Cross-laboratory evaluation of UDP-glucuronosyltransferase (UGT) abundances and correlation with catalytic activity

Background Quantification of UDP-glucuronosyltransferase (UGT) enzymes is of considerable value in phenotyping of glucuronidation pathways in drug development and predicting metabolic clearance and drug-drug interactions using IVIVE-PBPK models. Various quantitative proteomic methodologies have been used to quantify these enzymes in various systems, with reports pointing out a large level of variability in expression, which cannot be attributed to biological inter-individual variability alone. While the factors affecting this variability are still unclear, methodological procedures are expected to be one of the major sources of variability. Methods To evaluate the effect of methodological differences on UGT abundance quantification, eight UGT enzymes were quantified in matched liver samples (n=24) by two independent laboratories (using stable isotope-labeled (SIL) peptide standards or a QconCAT standard), and differences in abundance were assessed by correlation with catalytic activity in seven enzymes (n=59) generated by a third laboratory. Results There was little agreement between individual abundance levels generated by the two methods, with only UGT1A1 showing strong correlation (Rs=0.73, p<0.0001; R2=0.30; n=24). Correlation of abundance with catalytic activity revealed moderate to strong relationships in the SIL-based dataset, with UGT1A1, 1A3, 1A4, and 2B7 showing strong correlations (Rs=0.79-0.90, p<0.0001; R2=0.70-0.80; n=59), whereas the QconCAT-based dataset showed poor correlation with activity, with moderate correlations identified for UGT1A1, 1A3 and 2B7. In addition, spurious abundance-activity correlations were identified in the cases of UGT1A4 (with 2B4), 2B7 (with 2B15) and 2B15 (with 2B7), which can be explained by correlations of protein expression between these enzymes. Conclusions Two important points can be raised: protein quantification data should be validated using catalytic activity; and caution should be exercised with spurious abundance-activity correlations in metabolic phenotyping to avoid misleading conclusions.