We have examined the effects of genetic polymorphisms of CYP2C9 (an important drug metabolizing enzyme (CYP2C9) and polymorphisms in the vitamin K epoxide reductase complex 1 on the dose of warfarin required to control cardiovascular (thromboembolytic) disease and risk for life-threatening side effects in a large prospective epidemiology study of 302 Caucasian Americans and 273 African-American patients treated with warfarin in over a 2 year period, in a collaboration with Limdi et al (University of Alabama)(1-5). We developed new rapid pyrosequencing genotyping tests for known or suspected defective alleles of human CYP2C9 alleles (*2, *3, *5, *6, 10, and 11 variants) which occur in Caucasians and African Americans. Individuals with aberrant CYP2C9 alleles required lower doses of warfarin. CYP2C9 and VKORC1 accounted for 30% and 18% of the variability of required warfarin dose in American Caucasians and 10% in African-Americans (1). Among VKORC haplotypes, VKORC1 was the most important. The risk of overcoagulation was higher among Caucasian Americans with variant VKORC1 allele and marginally higher among those with variant CYP2C9 genotypes. We then examined the effects of polymorphisms of CYP2C9 on serious and life threatening side effects (serious hemorrhage and death). The presence of variant alleles of CYP2C9 conferred a 5-fold increase in risk for major hemorrhage before stabilization of dose, and a 2-fold risk even after stabilization of the dose (3). The variant VKORC1 genotype did not confer a significant increase in risk for major hemorrhages. Chronic kidney disease (CKD) was found to significantly lower required warfarin dosages and these individuals were at high risk for overcoagulation (5). We also identified a new coding allele of CYP2C9 (Asp360Glu) in an African-American hospitalized for life-threatening hemorrhage (6). CYP2C9 metabolizes losartan which is used for anti-proteinuric and blood pressure control in chronic kidney disease (CDK). In another collaborative clinical study (7) we found patients with CDK who had defective CYP2C9 genotypes had less favorable lowering of urinary protein and less favorable effects on diastolic blood pressure. In studies of the regulation of the CYP2C enzymes, we studied the ability of retinoid-related orphan receptors (ROR) on transcriptional up regulation of the human CYP2Cs in liver (8). Three ROR genes, alpha, beta, and gamma, are differentially expressed in liver, kidney, lung, muscle, brown fat, thymus, and brain where they are involved in the regulation of many physiological processes such as immune function, brain development, circadian rhythm, and lipid metabolism. ROR alpha1, 2, and gamma1 are expressed in liver. We examined whether RORs activate the upstream promoter regions of the human CYP2C drug metabolizing enzymes in HepG2 cell (liver cell line) and a human colon carcinoma cell line (Caco 2 cells) and whether they up regulate the mRNAs for the CYP2Cs. Overexpression of ROR and ROR in HepG2 cells increased luciferase activity of a 3kb CYP2C8 promoter construct, but not that of CYP2C9 or CYP2C19 promoter constructs. CYP2C8 is important in the metabolism of clinical drugs such as paclitaxel (breast cancer), certain antidiabetic drugs for type II diabetes, and antimalarial drugs as well as metabolizing arachidonic acid to active metabolites which are important in cardiovascular regulation. We identified the regulatory sites for ROR to one essential site in the CYP2C8 promoter using mutational assays, showed binding of RORs in gel-shift assays. Overexpression of RORs using adenoviral constructs increased CYP2C8 mRNA in human primary hepatocytes as well as HepG2 cells. siRNA to ROR alpha and gamma decreased CYP2C8 mRNA expression in HepG2 cells. These data show a role for RORs in upregulation of CYP2C8 expression in liver and possibly extrahepatic tissues in response to stress, hypoxia and diurnal rhythm. In other studies exposure of patients or human liver cells to clinical drugs transcriptionally has been shown to upregulate CYP2C8, 2C9, and 2C19 enzymes increasing metabolism of drugs metabolized by these enzymes in vivo in man. This could produce tolerance to drugs or drug-drug interactions complicating patient therapy. In liver and intestine the CYP2Cs can be increased >2-8 fold by prior administration of drugs, producing higher metabolism in exposed individuals leading to tolerance. We have previously shown that the promoter regions of the human CYP2C genes are regulated by elements which bind the nuclear receptors CAR (constitutive androstane receptor), PXR (pregnane X receptor), and liver-enriched receptors such as HNF4 alpha. New data shows that HNF4 sites in the promoter enhance inducibility by CAR or PXR, and HNF4 and CAR act synergistically to increase CYP2C9 in liver cell lines. We theorize that various coactivators may be involved in forming a bridge between the distal CAR site and the proximal HNF4 site in the CYP2C9 promoter. Results of yeast two hybrid screens identified NCOA6 as a new HNF4 interacting protein as well known coactivators such as PGC-1(10). We overexpressed CAR and HNF4 in adenoviral vectors and performed pull downs with GST-HNF4 and GST-CAR. CAR was identified as part of a complex with GST-HNF4 complex in nuclear extracts of HepG2 cells by Western blotting and by mass spectrometric analysis. Nuclear cofactors such as NCOA6 and PGC-1 were also identified in the complex. When NCOA6 and PGC-1 were translated in vitro, they could also interact with HNF4-GST and were identified by Western blotting showing direct interaction. Using promoter assays and mRNA measurement, the synergy between HNF4 and CAR in the presence of their respective drug ligands could be blocked by adenoviral constructs to siRNAs for NCOA6, the new HNF4 interacting protein. In chromatin immunoprecipitation (ChIP) assays antibodies to CAR pulled down both the CAR binding site and the HNF4 binding site of the CYP2C9 promoter showing these factors bind to the promoter in vivo. NCOA6 was pulled down at the HNF4 site and the CAR site. Adenoviral constructs containing siRNA to NCOA6 did not affect HNF4 binding to its site but blocked NCOA6 binding and binding of other coregulators such as PGC-1, and PMT (an O-methyltransferase which repairs ASN and ASP residues). Therefore NCOA6 appears to be an essential cofactor that helps form bridge between CAR and HNF4 (the bridge may contain other coregulators as mentioned). We recently reviewed all studies available on regulation of the CYP2C enzymes (11). Recently we have shown in cultured primary human hepatocytes have shown that binding sites for liver-specific HNF4 sites are necessary for the upregulation of the CYP2C8 and CYP2C9 promoter of the CYP2Cs genes. Mutation of these sites prevents induction by the PXR agonist rifampicin. Mutation of either the PXR sites or HNF4 sites abolishes induction. siRNA to HNF4 almost abolishes induction of CYP2C8, 2C9 or 2C19 mRNA in primary human hepatocytes which are the best in vitro model we have for human liver.
Showing the most recent 10 out of 29 publications
