Peroxisome proliferator-activated receptor alpha, PPARα, directly regulates transcription of cytochrome P450 CYP2C8

The cytochrome P450, CYP2C8, metabolizes more than 60 clinically used drugs as well as endogenous substances including retinoic acid and arachidonic acid. However, predictive factors for interindividual variability in the efficacy and toxicity of CYP2C8 drug substrates are essentially lacking. Recently we demonstrated that peroxisome proliferator-activated receptor alpha (PPARα), a nuclear receptor primarily involved in control of lipid and energy homeostasis directly regulates the transcription of CYP3A4. Here we investigated the potential regulation of CYP2C8 by PPARα. Two linked intronic SNPs in PPARα (rs4253728, rs4823613) previously associated with hepatic CYP3A4 status showed significant association with CYP2C8 protein level in human liver samples (N = 150). Furthermore, siRNA-mediated knock-down of PPARα in HepaRG human hepatocyte cells resulted in up to ∼60 and ∼50% downregulation of CYP2C8 mRNA and activity, while treatment with the PPARα agonist WY14,643 lead to an induction by >150 and >100%, respectively. Using chromatin immunoprecipitation scanning assay we identified a specific upstream gene region that is occupied in vivo by PPARα. Electromobility shift assay demonstrated direct binding of PPARα to a DR-1 motif located at positions –2762/–2775 bp upstream of the CYP2C8 transcription start site. We further validated the functional activity of this element using luciferase reporter gene assays in HuH7 cells. Moreover, based on our previous studies we demonstrated that WNT/β-catenin acts as a functional inhibitor of PPARα-mediated inducibility of CYP2C8 expression. In conclusion, our data suggest direct involvement of PPARα in both constitutive and inducible regulation of CYP2C8 expression in human liver, which is further modulated by WNT/β-catenin pathway. PPARA gene polymorphism could have a modest influence on CYP2C8 phenotype.

Gene copy number variation analysis reveals dosage-insensitive expression of CYP2E1

Gene copy number variants (CNVs) of CYP2E1 have been described but not functionally characterized. Here we investigated effects of CNVs on hepatic and lymphoblastoid CYP2E1 expression. Using available single-nuleotide polymorphism microarray data and quantitative PCR, CYP2E1 gene duplication and deletion carriers were identified. CYP2E1 mRNA, protein and enzyme activity (chlorzoxazone-6-hydroxylation) phenotypes of CYP2E1 were not associated with gene copy number. Analysis of gene expression in lymphoblastoid cell lines in relation to CNV confirmed this finding in an extrahepatic tissue and for other ethnicities. Further analyses identified a linked haplotype cluster with possible influence on gene expression. In summary, our data suggest a homeostatic, gene dosage-insensitive regulation of CYP2E1 expression by unknown gene dosage compensation mechanisms. This is in striking contrast to well-known structural variations of CYP2A6 and CYP2D6 that have a strong impact on expression and activity. These findings are important in the context of pharmacogenetic prediction.

Comparative Proteomic Analysis of Human Liver Tissue and Isolated Hepatocytes with a Focus on Proteins Determining Drug Exposure

Freshly isolated human hepatocytes are considered the gold standard for in vitro studies of liver functions, including drug transport, metabolism, and toxicity. For accurate predictions of the in vivo outcome, the isolated hepatocytes should reflect the phenotype of their in vivo counterpart, i.e., hepatocytes in human liver tissue. Here, we quantified and compared the membrane proteomes of freshly isolated hepatocytes and human liver tissue using a label-free shotgun proteomics approach. A total of 5144 unique proteins were identified, spanning over 6 orders of magnitude in abundance. There was a good global correlation in protein abundance. However, the expression of many plasma membrane proteins was lower in the isolated hepatocytes than in the liver tissue. This included transport proteins that determine hepatocyte exposure to many drugs and endogenous compounds. Pathway analysis of the differentially expressed proteins confirmed that hepatocytes are exposed to oxidative stress during isolation and suggested that plasma membrane proteins were degraded via the protein ubiquitination pathway. Finally, using pitavastatin as an example, we show how protein quantifications can improve in vitro predictions of in vivo liver clearance. We tentatively conclude that our data set will be a useful resource for improved hepatocyte predictions of the in vivo outcome.

Meta-analysis of expression of hepatic organic anion-transporting polypeptide (OATP) transporters in cellular systems relative to human liver tissue

Organic anion-transporting polypeptide (OATP)1B1, OATP1B3, and OATP2B1 transporters play an important role in hepatic drug disposition. Recently, an increasing number of studies have reported proteomic expression data for OATP transporters. However, systematic analysis and understanding of the actual differences in OATP expression between liver tissue and commonly used cellular systems is lacking. In the current study, meta-analysis was performed to assess the protein expression of OATP transporters reported in hepatocytes relative to liver tissue and to identify any potential correlations in transporter expression levels in the same individual. OATP1B1 was identified as the most abundant uptake transporter at 5.9 ± 8.3, 5.8 ± 3.3, and 4.2 ± 1.7 fmol/μg protein in liver tissue, sandwich-cultured human hepatocytes (SCHH), and cryopreserved suspended hepatocytes, respectively. The rank order in average expression in liver tissue and cellular systems was OATP1B1 > OATP1B3 ≈ OATP2B1. Abundance levels of the OATP transporters investigated were not significantly different between liver and cellular systems, with the exception of OATP2B1 expression in SCHH relative to liver tissue. Analysis of OATP1B1, OATP1B3, and OATP2B1 liver expression data in the same individuals (n = 86) identified weak (OATP1B1-OATP2B1) to moderately (OATP1B3-OATP2B1) significant correlations. A significant weak correlation was noted between OATP1B1 abundance and age of human donors, whereas expression of the OATPs investigated was independent of sex. Implications of the current analysis on the in vitro-in vivo extrapolation of transporter-mediated drug disposition using physiologically based pharmacokinetic models are discussed.