E is very busy and it is difficult to distinguish data points for diVCA wild type from diVCA mutants. The data might be more clearly illustrated in four separate graphs without taking up too much more space. Also, there appear to be outliers for which diVCA release occurs after daughter nucleation. These are not mentioned explicitly in the text, and it is unclear whether these outliers are the ones that fall within “experimental uncertainty”. Moreover, it is unclear what the error bars in this figure represent (and, more generally, the authors need to be careful to describe what error bars represent in all of the different graphs in the figures).</p><p>2) Although the data with diVCA wild type is pretty clear, the data with the diVCA mutants is sometimes a bit confusing. For example, the data in Figure 3C suggests that diVCA-C* dissociates more rapidly from Arp2/3 complex than diVCA-wt. However, the data in Figure 4E suggest that the release rate constant is similar for diVCA-C* and diVCA-wt. Might this is due to the difference between free versus filament-bound Arp2/3 complex? Also, the error bars for some of these panels are fairly large, and it isn't clear whether the observed parameters are statistically significantly different. The authors should therefore carry out statistical tests to determine whether there are significant differences in the measured parameters for wild type and mutant diVCA in Figure 4 (and elsewhere). Without that, it is hard to say whether the trend with the mutants fully supports the main conclusions.</p><p>3) Please also: A) reorganize the figures, and B) reframe the presentation of the central question under study to reflect previous literature precedents more accurately.</p><p>Below, we have raised a series of specific issues for your consideration. We acknowledge that authors have the right to write their own paper, but believe you will agree that the manuscript will be improved once these different issues are addressed, particularly those involving accurate literature citations.</p><p>Additional important points to consider:</p><p><italic>1) Figure organization</italic></p><p>The figures are currently organized based on the methodology rather than on logical flow. The panels in Figure 2 about mutant VCA's beg for the information in Figure 3. We therefore suggest that you either: A) Switch the figure order so that you present the nature of these mutations and fully explore the binding of VCA wild type and mutants before presenting what happens when a mother filament binds these mutant VCA's. Having Figure 2 after Figure 3 will also provide a better opportunity to integrate the two (currently neglected) panels in Figure 2 on mutant VCA's into the presentation of branching by mutant VCA's in Figure 4. B) Alternatively, consider merging Figure 2A,B (diVCAwt) with Figure 1 and merging the remaining panels in Figure 2B (diVCAmutants) with Figure 3.</p><p><italic>2) Presentation</italic></p><p><italic>Introduction</italic></p><p>A fundamental issue is that the study is presented as resolving the literature “paradox” that having WASP tethered to the plasma membrane by its activators (Rho-family GTPases, polyphosphoinositides and SH3 domain proteins) restricts filament growth. In fact, the present work confirms the existing hypothesis about the order of events. That is super, but it's not resolving a “paradox” or “conundrum”.</p><p>This concern is best illustrated by a review of the background leading up to this study. In 1999 Egile et al. and Higgs et al. found that the V motif binds the barbed end of actin, so the V-actin complex does not bind the pointed end of actin filaments and inhibits elongation at that end. Egile et al. imagined that VCA was a “motor of insertional polymerization” that would bind actin monomers and deliver them to the end of the filament and then dissociate rapidly. The current work disproves this hypothesis. On the other hand, Higgs concluded: “Since WA (VCA) only slightly inhibits barbed end elongation at the concentrations we tested, it must rapidly dissociate from the filament barbed end upon addition of WA-bound monomer. Profilin has a similar effect on polymerization; it inhibits nucleation and pointed end elongation but not barbed end elongation.” Thus the concept that rapid V dissociation is required for barbed end elongation is 14 years old.</p><p>Next, Marchand et al. (2001) showed directly that VCA regions of WASP family proteins bind to and dissociate from actin and Arp2/3 complex rapidly. They measured koff from actin monomers to be 3/s and from Arp2/3 complex to be 0.2/s.</p><p>Dayel (2001) discovered that hydrolysis of ATP bound to Arps promotes dissociation of VCA from Arp2/3 complex during branch formation. They concluded “binding N-WASP brings the actin monomer attached to the WH2 domain of N-WASP in contact with the Arp2/3 complex and this stimulates ATP hydrolysis. Step 2: Hydrolyzing ATP to ADP-Pi causes a conformational change on the complex, forming a stable nucleus among Arp3, Arp2, and the conventional actin monomer. Step 3: A new actin filament polymerizes from this nucleus. Step 4: Phosphate release from Arp2/3 complex decreases the affinity for N-WASP and allows the Arp2/3 complex to release its membrane-associated activator.” Dayel (2004) looked at the timing more carefully and concluded “it is very likely that ATP hydrolysis on Arp2, like actin, provides a timing signal to the system. ATP hydrolysis on Arp2/3 would promote release of VCA from the complex and allow a new actin branch to move away from the site of its creation (Dayel et al. 2001).” The enclosed drawing from Dayel (2004) summarizes their understanding of the process. This drawing is very similar to Figure 5 in the current paper. Neither Dayel paper is cited, and this seems inappropriate. [See attached PDF version for image of Figure 5B from Dayel (2004).]</p><p>A new paper from the Nolen lab (Hetrick et al, Chem Bio, 2013) has an updated version of the Dayel diagram in their Figure 6. Of course the present authors would not have known about this new paper, but it illustrates that the basic idea is central to the thinking in the field.</p><p>Although it is less well established that the C motif binds the barbed end of Arp3 and possibly Arp2, previous workers postulated that C might inhibit binding of the first two subunits of the daughter filament. Therefore, several investigators emphasized that early in the nucleation process VCA must inevitably dissociate from Arp2/3 complex and the rates of dissociation measured previously mean that the delay would be on the order seconds. A recent paper from the current authors (Smith et al.) reported the important insight that most interactions of Arp2/3 complex with the side of an actin filament are transient, explaining why Arp2/3 complexes associate stably with actin filaments so slowly (Beltzer; Ti).</p><p>An account of the background with these points would be an appropriate place to begin this paper. Rather than investigating a paradox, this work tests the main hypothesis in the field, which has been put forward by multiple labs.</p><p>Other aspects of the Introduction are a nice account of branching by Arp2/3 complex but several of the citations are misplaced. See the attached pdf with notes.</p><p><italic>Results</italic></p><p>It might be noted that the observation that most Arp2/3 complexes binding to filaments are associated with di-VCA is exactly what is expected, given the thermodynamic argument in Ti et al. (2011), namely that bound VCA increases the affinity of Arp2/3 complex for actin filaments about 30 fold.</p><p>The conclusion “daughter nucleation is essentially always accompanied by Arp2/3 complex retention and diVCA release” is exactly what is expected from previous work, although it is certainly valuable to confirm this order of events by these lovely single molecule observations. Congratulations on the lovely experiments.</p><p>“Thus, diVCA release may serve as the trigger for daughter nucleation.” The wording of this conclusion reinvents the concept of the actin nucleus, which according to historical precedent is the smallest actin oligomer that elongates like a filament. Classic simulations of the time course of spontaneous actin polymerization (labs of Frieden, Pollard and Korn) showed that the nucleus is a trimer. Sept confirmed this interpretation with Brownian dynamics simulations. Even though the trimer is unstable, it adds subunits like any longer filament. The nucleus in the case of Arp2/3 complex is therefore likely consists of the two Arps and one or two actin subunits. Therefore, the nucleus is formed when the first and second actin subunits bind Arp2/3 complex, which can surely happen prior to dissociation of V from the two actins. Thus what is observed in this experiment is not nucleation, but the beginning of elongation, which appears to be limited by dissociation of the diVCA as suggested by others. Therefore throughout the paper “nucleation” should be replaced with “elongation” of the daughter filament (which is what is observed).</p><p>“Most of the tethered Arp2/3 complexes (>80 %) were observed to bind diVCA.” Please explain. Does this mean that 20% never bound di-VCA? Were they dead?</p><p>Machesky et al. (PNAS, 1999) (not cited) showed that mother filaments are required for nucleation, not Mullins et al. (1998).</p><p>The authors should check carefully whether or not there is a precedent for their finding that “the affinity of diVCA for isolated Arp2/3 complex is high, whereas the diVCA affinity for Arp2/3 complex in the branch junction is comparatively low.” Regardless, that observation makes sense given that both the C and V binding sites are likely to be occluded in the branch junction, and since the A motif binds weakly to Arp2/3 complex (Kd ∼ 9 µM, Marchand, 2001).</p><p>The text states “The goal was to modestly perturb VCA interactions with Arp2/3 complex or actin without altering the mechanism by which diVCA stimulates branch formation.” This statement reveals a misunderstanding about the thermodynamics and should be rephrased. It is highly unlikely that one could change the affinity of VCA for its receptors without altering the activation process. The reason is that activation depends on the free energy from Arp2/3 complex binding its three ligands (VCA, actin monomers and actin filament), so if one perturbs any of these free energy changes, one inevitably perturbs activation.</p><p>“With the A* mutant dissociating from Arp2/3 complex more slowly than wild type.” Were you surprised? Why should this be true?</p><p>“Both wild type and mutant complexes displayed lifetime distributions that were fit well with two exponential components (Figure 3D, Figure 3–figure supplement 2D, Table 1), indicating that at least two distinct diVCA-Arp2/3 complex assemblies could form.” The semi-log plot makes it difficult to see the fast component in some of these plots. What is the behavior of each individual Arp2/3 complex? Were they consistently in one of these states or did one Arp2/3 complex behave differently over time? This behavior deserves a more detailed analysis and presentation.</p><p>The presence of multiple complexes is expected – really? Alternatively, given the bivalent nature of the ligand and the receptor, loss of function of one site on a subset of ligands or receptors could result in the observed two lifetimes. Can you rule out this explanation?</p><p>Figure 4A (and the associated supplemental figure) show that wild type diVCA boosted Arp2/3 complex-dependent polymerization only 1.5 fold (microscopy) to 2.0 fold (bulk biochemistry). This result agrees with previous work from the Rubenstein lab on bulk samples (not cited), but it deserves some comment. Describe what was observed by microscopy about branches formed by Arp2/3 complex in the absence of VCA, since this unusual feature of budding yeast Arp2/3 complex is poorly understood. The experiments in the magenta and green boxes in Figure 2 also need more detailed explanations.</p><p><italic>Discussion</italic></p><p>Arasada (2011) has another estimate of NPF density that is more relevant this work on budding yeast Arp2/3 complex than the other examples, namely between 16,000 and 45,000 NPF's per µm2 at sites of endocytosis where actin polymerization depends on Arp2/3 complex.</p><p>Only Ti et al. (2011) presented a molecular structure of Arp2/3 complex showing how the C-region of VCA may occupy a similar location on Arp3 that the V-region occupies on actin. The other cited work infers the location of the C binding sites from indirect data.</p></body></sub-article><sub-article article-type="reply" id="SA2"><front-stub><article-id pub-id-type="doi">10.7554/eLife.01008.023</article-id><title-group><article-title>Author response</article-title></title-group></front-stub><body><p><italic>1)</italic> <italic>Figure 1</italic><italic>: We would urge the authors to provide a more explanatory diagram at the beginning of the paper to draw in the outsider and help explain the points being made in the Introduction, and the key question that is being tested.</italic> <italic>Figure 1A</italic> <italic>does this to some extent, but it's small and could be much more explanatory</italic>.</p><p>We have added a new figure, which we cite in the Introduction. It is intended to illustrate the biological context and to define the question being addressed in the research.</p><p><italic>Figure 1E</italic> <italic>is very busy and it is difficult to distinguish data points for diVCA wild type from diVCA mutants. The data might be more clearly illustrated in four separate graphs without taking up too much more space</italic>.</p><p>We have replaced Figure 1E with a clearer panel (now <xref ref-type="fig" rid="fig2">Figure 2E</xref>) that shows the wild-type data only. Presentation of the mutant data is deferred until later in the paper. We think it important to show all mutant and wild type data superimposed (so that readers can see that the distribution of points is similar in all four cases), but we agree with the reviewers that such a presentation makes it difficult to evaluate which points are significant outliers. We therefore compromised by presenting both the original panel (now <xref ref-type="fig" rid="fig5">Figure 5B</xref>) and new panels that present the data for the four constructs individually (in <xref ref-type="fig" rid="fig2">Figure 2E</xref> and in <xref ref-type="fig" rid="fig5s3">Figure 5—figure supplement 3</xref>).</p><p><italic>Also, there appear to be outliers for which diVCA release occurs after daughter nucleation. These are not mentioned explicitly in the text, and it is unclear whether these outliers are the ones that fall within “experimental uncertainty”</italic>.</p><p>In the wild type data (<xref ref-type="fig" rid="fig2">Figure 2E</xref>) there are no significant outliers as can be seen by the fact all points are on or above the diagonal dashed line, or have error bars that cross it. In the mutant data (<xref ref-type="fig" rid="fig5s3">Figure 5—figure supplement 3</xref>) there is one point for V*, zero for C*, and one for A* that could be significant outliers, out of 41, 49, and 27 observations, respectively. We have marked these two points with asterisks and now explicitly discuss them in the figure caption.</p><p><italic>Moreover, it is unclear what the error bars in this figure represent (and, more generally, the authors need to be careful to describe what error bars represent in all of the different graphs in the figures)</italic>.</p><p>Error bars in this figure represent S.E. of the daughter nucleation time measurements derived from fits like those shown in <xref ref-type="fig" rid="fig2">Figure 2C</xref>, blue. This is now stated in the figure caption and described in the methods. We have reviewed all figure captions and specified the meaning of the error bars where this was not done before.</p><p><italic>2) Although the data with diVCA wild type is pretty clear, the data with the diVCA mutants is sometimes a bit confusing. For example, the data in</italic> <italic>Figure 3C</italic> <italic>suggests that diVCA-C* dissociates more rapidly from Arp2/3 complex than diVCA-wt. However, the data in</italic> <italic>Figure 4E</italic> <italic>suggest that the release rate constant is similar for diVCA-C* and diVCA-wt. Might this is due to the difference between free versus filament-bound Arp2/3 complex</italic>?</p><p>Yes, that is the likely explanation. Indeed, the reason for the measurements in Figure 4E (it is now <xref ref-type="fig" rid="fig6">Figure 6D</xref>) is to characterize the dissociation kinetics of VCA from filament–bound Arp2/3 complex. They do in fact turn out to be different from those with off-filament Arp2/3 complex. The text now states this clearly.</p><p><italic>Also, the error bars for some of these panels are fairly large, and it isn't clear whether the observed parameters are statistically significantly different. The authors should therefore carry out statistical tests to determine whether there are significant differences in the measured parameters for wild type and mutant diVCA in</italic> <italic>Figure 4</italic> <italic>(and elsewhere). Without that, it is hard to say whether the trend with the mutants fully supports the main conclusions</italic>.</p><p>There are two kinds of conclusions related to comparison of the wild type and mutant constructs (data that are now in <xref ref-type="fig" rid="fig6">Figure 6</xref> as well as in some of the figure supplements) that require statistical support. The first are pairwise comparisons (e.g., statements that a particular mutant is different than wild type with respect to a particular measurement). For each of the pairwise comparisons that we utilize, we have now performed a significance test and p values are given in the text or figure captions. The second type of conclusion is the apparent correlations (<xref ref-type="fig" rid="fig6">Figure 6E</xref> and <xref ref-type="fig" rid="fig6s1">Figure 6—figure supplement 1</xref>). For each of these we now report the correlation coefficients and further report a bootstrap calculation, which shows that it is unlikely (p = ∼0.005 and ∼0.03, respectively) that the high measured correlation coefficient was obtained by chance from two quantities that are in truth uncorrelated. This analysis is described in the Materials and methods section of the revised manuscript.</p><p><italic>3) Please also: A) reorganize the figures, and B) reframe the presentation of the central question under study to reflect previous literature precedents more accurately. Below, we have raised a series of specific issues for your consideration. We acknowledge that authors have the right to write their own paper, but believe you will agree that the manuscript will be improved once these different issues are addressed, particularly those involving accurate literature citations</italic>.</p><p>As described below, we have largely incorporated these suggestions. In a few cases we did not agree that the suggested changes would be improvements, and we explain why below.</p><p><italic>Additional important points to consider</italic>:</p><p>1) Figure organization</p><p><italic>The figures are currently organized based on the methodology rather than on logical flow. The panels in Figure 2 about mutant VCA's beg for the information in Figure 3. We therefore suggest that you either: A) Switch the figure order so that you present the nature of these mutations and fully explore the binding of VCA wt and mutants before presenting what happens when a mother filament binds these mutant VCA's. Having Figure 2 after Figure 3 will also provide a better opportunity to integrate the two (currently neglected) panels in Figure 2 on mutant VCA's into the presentation of branching by mutant VCA's in Figure 4. B) Alternatively, consider merging Figure 2A,B (diVCAwt) with Figure 1 and merging the remaining panels in Figure 2B (diVCAmutants) with Figure 3</italic>.</p><p>The principal concern seems to be that we originally had the data from the mutants and the wild type presented together in figures in a way that was inconsistent with the order of presentation in the text. We have now split out the mutant data so that all data are presented in the order described in the text. This is essentially adopting the order of presentation described above as alternative B, although we have grouped the panels slightly differently than what was suggested.</p><p>Specifically, we have:</p><p>• Added the new <xref ref-type="fig" rid="fig1">Figure 1</xref> (as described in response to referee comment 1).</p><p>• Replaced Figure 1E (now <xref ref-type="fig" rid="fig2">2E</xref>) with version showing wild type data only.</p><p>• Replaced Figure 2 (now <xref ref-type="fig" rid="fig3">Figure 3</xref>) with version showing wild type data only.</p><p>• Figure 3 (now <xref ref-type="fig" rid="fig4">Figure 4</xref>) is unchanged except for minor edits.</p><p>• Added <xref ref-type="fig" rid="fig5">Figure 5</xref> consisting of the mutant data removed from the earlier figures (old 4A, old 1E and old 2B).</p><p>• Old Figure 4B–F is now <xref ref-type="fig" rid="fig6">Figure 6</xref>.</p><p>• Old Figure 5 (now <xref ref-type="fig" rid="fig7">7</xref>) is edited to clarify but otherwise unchanged.</p><p>2) Presentation</p><p><italic>Introduction</italic></p><p><italic>A fundamental issue is that the study is presented as resolving the literature “paradox” that having WASP tethered to the plasma membrane by its activators (Rho-family GTPases, polyphosphoinositides and SH3 domain proteins) restricts filament growth. In fact, the present work confirms the existing hypothesis about the order of events. That is super, but it's not resolving a “paradox” or “conundrum”</italic>.</p><p>We did not intend to present our work as resolving a “literature paradox”. Rather, we are addressing the inherent conflict between the need for Arp2/3 complex to associate with membrane bound activators at one stage of the daughter filament formation process and to be distant from those activators at a subsequent stage. By directly observing the order of events by which the complex moves from one stage to the next, our study defines the mechanism by which this conflict is resolved.</p><p>In order to more clearly define the problem that is resolved by our study, and to more fully place the work in the context of the existing literature, we have slightly rephrased the Abstract and made substantial changes to the Introduction. The changes to the Introduction include:</p><p>• We added <xref ref-type="fig" rid="fig1">Figure 1</xref> and accompanying text in the Introduction. We feel that these work together to more clearly define the most important problem that is addressed by our study.</p><p>• We have expanded and reorganized the beginning of the Introduction to include additional literature citations (mostly ones suggested below by the reviewers) to more thoroughly place the work in the context of existing knowledge about how VCA proteins function to stimulate daughter filament initiation.</p><p>• We added a new paragraph that emphasizes (with numerous literature citations) that separation of the new branch junction from the membrane- or surface-bound activator is already well established in a variety of biological contexts and in vitro.</p><p>• We added a new penultimate paragraph to the Introduction that explicitly states (with citations) that VCA departure prior to initiation of daughter filament growth was previously hypothesized.</p><p>• We removed the words “paradox” and “conundrum” in an effort to avoid any implication that there are inconsistencies between different results in the existing literature, an implication that we did not intend.</p><p>We believe that these changes will largely address the concerns of the referees about the original introduction. We anticipate that they will also make the context of the work clearer to readers, particularly those who may be less familiar with prior work in the field.</p><p><italic>This concern is best illustrated by a review of the background leading up to this study…</italic>.</p><p>Although we concur with most of what is said below, there are a few stated conclusions and implications that we do not think are well supported by the literature or that are contradicted by it. However, since it is explicit that some of what follows is intended to be illustrative, we have responded only to parts that explicitly recommend changes to the manuscript.</p><p><italic>…In 1999 Egile et al. and Higgs et al. found that the V motif binds the barbed end of actin, so the V-actin complex does not bind the pointed end of actin filaments and inhibits elongation at that end. Egile et al. imagined that VCA was a “motor of insertional polymerization” that would bind actin monomers and deliver them to the end of the filament and then dissociate rapidly. The current work disproves this hypothesis. On the other hand, Higgs concluded: “Since WA (VCA) only slightly inhibits barbed end elongation at the concentrations we tested, it must rapidly dissociate from the filament barbed end upon addition of WA-bound monomer. Profilin has a similar effect on polymerization; it inhibits nucleation and pointed end elongation but not barbed end elongation.” Thus the concept that rapid V dissociation is required for barbed end elongation is 14 years old</italic>.</p><p><italic>Next, Marchand et al. (2001) showed directly that VCA regions of WASP family proteins bind to and dissociate from actin and Arp2/3 complex rapidly. They measured koff from actin monomers to be 3/s and from Arp2/3 complex to be 0.2/s</italic>.</p><p><italic>Dayel (2001) discovered that hydrolysis of ATP bound to Arps promotes dissociation of VCA from Arp2/3 complex during branch formation. They concluded “binding N-WASP brings the actin monomer attached to the WH2 domain of N-WASP in contact with the Arp2/3 complex and this stimulates ATP hydrolysis. Step 2: Hydrolyzing ATP to ADP-Pi causes a conformational change on the complex, forming a stable nucleus among Arp3, Arp2, and the conventional actin monomer. Step 3: A new actin filament polymerizes from this nucleus. Step 4: Phosphate release from Arp2/3 complex decreases the affinity for N-WASP and allows the Arp2/3 complex to release its membrane-associated activator.” Dayel (2004) looked at the timing more carefully and concluded “it is very likely that ATP hydrolysis on Arp2, like actin, provides a timing signal to the system. ATP hydrolysis on Arp2/3 would promote release of VCA from the complex and allow a new actin branch to move away from the site of its creation (Dayel et al. 2001).” The enclosed drawing from Dayel (2004) summarizes their understanding of the process. This drawing is very similar to Figure 5 in the current paper. Neither Dayel paper is cited, and this seems inappropriate. [See attached PDF version for image of Figure 5B from Dayel (2004).</italic>]</p><p>We have added a citation to both Dayel papers in the new section in the final paragraph of the Introduction. As an aside, we note here that the image provided by the referees shows a hypothesis that is not identical to our conclusions. First, it posits a key role for ATP hydrolysis by Arp2/3 complex, a question about which our work is silent but which is addressed in subsequent studies (Martin, 2006; Ingerman, 2013). (These subsequent studies suggest that hydrolysis does not, in fact, accompany initiation of the daughter filament, as suggested in Dayel (2004), but rather is related to debranching.) Second, it does not address the central concern of our paper, which is the order of VCA release and initiation of daughter filament polymerization. It shows the two lumped into a single, final step so that their order is not specified. In this respect the figure is more akin to our <xref ref-type="fig" rid="fig1">Figure 1</xref> (which illustrates the key question we are asking in our study) than it is to our <xref ref-type="fig" rid="fig7">Figure 7</xref> (which illustrates the answer).</p><p><italic>A new paper from the Nolen lab (Hetrick et al, Chem Bio, 2013) has an updated version of the Dayel diagram in their Figure 6. Of course the present authors would not have known about this new paper, but it illustrates that the basic idea is central to the thinking in the field</italic>.</p><p>We have added citations to the Hetrick paper at the relevant locations in the Introduction. We would respectfully disagree with the reviewers’ implication that the inclusion of this model in a single paper published a few months ago establishes that the idea is central to the thinking in the field. In our estimation, a more reasonable conclusion is that some features in the model published by Hetrick et al. were influenced by the as-yet unpublished work from the Nolen lab (we cite their submitted publication) and perhaps also by reports of our work reported in this paper, which has been presented at conferences.</p><p><italic>Although it is less well established that the C motif binds the barbed end of Arp3 and possibly Arp2, previous workers postulated that C might inhibit binding of the first two subunits of the daughter filament. Therefore, several investigators emphasized that early in the nucleation process VCA must inevitably dissociate from Arp2/3 complex and the rates of dissociation measured previously mean that the delay would be on the order seconds</italic>.</p><p>While there is a great deal of literature that touches on related points, we were unable to find any published studies that explicitly “postulated that C might inhibit binding of the first two subunits of the daughter filament”.</p><p><italic>A recent paper from the current authors (Smith et al.) reported the important insight that most interactions of Arp2/3 complex with the side of an actin filament are transient, explaining why Arp2/3 complexes associate stably with actin filaments so slowly (Beltzer; Ti)</italic>.</p><p><italic>An account of the background with these points would be an appropriate place to begin this paper. Rather than investigating a paradox, this work tests the main hypothesis in the field, which has been put forward by multiple labs</italic>.</p><p>As noted earlier, we have expanded the Introduction to incorporate additional background, including many of the references cited by the referees that we did not already include. We did not include all such material because we regard some of it as related but not essential, and we think that too much additional length in the Introduction would weaken the presentation. Also noted earlier is that we have added a section explaining that it is well established that VCA must depart before the daughter filament elongates too far. We do not agree (for reasons that are in part explained above) that the field has coalesced on a single hypothesized mechanism to explain this. For this reason, we did not (as the reviewers seem to be suggesting) adopt this hypothesis as a rhetorical device to introduce the question being asked in the paper. Instead, we used the new <xref ref-type="fig" rid="fig1">Figure 1</xref> and related discussion. We think this is a more effective presentation and will also make the paper more accessible to readers outside of the field. We realize that the reviewers may not agree with this decision, but we ask that some deference be given to authors’ choices on how they introduce their work.</p><p><italic>Other aspects of the introduction are a nice account of branching by Arp2/3 complex but several of the citations are misplaced. See the attached pdf with notes</italic>.</p><p>Thank you for the notes on the pdf. These suggestions have been incorporated by changing the references and/or citing text.</p><p><italic>Results</italic></p><p><italic>It might be noted that the observation that most Arp2/3 complexes binding to filaments are associated with di-VCA is exactly what is expected, given the thermodynamic argument in Ti et al. (2011), namely that bound VCA increases the affinity of Arp2/3 complex for actin filaments about 30 fold</italic>.</p><p>The observation that nearly all Arp2/3 complexes binding to filaments are accompanied by diVCA is not a reflection of the stability of the ternary complex. Rather, it is mainly a consequence of the fact that Arp2/3 complexes free in solution (i.e., not attached to a filament) are nearly saturated with diVCA under the conditions of the experiments.</p><p><italic>The conclusion “daughter nucleation is essentially always accompanied by Arp2/3 complex retention and diVCA release” is exactly what is expected from previous work, although it is certainly valuable to confirm this order of events by these lovely single molecule observations. Congratulations on the lovely experiments</italic>.</p><p>This comment seems to imply that expectations and hypotheses do not satisfy the need for direct experimental data. We agree.</p><p><italic>“Thus, diVCA release may serve as the trigger for daughter nucleation.” The wording of this conclusion reinvents the concept of the actin nucleus, which according to historical precedent is the smallest actin oligomer that elongates like a filament. Classic simulations of the time course of spontaneous actin polymerization (labs of Frieden, Pollard and Korn) showed that the nucleus is a trimer. Sept confirmed this interpretation with Brownian dynamics simulations. Even though the trimer is unstable, it adds subunits like any longer filament. The nucleus in the case Arp2/3 complex is therefore likely consists of the two Arps and one or two actin subunits. Therefore, the nucleus is formed when the first and second actin subunits bind Arp2/3 complex, which can surely happen prior to dissociation of V from the two actins. Thus what is observed in this experiment is not nucleation, but the beginning of elongation, which appears to be limited by dissociation of the diVCA as suggested by others. Therefore throughout the paper “nucleation” should be replaced with “elongation” of the daughter filament (which is what is observed)</italic>.</p><p>In the original manuscript, we used the term “nucleation” to refer both to the overall process shown in the final figure and to the specific step in which the daughter filament begins to grow. To minimize any possible confusion, we now refer to the latter as “initiation of daughter filament growth”. We think this terminology is consistent with the point made by the reviewers but is more apt than “elongation”.</p><p><italic>“Most of the tethered Arp2/3 complexes (>80 %) were observed to bind diVCA.” Please explain. Does this mean that 20% never bound di-VCA? Were they dead</italic>?</p><p>Yes, <20% never bound di-VCA. There are multiple possible explanations for the apparently inactive fraction, including 1) incompletely assembled “dead” Arp2/3 complexes, 2) fully assembled and active complexes bound to the surface in an orientation or local environment that precludes VCA binding, and 3) presence of a small number of the dye-biotin adducts used to label the Arp2/3 complex that never were or are no longer attached to protein.</p><p><italic>Machesky et al. (PNAS, 1999) (not cited) showed that mother filaments are required for nucleation, not Mullins et al. (1998)</italic>.</p><p>Thanks; we now cite the correct reference.</p><p><italic>The authors should check carefully whether or not there is a precedent for their finding that “the affinity of diVCA for isolated Arp2/3 complex is high, whereas the diVCA affinity for Arp2/3 complex in the branch junction is comparatively low.” Regardless, that observation makes sense given that both the C and V binding sites are likely to be occluded in the branch junction, and since the A motif binds weakly to Arp2/3 complex (Kd ∼ 9 µM, Marchand, 2001)</italic>.</p><p>We now say that the results are consistent with data from previous studies and cite two. We agree that the result makes sense given the Marchand result and we now point this out in the Discussion.</p><p><italic>The text states “The goal was to modestly perturb VCA interactions with Arp2/3 complex or actin without altering the mechanism by which diVCA stimulates branch formation.” This statement reveals a misunderstanding about the thermodynamics and should be rephrased. It is highly unlikely that one could change the affinity of VCA for its receptors without altering the activation process. The reason is that activation depends on the free energy from Arp2/3 complex binding its three ligands (VCA, actin monomers and actin filament), so if one perturbs any of these free energy changes, one inevitably perturbs activation</italic>.</p><p>We disagree that the statement reveals a misunderstanding about the thermodynamics. Rather, it arises from different meanings attached to the term “mechanism” by the authors and the reviewers. To lessen the possibility of being misunderstood, we have substituted the term “reaction pathway” for “mechanism” to clarify that we are making a statement about the sequence of chemical species in the reaction rather than about free energy changes or rate constants of the reaction steps.</p><p><italic>“With the A* mutant dissociating from Arp2/3 complex more slowly than wild type.” Were you surprised? Why should this be true</italic>?</p><p>The A* mutation was indeed designed with this goal in mind. In 2001, Zalevsky et al. reported that VCA peptides have an activation step following assembly of VCA, actin and Arp2/3 complex onto the mother filament. The apparent activation rate constant for this step differed between N-WASP, WASP and WAVE1. The origin of this difference was concluded to be in the length of the acidic region, with a longer acidic region (e.g. in N-WASP) giving faster rate and a shorter region (e.g. in WAVE1) giving slower rate. We wondered if the ‘activation step’ might be departure of VCA. If so, shortening the N-WASP acidic region might make it more WAVE1-like, slowing dissociation. We proceeded to essentially reverse the series of experiments used by Zalevsky et al. by deleting several acidic residues from the N-WASP acidic region. Multiple VCA modifications along these lines generated VCA dimers with lower activity in pyrene actin assays than our wild type diVCA. The “A*” mutant had the smallest perturbation (in terms of amino acid changes) and was therefore chosen for further analysis in the single molecule measurements. In the interest of keeping the manuscript focused, we have not included a description of the reasoning involved in this aspect of the experimental design.</p><p><italic>“Both wild type and mutant complexes displayed lifetime distributions that were fit well with two exponential components (Figure 3D, Figure 3–figure supplement 2D, Table 1), indicating that at least two distinct diVCA-Arp2/3 complex assemblies could form.” The semi-log plot makes it difficult to see the fast component in some of these plots. What is the behavior of each individual Arp2/3 complex? Were they consistently in one of these states or did one Arp2/3 complex behave differently over time? This behavior deserves a more detailed analysis and presentation</italic>.</p><p>We added a magnified inset so that the fast component of the distributions in Figure 3D (now 4D) is more visible. Individual Arp2/3 complexes show both short and long events. In a more detailed analysis with the V* mutant, which is the one for which the short and long components are most clearly resolved, we saw no evidence that the short and long lifetime components segregated into different subpopulations of individual Arp2/3 complexes. Both results are now stated in the Results section and further details are given in Methods. We agree with the reviewer that the multicomponent lifetime distributions are intriguing: they could potentially shed light on the different modes of interaction between VCA dimers and Arp2/3. However, such an investigation would likely require analysis of Arp2/3 complex mutations and of a more extensive panel of VCA mutations. Such work is beyond the scope of the present study.</p><p><italic>The presence of multiple complexes is expected - really? Alternatively, given the bivalent nature of the ligand and the receptor, loss of function of one site on a subset of ligands or receptors could result in the observed two lifetimes. Can you rule out this explanation</italic>?</p><p>We changed this passage to say that the result is “consistent” with the results of the previous studies cited, not “expected” from them. We cannot rule out the alternative explanation raised by the reviewers, but we see no reason to explicitly discuss it in the manuscript since we already give multiple examples of other possible explanations for the observed behavior and there is currently no basis to choose between them. Distinguishing between these alternatives is not essential to the conclusions of our paper.</p><p><italic>Figure 4A (and the associated supplemental figure) show that wild type diVCA boosted Arp2/3 complex-dependent polymerization only 1.5 fold (microscopy) to 2.0 fold (bulk biochemistry). This result agrees with previous work from the Rubenstein lab on bulk samples (not cited), but it deserves some comment</italic>.</p><p>We surmise that the reviewers feel that 1.5–2.0 fold activation is somehow anomalously low and, further, that this value is explained by the significant activity of budding yeast Arp2/3 complex in the absence of VCA. However, another factor is that the N-WASP diVCA construct we use was deliberately chosen because it is only a weak activator (this helped us to engineer mutants that increased activation). This is illustrated by data in <xref ref-type="fig" rid="fig2s2">Figure 2–figure supplement 2C,D</xref> which show for example, that N-WASP VVCA dimers are much better activators than N-WASP VCA dimers like the one we used as our wild type construct. We now mention both factors in the caption to that figure and we cite Wen & Rubenstein (2005) (although they observed high basal activity only with yeast actin and did not see it with rabbit muscle actin).</p><p><italic>Describe what was observed by microscopy about branches formed by Arp2/3 complex in the absence of VCA, since this unusual feature of budding yeast Arp2/3 complex is poorly understood. The experiments in the magenta and green boxes in Figure 2 also need more detailed explanations</italic>.</p><p>A thorough presentation of the microscopy observations without VCA was included in our previous publication (Smith et al., PNAS 2013), which we cite. We see no reason to recapitulate that in the present paper, particularly since the small subpopulation of Arp2/3 complexes that bind to filament unaccompanied by detectable di-VCA fluorescence are excluded from our analyses (e.g., in <xref ref-type="fig" rid="fig6">Figure 6</xref>). As described above, the data in the magenta and green boxes from Figure 2B have now been moved to <xref ref-type="fig" rid="fig5">Figure 5C,D</xref>; they are described in the caption.</p><p><italic>Discussion</italic></p><p><italic>Arasada (2011) has another estimate of NPF density that is more relevant this work on budding yeast Arp2/3 complex than the other examples, namely between 16,000 and 45,000 NPF's per µm2 at sites of endocytosis where actin polymerization depends on Arp2/3 complex</italic>.</p><p>The densities mentioned in Arasada (2011) appear to be densities of SH3 domains of F-BAR proteins, inferred from the electron microscopy work of Frost et al. Given the similar numbers of Wsp1 in close proximity to the F-BAR proteins, this data does (albeit somewhat indirectly) support the point being made in the manuscript, and we now cite it.</p><p><italic>Only Ti et al. (2011) presented a molecular structure of Arp2/3 complex showing how the C-region of VCA may occupy a similar location on Arp3 that the V-region occupies on actin. The other cited work infers the location of the C binding sites from indirect data</italic>.</p><p>We have added citation to the Ti paper. We think that the other references are also appropriate to support the statement made in the manuscript.</p></body></sub-article></article>