O=C(Cc1cccc(Cl)c1)N[C@@H]1CCCOC1
O=C(Cc1cccc(Cl)c1)N[C@H]1COCc2ccccc21
CN(C(=O)Cc1cccc(Cl)c1)[C@H]1CCOC1
O=C(Cc1cccc(Cl)c1)N[C@@H]1CCOC1
O=C(Cc1cccc(Cl)c1)N[C@H]1CCOC1
O=C(Cc1cccc(Cl)c1)N[C@H]1CCCOC1
CN(C(=O)Cc1cccc(Cl)c1)[C@H]1CCCOC1
O=C(Cc1cccc(Cl)c1)N[C@@H]1COCc2ccccc21
CN(C(=O)Cc1cccc(Cl)c1)[C@@H]1COCc2ccccc21
Replace pyridine substructure in 3-aminopyridine-like series with cyclic ether (this tactic has been used in design of kinase inhibitors). The natural conformational preference of these analogs places the ring plane of the cyclic ether in an orthogonal orientation with respect to the plane of the amide ring (bound conformation likely to be more stable than for 3-pyridinyl amides). Replacing aromatic nitrogen with ether oxygen is also likely to be beneficial if CYP inhibition is an issue. The designs have been submitted as 3-chlorobenzyl analogs since the different (with respect to pyridine substructure) requirements of the cyclic ether may require more flexibility.
This submission is the result of discussions with Ed Griffen on the DAV-ILL-89eb723b-1 submission. The S-enantiomers in the submission have been also been submitted as their N-methyl analogs because this is likely to stabilize their bound conformations. I recommend starting with design 1 (if using chiral starting material). only proceeding with other designs if interesting activity is observed. Benzannulation of design 1 (design 2) appears to compromise fit of the cyclic ether into the binding site and I would anticipate that it will actually exacerbate any metabolic problems associated with the isoquinoline (I've included it because it's an analog of the isoquinoline). Design 3 would also be worth looking at if interesting activity is observed for design 1. The X12207 crystal structure was used for modelling and the pdb file associated with the submission contains this protein structure, its crystallographic ligand (EDJ-MED-e4b030d8-13) and the designs (not in same order as submission).