Evaluation of plasma enzyme activities using gas chromatography–mass spectrometry based steroid signatures

The simultaneous quantification of 65 plasma steroids, including 22 androgens, 15 estrogens, 15 corticoids and 13 progestins, was developed using gas chromatography-mass spectrometry (GC–MS). The extraction efficiency of the catechol estrogens was improved by the addition of l-ascorbic acid in several steps. All steroids, as their trimethylsilyl derivatives, were well separated with good peak shapes within a 50 min run. The devised method provided good linearity (correlation coefficient, r2 > 0.993), while the limit of quantification ranged from 0.2 to 2.0 ng mL−1. The precision (% CV) and accuracy (% bias) were 2.0–12.4% and 93.5–109.2%, respectively. The metabolic changes were evaluated by applying this method to plasma samples obtained from 26 healthy male subjects grouped according to the pre- and post-administration of dutasteride, which inhibits 5α-reductase isoenzyme types 1 and 2. The levels of three plasma steroids, such as dihydrotestosterone, 5α-androstanedione and allotetrahydrocortisol, were decreased significantly after drug administration, while the levels of testosterone and 5β-androstane-3β,17α-diol were increased. In addition, the ratios of the steroid precursors and their metabolites, which represent the activities of the related enzymes, were z-score transformed for visualization in heat maps generated using supervised hierarchical clustering analysis. These results validated the data transformation because 5α-reductase is an indicator for the biological actions of dutasteride. GC–MS base quantitative visualization might be found in the integration with the mining biomarkers in drug evaluations and hormone-dependent diseases.