Desired Compounds Consultation

Published by pylioja on 8 June 2012 - 4:32am


Request for Help

We're looking to identify a new set of compounds for the next round of optimisation. This is happening in addition to sourcing of commercially available analogues that will fill a bit more of the SAR space but aren't necessarily exactly what we want.

We've posted a list of compounds on OpenWetWare that contains a range of the things were after, along with a SMARTS filter summary.  We've had feedback and ideas submitted before here on TSL and some of those ideas have been included in this list. Below I've posted the 10 "priority" compounds from the list. It's very much open to debate so get involved. Once we've had a bit of feedback, we'll settle on a definitive list and go after them by any means necessary. The plan was to mostly concentrate on the side-chain, leaving the aryl pyrrole unit mostly untouched for the moment.


[Edit 0905 AEST 15 June 2012: Added letter identifiers for compounds to aid discussion.]


MatTodd's picture

Great. So this starts a 2-week or so period of consultation, after which we will have a list of the 10 Most Wanted - one list for synthesis, one for sourcing from suppliers. We'll take some, and then put out to tender for the others. Final list to be ready by about June 22. Any comments on why/why not compounds should be on this list would be gratefully received. Some are simple explorations of isosteres. Some are curveballs.

Cool way of choosing which compounds to screen.
As a matter of interest, have you thought any more about exploring a different series of compounds considering the lack of efficacy in vivo? Or is the idea that lack of efficacy in vivo is due to poor pharmokinetics that can be optimised out?

MatTodd's picture

Hi Iain. Well we are indeed starting other series in the background. This is Jimmy Cronshaw's project. He's liking the theinopyrimidine set at the moment and has nearly cracked the parent:

But we're doing another round on the arylpyrroles. Firstly the compounds are very potent, and some hit gametocytes. While there was no oral efficacy there is the possibility that this was due to low solubility. We're hoping to acquire data from the in vivo screen that might confirm that. The aim of the above sets (particularly the set to synthesise) is to diversify the upper portion within sensible physicochemical boundaries. If we don't ID a good-looking compound with good solubility then we'll be tempted to hop to a different series with no hard feelings, yes. So the purpose of this consultation (for about another week) is to make sure we're going after a good set of compounds and don't miss something obvious, while throwing in a few non-obvious analogs.

The poor bioavailability is likely due to a combination of poor solubility, low permeability and in some cases poor metabolic stability. Given the exciting activity of the series against blood stage and gametocytes, I feel a final round of optimization to try and introduce bioavailability is justified. If this fails we will park the series and look at alternate leads.

I would support targets A-F as priority targets but would also make a secondary and tertiary amide alongside the CONH2 analogue - primary amides often have low bioavailability. For related reasons it may be worth trying the secondary and tertiary amide analogues of the original hit PMY 10-6. We are currently measuring the plasma stability of this ester. I would be pragmatic over the choice of targets, if the sulphonamide E is challenging to make I would not prioritize.

I think targets G, I and J are highly speculative so I would not list them as essential targets, if someone has the resource and desire to explore these, that’s great but I would not say they are on a list of must have targets

I feel target H is too close to the compounds that already have poor oral efficacy (indeed the OH analogue introduces an additional H donor) so I would not prepare this compound

I look forwards to reading other views


[Further information submitted by Paul Willis via email, included below by pylioja 16.13 AEST 22/06/12]

I quite like the look of targets like this where X is CH2, O or possible N since they put an aromatic group in a similar position to that occupied by current best leads- what heterocycles for Ar are synthetically achievable?

Similarly, if there is an intermediate that is easily accessible that allows us to rapidly access a series of analogues of interest, that would be of interest, for example

X = halogen use TM coupling

X = halogen,  displacement reaction to make O, N  linked compounds

Good to hear you agree! I've added a further list of commercially available analogues of the above on OWW. They'll definitely help fill some of the space while saving time. The secondary/tertiary amides of the original TCMDC-123812 hit are commercially available so that's good. Some amide (A) and amide side-chain analogues are also commercially available. This should give us a good idea of what we can tolerate and where on the side chain without having to make each one.
I haven't found any oxazole (B) compounds for sale and I've had a couple of attempts at making these but no luck yet so they are firmly on the synthesis list. C should be easy enough (I have already made the relevant acid). Ether linked analogues (D) are not very available commercially and there don't seem to be too many syntheses in the literature. I've had a bash at making them via various substitutions but nothing has worked so far. I also didn't find any 1,2,4-oxadiazole (F) compounds so it looks like any heterocycle linked compounds will have to be made.
I like the sulphonamide E but I haven't yet had a serious think about how to make it. If we're lucky it'll just be a case of dumping in some chlorosulfonic acid onto a pyrrole to get to the sulfonylchloride, then add the relevant amine side-chain. 
As for H, it was mostly put it in there because it would be a very easy compound to make and should be a bit more soluble. I've infact *probably* made it (pending characterisation) today but its just as tedious as the rest of the near-neighbours with solubility etc.
G, I and J were in the curveball region so yes, only if someone fancies it.

Further comments since the further details provided by Paul Willis:
The 3-boronic acid/halide pyrroles are on the cards. Bromination with NBS of the plain pyrrole should be do-able and would allow someone with a good range of heterocyclic coupling partners make a sub-series quite quickly. We don't have a huge range of boronic acids in our existing inventory, so this is where a CRO or a cross-coupling methodology group would be a useful partner. 
As for the halide displacements to make ether and amine linked compounds, I've tried this briefly a while back but no initial success (e.g. and and but these need revisiting.

cdsouthan's picture

There is considerable utility  if SMILES and/or IUPAC lists surfaced anywhere in the project, can be "chemicalized"  via (or the Safari plug-in).   I tried this with the desired compounds list
As you can see  (if you just paste this in and click "webpage view") the conversion works but is patchy, mainly because of the % characters in the screen renderings.  
Would it be possible to clean this SMILES list up as a pilot ?
If there are residual failiures I can take a look at them.

MatTodd's picture

What's the issue with % in SMILES? Are those non-standard? And the advantage in listing SMILES I had thought was that the page can be found, if people are searching for certain structures. Are you seeking a SMILES list for some other function that needs Chemicalize? Want us to add SMILES for the new sets at

cdsouthan's picture

Yes I know the % is for ring closures so I'll  follow up with ChemAxon. Might be just some quirk because in this case each individual string converts OK (but changes the SMILES) when pasted over but not direct from urls or paste > PDF upload.  
(Update 4 Jul - ChemAxon developers have already done a fix but may take a while to release)
For surfacing project lists of structures as open urls  in general you could do belt-and-braces by adding the IUPACs as well as SMILES  and then  trying chemicalize off the bat (they welcome feedback).  Note if you can also add InChI keys they should get scraped into Google.  The keys on my blog seem to get indexed in minutes.   I see no poblem with incuding a few virtuals (what you wanna make next) so long as this is marked clearly on the lists.
Once  the pages here are converting OK I could write a few tips but you can just launch structure and web searches from the chemicalize record,  its fairly intuative.  There are also a few tips in this vid.