Topic automatically created for discussing the designs at:
https://covid.postera.ai/covid/submissions/PET-UNK-615ad708
Hi @edgriffen @mc-robinson @alphalee
The designs in this submission are intended to address potential ADME issues for MAT-POS-e194df51-1.
Design 1 replaces the P1-isoquinoline isosterically with 6-azabenzothiophene which is likely to result in a marginal increase in potency against the enzyme (compare JIN-POS-6dc588a4-22 with ADA-UCB-6c2cb422-1). This may (or may not) lead to a decrease in clearance but the isosteric nature of isoquinoline replacement does increase confidence that the noncovalent stuff (e.g. antiviral activity) will be unchanged.
Design 2 replaces the cyano group of MAT-POS-e194df51-1 with chloro. The substructural alert for tertiary alkyl halide is not appropriate for cyclopropyl (carbocation instability) and example of its use is given in this article. I see chloro as more similar to cyano than any of the other substituents that have been deployed on the pendant cyclopropyl (in the context of the binding mode) and it has the advantage of not positioning potential metabolic liabilities like methoxy and methyl at the periphery of the molecular structure. The binding mode for MAT-POS-e194df51-1 indicates a potential desolvation penalty for the nitrogen of the cyano substituent.
Design 3 caps the amide nitrogen of MAT-POS-e194df51-1 with cyano (I would not expect this structural transformation to invert the trans/cis geometric preference of the amide as is the case for methyl substitution). Covalent bond formation between the cyano carbon and the catalytic cysteine sulfur could conceivably occur (and lead to improved potency) the primary rationale for this design is that capping the amide NH is potentially advantageous (the rationale for the spirocycles). Specifically, the amidic hydrogen bond donor is eliminated and concerns about generation of the P1-amine in vivo are addressed. Capping the amide NH with cyano could also be advantageous in situations where spirocyclization results in reduced potency and the N-cyano analogs are likely to be synthetically more accessible (e.g. with cyanogen bromide) than the spirocycles. There are four substructural alerts linked to the N-cyano group (would exclude cimetidine) and one for ‘too many cyano groups’ (which would exclude anastrozole).