I've come up with a synthetic route to amide-substituted triazolopyrazines - what do you think?
Description cross-pasted from my blog:
Compound 13 seems tolerably cheap but that's an anomaly - all other suppliers list it as $200/g and it may not even exist. Given the and budget of the project, it's probably worth a week of time to synthesise it from the readily available 9, which can be esterified using standard conditions (many literature examples available) to 10, which is converted to the known 13 quickly with only one chromatography step in this 2011 J Med Chem paper.
Saponifying 13 to the acid earlier in the synthesis seems like a better decision in this case - otherwise, the hydrazine of step "e" might also displace methanol from the ester to make the undesired double hydrazide. The free acid should also not react with the PIDA of step "g" or the aldehyde of step "f". It also shouldn't self-condense with its own hydrazide. The acid 17 can then be subjected to standard amide-bond forming techniques.
When I say the double hydrazide formed by hydrazine reacting with 13 is "undesired", I recognise that it could potentially be a useful linker as well.
My suggestion would be to use the same Cl-intermediate (Compound 2 in your blog post) and do a Pd mediated carbonylation to give you the methyl ester. This way you will utalize the same common intermediate necessary for the synthesis of the ethers. Once the ester is in hand than the amides are straight forward to prepare.
Yeah, I thought about carbonylation - This is somewhat of a "balance" issue where I've picked the old-fashioned heterocyclic bucket chemistry since I think my project students would rather appreciate a few weeks of exact preps and literature data to follow, but it would be more generally useful for the project as a whole.
 and I'm not going to let my students play with carbon monoxide yet!
Carbonylation seems fairly well-explored - plenty of literature for 2-chloropyrazine, high yielding e.g. this one. There's also a few procedures on 5,6-bicycles with a bridgehead nitrogen, for example patent WO2013059594 page 81, here (link may not work) but nothing on our exact core that I've found so far. I've not done very much digging, this isn't meant to be exhaustive.
Further to my earlier grousing at the entirely sensible suggestion of @JoieG, there's an industrial chemistry group in my university that's set up to do high pressure carbon monoxide reactions who are willing to lend a hand. I shall explore this part myself once we have a decent lump of the chloride in hand (and if we can come up with a way of purifying it that doesn't involve chromatography, so much the better!)
@PatrickThomson Purification of the ester without chromatography may be difficult because most carbonylations use PdCl2(PPh3)2 as the catalyst, which leads to PPh3 by-products which may be hard to crystalize away from your compound. I think it is a wonderful idea for you to utilize a high pressure facility for these reactions. They will have all the appropriate facilities for detection and safety.
I'm sure carbonylation is the way to go for this synthesis and it would be good to have a facility on board so we should certainly embark on this route. However, I also think its worth pursing a route similar to the one that @PatrickThomson suggests if we want to get some more students involved in the synthesis, particularly if they are in undergraduate labs. A summary is coming up....
Alternatively, you can oxidize 2-chloro-6-methylpyrazine (30£/g, Apollo) with sulphuric acid / potassium dichromate to obtain intermediate 14. It works well with 3-chloropyridazines. Good "old-fashioned bucket chemistry".
I've summarised the ideas from @sabinllm @PatrickThomson and @JoieG here: http://malaria.ourexperiment.org/uri/44b so that we can invite others to contribute suggestions/comments. Thanks for your ideas everyone. I favour @JoieG's approach but think if @sabinllm's idea works we could also start a synthesis down this route.
I have a student making 13 from 9 now, but I do also favour the carbonylation.
I have posted these comments at the bottom of the blog article and will also post them here so that people don't miss them...
Comments from Jochen Robert Brandt (PhD) who posted to my personal facebook page when I shared the OSM post:
'Compound 11 has a Texas nitrogen ;)'
'It won't let me comment below your post, so just a quick write-up from me: I would try to get to compounds 2/17/19 without the Ar group; that could make the synthesis easier. I bet that Baran's radical Minisci conditions would arylate in the desired position. Look at Fionn's (AEW edit - Fionn O'Hara, a postdoc that both Jochen and I completed our PhD with in the Gaunt Group and this is a paper that she published with Phil Baran during her postdoc) recent JACS paper for details on influencing the selectivity. http://pubs.acs.org/doi/pdf/10.1021/ja406223k Alternatively, using the Ester of 19 to direct a cyclometalation could also work.
'Sorry for spamming: I don't see the transformation from 16 to 17. You break a C-N bond (N-Ar) and then form a C-C bond (C-Ar) using DCM (the only C1 source in the conditions)? There's no citation for that step in the blog post...'
11, there's two ways of drawing N-Oxides - as the 4-coordinate zwitterion, or the pentavalent N=O - both are accurate. As for 16-17, it's our oxidation but I got sloppy, it should be Pyrazine-NH-N=CH-Ar - don't have the editing tools around just now.
@JoieG this is where we're up to (combined with http://malaria.ourexperiment.org/the_osm_blog/8173/post.html) on planning/doing the amides. Correct me if I'm wrong, but we don't have details on how the CRO made the amides yet, correct? Is there a way of finding that info Joie? At least the key step(s)?
Just to clarify for those such as @JoieG not currently doing the synthesis, could @alintheopen and @PatrickThomson just quickly clarify what their (or their students' !) short-term synthetic goals are and an approximate timeline for when they might be done? I know it can be difficult to say since there is methodology to work out.
Update on progress: Eduvie has made 11 and will have gram quantites of 13/14 in hand by the 18th or so, and will synthesise some amides from it [edit: 4-aminopyridine as a model, then the CF3 derivative as per this plan) Devon is making the same chloride as @alintheopen and will have hundred milligram quantities in hand by the 18th or so, which will be coupled with various alcohols. Once an optimised procedure for the chloride is in place, we'll subject some to carbonylation.
If the carbonylation route is the preferred one, I would suggest exploring reaction conditions that do not involve the use of a high pressure reactor, since not many people have access to this kind of equipment and will prevent other project collaborators from following the procedure. Pressure tubes or microwave vials should be OK.
I just noticed that the excellent carbonylation paper @PatrickThomson found (http://pubs.acs.org/doi/abs/10.1021/ol0498287) comes from Merck. Any chance of some advice from that group, do you think @JoieG ? Naturally we can just have a go, but it would be handy to have someone well versed in this reaction cast their eye over our specific case.
Update 14th Nov 2013
Joie Garfunkle analysed the weekly reports from the CRO and deduced a common route used for the synthesis of amide compounds, see below:
This route contains several analogies to a suggested improvement to the synthesis of the triazolopyrazine core by Stefan Debbert when he cited an ACIE paper that employs Chloromamine-T as oxidant in combination with 2-MeTHF solvent rather than CH2Cl2 and PIDA (DOI: 10.1002/anie.201001999). The ACIE paper also employs a Pd-catalysed coupling of an aldehyde derived hydrazone with a 2-chloropyrazine prior to oxidative cyclisation.
This route also demonstrates the importance of starting material 14 - a commercial source or cheap/quick synthesis is required. Additionally, it suggests that Patrick's route could be problematic as the hydrazine may react with the carboxylic acid in preference to displacement of the chloride. References for each of the steps need to be found but so far this route looks quite robust.
I still think that carbonylation is the most attractive route long term as it makes use of common intermediates. However, if optimisation is required then it might be quicker to go for a route involving more steps if they can a) be performed by undergraduates and b) require less time spent on reaction improvement.
*The OSM Blog has also been updated (http://malaria.ourexperiment.org/uri/44b)
My undergrad research student and I found a cheap source of the 2,6-dichloropyrazine ($60/100 g), and the material looked good when we received it. We're going to try and replace a chlorine with a cyano group - there are some Pd-cat. methods that use Zn(CN)2 that look straightforward -- and then hydrolyze the CN to a COOH. We'll keep the notebook posted as we go.
@sdebbert awesome! If you could let us know the 2,6-dichloropyrazine source that would be great as ours was still a bit pricey. Join the notebook http://malaria.ourexperiment.org/triazolopyrazine_se when you get chance, good luck. Alice
Sure! Here's the link:
It's a bit wet, but the NMR otherwise looks fine. I think they've got quite a few pyrazines cheap.
I'll post as soon as we can make some headway on the synthesis, our materials just came in!
OSM Blog post coming very soon but....which amides should we make next? @JoieG highlighted two structures (http://malaria.ourexperiment.org/uri/41f) that we are going to target but which others should we make and why?
Does anybody (perhaps a chemist not based in a lab) have time to do a search of commercially available compounds on e-molecules or pubchem for example? Would be useful to know what we can buy, preferably to be bought and tested inside Europe to save time/import costs.
Apologies if I have missed something, but why is there still interest in preparing isoindoline carboxamides? It looks like the carbonyl in MMV669000 is detrimental for activity since it induces a different torsion between the "rigid" isoindoline and the neighbouring RHS benzene ring. On the other hand, the benzylic linker in MMV668958 confers some flexibility and perhaps allows the LHS benzene ring to adopt the preferred conformation seen in MMV669848, and despite the carbonyl, activity is retained. Same story with MMV670246: a bit more rigid, inactive.
If N-dealkylation /ring opening is believed to be the main reason for the poor microsomal half life, and we want to keep a nitrogen atom in that part of the molecule, we could explore this hypothesis by: a) changing the hybridization and lowering pKa (compounds 1 & 2, X= CH or N), b) lowering the pKa (compound 3), c) changing the shape of the surroundings and hope for the best (compound 3, forgive the un-substituted thiophene, but you get the idea...) c) oxidise the nitrogen atom itself (compound 4).
If this is not the right thread for discussing this topic, please could you point me in the right direction @alintheopen ?
Alice @alintheopen if you point me to exactly which set of structures (SMILES and InChIs) you want to explore for commercial analogs (as close 2D Tanimoto neigbours first and maybe PubChem 3D) I'll take a look around
@sabinllm I believe that the amides improved the metabolic stability, which the other high-potency compounds didn't do. I'm kind of guessing though.
@patrickthomson Yes, that's the hypothesis. I was just suggesting that isondolinone carboxamides might not be the best option and the above-mentioned compounds could be an interesting alternative if they turn out to be inactive. Other flexible amides should be alright too.
@sabinllm You make some very good points about the proposed amide compounds. In general the linker to the pyrazine ring plays a crucial role in activity, and conformational restrictions within the side-chain did lower potency. The previous amides have shown improved metabolic stability despite decreased potency. The newly proposed amides were to capitalize on the improved RLM stability with an eye towards building back potency through other modifications. However, modifying the amine linker by altering basicity of the amine and steric bulk could also be used to effect the microsomal stability. However, until we have the results from the MetID, we are unsure of the route of metabolism. I would also keep an eye on the lipophilicity of some of the proposed compounds; such as, the CF3 indolines and the thiophene since the solubility may suffer with these analogues.
Discussion of the amide synthesis has moved to #121
Carbonylation has also been updated: #126
Keeping this issue open because of the non-amide structures proposed by Sabin that are relevant to the amide design: #123
Being discussed at July 2014 online meeting http://malaria.ourexperiment.org/osddmalaria_meeting_/10142
Since the structures were covered in the July meeting, and since the Series 4 wiki has been updated with this Issue, it's being closed.