UHPLC-Q-Exactive orbitrap MS analysis for metabolite identification of doping drug bolasterone in vitro and in rats

Abstract

Purpose. This study was aimed to identify metabolites of anabolic steroid bolasterone by UHPLC-Q-Exactive Orbitrap MS, through in vitro (rat liver microsomes) and in rats after oral administration of bolasterone. Methods. The in vitro metabolites (phases I and II) were obtained by incubation of bolasterone with rat liver microsomes, NADPH/UDPGA, and MgCl2 (for the glucuronidation) in phosphate buffer at 37°C for 1 hr. The reaction was terminated by the addition of ice-cold acetonitrile, vortexed, centrifuged and the supernatant was collected. In vivo experiments were carried out using urine samples of rats orally administered bolasterone (40 mg/kg). The samples were subjected to liquid-liquid extraction. The identification of bolasterone metabolites was accomplished by using an ultra high-performance liquid chromatography/Orbitrap mass spectrometer (UHPLC-Q-Exactive Orbitrap MS). The mobile phase of LC consisted of 0.1% formic acid in water and 0.1% formic acid in methanol and used in a gradient elution mode. A full MS followed by data dependent-MS/MS scan modes was used to obtain mass spectra for structural information of the metabolites. The ionization was carried in positive mode of ESI. Results. Detected were bolasterone, 16 hydroxylated metabolites (mono- for M2-M7; di- for M8-M13; tri-for M14-M16), 1 glucuronic acid conjugated metabolite (M17) and 1 metabolite from the reduction of the 4-ene and 3-keto moieties (M18). The M17 and M18 were confirmed with available reference and authentic standards, respectively. Metabolic modifications in the structure of the parent bolasterone resulted in different fragmentation patterns. The structures of the metabolites were established based on the interpretation of the collision-induced dissociation (CID) fragmentation of the mass spectra, where the unique characteristic ions were correlated to the possible structural fragments. The characteristic ions such as m/z 121.0648 (C8H9O) generated from ring A of the mono-hydroxylated metabolites and 121.1016 (C9H13) from ring D of the di-hydroxylated metabolites were observed, indicating that these characteristic ions can be distinguished only by using a high resolution MS. Conclusion. For the metabolite identification of bolasterone, the methods used in this study could give a better understanding of the differences in metabolism between the in-vitro and in rats, and the different fragmentation patterns (pathways) between mono-, di-, and tri-hydroxylated metabolites. Due to no availability of authentic standards of most metabolites for applying to doping tests, these results are useful to provide structural information of metabolites.