Background: Quantitative data of cytochrome P450 (CYP) and UDP-glucuronosyltransferase (UGT) enzymes are important in identifying metabolic pathways and predicting drug clearance and drug-drug interactions using physiologically-based pharmacokinetic models. Various proteomic methodologies are used to quantify these enzymes, with reports pointing out a large level of variation in expression, attributed to different sources. This study aimed to evaluate the quality of CYP and UGT quantitative proteomic data used in pharmacokinetic research against specific catalytic activity. Methods: Four CYPs (CYP1A2, 2B6, 2D6, 3A4) and seven UGTs (UGT1A1, 1A3, 1A4, 1A6, 1A9, 2B7, 2B15) were quantified using distinct proteomic methods and their catalytic activities were assessed in matching human liver samples (n=24). For UGT enzymes, the abundances were generated by two independent laboratories (stable isotope-labeled (SIL) standards or QconCAT). For CYP enzymes, two (QconCAT or label-free) methodologies were employed in the same laboratory. Catalytic activities to specific CYP and UGT substrates were used to validate abundance measurements. Several criteria were considered to improve abundance measurements: peptides/fragments selected for monitoring, total protein measurement. Results: Unlike CYPs, there was little agreement between UGT abundance levels generated by different proteomic methods, except UGT1A1 (Rs=0.73, p<0.001; R2=0.30). Significant correlations between abundance and activity were demonstrated by SIL-based data, particularly for UGT1A1, 1A3, 1A4, and 2B7 (Rs=0.79-0.90, p<0.001; R2=0.69-0.79; n=59). For QconCAT, although correlations of abundance with activity for CYP enzymes were strong (Rs=0.65-0.82, p<0.001; R2=0.49-0.79; n=24), they were poor for UGTs, with moderate levels identified for UGT1A1, 1A3 and 2B7. Applying several optimization steps (selection of standard peptides, monitored fragments, optimization of total protein measurement) significantly improved correlations in QconCAT data for six UGT enzymes (Rs=0.55-0.87, p<0.01; R2=0.48-0.73). Conclusions: Enzyme abundance measurements should be validated against catalytic activity if available; several criteria can be considered to improve proteomic quantification.
19 Sep 2017
|Title||16th Human Proteome Organization World Congress|
|Period||17/09/17 → 20/09/17|
|Web address (URL)|
|Location||Dublin Convention Centre (CCD)|
|Degree of recognition||International event|