Near-Neighbour Round 2 SAR

12 Mar
Published by pylioja

 The first results of bio-data for the 2nd round of compounds is back from the Avery group and Mat has summarised the results in his post here on synaptic leap and on Google+. We are seeing some very active compounds but as many people have pointed out, all of the highly active compounds have high logP value (ChemDraw calculated). This is something we've definitely noticed when synthesising the compounds themselves. 
SAR summary of thiazolidinone aryl pyrroles
It's fairly clear from the effect of the aryl pyrrole substituent that as the LogP drops so does the 3D7 IC50. It also looks like we can change the N-subtituent quite readily (Ac, cyclopentyl, phenyl) but we do see a drastic drop if it is changed to an acetonitrile subsituent. The crystals structures of these compounds show the nitrile pointing directly out of the plane of the ring systems and maybe this gives us an idea of the amount of space that we can play with?
Crystal structure of ZYH 22-3Crystal structure of ZYH 10-2-A
Right now, we don't have enough analogues to build a comprehensive picture of any SAR but we're starting to get a feel for it. The next round will concentrate on making these more drug-like. It would be great to access some of the commercial and academic analogues in this series to save some synthetic effort. If anyone has any compound spare (mg) then it would be fantastic to get that tested in our assays.
Mat's summary of the data on TCMDC-123812/123794 showed that they are more sensitive to changes in the structure, but I'm still working on the ether and amine linked analogues of TCMDC-123812 to help us complete this picture. As well as these, we will look at heterocycles as alternatives to the linkers/side-chains and replacements for the aryl-pyrrole. We have no idea of the mode of action at the moment so it would be interesting to see how the activities change as we make compounds that are something in between TCMDC-123* and the near-neighbours. Are we working with separate modes of action for the 2 sub-series of aryl-pyrroles?


MatTodd's picture

Interesting when you sort the data like that. There does seem to be a clear, and rather depressing, trend of increasing potency with increasing logP, yet that's not the only thing going on. In the change from R=F to R=Me for the middle series (N-acetyl) there's little difference in logP, but a substantial difference in potency. Likewise ZYH3-1 and PMY 14-3 in the regular NN series.
But I wonder if we shouldn't put a Lipinski-respectful cap of 5 on cLogP for the moment during the planning stages? These are just from Chemdraw. Have you correlated that with other software for a couple of cases to make sure Chemdraw is on target?
(and those beautiful xstal structures need CCDC numbers for the paper)

The logP varies quite a bit depending on how it is calculated. Taking ZYH 3-1 as an example for logP/tPSA (Å2):
ChemDraw: 4.97/44.7 
Molinspiration: 5.27/50.2
gCDK ALogP: 3.09/75.2
gCDK XLogP: 3.15 
ACD/labs via ChemSpider: 5.33/71.7
Obviously gCDK appears to give values that are way lower than the others. Otherwise not too bad a correlation. tPSA varies quite a bit though.
Yes, the N-substituted compounds do not have as clear a trend in terms of activity vs LogP. Certainly an approximate cap of 5 would be good but using which calculation method?