RESUMEN
We report on a new preclinical drug optimization strategy that measures drug candidates' binding affinity with human serum albumin (HSA) as an assessment of increasing or decreasing serum T1/2. Three common scaffolds were used as drug prototypes. Common polar and nonpolar substituents attached to the scaffolds have been identified as opportunities for increasing or decreasing the HSA binding affinity. This approach of adjusting HSA binding could be proactively established for preclinical drug candidates by appending functionality to sites on the drug scaffold not involved in biological target interactions. This strategy complements other medicinal chemistry efforts to identify longer or shorter human dosing regimens.
RESUMEN
Tuberculosis (TB) is one of the world's most deadly infectious diseases resulting in nearly 1.3 million deaths annually and infecting nearly one-quarter of the population. para-Aminosalicylic acid (PAS), an important second-line agent for treating drug-resistant Mycobacterium tuberculosis, has moderate bioavailability and rapid clearance that necessitate high daily doses of up to 12 g per day, which in turn causes severe gastrointestinal disturbances presumably by disruption of gut microbiota and host epithelial cells. We first synthesized a series of alkyl, acyloxy and alkyloxycarbonyloxyalkyl ester prodrugs to increase the oral bioavailability and thereby prevent intestinal accumulation as well as undesirable bioactivation by the gut microbiome to non-natural folate species that exhibit cytotoxicity. The pivoxyl prodrug of PAS was superior to all of the prodrugs examined and showed nearly quantitative absorption. While the conceptually simple prodrug approach improved the oral bioavailability of PAS, it did not address the intrinsic rapid clearance of PAS mediated by N-acetyltransferase-1 (NAT-1). Thus, we next modified the PAS scaffold to reduce NAT-1 catalyzed inactivation by introduction of groups to sterically block N-acetylation and fluorination of the aryl ring of PAS to attenuate N-acetylation by electronically deactivating the para-amino group. Among the mono-fluorinated analogs prepared, 5-fluoro-PAS, exhibited the best activity and an 11-fold decreased rate of inactivation by NAT-1 that translated to a 5-fold improved exposure as measured by area-under-the-curve (AUC) following oral dosing to CD-1 mice. The pivoxyl prodrug and fluorination at the 5-position of PAS address the primary limitations of PAS and have the potential to revitalize this second-line TB drug.