Author(s): Markus Schirmer 1 , Albert Rosenberger 2 , Kathrin Klein 3 , Bettina Kulle 4 , Mohammad R Toliat 5 , Peter Nürnberg 5 , Ulrich M Zanger 3 , Leszek Wojnowski [[dagger]] 6
CYP3A4; drug metabolism; genetic makeup; gene variant; sex; single nucleotide polymorphism
Cytochrome P450 (CYP)3A4 plays a unique role in drug metabolism, since it metabolizes every second drug currently in use  . The substantial interindividual differences in the hepatic expression level are bound to affect pharmacokinetics of CYP3A4 substrates, which generated a strong interest in factors underlying this variability. It is well established that CYP3A4 activity may be affected by environmental factors, including drugs capable of inducing or inhibiting the enzyme. For instance, the induction by rifampin of the metabolism of the CYP3A4 substrate ciclosporine resulted in organ graft rejections  , whereas rifampin and the widely used herbal antidepressant St John's wort attenuated the efficacy of HIV protease inhibitors  . Furthermore, the hepatic expression of CYP3A4 has been shown repeatedly to be approximately twofold higher in women than in men [4-10] . However, sex explains only a portion of the CYP3A4 protein expression level, which may vary up to 100-fold between two liver samples. A meta-analysis of longitudinal clinical studies with CYP3A substrates indicated that the individual genetic makeup accounts for as much as 90% of variability in hepatic CYP3A4 activity. However, several extensive screens failed to identify the underlying gene variants in CYP3A4 or in its major regulator, pregnane X receptor. The only exception was the CYP3A4*1B allele in the proximal promoter of the gene, but its effect on protein expression was small (less than twofold) [11-17] , so that most CYP3A4 variability remained unexplained.
The linkage disequilibrium (LD) structure within the CYP3A locus in major ethnic groups was recently investigated by ourselves  and others [18,19] . These investigations revealed extensive linkage disequilibrium, which was especially pronounced in non-African populations. Based on these results, we conducted another association study between CYP3A phenotypes and genotypes in a large bank of liver samples. Single nucleotide polymorphisms (SNPs) were selected from public resources considering the LD structure. We strove to achieve a good representation of regions of low LD. As gene variants can affect gene expression in a sex-specific way [20,21] , we hypothesized that the expression of CYP3A4 may be influenced by gene variants in a sex-dependent manner.
Genotyping & phenotyping of liver samples
Samples of liver tissues were collected from 131 patients of European Caucasian origin during surgical interventions conducted at the Department of Surgery, Campus Virchow, University Medical Centre Charité, Humboldt University (Berlin, Germany), as described  . Written informed consent was obtained from all donors. Normal liver tissue surrounding primary liver tumors or liver metastases was resected and used to prepare microsomes and total RNA. Genotyping was performed by Pyrosequencing ® or by predeveloped assays (Applied Biosystems, CA, USA) on genomic DNA isolated from the corresponding blood samples. The genotyping assays are available upon request. Clinical data were assessed for exposure to CYP3A inducers. All genotyping and phenotyping experiments have been approved by the responsible...