Characterization of Atrasentan Metabolic Pathway in Human Liver Microsomes Using Feature-Based Molecular Networking

Background/Objectives: Atrasentan is a selective endothelin A receptor antagonist (SERA) developed as a potential therapy for chronic renal diseases, including diabetic nephropathy and immunoglobulin A nephropathy. Despite this potential, understanding its metabolic bioactivation is essential for assessing the risks of drug-induced liver injury (DILI). However, the metabolic profile of atrasentan remains poorly characterized, and the mechanisms underlying its potential hepatotoxicity remain underexplored. Therefore, this study aims to investigate the metabolic pathways of atrasentan in human liver microsomes (HLMs) in the presence of nicotinamide adenine dinucleotide phosphate (NADP+), uridine diphosphate glucuronic acid (UDPGA), or glutathione (GSH). Methods: A liquid chromatography–high resolution mass spectrometry (LC-HRMS) coupled with a feature-based molecular networking approach was used to characterize metabolites. Characterization of the major metabolites was achieved through cytochrome P450 (P450) phenotyping with human recombinant P450 isoforms. Results: A total of eighteen metabolites were characterized through phase I and II metabolic reactions, including demethylenation, N-dealkylation, O-demethylation, hydroxylation, dehydrogenation, and glucuronidation. Atrasentan acyl glucuronide (M8) was confirmed as the predominant metabolite, and we also putatively annotated a catechol intermediate (M5) and its corresponding GSH conjugate (M15). Characterizing the GSH conjugate (M15) indicates that catechol intermediate (M5) can be further oxidized to a reactive ortho-quinone intermediate, which is subsequently trapped by GSH, suggesting the potential for a bioactivation mechanism. Reaction phenotyping demonstrated that the formation of M5 is catalyzed almost exclusively by the CYP3A subfamily. However, its direct translation to in vivo oxidative stress or covalent protein binding requires further studies. Conclusions: These findings demonstrate that feature-based molecular networking is a valuable strategy for metabolite characterization, underscoring the urgent need for further in vivo metabolism studies to definitively assess hepatotoxic risks associated with these reactive metabolites.