Microsomal Epoxide Hydrolase Gene Polymorphisms and Susceptibility to Chronic Obstructive Pulmonary Disease in the Tunisian Population

Abstract

It is well known that cigarette smoking is the major risk factor for chronic obstructive pulmonary disease (COPD). However, only 10%-20% of chronic heavy cigarette smokers develop symptomatic disease, which suggests the presence of genetic susceptibility. Microsomal epoxide hydrolase (EPHX1) is an enzyme involved in the protective mechanism against oxidative stress. It has been reported that gene polymorphisms of this enzyme may be associated with variations in EPHX1 activity. In this study, we aimed at investigating the relationship between EPHX1 polymorphisms and susceptibility to COPD in the Tunisian population. EPHX1 exon 3 (rs1051740, Tyr113His) and exon 4 (rs2234922, His139Arg) polymorphisms were genotyped by polymerase chain reaction followed by restriction fragment length polymorphism analysis. These techniques were used to examine a total of 416 Tunisian individuals, including 182 blood donors and a group of 234 COPD patients. All subjects were not related. An increased risk for COPD was observed in subjects with EPHX1 His113-His113 genotype (odds ratio = 2.168; confidence interval 1.098-4.283; p = 0.02386). However, multivariate logistic regression analysis showed no significant relationship between the mutant genotype and the disease after adjustment for sex, age, body mass index, smoking status, and pack-year smoking (odds ratio = 1.524; confidence interval, 0.991-6.058; p = 0.06137). Regarding the two subtypes of COPD, our investigations demonstrated that there is no significant correlation between exon 3 polymorphism and the chronic bronchitis subgroup (p = 0.09034). The relation between exon 3 polymorphism and emphysema was significant in the univariate analysis (p = 0.02257), but no association was found after controlling for classic risk factors (p = 0.06273). In conclusion, our results showed that there is a weak relation between 113His genotype and COPD, and no apparent relation between 139Arg and COPD in the studied Tunisian population.

Introduction

Chronic obstructive pulmonary disease (COPD) is a major and increasing cause of morbidity and mortality worldwide (Celli et al., 2004). Clinically, COPD is usually diagnosed on the basis of airflow obstruction that is not fully reversible after the administration of inhaled bronchodilators (Rabe et al., 2007). The clinical diagnosis of COPD includes chronic bronchitis, emphysema, and small airway disease (DeMeo et al., 2007). The main etiologic factor contributing to the development of this disease is cigarette smoking. Nevertheless, only a minority of individuals who smoke cigarettes develop COPD. Thus, there is evidence of an important role of genetic variability. A number of genetic association studies have been performed to find susceptibility genes of COPD. The only well-established genetic cause of COPD is α1-antitrypsin deficiency but its prevalence is rare (Black and Kueppers, 1978).

Several antioxidant genes such as glutathione S-transferase M1 (GSTM1), GSTT1, GSTP1, heme oxygenase-1, cytochrome P450, and microsomal epoxide hydrolase (EPHX1) have been proposed to contribute to the development of the disease. In fact, oxidative stress and reactive oxygen species (ROS), resulting from an oxidant/antioxidant imbalance, are believed to play an important role in the pathogenesis of COPD (Church and Pryor, 1985; Farber, 1994), and a defect in the detoxification of reactive species produced by cigarette smoking may predispose smokers to airflow obstruction and emphysema. In this study, we focused on the relation between the microsomal epoxide hydrolase EPHX1 gene and pulmonary chronic obstructive disease. The EPHX1 enzyme is an important phase II biotransformation enzyme that catalyzes the hydrolysis of various epoxides and reactive epoxide intermediates into less-reactive and more water-soluble dihydrodiols (Omiecinski et al., 1993; Archelas and Furstoss, 1998; Curtis et al., 2000). Its activity has been detected in all tissues, and the highest concentrations have been found in lung, liver, kidney, gonads, and epithelial cells (Omiecinski et al., 1993; Oesch and Schmassmann, 1997; Seidegard and Ekstrom, 1997). The gene of human EPHX1 enzyme is located on chromosome 1q42.1 and consists of nine exons. Within the coding regions, the T to C transition in exon 3 changes tyrosine (Tyr) residue 113 to histidine (His), and the A to G transition in exon 4 changes His residue 139 to arginine (Arg) (Hassett et al., 1994). Based on these two different single-nucleotide polymorphisms (SNPs), the population can be classified into four groups of putative EPHX1 phenotypes having different activities (fast, normal, slow, and very slow) (Smith and Harrison, 1997). Many studies have investigated the relationship between the EPHX1 genotypes and phenotypes with COPD susceptibility in different populations. It has been found that the slow-metabolizing form of EPHX1 is significantly higher in the COPD cases than in healthy controls (Smith and Harrison, 1997). However, the results remain controversial (Takeyabu et al., 2000; Yim et al., 2000; Yoshikawa et al., 2000; Sandford et al., 2001; Rodriguez et al., 2002; Zhang et al., 2002; Budhi et al., 2003; Cheng et al., 2004; Xiao et al., 2004; Hersh et al., 2005; Park et al., 2005; Zidzik et al., 2008). To confirm this susceptible association, more data from these polymorphisms by means of well-conducted studies (e.g., case-control, family-based linkage) remain necessary.

The purpose of our present case-control study was to investigate the effects of EPHX1 gene polymorphisms on COPD risk in a sample of a Tunisian population. Therefore, we attempted to evaluate the frequency of EPHX1 Tyr113His and His139Arg polymorphisms in the general population of our area and compare it with that of a population of individuals with COPD. We further analyzed the role of EPHX1 genotypes on the susceptibility to COPD subtypes (emphysema and chronic bronchitis). EPHX1 putative phenotype distribution was also compared between the different groups.

Material and Methods

Study design and inclusion criteria

Our hospital-based case-control study was carried out during the last 2 years among groups of patients with the diagnosis of COPD and in controls of Tunisian origin. The involved cases and healthy controls are not related.

The baseline demographic characteristics of the 416 study subjects are shown in Table 1.

Table 1.

Demographic Characteristics of Chronic Obstructive Pulmonary Disease Patients and Healthy Controls

Demographic characteristic	COPD patients (n = 234)	Controls (n = 182)	p-Value
Age (years, means ± SD)	61.75 ± 13.96	56.43 ± 7.03	0.18^a
Male:female	222:12	173:9	0.92^b
BMI (means ± SD)	25.3 ± 6.7	24.1 ± 5.2	0.48^a
Smoking status
Never	15	13	0.17^b
Ex	102	63
Current	117	106
Pack-years (means ± SD)	51.76 ± 29.35	47.58 ± 26.42	0.11^a
Geographic origin in Tunisia
North	0	4
Center	234	178	0.21^b
South	0	0

Body mass index (BMI) = weight (kg)/[height (m)]².

Pack years = (number of cigarettes smoked per day × number of years smoked)/20.

Student's t-test.

Pearson's χ² test.

COPD, chronic obstructive pulmonary disease.

Two hundred thirty-four patients were recruited from the service of pneumology in CHU Tahar Sfar in Mahdia, Tunisia. All patients satisfied the clinical criteria of COPD set down in the global strategy for obstructive lung disease guidelines. Inclusion criteria for COPD are the following: chronic airway symptoms and signs such as coughing, breathlessness, wheezing, and chronic airway obstruction (a forced expiratory volume in 1 s [FEV₁]/forced vital capacity [FVC] <70%, FEV₁ <80% of the predicted values from spirometric data, and FEV₁ reversibility <12% of prebronchodilator FEV₁ after inhalation of 200 mg of salbutamol). COPD phenotype exploration was based on chest radiographic and high-resolution computerized tomography density findings. The patients with bronchial asthma were excluded on the basis of reversibility of airflow obstruction. Information about patients is listed in Table 2.

Table 2.

Clinical Characteristics of Chronic Obstructive Pulmonary Disease Patients

Clinical characteristics	Patients (n = 234)
COPD phenotype
CB	127 (54.27%)
CLE + PLE	107 (45.72%)
Spirometry
FEV₁	1.19 ± 0.78
FEV₁% pred	46.7 ± 19.2
FVC	1.65 ± 0.83
FVC% pred	50.9 ± 16.4
FEV₁/FVC% pred	69.36 ± 0.66
RV%	4.3 ± 3.2
COPD stage
I	6 (2.56%)
II	40 (17.09%)
III	113 (48.29%)
IV	75 (32.05%)

Data are reported as number (percentage in parentheses) or as means ± SD.

CB, chronic bronchitis; CLE, centrolobular emphysema; PLE, panlobular emphysema; FEV₁, forced expiratory volume in 1 s (L); FVC, forced vital capacity (L); FEV₁% pred: percentage of the predicted FEV₁ value adjusted for age, height, and weight; FVC% pred, percentage of the predicted FVC value adjusted for age, height, and weight; RV, reversibility of baseline FEV₁ following bronchodilator inhalation.

The control group contains 182 healthy subjects. All were recruited from a blood donors' cohort from the central area of Tunisia. Individuals with respiratory diseases or any familial history of lung disease were excluded. All control subjects exhibited normal pulmonary function (FEV₁/FVC >70% and FEV₁ >80% of predicted value). Hospital ethics committee approval and informed consent were obtained.

DNA preparation

Venous blood was collected in EDTA tubes and stored at −20°C until DNA extraction was carried out. The standard proteinase K-phenol-chloroform protocol was used to isolate genomic DNA from the frozen specimens (Dsavis et al., 1986).

Polymerase chain reaction-restriction fragment length polymorphism analysis of EHPX1

Two separate polymerase chain reaction (PCR)-based restriction fragment length polymorphism assays were used to examine EHPX1 gene polymorphisms. Genotyping was carried out using the method of Smith and Harrison (1997). For both exon 3 and exon 4, genomic DNA (100 ng) was amplified using a Mastercycler personal thermal cycler (Eppendorf ) in 25 μL reaction mixture containing 1.5 mM MgCl₂, 1 μM of each primer, 0.2 mM deoxyribonucleoside triphosphate, and 0.5 IU of GoTaq^® Flexi DNA polymerase (Promega). The primers EPO1 (5′-GATCGATAAGTTCCGTTTCACC-3′) and EPO2 (5′-ATCCTTAGTCTTGAAGTGAGGAT-3′) were used for the amplification of exon 3. SNP genotypes of the EHPX1 Tyr113-His113 substitution were determined by carrying out an initial denaturation for 10 min at 95°C, followed by 40 cycles of denaturation at 95°C for 30 s, annealing at 56°C for 30 s, and extension at 72°C for 1 min. The final step of elongation was performed at 72°C for 10 min. Negative controls were included in every PCR run.

After enzyme digestion with EcoRV (Fermentas) overnight at 37°C, the homozygous wild genotype yielded two bands, 140 and 20 bp, whereas the homozygous mutant genotype yielded only one band of 162 bp. The fragments were resolved on ethidium bromide-stained 3% agarose gel and transilluminated with UV light. For exon 4, the two variant alleles were also differentiated by restriction fragment length polymorphism. The DNA of each sample was subjected to PCR amplification with the following primers: EPO3 (5′-ACATCCACTTCATCCACGT-3′) and EPO4 (5′-ATGCCTCTGAGAAGCCAT-3′).

The PCR program was as follows: an initial denaturation step at 95°C for 10 min, 35 cycles of 95°C for 30 s, 56°C for 30 s, and 72°C for 1 min, followed by an elongation step at 72°C for 5 min.

The restriction enzyme RsaI (Fermentas) was then used to digest the PCR product, which was then, together with the DNA ladder, visualized on a 3% agarose gel. The wild-type genotype showed a fragment of 210 bp, whereas the mutant genotype yielded two fragments of 164 and 46 bp.

Statistical analysis

All the statistical analyses were undertaken using the Statistical Package for the Social Sciences (SPSS), version 10 software. Hardy-Weinberg equilibrium test of genotype distribution in cases and controls was evaluated by calculating χ² and p-values. Genotypes and allele distribution differences among the groups were examined for statistical significance by Pearson's χ² test. Odds ratios (OR) with their corresponding 95% confidence intervals (CIs) were calculated. The χ² test was also used to compare gender, smoking status, and geographic origin between the two groups. Demographic data and spirometric parameters are presented as means ± standard deviation. Age and smoking index expressed as pack-years [(number of cigarettes smoked per day × number of years smoked)/20] were compared using the unpaired Student's t-test. Frequencies, OR, and CI of each phenotype (normal, fast, slow, and very slow) versus all other phenotypes described by Smith and Harrison were determined. A multivariate logistic regression model was constructed to confirm the association found between EHPX1 exon 3 polymorphism and COPD as well as emphysema, by adjustment for covariates, including age, sex, body mass index (BMI), smoking status, and pack-years of smoking. Analysis of variance was carried out to evaluate continuous variables distribution according to different genotypes of EHPX1 exon 3 polymorphism. p < 0.05 was considered as statistically significant.

Results

Characteristics of the studied population

Table 1 shows the demographic characteristics of cases and controls.

There was no significant difference between patients and controls in terms of gender (male:female ratio, 222:12 vs. 173:9) and age (mean age, 61.75 ± 13.96 vs. 56.43 ± 7.03). BMI was similar in both groups (p = 0.48), as well as pack-years of cigarette smoking (p = 0.11). Clinical characteristics of patients are given in Table 2.

EPHX1 alleles in the different study groups

There was no deviation in the distribution of EPHX1 exon 3 and exon 4 genotypes from Hardy-Weinberg equilibrium in the groups studied. Genotype and allele frequencies by case-control status are given in Table 3.

Table 3.

Distribution of Epoxide Hydrolase Exon 3 and Exon 4 Polymorphisms in Chronic Obstructive Pulmonary Disease Patients and Healthy Controls

Genotype	COPD cases No. (%)	Controls No. (%)	OR (95% CI)	p-Value
EPHX1 exon 3
Tyr113/Tyr113	102 (43.58)	86 (47.25)
Tyr113/His113	96 (41.02)	82 (45.05)
His113/His113	36 (15.38)	14 (7.69)	2.168 (1.098-4.283)	0.02386
His113 allele frequency	0.36	0.3	1.293 (0.965-1.733)	0.08500
EPHX1 exon 4
His139/His139	142 (60.68)	119 (65.38)
His139/Arg139	86 (36.75)	59 (32.41)
Arg139/Arg139	6 (2.56)	4 (2.19)	1.257 (0.347-4.559)	0.72733
Arg139 allele frequency	0.21	0.18	1.174 (0.831-1.660)	0.36324

OR, odds ratio; CI, confidence interval; EPHX1, epoxide hydrolase; Tyr, tyrosine; His, histidine; Arg, arginine.

In the present control group, a total of 14 persons (7.69%) and 86 persons (47.25%) from the general population were homozygous for the EHPX1 113His and 113Tyr alleles, respectively.

In our sample, only four subjects exhibited the homozygous mutant variant involving the EHPX1 exon 4 Arg139Arg SNP (Table 3).

The frequencies of His139/His139 (60.68% vs. 65.38%), His139/Arg139 (36.75% vs. 32.41%), and Arg139/Arg139 (2.56% vs. 2.19%) genotypes were similar between COPD patients and controls. There was also no significant difference in mutant allele frequency among the two groups (OR = 1.174; CI, 0.831-1.660; p = 0.36324).

In contrast, for the exon 3 genotypes, we found that the frequency of the mutant genotype was twofold higher in the patient group than in controls (15.38% vs. 7.69%). The statistical calculations showed a significant difference between cases and controls for the homozygous mutation (OR = 2.168; CI, 1.098-4.283; p = 0.02386).

A logistic regression model was used to assess the contribution of predictors of the disease, with COPD as the dependent variable and sex, age, BMI, smoking status, pack-year smoking, and EHPX1 exon 3 His113/His113 genotype as the independent, potentially confounding variables. After adjustment, the OR changed, the p-value was somewhat attenuated (p = 0.06137), and the association found before the adjustment disappeared (Table 4). Pack-year smoking showed a significant association with the disease in the regression model (p = 0.03291), which would indicate that the number of pack cigarettes per year might be in correlation with the EHPX1 exon 3 His113/His113 genotype in the studied sample. This was confirmed by analyzing pack-year cigarette mean distribution between EHPX1 exon 3 genotypes. Analysis of variance demonstrated that subjects carrying His113/His113 genotype showed a significantly higher pack-year smoking mean than Tyr113/His113 and Tyr113/Tyr113 subjects (57.76 ± 23.64 vs. 44.66 ± 24.31 and 51.88 ± 25.65, respectively; p = 0.0311).

Table 4.

Logistic Regression Analysis for Chronic Obstructive Pulmonary Disease and Emphysema Risk Factors

	COPD			Emphysema
Factors	OR	95% CI	p-Value	OR	95% CI	p-Value
His113/His113	1.524	0.991-6.058	0.06137	1.937	0.876-7.854	0.06273
Age	1.295	0.582-4.156	0.33956	0.876	0.486-1.678	0.08613
Sex	1.315	0.117-15.718	0.82306	1.312	0.078-12.989	0.86249
BMI	0.912	0.265-4.657	0.55008	0.585	0.139-0.826	0.03526
Smoking status	4.643	0.684-31.649	0.09309	2.189	0.315-15.281	0.42371
Pack-years	1.787	1.455-9.617	0.03291	1.246	0.441-18.693	0.16725

To investigate the relation between EPHX1 alleles and emphysema as well as chronic bronchitis susceptibility, we studied the frequency of each genotype in the two subgroups of COPD patients.

The results obtained by the comparison of both EPHX1 exon 3 and exon 4 variants between emphysema, chronic bronchitis groups, and controls are presented in Table 5.

Table 5.

Distribution of Epoxide Hydrolase Exon 3 and Exon 4 Polymorphisms in Patients with Emphysema and Chronic Bronchitis Compared with Healthy Controls

Genotype	E (n = 107) No. (%)	OR (95% CI)	p-Value	CB (n = 127) No. (%)	OR (95% CI)	p-Value
EPHX1 exon 3
Tyr113/Tyr113	45 (42.05)			57 (44.88)
Tyr113/His113	44 (41.12)			52 (40.94)
His113/His113	18 (16.82)	2.457 (1.120-5.393)	0.02257	18 (14.17)	1.940 (0.894-4.209)	0.09034
Mutant allele frequency	0.37	1.379 (0.966-1.968)	0.07668	0.35	1.224 (0.870-1.723)	0.24600
EPHX1 exon 4
His139/His139	57 (53.27)			85 (66.92)
His139/Arg139	46 (42.99)			40 (31.49)
Arg139/Arg139	4 (3.73)	2.088 (0.504-8.649)	0.30067	2 (1.57)	0.700 (0.125-3.909)	0.68297
Mutant allele frequency	0.25	1.496 (0.996-2.247)	0.06141	0.17	0.929 (0.611-1.413)	0.72985

E, emphysema.

The frequency of the homozygous mutant His113/His113 variant was slightly higher in the group of patients with emphysema than those with chronic bronchitis (16.82% vs. 14.17%). Our results showed that there was no significant relationship between the homozygous mutation and chronic bronchitis (OR = 1.940; CI, 0.894-4.209; p = 0.09034). However, we found that the association between the mutant variant and emphysema susceptibility was significant (OR = 2.457; CI, 1.120-5.393; p = 0.02257). After controlling for sex, age, BMI, smoking status, and pack-year smoking, no relationship was found between the homozygous mutant and this group of patients (OR = 1.937; CI, 0.876-7.854; p = 0.06273). BMI (p = 0.03526) was the only variable found in correlation with emphysema (Table 4).

Regarding the exon 4 polymorphism, the mutant allele frequency was higher in cases with emphysema than those with chronic bronchitis (0.25 and 0.17, respectively). The statistical results demonstrated that there was no correlation between the two subgroups of patients and the Arg allele when considered as an allele of risk (p = 0.30067and p = 0.68297 for emphysema cases and chronic bronchitis, respectively) (Table 5).

Determination of EPHX1 phenotypes

On the basis of the assumption that Tyr allele at exon 3 and His allele at exon 4 confer normal activity, whereas His allele at exon 3 confers low activity and Arg allele at exon 4 confers high activity; Smith and Harrison (1997) classified predicted EPHX1 activity as normal, fast, slow, and very slow [normal: no mutation in the gene, or heterozygotes for both exon 3 and exon 4 mutations; fast: at least one fast mutation (exon 4) and no exon 3 mutations; slow: one slow (exon 3) allele; very slow: two slow alleles] on the presence or absence of the two polymorphisms.

The distribution of putative EPHX1 phenotypes in COPD, emphysema, chronic bronchitis patients, and healthy controls is presented in Table 6.

Table 6.

Distribution of Putative Epoxide Hydrolase Phenotypes in Patients and Healthy Controls

	Number of individuals (% of group) with phenotype ^a
	Normal	Fast	Slow	Very slow	OR (95% CI) ^b
Controls (n = 182)	77 (42.30)	45 (24.72)	46 (25.27)	14 (7.69)
COPD (n = 234)	54 (23.07)	53 (22.64)	91 (38.88)	36 (15.38)	2.168 (1.098-4.283)
					p < 0.005
E (n = 107)	24 (22.42)	23 (21.49)	42 (39.25)	18 (16.82)	2.457 (1.120-5.393)
					p < 0.005
CB (n = 127)	38 (29.92)	34 (26.77)	37 (29.13)	18 (14.17)	1.940 (0.894-4.209)
					p > 0.005

Classification based on Smith and Harrison.

OR calculated for the risk of very slow versus all other phenotypes.

Our results showed that the frequency of normal phenotype was lower in COPD, emphysema, and chronic bronchitis cases compared with controls (23.07%, 22.42%, and 29.92% vs. 42.30%, respectively). In contrast, the slow phenotype was found higher in the emphysema subgroup (39.25%) and COPD patients (38.88%) than in controls (25.27%). The frequency of the fast phenotype was similar in all groups. These phenotypes did not show any significant association with the disease (data not shown). However, the lower enzyme activity was found in association with COPD (OR = 2.168; CI, 1.098-4.283) and emphysema (OR = 2.457; CI, 1.120-5.393) as previously demonstrated in the univariate analysis (Table 6).

Discussion

In the present work, we examined EPHX1 exon 3 and exon 4 polymorphisms in COPD patients and healthy controls to demonstrate if the EPHX1 alleles are associated with COPD susceptibility. We provided new genotyping data as well as putative phenotypes of these polymorphisms in the Tunisian population.

After comparing patients and controls, we found that the frequencies of EPHX1 exon 3 polymorphisms in COPD patients differ significantly from the control group. Nevertheless, adjustment for COPD classic risk factors showed that the relationship between the EPHX1 His/His genotypes of Tyr113His polymorphism and COPD seems not to be independent. Our findings were in agreement with several published works (Smith and Harrison, 1997; Sandford et al., 2001; Cheng et al., 2004; Hersh et al., 2005). However, the implication of this enzyme in COPD susceptibility was not yet confirmed (Takeyabu et al., 2000; Yim et al., 2000; Matheson et al., 2006; Zidzik et al., 2008). Indeed, the results were different from one study to another. This nonreplication is due to many reasons such as small sample sizes, population stratification, and genetic heterogeneity (Silverman and Palmer, 2000). COPD is a heterogeneous disease, and its definition has been changing (Celli et al., 2004; Rabe et al., 2007). Phenotypic heterogeneity may be an important reason for nonreplication. Indeed, the nonreproducible COPD phenotypes used may lead to inconsistent results. Further, the distribution of genotypes in some studies deviated from Hardy-Weinberg equilibrium (Smith and Harrison, 1997; Yim et al., 2000; Cheng et al., 2004). A perfect COPD genetic association study would be large and longitudinal and would use the cumulative reduction in lung function in relation to cumulative smoking as a main outcome measure. Such a study would take decades to be completed and require tens of thousands of subjects. Thus, to avoid these error sources, meta-analysis studies seem to be an essential tool for summarizing case-control studies.

On the other hand, some studies reported a relationship between EPHX1 polymorphisms and COPD susceptibility when EPHX1 polymorphisms are combined with other enzyme polymorphisms such as GSTM1 and GSTP1 (Cheng et al., 2004; Zidzik et al., 2008). As the ability to decrease oxidative stress is mediated in part through several pathway enzymes, it is likely that the operation of multiple genes is necessary and that susceptibility to disease depends on the coincident actions of several genetic events. Polymorphism of each gene may impart only a small relative risk of COPD (Cheng et al., 2004).

In this work, we further investigated the relationship between EPHX1 polymorphisms and COPD subtypes, emphysema, and chronic bronchitis susceptibility. Our results suggest that the homozygous mutant genotypes of both Tyr113His and His139Arg are, so far, not associated with any subgroup of the disease although the relationship between emphysema and the mutant variant in the exon 3 His113His113 was found significant in the univariate analysis. This slight association was attenuated after controlling for other COPD classic risk factors.

The obtained results concerning EPHX1 exon 3 SNP are not in agreement with some previous studies (Smith and Harrison, 1997; DeMeo et al., 2007). Indeed, in our study, after patients classification according to COPD phenotypes, the size of the subgroups became small and this may affect the statistical power and also the findings. Further, this inconsistency may be due to population stratification. The frequency distribution in the control sample investigated by Smith and Harrison (1997) deviated from Hardy-Weinberg equilibrium.

It is well known that ethnic difference may be a cause of the discrepancy between works because of the difference in the distribution of polymorphisms and allele frequencies among races. Nevertheless, we notice that our results are in line with those of Takeyabu et al. (2000), who studied an Asiatic population.

Regarding the genotype analysis, our results suggest that the exon 3 EPHX1 mutant variant may contribute to the susceptibility to COPD and emphysema in the Tunisian population. However, the significant association of the exon 3 EPHX1 mutant homozygotes disappeared after adjustment for classic covariates. These results might be explained by the fact that the effects of the genetic variants on COPD risk could be exerted through clinical characteristics such as pack-year smoking and BMI, which imply that including both genes and intermediate factors in the regression model will reduce the effect of the gene. Indeed, it is generally held that the disease is frequently associated with smoking, and COPD prognosis is improved for patients who cease smoking. Further, several investigations demonstrated that emphysema risk was negatively associated with BMI (Ogawa et al., 2009). Moreover, a second reason that may explain the results found is that the contribution of the exon 3 EPHX1 mutant variant may be outweighed by the effects of pack-year smoking and BMI on COPD and emphysema risk in our population.

As for the association between putative phenotypes and COPD, emphysema, and chronic bronchitis, our results showed that normal, fast, and slow phenotypes seem not to have a strong relationship with the disease. Moreover, the prevalence of very slow phenotype, conferring lower enzymatic activity, was higher in all patients than in controls. Subjects with very slow phenotype in EPHX1 gene are likely more susceptible to COPD, even if it was not an independent risk factor for the disease as suggested by the findings of the present study. Our data, before controlling for classic COPD risk factors, were in line with some published studies in the Caucasian population (Smith and Harrison, 1997; Sandford et al., 2001; Park et al., 2005) and the Asiatic population (Fu et al., 2007) but inconsistent with some others works from East Asia (Takeyabu et al., 2000; Yim et al., 2000; Zhang et al., 2002). Ethnicity and different criteria of patient selection may explain the controversy. Genotyping errors may also explain this discrepancy. Indeed, it has been reported that the results with PCR-RLFP performed in exon 3 polymorphism genotyping of most researches (Seidegard and Ekstrom, 1997; Takeyabu et al., 2000; Sandford et al., 2001; Rodriguez et al., 2002; Zhang et al., 2002; Budhi et al., 2003; Cheng et al., 2004; Park et al., 2005) were not consistent with those by direct sequencing (Keicho et al., 2001). The obtained frequencies from the conventional PCR-based method might induce some wrong conclusions.

Several limitations remain in our present investigation. First, our studied sample was moderate. This may affect the potential to detect and replicate the previous associations because, as reported (Brogger et al., 2006), the size of the population may be a cause of nonreplication. Second, in genotyping, we followed the PCR-RLFP method described by Smith and Harrison. A Japanese team have demonstrated that a pitfall for PCR-based association studies on EPHX1 gene polymorphisms can occur from the interference of the nearby SNP (Lys119) with primer hybridization. Thus, the distribution of polymorphisms should be tested for Hardy-Weinberg equilibrium (Keicho et al., 2001). In our work, the statistical calculations showed that the distribution of EPHX1 gene polymorphisms was in Hardy-Weinberg equilibrium. Therefore, our results are likely unaffected by the potential bias resulting from the conventional way of genotyping EPHX1 gene.

In summary, we have demonstrated, after adjustment for COPD classic risk factors, that the genetic polymorphisms of EPHX1 did not seem to be an independent risk factor for either COPD or emphysema susceptibility in the Tunisian population.

We believe that further studies using larger population and investigating other candidate genes in addition to the previously known in combination and using different approaches, such as genome-wide scanning method, are needed to confirm the association of EPHX1 gene polymorphisms with susceptibility to pulmonary emphysema or COPD and to find a novel genetic link with COPD.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Archelas

, Furstoss

. 1998. Epoxide hydrolases: new tools for the synthesis of fine organic chemicals. Trends Biotechnol, 16:108-116.

Black

, Kueppers

. 1978. α1 antitrypsin deficiency in nonsmokers. Am Rev Respir Dis, 17:421-428.

Brogger

, Steen

, Eiken

et al. 2006. Genetic association between COPD and polymorphisms in TNF, ADRB2 and EPHX1. Eur Respir J, 27:682-688.

Budhi

, Hiyama

, Isobe

et al. 2003. Genetic susceptibility for emphysematous changes of the lung in Japanese. Int J Mol Med, 11:321-329.

Celli

, Mac

. ATS/ERS Task Force. 2004. Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper. Eur Respir J, 23:932-946.

Cheng

, Yu

, Chen

et al. 2004. Genetic polymorphism of epoxide hydrolase and glutathione S-transferase in COPD. Eur Respir J, 23:818-824.

Church

, Pryor

. 1985. Free radical chemistry of cigarette smoke and its toxicoligical implications. Environ Health Perspect, 64:111-126.

Curtis

, Hassett

, Hosagrahara

. 2000. Epoxide hydrolase-polymorphism and role in toxicology. Toxicol Lett, 112-113:365-370.

DeMeo

, Hersh

, Hoffman

et al. 2007. Genetic determinants of emphysema distribution in the national emphysema treatment trial. Am J Respir Crit Care Med, 76:42-48.

10.

Dsavis

, Dibner

, Batte

. 1986. Basic Methods in Molecular Biology. Elsevier: New York, 44-87.

11.

Farber

. 1994. Mechanisms of cell injury by activated oxygen species. Environ Health Perspect, 102:17-24.

12.

, Sun

, Dai

et al. 2007. Relationship between COPD and polymorphisms of HOX-1 and mEPH in a Chinese population. Oncol Rep, 17:483-488.

13.

Hassett

, Aicher

, Sidhu

et al. 1994. Human microsomal epoxide hydrolase: genetic polymorphism and functional expression in vitro of amino acid variants. Hum Mol Genet, 3:421-428.

14.

Hersh

, De Meo

, Lange

et al. 2005. Attempted replication of reported chronic obstructive pulmonary disease candidate gene associations. Am J Respir Cell Mol Biol, 33:71-78.

15.

Keicho

, Emi

, Kajita

et al. 2001. Overestimated frequency of a possible emphysema-susceptibility allele when microsomal epoxide hydrolase is genotyped by the conventional polymerase chain reaction-based method. J Hum Genet, 46:96-98.

16.

Matheson

, Raven

, Walters

et al. 2006. Microsomal epoxide hydrolase is not associated with COPD in a community-based sample. Hum Biol, 78:705-717.

17.

Oesch

, Schmassmann

. 1979. Species and organ specificity of the transstilbene oxide induced effects on epoxide hydratase and benzo(a)pyrene monooxygenase activity in rodents. Biochem Pharmacol, 28:171-176.

18.

Ogawa

, Nakano

, Ohara

et al. 2009. Body mass index in male patients with COPD: correlation with low attenuation areas on CT. Thorax, 64:20-25.

19.

Omiecinski

, Aicher

, Holubkov

et al. 1993. Human peripheral lymphocytes as indicators of microsomal epoxide hydrolase activity in liver and lung. Pharmacogenetics, 3:150-158.

20.

Park

, Chen

, Wadhwa

et al. 2005. Polymorphisms for microsomal epoxide hydrolase and genetic susceptibility to COPD. Int J Mol Med, 15:443-448.

21.

Rabe

, Hurd

, Anzueto

et al. 2007. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med, 176:532-555.

22.

Rodriguez

, Jardi

, Costa

et al. 2002. Detection of polymorphisms at exons 3 (Tyr113→His) and 4 (his139→Arg) of the microsomal epoxide hydrolase gene using fluorescence PCR method combined with melting curves analysis. Anal Biochem, 308:120-126.

23.

Sandford

, Chagani

, Weir

et al. 2001. Susceptibility genes for rapid decline of lung function in the lung health study. Am J Respir Crit Care Med, 163:469-473.

24.

Seidegard

, Ekstrom

. 1997. The role of human glutathione transferases and epoxide hydrolases in the metabolism of xenobiotics. Environ Health Perspect, 105:791-799.

25.

Silverman

, Palmer

. 2000. Case-control association studies for the genetics of complex respiratory diseases. Am J Respir Cell Mol Biol, 22:645-648.

26.

Smith

, Harrison

. 1997. Association between polymorphism in gene for microsomal epoxide hydrolase and susceptibility to emphysema. Lancet, 350:630-633.

27.

Takeyabu

, Yamaguchi

, Suzuk

et al. 2000. Gene polymorphism for microsomal epoxide hydrolase and susceptibility to emphysema in a Japanese population. Eur Respir J, 15:891-894.

28.

Xiao

, Wang

, Du

et al. 2004. Relationship between polymorphisms of genes encoding microsomal epoxide hydrolase and glutathione S-transferase P1 and chronic obstructive pulmonary disease. Chin Med J, 117:661-667.

29.

Yim

, Park

, Lee

et al. 2000. Genetic susceptibility to chronic obstructive pulmonary disease in Koreans: combined analysis of polymorphic genotypes for microsomal epoxide hydrolase and glutathione S-transferase M1 and T1. Thorax, 55:121-125.

30.

Yoshikawa

, Hiyama

, Ishioka

et al. 2000. Microsomal epoxide hydrolase genotypes and chronic obstructive pulmonary disease in Japanese. Int J Mol Med, 5:49-53.

31.

Zhang

, Zhang

, He

et al. 2002. Microsomal epoxide hydrolase gene polymorphism and susceptibility to chronic obstructive pulmonary disease in Han nationality North China (in Chinese) Zhonghua Nei Ke Za Zhi, 41:11-14.

32.

Zidzik

, Slabá

, Joppa

et al. 2008. Glutathione S-transferase and microsomal epoxide hydrolase gene polymorphisms and risk of chronic obstructive pulmonary disease in Slovak population. Croat Med J, 49:182-191.