Glucuronidation of psilocin and 4-hydroxyindole by the human UDP-glucuronosyltransferases

Abstract

We have examined the glucuronidation of psilocin, a hallucinogenic indole alkaloid, by the 19 recombinant human UDP-glucuronosyltransferases (UGTs) of subfamilies 1A, 2A, and 2B. The glucuronidation of 4-hydroxyindole, a related indole that lacks the N,N-dimethylaminoethyl side chain, was studied as well. UGT1A10 exhibited the highest psilocin glucuronidation activity, whereas the activities of UGTs 1A9, 1A8, 1A7, and 1A6 were significantly lower. On the other hand, UGT1A6 was by far the most active enzyme mediating 4-hydroxyindole glucuronidation, whereas the activities of UGTs 1A7-1A10 toward 4-hydroxyindole resembled their respective psilocin glucuronidation rates. Psilocin glucuronidation by UGT1A10 followed Michaelis-Menten kinetics in which psilocin is a low-affinity high-turnover substrate (K(m) = 3.8 mM; V(max) = 2.5 nmol/min/mg). The kinetics of psilocin glucuronidation by UGT1A9 was more complex and may be best described by biphasic kinetics with both intermediate (K(m1) = 1.0 mM) and very low affinity components. The glucuronidation of 4-hydroxyindole by UGT1A6 exhibited higher affinity (K(m) = 178 microM) and strong substrate inhibition. Experiments with human liver and intestinal microsomes (HLM and HIM, respectively) revealed similar psilocin glucuronidation activity in both samples, but a much higher 4-hydroxyindole glucuronidation rate was found in HLM versus HIM. The expression levels of UGTs 1A6-1A10 in different tissues were studied by quantitative real-time-PCR, and the results, together with the activity assays findings, suggest that whereas psilocin may be subjected to extensive glucuronidation by UGT1A10 in the small intestine, UGT1A9 is likely the main contributor to its glucuronidation once it has been absorbed into the circulation.

Fig. 1.

Chemical structures of psilocin, 4-hydroxyindole, and serotonin.

Fig. 2.

Influence of DTT concentration on psilocin (A) and β-estradiol (B) glucuronidation activity by UGT1A10. The results (mean ± S.E., n = 3) are presented as the relative peak area (in percent) under the curve, for both substrate and glucuronide peaks, compared with control samples without DTT.

Fig. 3.

Glucuronidation of psilocin by 19 human recombinant UGTs. The screening assay was performed at three psilocin concentrations (100, 500, and 1000 μM). The bars represent mean (n = 3) ± S.E. The results are presented as actual glucuronidation rates (measured rates). For the active UGTs, the results are presented as normalized rates (corrected for relative expression level). The expression level of UGT1A10 was used as 1.0 for normalization. See Materials and Methods for additional details.

Fig. 4.

Glucuronidation of 4-hydroxyindole by 19 human recombinant UGTs. The screening assay was performed at two 4-hydroxyindole concentrations (100 and 500 μM). The results are presented as both measured and normalized rates (see legend to Fig. 3).

Fig. 5.

Glucuronidation of psilocin (A) and 4-hydroxyindole (B) by HLM and HIM. The assays were performed at three psilocin concentrations (100, 500, and 1000 μM) and two 4-hydroxyindole concentrations (100 and 500 μM). The bars represent an average of three samples ± S.E.

Fig. 6.

Enzyme kinetics of psilocin glucuronidation by UGT1A10 (A) and UGT1A9 (B). The points represent an average of three samples ± S.E. Glucuronidation rates are presented as actual (measured) rates in nmol/min/mg recombinant protein. The derived kinetic constants and normalized glucuronidation values are presented in Table 1. For psilocin glucuronidation by UGT1A10, the data were fitted to the Michaelis-Menten equation. For UGT1A9, data were fitted to the two-site biphasic equation. The Eadie-Hofstee transforms of the data are presented as insets.

Fig. 7.

Enzyme kinetics of 4-hydroxyindole glucuronidation by UGT1A6 (A), UGT1A10 (B), and UGT2A1 (C). The points represent an average of three samples ± S.E. Glucuronidation rates are presented as actual (measured) rates in nmol/min/mg recombinant protein. The derived kinetic constants and normalized glucuronidation values are presented in Table 1. For 4-hydroxyindole glucuronidation by UGT1A6, the data were fitted to substrate inhibition equation. For 4-hydroxyindole glucuronidation by UGT1A10, the data were fitted to the substrate inhibition equation, and in the case of UGT2A1 it was fitted to the Michaelis-Menten equation. The Eadie-Hofstee transforms of the data are presented as insets.

Fig. 8.

Inhibition of UGT1A6-catalyzed 4-hydroxyindole glucuronidation by psilocin (A) and 1-naphthol (B). The concentration of 4-hydroxyindole was 200 μM. The glucuronidation rate of the uninhibited reaction was 5.29 ± 0.19 nmol/min/mg (mean ± S.E.). Results are presented as percentage of activity relative to control incubations without inhibitor. Bars represent an average of three samples ± S.E.