Investigation of Relative Contribution of Intestinal and Hepatic UGT2B17 on Testosterone First‐Pass Metabolism

Background and Objective Oral testosterone show poor and unpredictable bioavailability due to high and variable first‐pass metabolism. We have demonstrated that testosterone glucuronidation is mediated by UGT2B17 (major) and UGT2B15 (minor) in human liver1. In this study, we compared 1) UGT2B17 protein abundance in microsomes and primary cells isolated from human intestine vs. liver, and 2) testosterone glucuronidation activity in primary human enterocytes vs. hepatocytes. Methods Human intestinal microsomes (HIMs) were isolated from different sections of intestine, i.e., duodenum (D), jejunum (J1–J3), ileum (I), and colon (C), from individual donors (n=7). Total membrane fractions were isolated from hepatocytes (n=5, Lonza) and enterocytes (n=16, IVAL). HIM, membrane fractions, and pooled liver and intestinal S9 fractions (Xenotech) were digested using optimized methods. Surrogate peptides of UGT2B17 and/or UGT2B15 were quantified by a validated LC‐MS/MS method. Testosterone glucuronidation activity was assessed in suspended hepatocytes and enterocytes as well as in S9 fractions at 5–50 μM concentrations at 37 °C. Reactions were quenched after 30–120 min, supernatants were analyzed via LC‐MS/MS, and data were normalized to total protein. Data comparisons were done using non‐parametric tests. Results UGT2B17 abundance was highly variable in both HIMs (>9‐fold) and human liver microsomes (HLMs) (>170‐fold), and was undetectable in 17% of HIMs and 45% of HLMs. When detected, UGT2B17 protein abundance (pmol/mg protein, mean ±SD) was 4.6‐fold higher (p<0.001) in entire small intestine (8.9 ± 3.0) vs. liver (1.9 ± 2.6)1. Noticeably, we observed an increasing trend along the gastrointestinal (GI) segments (D, 7.2 ± 8.6; J1, 5.9 ± 4.7; J2, 7.2 ± 4.3; J3, 11.1 ± 7.2; I, 12.9 ± 10.5; and C, 19.4 ± 9.1). Mean UGT2B17 protein abundance in UGT2B17‐expressing enterocytes (0.47 ± 0.52) was higher compared to hepatocytes (0.22 ± 0.27). UGT2B15 protein abundance in hepatocytes was 3.0 ± 1.7 pmol/mg protein but undetectable in enterocytes. Similar trend was observed in testosterone glucuronidation activity (pmol/min/mg protein) in enterocytes vs. hepatocytes (32.3 ± 25.0 vs. 18.6 ± 10.3). Higher activity and/or expression of UGT2B17 were consistent in the pooled intestine vs. liver (S9 and microsomes). Discussion Unavailability of oral testosterone formulation is an unmet need in the growing US market for testosterone replacement therapy. Overall higher UGT2B17 abundance in the GI compared to liver indicate a greater magnitude of testosterone first‐pass metabolism in the GI vs. liver, as reflected by the higher activity in enterocytes. Fractional contribution of UGT2B17 in testosterone first‐pass metabolism is predicted to be much greater in the GI vs. liver due to undetectable UGT2B15 in the GI, which postulates that variable testosterone bioavailability2 is likely primarily due to the intestinal UGT2B17. Consideration of differential UGT2B17 expression along GI tract is important in designing testosterone oral formulation for targeted absorption. Further studies are warranted to investigate physiological relevance of high UGT2B17 abundance in distal GI segments. Nevertheless, these observational data are crucial in constructing a PBPK model of oral testosterone and other UGT2B17 substrates. Support or Funding Information R01 HD081299 This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .