Materials and methods

PATIENTS

The study group consisted of 181 white Dutch women with a history of pre-eclampsia with (n=96) or without (n=87) the HELLP syndrome, who have been described previously.13 The institutional review board approved the experimental protocol. Pre-eclampsia was defined as the occurrence after 20 weeks' gestation of a diastolic blood pressure greater than 90 mm Hg (phase 5 Korotkoff sound) on two or more occasions at least four hours apart with proteinuria (urinary protein greater than 0.3 g/l in a 24 hour urine collection period). The HELLP syndrome was defined as the simultaneous occurrence of a platelet count less than 100 × 10⁹/l, serum aspartate aminotransferase and serum alanine aminotransferase concentrations greater than 30 IU/l, and haemolysis established by an abnormal blood smear. A group of 151 randomly selected, non-pregnant, white Dutch women without underlying diabetes or renal, hepatic, haematological, pre-existing hypertensive, or cardiovascular disease who had experienced normal normotensive pregnancies only (median number of pregnancies 2, range 1 to 4) served as controls. General population characteristics of controls and cases in the period that they were admitted for pre-eclampsia are given in table 1.

MUTATION ANALYSIS

After informed consent was given, blood sampling took place. Blood was collected by venepuncture in sterile, siliconised, ethylenediaminetetra-acetic acid (EDTA) 4 ml vacutainer tubes (Becton Dickinson, Meylan cedex, France). Immediately after collection, blood was stored at −20°C until analysis. Genomic DNA was isolated from whole blood using the Wizard^TM genomic DNA purification kit, according to the manufacturer's instructions (Promega, Madison, WI, USA).

Two separate polymorphisms in EPHX, namely 113Tyr→His in exon 3 and 139His→39Arg in exon 4, were examined. First, DNA samples were amplified by a polymerase chain reaction (PCR) using a modification of the method described by Hassettet al.12 PCR reaction mixture contained approximately 100 ng of genomic DNA, 50 mmol/l KCl, 10 mmol/l Tris HCl, pH 8 .0, 2.0 mmol/l MgCl₂, 1.5 U ofTaq polymerase, and 1 μmol/l each of the specific forward (sense) and reverse (antisense) amplification primers in a total volume of 50 μl. DNA was denatured at 94°C for five minutes. Thirty cycles of amplification began with denaturation at 93°C, primer annealing at 58°C for exon 3 and 65°C for exon 4, and extension at 72°C, each step for one minute. To detect the 113Tyr→His variant, the PCR product was digested withTth111 and then subjected to electrophoresis in a 3% agarose gel. The 113Tyr allele was identified by an undigested single DNA band of 231 base pairs (bp), whereas the 113His coding allele gave two DNA bands of 209 bp and 22 bp. To detect the 139His→Arg variant, the PCR product was digested withRsaI and then run on a 3% agarose gel. The 139His coding allele was identified by two DNA bands (295 and 62 bp), whereas the 139Arg allele resulted in three DNA bands after digestion (174, 121, and 62 bp). Pharmacia Biotech (Roosendaal, The Netherlands) synthesised all primers, and all chemicals needed for PCR were purchased from Promega (Madison, WI, USA).

STATISTICAL ANALYSIS

Clinical characteristics between groups were compared using the Mann-Whitney U test. The statistical significance of differences for individual polymorphisms between different groups and the predicted enzyme activity (high, intermediate, or low activity) was tested with chi-square (χ²) analysis with Yates's correction in 2 × 3 contingency tables. Statistical significance was taken as p<0.05. All statistical analyses were performed with the SPSS statistical software package (SPSS Inc, Chicago, USA).

Results

The median age between cases and controls did not differ significantly. There were no significant differences in gestational age, parity, maternal age, or diastolic blood pressure between the pre-eclamptic groups with or without the HELLP syndrome.

The distribution of polymorphic variants in both exon 3 and exon 4 of the EPHX gene in patients and controls is summarised in table 2.

In the control group, the 113His coding allele in exon 3 corresponding to a low enzyme activity was more common than the 113Tyr coding allele. Sixty seven (44%) of the controls were homozygous for the exon 3 polymorphism (His113 His113) and 60 (40%) were heterozygous (Tyr113 His113). We found a statistically significantly higher rate of the fast Tyr113 Tyr113 genotype in exon 3 among women with a history of pre-eclampsia (29%) compared with controls (16%, odds ratio (OR) 2.0, 95% confidence interval (CI) 1.2-3.7) (χ²=9.99, p=0.0068) comparing the Tyr113 Tyr113 genotype versus the Tyr113 His113 and His113 His113 genotype. The subgroup of women with a history of pre-eclampsia with HELLP, as compared to the pre-eclamptic group without HELLP, showed no significant differences in the occurrence of polymorphic variants in exon 3 (OR 1.4, 95% CI 0.7-2.8) (χ²=0.83, p=0.66). There was no significant difference in exon 3 polymorphic rates between pre-eclamptic women with (n=98) or without (n=85) intrauterine growth retardation (IUGR, defined as a birth weight below the 10th centile). Twenty nine percent of the cases with IUGR had the Tyr113 Tyr113 genotype compared with 22% of the cases without IUGR (OR 1.2, 95% CI 0.6-2.4, p=0.58).

Only 3% of the controls were homozygous for the exon 4 polymorphism (Arg139 Arg139) and 33% of 151 subjects were heterozygous (His139 Arg139), compared to 7% and 37%, respectively, of the pre-eclamptic cases with or without the HELLP syndrome. This resulted in a non-significant OR of 2.2 (95% CI 0.7-7.4, p=0.20) comparing the Arg139 Arg139 genotype versus the His139 Arg139 and His139 His139 genotype. Again, the subgroup of women with a history of pre-eclampsia with HELLP, as compared to the group without HELLP, showed no significant differences in the rates of polymorphic variants in exon 4 (OR 0.9, 95% CI 0.3-3.3, p=0.85).

Allele frequencies for both exon 3 and exon 4 were not different between cases and controls (χ²=1.90, p=0.17 for exon 3; χ²=2.88, p=0.10 for exon 4, respectively).

The distribution of subjects with predicted high, intermediate, and low EPHX activity were 16.0, 26.9, and 57.1%, respectively, among pre-eclamptic women and 4.5%, 18.2%, and 77.3%, respectively, among controls. A significant association was found between predicted enzyme activity levels and pre-eclampsia (p=0.0009), resulting in a higher risk at higher predicted enzyme levels.

Discussion

An imbalance between toxic compounds and detoxifying substances may play a role in the pathophysiology of pre-eclampsia. Microsomal epoxide hydrolase is one of the enzymes playing a prominent role in a wide variety of detoxification processes and in the metabolism of endogenous and exogenous compounds. We tried to elucidate in this case-control study whether variations in rates of genetic polymorphisms in the EPHX gene may correlate with susceptibility to pre-eclampsia with or without the HELLP syndrome.

A significant difference in the frequencies of the Tyr113 Tyr113 genotype of EPHX was observed between the pre-eclamptic and control groups. Hassett et al 12 showed that allele substitution of histidine 113 for tyrosine 113 diminished the activity of EPHX by approximately 40%, whereas substitution of arginine 139 for histidine 139 increased the enzyme activity by 25%. Our study shows that the homozygous genotype in exon 3 of the EPHX gene resulting in Tyr113 Tyr113 and corresponding to high enzyme activity is associated with an increased incidence of pre-eclampsia. Also, a predicted high EPHX activity level, assessed from the polymorphisms in exons 3 and 4 together, was associated with pre-eclampsia. The association between a higher incidence of the high EPHX enzyme activity genotype Tyr113 Tyr113 in the pre-eclamptic group and the risk for pre-eclampsia may reflect enhanced activation by either endogenous or exogenous substrates to more reactive mutagenic diol derivatives in these patients.

Until now, only a few studies on the role of EPHX in the reproductive system have been performed. Wang et al 14 showed recently that polymorphisms in theEPHX gene might influence the susceptibility to spontaneous abortion. They showed an association between the His113 His113 genotype and an increased risk of spontaneous abortion. In contrast, Lancaster et al 15found an association between the Tyr113 Tyr113 genotype in exon 3 and ovarian cancer. These various results might be because under most circumstances EPHX plays a detoxifying role by preventing highly reactive epoxides from modifying essential cellular (macro)molecules. However, under certain conditions, EPHX can contribute in activating other compounds, resulting in toxification instead of detoxification.8 11 16 17 Furthermore, EPHX may be involved in steroidogenesis reactions that could explain the observation of an association between the slow genotype and protection against ovarian cancer.15

Polymorphisms in the EPHX gene are also associated with hepatocellular cancer, emphysema, and lung cancer.18-20 These results may represent an example of organ specific differences in disease susceptibility as a result of genetic polymorphisms, since the high activity EPHX genotype Tyr113 Tyr113 may protect against spontaneous abortion14 and hepatocellular cancer,18 whereas susceptibility for ovarian cancer,15 lung cancer,20 and pre-eclampsia may be increased.

The frequencies of the polymorphic variants in exon 4 of theEPHX gene do not vary much in several study populations.12 14 19 20 However, remarkable variation in polymorphic rates in exon 3 have been reported for different ethnic and geographical populations; ranges between 26-69% and 22-56% have been reported for the Tyr113 Tyr113 and the Tyr113 His113 genotypes, respectively.12 14 15 18-20 Reported frequencies of genetic polymorphisms in the EPHX gene are listed in table 3. It is striking that our data on the exon 3 polymorphisms in white subjects are more similar to the published data in Chinese controls than those of other white populations.14 18

These data support the hypothesis that genetic variations in genes coding for enzymes involved in metabolism of exogenous or endogenous substances may influence the risk for diseases of pregnancy such as spontaneous abortion and pre-eclampsia.