Submission Details

Design Rationale:

1) obviously, no potent inhibitors for Mpro are known 2) searched PDB for related enzymes and PDBbind for potent ligands binding to these enzymes 3) shortlisted 4YoJ, inhibitor is with 0.4 uM one of the most potent inhibitors for related enzymes, synthesis appears to be pretty straight forward (Targeting zoonotic viruses: Structure-based inhibition of the 3C-like protease from bat coronavirus HKU4—The likely reservoir host to the human coronavirus that causes Middle East Respiratory Syndrome (MERS), Sarah E.St. JohnabcSakshiTomaracShaun R.StaufferdeAndrew D.Mesecarabc ) 4) The core (amid bond) of the ligand in 4YOJ also superimposes very well with X_0434 and X_0678 5) P1 is occupied by benzotriazol in 4YOJ, should also fit in Mpro 6) pyridine (found in X_0107, X_0434, X_0678) should also fit in this pocket 7) X_0995 has a pyrimidine, but this can not easily be attached to the core 8) other heterocycles which have an acceptor group overlapping with the nitrogen atom in X_0107 could also be explored 9) P2-P4 pocket is occupied with a thiophenyl group in 4YOJ 10) this area is not well explored by the fragments. The fragments that bind there don't have a vector that could be joined with the core of the ligand in 4YOJ 10) based on synthetic route described in the paper mentioned above, Enamine building block library was searched for in stock compounds which have an aldehyde group directly attached to an aromatic ring => 429 this 11) Used RDkit to replace the part of the ligand in 4YOJ that goes into this subpocket and docked the resulting compounds into the X_0107 binding site using FlexX, used parts of ligand in 4YOJ as template, rescoring was done with Hyde 12) This pocket is very tight and only three reasonable hits very obtained (predicted affinity low micromolar, N-methyl-pyrol, Fluro-thiophen, Fluro-furan) 13) other five-membered rings could also be explored 14) P2-P1' pocket is rather hydrophobic, X_1334, X_1412, X_1385, X_1380, X_1336, X_0774, but the vectors in these fragments are not suited to link them with the core of the ligand in 4YOJ 15) searched Enamine building blocks library for aromatic and aliphatic amines which are in stock => 549 hits 16) Used RDkit to replace the part of the ligand in 4YOJ that goes into this subpocket and docked the resulting compounds into the X_0107 binding site using FlexX, used parts of ligand in 4YOJ as template, rescoring was done with Hyde 17) as expected, many hydrophobic substituents scored well, some hits were selected based on visual exception Based on these results, I suggest to 1) synthesize and test the ligand that is contained in 4YOJ 2) replace each of the substituents in this ligand addressing the different subpockets (3 for P1 based on the fragment hits, 4 for P2-P4 and 11 for P2-P1'). Depending on the synthetic and testing capacities, either only one of the substituents can be exchanged at the time, resulting in 18 different ligands or combinations can be explored, resulting in max 240 ligands (if the substituents of the ligand in 4YOJ are also considered).

Other Notes:

The compounds are based on an inhibitor of a related protease and the fragment hits. Except of the P1, they only contain R-groups that are in stock at Enamine. (I did not check the P1 substituent, so they could be available as well). The general synthesis is described in the literature (see design rationale). Therefore, I expect that synthesis is straight forward.

Inspired By:

ALE-HEI-f28a35b5-9

AAR-POS-d2a4d1df-11