Multisite kinetic modelling of interactions between UDP-glucuronosyltransferase 2B7 (UGT2B7) substrates

There is increasing evidence that drug glucuronidation reactions in vitro may exhibit �atypical' or non-Michaelis-Menten kinetics. For example, 4-methylumbelliferone (4MU) and 1-naphthol (1NP) glucuronidation by UGT2B7 exhibit sigmoidal kinetics, characteristic of homotropic cooperativity. In contrast, zidovudine (AZT) glucuronidation follows Michaelis-Menten kinetics. This study investigated mutual interactions of the three commonly used UGT2B7 substrates, namely AZT, 4MU and 1NP, due to their distinctive kinetic properties. In addition to the empirical models (i.e., Michaelis-Menten and Hill equations), multisite kinetic models were employed to determine the Ki values and provide mechanistic insights into the interactions. AZT inhibited 4MU and 1NP glucuronidation by UGT2B7, with Ks (binding affinity constant) values of both substrates increasing with increasing AZT concentration. However, Vmax and the sigmoidal properties of the substrate showed no significant change, suggesting that AZT may inhibit at a distinct effector site. The Ki values for AZT inhibition of 4MU and 1NP glucuronidation were 176 and 379 �M, respectively. Both 4MU and 1NP inhibited AZT glucuronidation; however, AZT initial Michaelis-Menten kinetic properties change into sigmoidal at high concentrations of the modifiers ([4MU]>300 �M and [1NP]>75 �M). 1NP was a 2.5-fold more potent inhibitor than 4MU; the Ki values for 1NP and 4MU inhibition of AZT glucuronidation were 145 and 369 �M, respectively. 1NP inhibited 4MU glucuronidation, causing a 5-fold decrease in Vmax but no significant changes in the binding affinity and the extent of sigmoidicity, indicating that 1NP may inhibit via a separate effector site (Ki 80 �M). In contrast to all the interactions investigated, 4MU activated 1NP glucuronidation by UGT2B7. The heteroactivation was most pronounced at low substrate concentrations (maximal activation 400% of control activity), whereas no effect was observed at [1NP]=1000 �M. In conclusion, the current analysis provides further kinetic evidence of distinct substrate and effector binding domains within the UGT2B7 active site.