Metabolic study of androsta-1,4,6-triene-3,17-dione in horses using liquid chromatography/high resolution mass spectrometry.

Androsta-1,4,6-triene-3,17-dione (ATD) is an irreversible steroidal aromatase inhibitor and is marketed as a supplement. It has been reported to effectively reduce estrogen biosynthesis and significantly increase the levels of endogenous steroids such as dihydrotestosterone and testosterone in human. ATD abuses have been reported in human sports. Its metabolism in human has been studied, and the in vitro metabolic study of ATD in horses has been reported, however, little is known about its biotransformation and elimination in horses. This paper describes the in vitro and in vivo metabolism studies of ATD in horses, with an objective of identifying the target metabolites with the longest detection time for controlling ATD abuse. In vitro metabolism studies of ATD were performed using homogenized horse liver. ATD was found to be extensively metabolized, and its metabolites could not be easily characterized by gas chromatography/mass spectrometry (GC/MS) due to insufficient sensitivity. Liquid chromatography/high resolution mass spectrometry (LC/HRMS) was therefore employed for the identification of in vitro metabolites. The major biotransformations observed were combinations of reduction of the olefin groups and/or the keto group at either C3 or C17 position. In addition, mono-hydroxylation in the D-ring was observed along with reduction of the olefin groups and/or the keto group at C17 position. Fourteen in vitro metabolites, including two epimers of androsta-1,4,6-trien-17-ol-3-one (M1a, M1b), androsta-4,6-diene-3,17-dione (M2), boldione (M3), androsta-4,6-diene-17β-ol-3-one (M4), androsta-4,6-diene-3-ol-17-one (M5), boldenone and epi-boldenone (M6a, M6b), four stereoisomers of hydroxylated androsta-1,4,6-trien-17-ol-3-one (M7a to M7d), and two epimers of androsta-1,4-diene-16α,17-diol (M8a, M8b), were identified. The identities of all metabolites, except M1a, M5, M7a to M7d, were confirmed by matching with authentic reference standards using LC/HRMS. For the in vivo metabolism studies, two thoroughbred geldings were each administered with 800 mg of ATD by stomach tubing. ATD, and twelve out of the fourteen in vitro metabolites, including M1a, M1b, M2, M4, M5, M6, M7a to M7d, M8a and M8b, were detected in post-administration urine. Two additional urinary metabolites, namely stereoisomers of hydroxylated androsta-4,6-dien-17-ol-3-one (M9a, M9b), were tentatively identified by mass spectral interpretation. Elevated level of testosterone was also observed. In post-administration blood samples, only the parent drug, M1b and M2 were identified. This study showed that the detection of ATD administration would be best achieved by either monitoring the metabolites M1b (androsta-1,4,6-trien-17β-ol-3-one) or M4 (both excreted as sulfate conjugates) in urine, which could be detected for up to a maximum of 77 h post-administration. The analyte of choice for plasma is M1b, which could be detected for up to 28 h post administration.