Identification of rat urinary metabolites of rifabutin using LC-MSn and LC-HR-MS

Rifabutin, an anti-mycobacterial agent, is reported to be extensively metabolized in vivo into more than 20 biotransformation products, with similar profile both in humans and rats. Among the metabolites formed, only seven have been characterized, the remaining are unknown. Hence, the purpose of the present study was to fill this gap by using modern in silico tools combined with advanced liquid chromatography-mass spectrometry (LC-MS) techniques. Initially a comprehensive mass fragmentation pattern for rifabutin was established using Frontier (TM) 5.1 software coupled with the data collected from multiple-stage MS (MSn), high resolution MS (HR-MS) and hydrogen/deuterium exchange MS (HDE-MS) experiments. The metabolites were then predicted in silico by using different software like MetaSite (TM), Metabolite Predict (TM) and MetWorks (TM). The in silico results were verified through in vivo studies by administration of 20 mg/kg drug to rats followed by LC-MS analyses. The urine was collected post-dose at different time intervals, and subjected to sample preparation involving sequentially protein-precipitation, liquid-freeze separation, and solid-phase extraction. The drug and metabolites were separated on an HPLC column followed by LC-MS studies. The difference of accurate masses of the drug and metabolites, and differences in their mass fragmentation pattern helped to assign structures to the metabolites and define the site of change. Here also, in silico detection tools were used, which provided complementary information. Using this strategy, 23 metabolites were detected and identified in rat urine without their isolation. The new sixteen metabolites were monohydroxy (05), dihydroxy (04), N-dealkyl (01), 25-O-desacetyl-27-O-demethyl (01), 25-O-desacetyl-23-O-acetyl (01), 25-O-desacetyl-monohydroxy (01), 27-O-demethyl-monohydroxy (01) and dehydrogenated (02) rifabutin. (C) 2010 Elsevier B.V. All rights reserved.