h2no2s h2no2s h2no2s
Replacing a phenyl ring with a thiophene system has been, and still is, one of the most classic examples of bioisosteric replacement. There are many reports of lead optimization using this technique. A series of 3-substituted-3,4-dihydroisoquinolinamines was replaced by 4,5-tetrahydrothieno[2,3-c]pyridines or 6,7-dihydrothieno[3,2-c]pyridines as nitric oxide synthase inhibitors (63). Not surprisingly, these analogs showed rather different activity and selectivity profiles. Similar heterocyclic replacements were used for the antagonists of the adenosine A1 receptor (64). The isoquinoline core was replaced with a thiadiazole or a thiazoie. The resulting compounds showed quite different activities and selectivities.
Bioisosteric replacement of 4-quinolones with 2-pyridones for the bacterial topoisomerase inhibitors resulted in several highly potent agents (65). The comparison of biological activities of these two series of compounds was recently reviewed (66). Although no marketed drugs emerged from this highly promising series, the concept of introducing a bridgehead nitrogen to replace a carbon atom in a heterocyclic ring is a well validated bioisosteric replacement. The recently discovered vardenafil 81, awaiting FDA approval, is a creative and successful way to apply this strategy (67,68). While vardenafil and sildenafil are structurally similar, the two compounds show rather different PDE subtype selectivities and PK profiles. Vardenafil is slightly more potent than sildenafil and has a faster on-set of action. The IC50 ratios of PDE6/PDE5 are 7-fold and 160-fold for sildenafil and vardenafil, respectively (69). A larger activity separation of PDE6/PDE5 could potentially reduce the side effect of blue vision in some patients.
Another example is shown in the attempted replacement of the Indole ring with pyrrolo[2,3-b]pyridine 83, pyrazolo[1,5-a]pyridine 84, tetrahydropyrazolo[1,5-ajpyridine 85, and imidazo[1,2-a]pyridine 86 in the search for dopamine D4 receptor ligands (70). Replacement with these heterocycles did not yield compounds of similar potency compared with the indole lead. An electrostatic potential map was calculated to further rationalize the molecular property and activity differences against respective receptor subtypes. This helps to gain some understanding of the electrostatic differences and binding requirements for these apparently similar heterocyclic rings.
Was this article helpful?