The expression of human microsomal epoxide hydrolase is predominantly driven by a genetically polymorphic far upstream promoter

Abstract

Microsomal epoxide hydrolase (EPHX1) biotransforms epoxide derivatives of pharmaceuticals, including metabolites of certain antiepileptic medications, such as phenytoin and carbamazepine, and many environmental epoxides, such as those derived from butadiene, benzene, and carcinogenic polyaromatic hydrocarbons. We previously identified a far upstream promoter region, designated E1-b, in the EPHX1 gene that directs expression of an alternatively spliced EPHX1 mRNA transcript in human tissues. In this investigation, we characterized the structural features and expression character of the E1-b promoter region. Results of quantitative real-time polymerase chain reaction analyses demonstrated that the E1-b variant transcript is preferentially and broadly expressed in most tissues, such that it accounts for the majority of total EPHX1 transcript in vivo. Comparative genomic sequence comparisons indicated that the human EPHX1 E1-b gene regulatory region is primate-specific. Direct sequencing and genotyping approaches in 450 individuals demonstrated that the E1-b promoter region harbors a series of transposable element cassettes, including a polymorphic double Alu insertion. Results of reporter assays conducted in several human cell lines demonstrated that the inclusion of the Alu(+/+) insertion significantly decreases basal transcriptional activities. Furthermore, using haplotype block analyses, we determined that the E1-b polymorphic promoter region was not in linkage disequilibrium with two previously identified nonsynonomous single nucleotide polymorphisms (SNPs) in the coding region or with functional SNPs previously identified in the proximal promoter region of the gene. These results demonstrate that the upstream E1-b promoter is the major regulator of EPHX1 expression in human tissues and that polymorphism in this region may contribute an interindividual risk determinant to xenobiotic-induced toxicities.

Fig. 1.

Quantification of the EPHX1 E1 and E1-b transcripts expression in different tissues. Real-time PCR was performed with RNA obtained from 20 human tissues; each sample represents a pool of at least three individuals. E1-b (black bar) and E1 (white bar) transcript copy numbers were quantified by absolute quantification based on standard curves determined using plasmid DNA templates.

Fig. 2.

Identification of Alu-insertion polymorphisms in the 5′-flanking region of E1-b. A, PCR analysis of polymorphic DNA fragments ∼2.8 kb upstream of E1-b. The genomic DNA (20 ng) of three individuals (nos. 10, 77, and 222 of DNA Polymorphism Discovery Resource, Coriell Institute) were subsequently amplified using two primer sets as described under Materials and Methods. B, schematic representation of the polymorphic Alu (Ya5) insertions upstream of E1-b. Also indicated are the TE cassettes formed through combinations of different types of TEs that belong to SINE (AluSp, AluSx, AluJo), long interspersed element (L1MC4, L1MB7), long terminal repeat (MER66B), tandem repeat ((Ca)n), or DNA transposon (MADE1, MER5C) classifications.

Fig. 3.

Analysis of promoter activities of the polymorphic E1-b 5′-flanking region. Luciferase-based promoter activities were analyzed by cloning the E1-b 5′-flanking DNA region (-2763/+6) for different haplotypes into the pGL3-basic vector. The data shown depict means and S.D. values derived from five separate experiments, each performed in duplicate. Inclusion of the 2xAlu element results in a statistically significant decreased promoter activity in A549, HepG2, and 293A cells (*, p < 0.05; **, p < 0.01).

Fig. 4.

Characterization of the E1-b promoter region and EPHX1 expression in human cell lines. A, PCR genotyping of Alu insertion polymorphism in A549, HepG2, and 293T cells. The DNA fragment between -2428 and -1305 of E1-b was specifically amplified from genomic DNA demonstrating the presence of the 2xAlu insertion in 293T cells but not A549 or HepG2 cells. B, real-time PCR was performed with RNA obtained from the respective cell lines. The data shown depict means and S.D. values derived from two separate experiments, each performed in triplicate. The 293T cells exhibited a significantly lower level of EPHX1 E1-b transcript expression compared either with HepG2 or A549 cells (**, p < 0.01). C, Western immunoblot blot analysis assessing corresponding EPHX1 protein levels.

Fig. 5.

LD structure of a 52-kb region spanning EPHX1 and extended promoter in European Americans. EPHX1 genotypes from HapMap (http://www.hapmap.org/) were downloaded, and the LD was determined using Haploview software. D′ values are displayed in the squares (empty squares have a pairwise D′ 1.00). Red squares show high pairwise LD, gradually coloring down to white squares of low pairwise LD. The black triangles indicate the haplotype blocks. There are three haplotype blocks in this region (one block in the large promoter region that spans into the first two exons of the gene, one block that spans exon 3 and nearby intronic regions, and one block that spans exons 4–9). At the upper portion gene structure region, red ovals indicate the coding polymorphisms in exons 3 (rs1051740) and 4 (rs2234922), which are not in LD with each other, nor with the promoter region.