N-glycosylation and residue 96 are involved in the functional properties of UDP-glucuronosyltransferase enzymes

The recent cloning of several human and monkey UDP-glucuronosyltransferase (UGT) 2B proteins has allowed the characterization of these steroid metabolic enzymes. However, relatively little is known about the structure-function relationship, and the potential post-translational modifications of these proteins. The mammalian UGT2B proteins contain at least one consensus asparagine-linked glycosylation site NX(S/T). Endoglycosidase H digestion of the human and monkey UGT2B proteins demonstrates that only UGT2B7, UGT2B15, UGT2B17, and UGT2B20 are glycosylated. Although UGT2B 15 and UGT2B20 contain three and four potential glycosylation sites, respectively, site-directed mutagenesis revealed that both proteins are glycosylated at the same first site. In both proteins, abolishing glycosylation decreased glucuronidation activity; however, the K-m values and the substrate specificities were not affected. Despite the similarities between UGT2B 15 and UGT2B20, UGT2B20 is largely more labile than UGT2B15. Treating HK293 cells stably expressing UGT2B20 with cycloheximide for 2 h decreased the enzyme activity by more than 50%, whereas the activity of UGT2B 15 remained unchanged after 24 h. The UGT2B20 protein is unique in having an isoleucine at position 96 instead of an arginine as found in all the other UGT2B enzymes. Changing the isoleucine in UGT2B20 to an arginine stabilized enzyme activity, while the reciprocal mutation in UGT2B15 R96I produced a more labile enzyme. Secondary structure predictions of UGT2B proteins revealed a putative a-helix in this region in all the human and monkey proteins. This alpha-helix is shortest in UGT2B20; however, the helix is lengthened in UGT2B20 I96R. Thus, it is apparent that the length of the putative alpha-helix between residues 84 and 100 is a determining factor in the stability of UGT2B enzyme activity. This study reveals the extent and importance of protein glycosylation on UGT2B enzyme activity and that the effect of residue 96 on UGT2B enzyme stability is correlated to the length of a putative alpha-helix.