Hepatic transcriptional up-regulator of the rat microsomal epoxide hydrolase gene

Tissue-specific expression and alternative splicing of human microsomal epoxide hydrolase

Human microsomal epoxide hydrolase (HYL1) plays an important role in the detoxification of environmental compounds and drugs, such as the aromatic anticonvulsants phenytoin, carbamazepine, and phenobarbital, by converting their P450-generated epoxide metabolites into nontoxic diols. Recently, we have shown that a genetic defect altering the structure and function of the HYL1 protein is unlikely to be responsible for predisposing individuals to idiosyncratic hypersensitivity reactions from anticonvulsants. To evaluate the possible involvement of regulatory mechanisms, we used 5' rapid amplification of cDNA ends (RACE) and reverse transcription polymerase chain reaction (RT-PCR) to identify and characterize HYL1 5' cDNA ends. In addition to exon 1 (E1) previously isolated from a liver cDNA library, we isolated four new exons (E1-a, E1-c, E1-d, and E1-e) from various tissues. E1 was always directly connected to exon 2 (E2) where the translation start codon is located. E1-a, E1-c E1-d, and E1-e are alternatively spliced to E2, having either E1-a or E1-a' (a truncated form of E1-a) at the 5' end of their respective transcript. Genomic data indicate that exons E1-a and E1-c are located at least 7 kb upstream from E1. Furthermore, we demonstrated a tissue-specific expression pattern for E1-containing mRNA species, whereas E1-a-containing transcripts appear to be expressed ubiquitously. Our results provide evidence that microsomal epoxide hydrolase is regulated by multiple untranslated exons flanked by tissue-specific promoters.

HNF1 alpha activates the rat UDP glucuronosyltransferase UGT2B1 gene promoter

The rat UDP glucuronosyltransferase UGT2B1 is expressed mainly in the liver where it glucuronidates steroids and environmental toxins and carcinogens. A region between -42 and -55 bp upstream from the UGT2B1 gene transcription start site was previously identified as sharing sequence similarity with the hepatocyte nuclear factor 1 (HNF1) consensus binding site. In this study, the importance of this region in the regulation of the UGT2B1 gene was confirmed by functional and DNA binding assays. A minimal UGT2B1 gene promoter containing the putative HNF1 binding site was fused to the CAT reporter gene and transfected into HepG2 cells. Only low levels of CAT activity were detected. This activity was increased 50-fold when an HNF1 alpha expression vector was co-transfected with the UGT2B1 promoter CAT construct but was not altered when a HNF1 beta expression vector was used. A UGT2B1 promoter construct with the HNF1-like region deleted was not activated by either co-transfected HNF1 expression vector. DNase 1 footprinting and gel-shift analysis demonstrated that nuclear proteins present in both HepG2 cells and rat liver bind to the HNF1-like element. The presence of HNF1 alpha in these nuclear proteins that bind to the HNF1-like element was confirmed by supershift analysis with antisera to HNF1 alpha. Specific binding of nuclear proteins to the HNF1-like element was not seen in extracts from three cell lines derived from nonhepatic tissues. These data strongly suggest that the liver-enriched factor HNF1 alpha binds to, and activates, the UGT2B1 gene promoter

Immunohistochemical assessment of human microsomal epoxide hydrolase in primary and secondary liver neoplasm: a quantitative approach

1. Epoxide hydrolases form an enzyme family involved in the metabolism of a variety of xenobiotics including cytostatic drugs and carcinogens. Whether human microsomal epoxide hydrolase--one of the main members of the epoxide hydrolase family--is expressed in neoplasia of the liver has been the subject of a controversial discussion. 2. We therefore developed a quantitative immunohistochemical assay and monitored epoxide hydrolase expression in hepatocellular carcinomas (HCC, n = 20), cholangio-cellular carcinomas (CCC, n = 2) and liver metastases (n = 57) of tumours of various origins, and compared the expression intensities and patterns to normal liver tissue. 3. In normal liver tissue microsomal epoxide hydrolase displays expression of the constitutive type with non-zonal staining of all hepatocytes. 4. When using a quantitative immunohistochemical approach statistically significant differences in microsomal epoxide hydrolase expression were observed between normal tissue, hepatocellular carcinoma and liver metastases (mean optical density 2.35, 1.63 and 0.21 respectively, p = 2.9, 6.3 and 18.9). These data indicate differential expression in different types of liver neoplasm. 5. As microsomal epoxide hydrolase is involved in metabolism of different xenobiotics our findings may have implications for tumour progression.