- Email: [email protected]
Association between polymorphism in gene for microsomal epoxide hydrolase and susceptibility to emphysema
THE LANCET
Early report
Association between polymorphism in gene for microsomal epoxide hydrolase and susceptibility to emphysema Christopher A D Smith, David J Harrison
Summary Background The first-pass metabolism of foreign compounds in the lung is an important protective mechanism against oxidative stress. We investigated whether polymorphisms in the gene for microsomal epoxide hydrolase (mEPHX), an enzyme involved in this protective process, had any bearing on individual susceptibility to the development of chronic obstructive pulmonary disease (COPD) and emphysema. Methods We designed PCR-based genotyping assays to detect variant forms of mEPHX that confer slow and fast activity. We used these assays to screen 203 blood-donor controls and groups of patients with asthma (n=57), lung cancer (n=50), COPD (n=68), and emphysema (n=94), who were attending specialised clinics in Edinburgh, UK. Findings The proportion of individuals with innate slow mEPHX activity (homozygotes) was significantly higher in both the COPD group and the emphysema group than in the control group (COPD 13 [19%] vs control 13 [6%]; emphysema 21 [22%] vs 13 [6%]). The odds ratios for homozygous slow activity versus all other phenotypes were 4·1 (95% CI 1·8–9·7) for COPD and 5·0 (2·3–10·9) for emphysema. Interpretation Genetic polymorphisms in xenobiotic enzymes may have a role in individual susceptibility to oxidant-related lung disease. Epoxide derivatives of cigarette-smoke components may be the cause of some of the lung damage characteristic of these diseases.
Lancet 1997; 350: 630–33
Introduction The lungs are subject to oxidative stress from cigarette smoke, occupational exposure to solvents, and other chemicals and environmental pollutants. All these potential hazards contain factors (xenobiotics and oxidants) that can induce severe macromolecular, cellular, and tissue damage through direct cytotoxic effects, promotion of primary genotoxic events, or generation of reactive oxygen intermediates. Chronic obstructive pulmonary disease (COPD) is characterised by fixed and irreversible air-flow limitation in the lungs, and emphysema is distinguished by abnormal, permanent distal air-space enlargement Department of Pathology, University of Edinburgh, Edinburgh, UK (C A D Smith PhD, D J Harrison MD) Correspondence to: Dr Christopher A D Smith, Genovar Diagnostics, Room LF6, South Academic Block, Southampton General Hospital, Southampton SO16 6YD, UK
630
accompanied by destruction of lung parenchyma. Two lines of research are relevant to the study of emphysema in particular. First, some findings suggest that lung damage can be attributed to an imbalance in the endogenous protease/antiprotease equilibrium, promoting tissue hydrolysis;1–3 examples of such imbalance include the genetically determined ␣1-antitrypsin deficiency,3,4 and the overexpression of elastases and collagenases that mimics the development of pulmonary emphysema.7,8 Second, the oxidant/antioxidant theory postulates that an excess of oxidants and free radicals in the lung promotes cellular and tissue damage and is the major initiator of the disease process.9,10 Cigarette smoke, oxidants, xenobiotics, and reactive oxygen intermediates directly inhibit antiproteases and promote cell and tissue proteolysis.1,11,12 Tobacco smoke is the commonest identifiable risk factor for both emphysema and COPD. Combustion products released into the respiratory tract include various highly reactive oxygen and carbon-centred species, free radicals in the vapour phase, and stable radicals in particulate matter. Microsomal epoxide hydrolase (mEPHX) is an enzyme involved in the first-pass metabolism of highly reactive epoxide intermediates. We investigated whether genetic polymorphisms in the gene for this enzyme have any bearing on the development of oxidant-related lung disease. Epidemological studies show that mEPHX activity in the liver, lung, and peripheral blood leucocytes varies as much as 50-fold in white populations.13 Two common aberrant alleles can be detected, which confer slow and fast enzyme activity.14 We have designed PCR assays to detect these mutations and have used them to study the role of the gene in protection against oxidative damage to the lung.
Patients and methods 203 control blood samples were obtained from the Scottish National Blood Transfusion Service. These anonymous donors were all white individuals aged between 18 and 65 years; roughly equal numbers were male and female. Because emphysema is a morphological diagnosis, we cannot be certain that some controls did not have early emphysema; however, all were clinically healthy on routine questioning and therefore met the usual criteria for blood donation. All patients were fully investigated clinically and were attending specialised clinics. We obtained blood samples from 57 patients with chronic asthma attending a clinic at the Western General Hospital, Edinburgh, and a cohort of 68 patients with non-reversible COPD associated with cigarette smoking, who were attending a respiratory failure clinic at the Royal Infirmary of Edinburgh. All the COPD patients (average age 63 years) had smoked (mean pack years 35 [range 13–96]), and all had forced expiratory volume in 1 s of less than 1·5 L (mean 1·05 [0·7–1·4]). Patients with ␣1-antitrypsin deficiency were excluded on the results of isoelectric focusing of serum for variant phenotypes.
Vol 350 • August 30, 1997
THE LANCET
PCR-based genotyping assays for mEPHX polymorphism Top panels show arrangement of exons 3 and 4. Middle panels show fragment patterns expected in individuals homozygous or heterozygous for mEPHX polymorphisms. Lower panels show typical electrophoresis gels with band visualisation; only larger diagnostic bands shown.
144 consecutive lung-cancer resection samples were obtained from the archives of the Department of Pathology, University of Edinburgh. All the patients were smokers, but precise data on smoking duration were not available. Each resection sample was macroscopically and microscopically assessed for presence of tumour and extent and type of emphysema.15 All emphysema patients had mild to moderate disease; those with severely compromised lung function would not have been candidates for surgical lung resection. In total we identified 94 samples from patients with emphysema and lung cancer (age range 34–82 years; 25 women, 69 men) and 50 from patients with lung cancer and no evidence of emphysema. All these patients were white. DNA was obtained16,17 from sections of archival tissue samples without neoplastic changes, from fresh peripheral blood leucocytes, or from both types of sample. The study design was approved by the Lothian Health Board Ethical Review Committee, and informed consent was obtained from the patients attending the respiratory clinics. There are four distinct mEPHX alleles, which arise because of the presence or absence of two point mutations in the gene. An exon-3 T to C mutation changes tyrosine residue 113 to histidine, and enzyme activity is reduced by at least 50% (slow allele). The second, an A to G transition mutation in exon 4 of the gene, changes histidine residue 139 to arginine, and produces an enzyme with an activity increased by at least 25% (fast allele).14 The wild-type allele has neither of these changes. The rare occurrence of both mutations together also produces a hydrolase with normal activity. Two separate PCR assays are used to detect the two mutations. The assay for the exon-3 variant uses the primer pair EPO1 (5'-GATCGATAAGTTCCGTTTCACC, starting at bp 321 in mEPHX cDNA21 and EPO2 (5'-ATCCTTAGTCTTGAAGTGAGGAT, starting at bp 461). The downstream primer abuts directly onto the mutation site, and an
Vol 350 • August 30, 1997
engineered base change (shown in bold type; A for G) produces an EcoR V restriction enzyme site (GATATC) in the wild-type only. The exon-4 mutation produces an Rsa I restriction-fragment length polymorphism in the mutant (fast) allele (ATAC to GTAC); a second primer pair EPO3 (5'-ACATCCACTTCATCCACGT, bp 494) and EPO4 (5'-ATGCCTCTGAGAAGCCAT, bp 685) is used to test for this mutation. The figure shows diagrammatically the PCR reactions and typical results of the genotyping assays. PCRs were carried out with a Omnigene (Hybaid) thermocycler in PCR buffer (1·5 mmol/L magnesium chloride, 100 ng primers, 4% dimethyl sulphoxide, 200 mmol/L of each dNTP, and 1 unit of Taq polymerase [Promega]) in a final volume of 50 L. The cycles consisted of steps of 56°C for 20 s, 72°C for 20 s, and 94°C for 30 s (38 cycles). The products were then digested with EcoR V (exon 3) and Rsa I (exon 4), which produce diagnostic fragments of the sizes shown in the figure. The fragments were separated by electrophoresis through a non-denaturing 8% polyacrylamide gel containing 0·09 mol/L Tris-borate and 0·002 mol/L edetic acid. The gels were stained with ethidium bromide and transilluminated with ultraviolet light. Three representative fragments corresponding to each allele were cloned and sequenced to confirm the specificity of the PCR (results not shown). Associations between disease groups and specific genotypes and phenotypes were analysed for significance by the two-tailed 2 test. Probability values were corrected by the Bonferroni method. Odds ratios and 95% CI were calculated to assess relative risk of disease conferred by a particular allele, genotype, and implied phenotype.
Results The PCR produced amplimers of the expected size and differentiated between individual mEPHX genotypes in the study populations (figure). In the control group the 631
THE LANCET
Number of individuals (% of group)
Odds ratio (95% CI) for homozygous mutations
Mutant allele frequency
Odds ratio (95% CI) for carriage of mutant allele
13 (6%) 5 (10%) 4 (7%) 13 (19%)† 26 (28%)‡
1·0* 1·6 (0·6–4·8) 1·1 (0·3–3·5) 3·5 (1·5–8·0) 5·6 (2·7–11·5)
0·31 0·30 0·34 0·40 0·52¶
1·0* 1·0 (0·6–1·6) 1·1 (0·7–1·7) 1·5 (0·7–2·2) 2·4 (1·5–4·0)
3 (2%) 1 (2%) 1 (2%) 2 (3%)§ 4 (4%)
1·0* 1·4 (0·1–13·3) 1·2 (0·1–11·7) 2·0 (0·3–2·4) 2·8 (0·6–12·9)
0·15 0·18 0·15 0·24¶ 0·19
1·0* 1·3 (0·7–2·3) 1·0 (0·6–1·8) 1·9 (1·2–3·0) 1·3 (0·8–2·1)
Homozygous wild-type Heterozygous
Homozygous mutant
Exon-3 polymorphism (slow) Controls (n=203) Lung cancer (n=50) Asthma (n=57) COPD (n=68) Emphysema (n=94)
91 (45%) 25 (50%) 24 (42%) 27 (40%) 23 (24%)
99 (49%) 20 (40%) 29 (51%) 28 (41%) 45 (48%)
Exon-4 polymorphism (fast) Controls (n=203) Lung cancer (n=50) Asthma (n=57) COPD (n=68) Emphysema (n=94)
147 (72%) 33 (66%) 41 (73%) 37 (54%) 63 (67%)
53 (26%) 16 (32%) 15 (26%) 29 (43%) 27 (28%)
*Reference group. For differences from controls (2 test): †p<0·01, ‡p<0·0001, §p<0·025. For significance of odds ratios ¶[<0·0001.
Table 1: Distribution of mEPHX genotypes in control and disease groups
slow allele was three times more common than the fast variant—13 (6%) of 203 individuals were homozygous for the exon-3 mutation (slow) and 99 (49%) were heterozygous. In comparison, the exon-4 polymorphism (fast) was detected in only 56 (28%) individuals, and only three individuals were homozygous for this variant (table 1). The distributions of mEPHX genotypes and predicted phenotypes in the lung-cancer and asthma cohorts did not differ significantly from those in the control group (tables 1 and 2). By contrast, the proportion of individuals homozyous for the exon-3 (slow) mutation was significantly higher in the COPD group than in the control group (13 [19%] vs 13 [6%], p<0·01; table 1). There was also a significant difference in the distribution of exon-4 (fast) mutant alleles between this group and the controls; 29 (43%) COPD patients were heterozygous for the fast allele compared with 53 (26%) controls (p<0·025). Only two of the 68 COPD patients were homozygous for the exon-4 mutation. In the analysis of putative phenotype (table 2), the COPD group had a significantly higher proportion of the predicted very slow mEPHX activity than the control group (13 [19%] vs 11 [5%]) and a slightly higher proportion with the fast phenotype (16 [24%] vs 35 [17%]). Consequently, the proportion of individuals with normal mEPHX activity was lower in the COPD group than in the control group (10 [15%] vs 76 [37%], p<0·001). There were significant differences between the emphysema group and the control group in the proportions with homozygous slow (26 [28%] vs 13 [6%]) and homozygous wild-type genotypes (23 [24%] vs 91 [45%]). The difference in distribution was highly significant (p<0·001, odds ratio 5·6 [95% CI 2·7–11·5]). Number of individuals (% of group) with phenotype*
Controls (n=203) Lung cancer (n=50) Asthma (n=57) COPD (n=68) Emphysema (n=94)
Slow
Very slow
Odds ratio (95% CI) for risk of very slow vs all other phenotypes
Normal
Fast
76 (37%) 17 (34%)
35 (17%) 81 (40%) 11 (5%) 9 (18%) 19 (38%) 5 (10%)
18 (32%) 10 (15%) 24 (25%)
7 (12%) 28 (43%) 4 (7%) 1·3 (0·4–4·3) 16 (24%) 29 (43%) 13 (19%)† 4·1 (1·8–9·7) 13 (14%) 36 (38%) 21 (22%)† 5·0 (2·3–10·9)
1·0 1·9 (0·6–5·9)
*Normal=no mutation in gene, or heterozygote 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. †p<0·001 for very slow vs all other phenotypes (2 test).
Table 2: Distribution of putative mEPHX phenotypes in control and disease groups
632
The exon-3 variant accounted for more than 50% of all alleles in the emphysema group compared with only 31% of the control group (p<0·0001; odds ratio 2·4). By contrast, although the odds ratio for fast genotype in the emphysema group was 2·8, the distribution of genotypes with respect to this polymorphism did not differ significantly from the control distribution (table 1). As in the COPD group, the distribution of implied phenotypes in the emphysema group varied significantly from that in the control population (for very slow phenotype p<0·001; odds ratio 5·0). Of the 94 emphysema samples, 32 were assessed as centriacinar type, 18 panacinar, and 32 a mixture of both types of emphysema; a further 12 were classified as mixed centriacinar, paraseptal, or bullous type. Overall, these subsets showed no difference in the distributions of mEPHX genotype (data not shown), and the proportion of very slow phenotype was similar (23%) to that in the whole emphysema group (22%). Comparisons of the observed distributions of mEPHX genotypes and those predicted by allele frequencies by 2 analysis showed that the populations studied were in Hardy-Weinberg equilibrium, indicating that the control and study groups were sufficiently random and representative (data not shown).
Discussion Our demonstration of an association between genetically defined polymorphisms in mEPHX activity and susceptibility to COPD and emphysema suggests that highly reactive epoxide intermediates may have a role in the initiation and progression of the characteristic tissue abnormalities seen in emphysema. The very slow phenotype was four to five times more common in both the COPD and emphysema groups than in controls. Also, the proportions of the emphysema group with normal and fast mEPHX phenotypes were significantly lower than those of the control group (table 2). Individuals with a genetically determined slow mEPHX activity may be more susceptible to COPD and emphysema than those who inherit alleles for normal enzyme activity. Interestingly, the COPD group also showed an excess of fast mEPHX alleles, not seen in the emphysema group, due mainly to higher proportion of heterozygotes (table 1); there was, however, no difference between this group and the controls in the proportion of individuals homozygous for the fast allele or the proportion with the fast phenotype (table 2). The importance of this finding in a diverse disease group such as COPD requires further study. The associated risk of slow and fast hydrolase activity was the
Vol 350 • August 30, 1997
THE LANCET
same for all types of emphysema, and no association was seen between mEPHX genotype and lung cancer or asthma. We believe, therefore, that polymorphism of the mEPHX gene may be an important risk factor in lung disease associated with oxidative stress consistent with the direct effects of cigarette-smoke components. In comparison, individuals who carry debilitating mutations (null) that lead to genetically determined deficiency of ␣1-antitrypsin (only about 2% of patients) are at a 40-fold increased risk of developing COPD and emphysema.18,19 This susceptibility is chiefly manifest in patients with early age of onset of disease and is associated generally with panacinar emphysema at the base of the lung. Tobacco smoking is the main factor involved in these types of lung disease. Cigarette smoke contains about 1015–1017 free radicals per puff and an array of other highly toxic electrophiles.9,20 Many of these compounds are, or are capable of generating, reactive epoxides by direct combustion or metabolic conversion. Generally, the disposition of toxic compounds is governed by metabolism in the liver, but the amount and site of exposure and the rate of metabolism all have a direct bearing on activity.21 mEPHX catalyses the detoxification (hydration) of many reactive epoxides, such as cytotoxic and genotoxic derivatives of naphthalene,22 benzo-a-pyrene,23 and hepatotoxic aflatoxin B1.24 The enzyme is strongly expressed in bronchial epithelial cells.13 Many similar compounds are found in cigarette smoke, and highly reactive intermediates that are not deactivated can interact with DNA, RNA, protein, and lipids and elicit serious cytotoxic and genotoxic effects.20 Since reactive xenobiotics directly inhibit antiproteases and increase protease secretion from neutrophils, a combination of oxidative attack and changes in the activity of antiproteases could amplify the characteristic tissue disruption seen in emphysema. We have previously shown that the polymorphisms in the cytochrome P450 CYP1A1 gene and the GST M1 enzyme are weakly associated with susceptibility to emphysema.15,17 Taken together with the findings of this study, these data suggest that first-pass metabolism and systemic metabolism of cigarette-smoke components significantly influences the progression of diseases associated with oxidative stress and chemical exposure.
References 1
2 3 4
5
6 7
8
9 10 11
12 13
14
15
16
17
18
19
20
Contributors
21
Christopher Smith developed the molecular genetic methods for the study, carried out or supervised all the molecular biological parts of the research, and was responsible for the data analysis and manuscript preparation. David Harrison was responsible for clinical matters, including the review of disease pathology and diagnosis, and for analysis of results and manuscript preparation.
22
23
Acknowledgments This study was supported by the Norman Salvesen Emphysema Research Trust. We thank David Lamb (Department of Pathology), the respiratory physicians at the Western General Hospital and the Royal Infirmary of Edinburgh for samples, and Frances Rae for technical assistance.
Vol 350 • August 30, 1997
24
Carp H, Miller F, Hoidal JR, Janoff A. Potential mechanism of emphysema: alpha 1-proteinase inhibitor recovered from lungs of cigarette smokers contains oxidised methionine and has decreased elastase inhibitory capacity. Proc Natl Acad Sci USA 1982; 79: 2041–45. Aaron J. Elastases and emphysema: current assessment of the proteaseantiprotease hypothesis. Am Rev Respir Dis 1985; 132: 417–33. Tetley TD. Proteinase imbalance: its role in lung disease. Thorax 1993; 48: 560–65. Garver RI, Mornex JF, Nukiwa T, et al. Alpha 1-antitrypsin deficiency and emphysema caused by homozygous inheritance of non-expressing alpha 1-antitrypsin gene. N Engl J Med 1986; 314: 762–66. Cox DW, Levison H. Emphysema of early onset associated with a complete deficiency of alpha 1-antitrypsin (null homozygotes). Am Rev Respir Dis 1988; 137: 371–75. Martorana PA, Brand T, Gardi C, et al. The pallid mouse: a model of genetic alpha1-antitrypsin deficiency. Lab Invest 1993; 68: 233–41. Foster JA, Rich CB, Miller MF. Pulmonary fibroblasts: an in vitro model of emphysema—regulation of the elastin gene expression. J Biol Chem 1990; 265: 15544–49. D’Armiento J, Dalal SS, Okada Y, Berg RA, Dhada K. Collagenase expression in the lungs of transgenic mice causes pulmonary emphysema. Cell 1992; 71: 955–61. Church DF, Pryor WA. Free radical chemistry of cigarette smoke and its toxicoligical implications. Environ Health Perspect 1985; 64: 111–26. Farber JL. Mechanisms of cell injury by activated oxygen species. Environ Health Perspect 1994; 102: 17–24. Ogushi F, Hubbard RC, Vogelmeier C, Fells GA, Crystal RG. Risk factors for emphysema: cigarette smoking is associated with a reduction in the association rate constant of lung alpha 1-antitrypsin for neutrophil elastase. J Clin Invest 1991; 87: 1060–65. Evans MD, Pryor WA. Cigarette smoking, emphysema and damage to ␣1-proteinase inhibitor. Am J Physiol 1994; 266: L593–611. Omiecinski CJ, Aicher L, Holubkov R, Checkoway H. Human peripheral lymphocytes as indicators of microsomal epoxide hydrolase activity in liver and lung. Pharmacogenetics 1993; 3: 150–58. Hassett C, Aicher L, Sidhu LA, Omiecinski CJ. Human microsomal epoxide hydrolase: genetic polymorphism and functional expression in vitro of amino acid variants. Hum Mol Genet 1994; 3: 421–28. Cantlay AM, Smith CAD, Wallace WA, Yap P-L, Lamb D, Harrison DJ. Heterogeneous and polymorphic expression of glutathione S-transferases in human lung. Thorax 1994; 49: 1010–14. Smith CAD, Gough AC, Leigh PN, et al. Debrisoquine hydroxylase gene polymorphism and susceptibility to Parkinson’s disease. Lancet 1992; 339: 1375–77. Cantlay AM, Lamb D, Gillooly M, et al. Association between the CYP1A1 gene polymorphism and susceptibility to emphysema and lung cancer. J Clin Mol Pathol 1995; 48: M210–14. Kalsheker NA, Hodgson IJ, Watkins GL, White JP, Morrison HM, Stockley RA. Deoxyribonucleic acid (DNA) polymorphism of the alpha 1-antitrypsin gene in chronic lung disease. BMJ 1987; 294: 1511–14. Morgan K, Scobie G, Kalsheker NA. The characterisation of a mutation of 3' flanking sequence of the ␣1-antitrypsin gene commonly associated with chronic obstuctive airways disease. Eur J Clin Invest 1992; 22: 134–37. Hoffman D, Hecht SS. Advances in tobacco carcinogenesis. Handbook of Exp Pharmacol 1990; 94: 63–102. Smith CAD, Smith G, Wolf CR. Genetic predisposition to cancer: polymorphisms in xenobiotic metabolising enzymes. Eur J Cancer 1994; 30A: 1921–36. Tingle MD, Pirmohamed M, Templeton E, et al. An investigation of the formation of cytotoxic, genotoxic, protein-reactive and stable metabolites from naphthalene by human liver microsomes. Biochem Pharm 1993; 46: 1529–38. Bartsch H, Petruzzelli S, De Flora S, et al. Carcinogen metabolism in human lung tissues and the effect of tobacco smoking: results from a case-control multicenter study on lung cancer patients. Environ Health Perspect 1992; 98: 119–24. McGlynn KA, Rosvold EA, Lustbader EA, et al. Susceptibility to hepatocellular carcinoma is associated with genetic variation in the enzymatic detoxification of Aflatoxin B1. Proc Natl Acad Sci 1995; 92: 2384–87.
633
Early report
Association between polymorphism in gene for microsomal epoxide hydrolase and susceptibility to emphysema Christopher A D Smith, David J Harrison
Summary Background The first-pass metabolism of foreign compounds in the lung is an important protective mechanism against oxidative stress. We investigated whether polymorphisms in the gene for microsomal epoxide hydrolase (mEPHX), an enzyme involved in this protective process, had any bearing on individual susceptibility to the development of chronic obstructive pulmonary disease (COPD) and emphysema. Methods We designed PCR-based genotyping assays to detect variant forms of mEPHX that confer slow and fast activity. We used these assays to screen 203 blood-donor controls and groups of patients with asthma (n=57), lung cancer (n=50), COPD (n=68), and emphysema (n=94), who were attending specialised clinics in Edinburgh, UK. Findings The proportion of individuals with innate slow mEPHX activity (homozygotes) was significantly higher in both the COPD group and the emphysema group than in the control group (COPD 13 [19%] vs control 13 [6%]; emphysema 21 [22%] vs 13 [6%]). The odds ratios for homozygous slow activity versus all other phenotypes were 4·1 (95% CI 1·8–9·7) for COPD and 5·0 (2·3–10·9) for emphysema. Interpretation Genetic polymorphisms in xenobiotic enzymes may have a role in individual susceptibility to oxidant-related lung disease. Epoxide derivatives of cigarette-smoke components may be the cause of some of the lung damage characteristic of these diseases.
Lancet 1997; 350: 630–33
Introduction The lungs are subject to oxidative stress from cigarette smoke, occupational exposure to solvents, and other chemicals and environmental pollutants. All these potential hazards contain factors (xenobiotics and oxidants) that can induce severe macromolecular, cellular, and tissue damage through direct cytotoxic effects, promotion of primary genotoxic events, or generation of reactive oxygen intermediates. Chronic obstructive pulmonary disease (COPD) is characterised by fixed and irreversible air-flow limitation in the lungs, and emphysema is distinguished by abnormal, permanent distal air-space enlargement Department of Pathology, University of Edinburgh, Edinburgh, UK (C A D Smith PhD, D J Harrison MD) Correspondence to: Dr Christopher A D Smith, Genovar Diagnostics, Room LF6, South Academic Block, Southampton General Hospital, Southampton SO16 6YD, UK
630
accompanied by destruction of lung parenchyma. Two lines of research are relevant to the study of emphysema in particular. First, some findings suggest that lung damage can be attributed to an imbalance in the endogenous protease/antiprotease equilibrium, promoting tissue hydrolysis;1–3 examples of such imbalance include the genetically determined ␣1-antitrypsin deficiency,3,4 and the overexpression of elastases and collagenases that mimics the development of pulmonary emphysema.7,8 Second, the oxidant/antioxidant theory postulates that an excess of oxidants and free radicals in the lung promotes cellular and tissue damage and is the major initiator of the disease process.9,10 Cigarette smoke, oxidants, xenobiotics, and reactive oxygen intermediates directly inhibit antiproteases and promote cell and tissue proteolysis.1,11,12 Tobacco smoke is the commonest identifiable risk factor for both emphysema and COPD. Combustion products released into the respiratory tract include various highly reactive oxygen and carbon-centred species, free radicals in the vapour phase, and stable radicals in particulate matter. Microsomal epoxide hydrolase (mEPHX) is an enzyme involved in the first-pass metabolism of highly reactive epoxide intermediates. We investigated whether genetic polymorphisms in the gene for this enzyme have any bearing on the development of oxidant-related lung disease. Epidemological studies show that mEPHX activity in the liver, lung, and peripheral blood leucocytes varies as much as 50-fold in white populations.13 Two common aberrant alleles can be detected, which confer slow and fast enzyme activity.14 We have designed PCR assays to detect these mutations and have used them to study the role of the gene in protection against oxidative damage to the lung.
Patients and methods 203 control blood samples were obtained from the Scottish National Blood Transfusion Service. These anonymous donors were all white individuals aged between 18 and 65 years; roughly equal numbers were male and female. Because emphysema is a morphological diagnosis, we cannot be certain that some controls did not have early emphysema; however, all were clinically healthy on routine questioning and therefore met the usual criteria for blood donation. All patients were fully investigated clinically and were attending specialised clinics. We obtained blood samples from 57 patients with chronic asthma attending a clinic at the Western General Hospital, Edinburgh, and a cohort of 68 patients with non-reversible COPD associated with cigarette smoking, who were attending a respiratory failure clinic at the Royal Infirmary of Edinburgh. All the COPD patients (average age 63 years) had smoked (mean pack years 35 [range 13–96]), and all had forced expiratory volume in 1 s of less than 1·5 L (mean 1·05 [0·7–1·4]). Patients with ␣1-antitrypsin deficiency were excluded on the results of isoelectric focusing of serum for variant phenotypes.
Vol 350 • August 30, 1997
THE LANCET
PCR-based genotyping assays for mEPHX polymorphism Top panels show arrangement of exons 3 and 4. Middle panels show fragment patterns expected in individuals homozygous or heterozygous for mEPHX polymorphisms. Lower panels show typical electrophoresis gels with band visualisation; only larger diagnostic bands shown.
144 consecutive lung-cancer resection samples were obtained from the archives of the Department of Pathology, University of Edinburgh. All the patients were smokers, but precise data on smoking duration were not available. Each resection sample was macroscopically and microscopically assessed for presence of tumour and extent and type of emphysema.15 All emphysema patients had mild to moderate disease; those with severely compromised lung function would not have been candidates for surgical lung resection. In total we identified 94 samples from patients with emphysema and lung cancer (age range 34–82 years; 25 women, 69 men) and 50 from patients with lung cancer and no evidence of emphysema. All these patients were white. DNA was obtained16,17 from sections of archival tissue samples without neoplastic changes, from fresh peripheral blood leucocytes, or from both types of sample. The study design was approved by the Lothian Health Board Ethical Review Committee, and informed consent was obtained from the patients attending the respiratory clinics. There are four distinct mEPHX alleles, which arise because of the presence or absence of two point mutations in the gene. An exon-3 T to C mutation changes tyrosine residue 113 to histidine, and enzyme activity is reduced by at least 50% (slow allele). The second, an A to G transition mutation in exon 4 of the gene, changes histidine residue 139 to arginine, and produces an enzyme with an activity increased by at least 25% (fast allele).14 The wild-type allele has neither of these changes. The rare occurrence of both mutations together also produces a hydrolase with normal activity. Two separate PCR assays are used to detect the two mutations. The assay for the exon-3 variant uses the primer pair EPO1 (5'-GATCGATAAGTTCCGTTTCACC, starting at bp 321 in mEPHX cDNA21 and EPO2 (5'-ATCCTTAGTCTTGAAGTGAGGAT, starting at bp 461). The downstream primer abuts directly onto the mutation site, and an
Vol 350 • August 30, 1997
engineered base change (shown in bold type; A for G) produces an EcoR V restriction enzyme site (GATATC) in the wild-type only. The exon-4 mutation produces an Rsa I restriction-fragment length polymorphism in the mutant (fast) allele (ATAC to GTAC); a second primer pair EPO3 (5'-ACATCCACTTCATCCACGT, bp 494) and EPO4 (5'-ATGCCTCTGAGAAGCCAT, bp 685) is used to test for this mutation. The figure shows diagrammatically the PCR reactions and typical results of the genotyping assays. PCRs were carried out with a Omnigene (Hybaid) thermocycler in PCR buffer (1·5 mmol/L magnesium chloride, 100 ng primers, 4% dimethyl sulphoxide, 200 mmol/L of each dNTP, and 1 unit of Taq polymerase [Promega]) in a final volume of 50 L. The cycles consisted of steps of 56°C for 20 s, 72°C for 20 s, and 94°C for 30 s (38 cycles). The products were then digested with EcoR V (exon 3) and Rsa I (exon 4), which produce diagnostic fragments of the sizes shown in the figure. The fragments were separated by electrophoresis through a non-denaturing 8% polyacrylamide gel containing 0·09 mol/L Tris-borate and 0·002 mol/L edetic acid. The gels were stained with ethidium bromide and transilluminated with ultraviolet light. Three representative fragments corresponding to each allele were cloned and sequenced to confirm the specificity of the PCR (results not shown). Associations between disease groups and specific genotypes and phenotypes were analysed for significance by the two-tailed 2 test. Probability values were corrected by the Bonferroni method. Odds ratios and 95% CI were calculated to assess relative risk of disease conferred by a particular allele, genotype, and implied phenotype.
Results The PCR produced amplimers of the expected size and differentiated between individual mEPHX genotypes in the study populations (figure). In the control group the 631
THE LANCET
Number of individuals (% of group)
Odds ratio (95% CI) for homozygous mutations
Mutant allele frequency
Odds ratio (95% CI) for carriage of mutant allele
13 (6%) 5 (10%) 4 (7%) 13 (19%)† 26 (28%)‡
1·0* 1·6 (0·6–4·8) 1·1 (0·3–3·5) 3·5 (1·5–8·0) 5·6 (2·7–11·5)
0·31 0·30 0·34 0·40 0·52¶
1·0* 1·0 (0·6–1·6) 1·1 (0·7–1·7) 1·5 (0·7–2·2) 2·4 (1·5–4·0)
3 (2%) 1 (2%) 1 (2%) 2 (3%)§ 4 (4%)
1·0* 1·4 (0·1–13·3) 1·2 (0·1–11·7) 2·0 (0·3–2·4) 2·8 (0·6–12·9)
0·15 0·18 0·15 0·24¶ 0·19
1·0* 1·3 (0·7–2·3) 1·0 (0·6–1·8) 1·9 (1·2–3·0) 1·3 (0·8–2·1)
Homozygous wild-type Heterozygous
Homozygous mutant
Exon-3 polymorphism (slow) Controls (n=203) Lung cancer (n=50) Asthma (n=57) COPD (n=68) Emphysema (n=94)
91 (45%) 25 (50%) 24 (42%) 27 (40%) 23 (24%)
99 (49%) 20 (40%) 29 (51%) 28 (41%) 45 (48%)
Exon-4 polymorphism (fast) Controls (n=203) Lung cancer (n=50) Asthma (n=57) COPD (n=68) Emphysema (n=94)
147 (72%) 33 (66%) 41 (73%) 37 (54%) 63 (67%)
53 (26%) 16 (32%) 15 (26%) 29 (43%) 27 (28%)
*Reference group. For differences from controls (2 test): †p<0·01, ‡p<0·0001, §p<0·025. For significance of odds ratios ¶[<0·0001.
Table 1: Distribution of mEPHX genotypes in control and disease groups
slow allele was three times more common than the fast variant—13 (6%) of 203 individuals were homozygous for the exon-3 mutation (slow) and 99 (49%) were heterozygous. In comparison, the exon-4 polymorphism (fast) was detected in only 56 (28%) individuals, and only three individuals were homozygous for this variant (table 1). The distributions of mEPHX genotypes and predicted phenotypes in the lung-cancer and asthma cohorts did not differ significantly from those in the control group (tables 1 and 2). By contrast, the proportion of individuals homozyous for the exon-3 (slow) mutation was significantly higher in the COPD group than in the control group (13 [19%] vs 13 [6%], p<0·01; table 1). There was also a significant difference in the distribution of exon-4 (fast) mutant alleles between this group and the controls; 29 (43%) COPD patients were heterozygous for the fast allele compared with 53 (26%) controls (p<0·025). Only two of the 68 COPD patients were homozygous for the exon-4 mutation. In the analysis of putative phenotype (table 2), the COPD group had a significantly higher proportion of the predicted very slow mEPHX activity than the control group (13 [19%] vs 11 [5%]) and a slightly higher proportion with the fast phenotype (16 [24%] vs 35 [17%]). Consequently, the proportion of individuals with normal mEPHX activity was lower in the COPD group than in the control group (10 [15%] vs 76 [37%], p<0·001). There were significant differences between the emphysema group and the control group in the proportions with homozygous slow (26 [28%] vs 13 [6%]) and homozygous wild-type genotypes (23 [24%] vs 91 [45%]). The difference in distribution was highly significant (p<0·001, odds ratio 5·6 [95% CI 2·7–11·5]). Number of individuals (% of group) with phenotype*
Controls (n=203) Lung cancer (n=50) Asthma (n=57) COPD (n=68) Emphysema (n=94)
Slow
Very slow
Odds ratio (95% CI) for risk of very slow vs all other phenotypes
Normal
Fast
76 (37%) 17 (34%)
35 (17%) 81 (40%) 11 (5%) 9 (18%) 19 (38%) 5 (10%)
18 (32%) 10 (15%) 24 (25%)
7 (12%) 28 (43%) 4 (7%) 1·3 (0·4–4·3) 16 (24%) 29 (43%) 13 (19%)† 4·1 (1·8–9·7) 13 (14%) 36 (38%) 21 (22%)† 5·0 (2·3–10·9)
1·0 1·9 (0·6–5·9)
*Normal=no mutation in gene, or heterozygote 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. †p<0·001 for very slow vs all other phenotypes (2 test).
Table 2: Distribution of putative mEPHX phenotypes in control and disease groups
632
The exon-3 variant accounted for more than 50% of all alleles in the emphysema group compared with only 31% of the control group (p<0·0001; odds ratio 2·4). By contrast, although the odds ratio for fast genotype in the emphysema group was 2·8, the distribution of genotypes with respect to this polymorphism did not differ significantly from the control distribution (table 1). As in the COPD group, the distribution of implied phenotypes in the emphysema group varied significantly from that in the control population (for very slow phenotype p<0·001; odds ratio 5·0). Of the 94 emphysema samples, 32 were assessed as centriacinar type, 18 panacinar, and 32 a mixture of both types of emphysema; a further 12 were classified as mixed centriacinar, paraseptal, or bullous type. Overall, these subsets showed no difference in the distributions of mEPHX genotype (data not shown), and the proportion of very slow phenotype was similar (23%) to that in the whole emphysema group (22%). Comparisons of the observed distributions of mEPHX genotypes and those predicted by allele frequencies by 2 analysis showed that the populations studied were in Hardy-Weinberg equilibrium, indicating that the control and study groups were sufficiently random and representative (data not shown).
Discussion Our demonstration of an association between genetically defined polymorphisms in mEPHX activity and susceptibility to COPD and emphysema suggests that highly reactive epoxide intermediates may have a role in the initiation and progression of the characteristic tissue abnormalities seen in emphysema. The very slow phenotype was four to five times more common in both the COPD and emphysema groups than in controls. Also, the proportions of the emphysema group with normal and fast mEPHX phenotypes were significantly lower than those of the control group (table 2). Individuals with a genetically determined slow mEPHX activity may be more susceptible to COPD and emphysema than those who inherit alleles for normal enzyme activity. Interestingly, the COPD group also showed an excess of fast mEPHX alleles, not seen in the emphysema group, due mainly to higher proportion of heterozygotes (table 1); there was, however, no difference between this group and the controls in the proportion of individuals homozygous for the fast allele or the proportion with the fast phenotype (table 2). The importance of this finding in a diverse disease group such as COPD requires further study. The associated risk of slow and fast hydrolase activity was the
Vol 350 • August 30, 1997
THE LANCET
same for all types of emphysema, and no association was seen between mEPHX genotype and lung cancer or asthma. We believe, therefore, that polymorphism of the mEPHX gene may be an important risk factor in lung disease associated with oxidative stress consistent with the direct effects of cigarette-smoke components. In comparison, individuals who carry debilitating mutations (null) that lead to genetically determined deficiency of ␣1-antitrypsin (only about 2% of patients) are at a 40-fold increased risk of developing COPD and emphysema.18,19 This susceptibility is chiefly manifest in patients with early age of onset of disease and is associated generally with panacinar emphysema at the base of the lung. Tobacco smoking is the main factor involved in these types of lung disease. Cigarette smoke contains about 1015–1017 free radicals per puff and an array of other highly toxic electrophiles.9,20 Many of these compounds are, or are capable of generating, reactive epoxides by direct combustion or metabolic conversion. Generally, the disposition of toxic compounds is governed by metabolism in the liver, but the amount and site of exposure and the rate of metabolism all have a direct bearing on activity.21 mEPHX catalyses the detoxification (hydration) of many reactive epoxides, such as cytotoxic and genotoxic derivatives of naphthalene,22 benzo-a-pyrene,23 and hepatotoxic aflatoxin B1.24 The enzyme is strongly expressed in bronchial epithelial cells.13 Many similar compounds are found in cigarette smoke, and highly reactive intermediates that are not deactivated can interact with DNA, RNA, protein, and lipids and elicit serious cytotoxic and genotoxic effects.20 Since reactive xenobiotics directly inhibit antiproteases and increase protease secretion from neutrophils, a combination of oxidative attack and changes in the activity of antiproteases could amplify the characteristic tissue disruption seen in emphysema. We have previously shown that the polymorphisms in the cytochrome P450 CYP1A1 gene and the GST M1 enzyme are weakly associated with susceptibility to emphysema.15,17 Taken together with the findings of this study, these data suggest that first-pass metabolism and systemic metabolism of cigarette-smoke components significantly influences the progression of diseases associated with oxidative stress and chemical exposure.
References 1
2 3 4
5
6 7
8
9 10 11
12 13
14
15
16
17
18
19
20
Contributors
21
Christopher Smith developed the molecular genetic methods for the study, carried out or supervised all the molecular biological parts of the research, and was responsible for the data analysis and manuscript preparation. David Harrison was responsible for clinical matters, including the review of disease pathology and diagnosis, and for analysis of results and manuscript preparation.
22
23
Acknowledgments This study was supported by the Norman Salvesen Emphysema Research Trust. We thank David Lamb (Department of Pathology), the respiratory physicians at the Western General Hospital and the Royal Infirmary of Edinburgh for samples, and Frances Rae for technical assistance.
Vol 350 • August 30, 1997
24
Carp H, Miller F, Hoidal JR, Janoff A. Potential mechanism of emphysema: alpha 1-proteinase inhibitor recovered from lungs of cigarette smokers contains oxidised methionine and has decreased elastase inhibitory capacity. Proc Natl Acad Sci USA 1982; 79: 2041–45. Aaron J. Elastases and emphysema: current assessment of the proteaseantiprotease hypothesis. Am Rev Respir Dis 1985; 132: 417–33. Tetley TD. Proteinase imbalance: its role in lung disease. Thorax 1993; 48: 560–65. Garver RI, Mornex JF, Nukiwa T, et al. Alpha 1-antitrypsin deficiency and emphysema caused by homozygous inheritance of non-expressing alpha 1-antitrypsin gene. N Engl J Med 1986; 314: 762–66. Cox DW, Levison H. Emphysema of early onset associated with a complete deficiency of alpha 1-antitrypsin (null homozygotes). Am Rev Respir Dis 1988; 137: 371–75. Martorana PA, Brand T, Gardi C, et al. The pallid mouse: a model of genetic alpha1-antitrypsin deficiency. Lab Invest 1993; 68: 233–41. Foster JA, Rich CB, Miller MF. Pulmonary fibroblasts: an in vitro model of emphysema—regulation of the elastin gene expression. J Biol Chem 1990; 265: 15544–49. D’Armiento J, Dalal SS, Okada Y, Berg RA, Dhada K. Collagenase expression in the lungs of transgenic mice causes pulmonary emphysema. Cell 1992; 71: 955–61. Church DF, Pryor WA. Free radical chemistry of cigarette smoke and its toxicoligical implications. Environ Health Perspect 1985; 64: 111–26. Farber JL. Mechanisms of cell injury by activated oxygen species. Environ Health Perspect 1994; 102: 17–24. Ogushi F, Hubbard RC, Vogelmeier C, Fells GA, Crystal RG. Risk factors for emphysema: cigarette smoking is associated with a reduction in the association rate constant of lung alpha 1-antitrypsin for neutrophil elastase. J Clin Invest 1991; 87: 1060–65. Evans MD, Pryor WA. Cigarette smoking, emphysema and damage to ␣1-proteinase inhibitor. Am J Physiol 1994; 266: L593–611. Omiecinski CJ, Aicher L, Holubkov R, Checkoway H. Human peripheral lymphocytes as indicators of microsomal epoxide hydrolase activity in liver and lung. Pharmacogenetics 1993; 3: 150–58. Hassett C, Aicher L, Sidhu LA, Omiecinski CJ. Human microsomal epoxide hydrolase: genetic polymorphism and functional expression in vitro of amino acid variants. Hum Mol Genet 1994; 3: 421–28. Cantlay AM, Smith CAD, Wallace WA, Yap P-L, Lamb D, Harrison DJ. Heterogeneous and polymorphic expression of glutathione S-transferases in human lung. Thorax 1994; 49: 1010–14. Smith CAD, Gough AC, Leigh PN, et al. Debrisoquine hydroxylase gene polymorphism and susceptibility to Parkinson’s disease. Lancet 1992; 339: 1375–77. Cantlay AM, Lamb D, Gillooly M, et al. Association between the CYP1A1 gene polymorphism and susceptibility to emphysema and lung cancer. J Clin Mol Pathol 1995; 48: M210–14. Kalsheker NA, Hodgson IJ, Watkins GL, White JP, Morrison HM, Stockley RA. Deoxyribonucleic acid (DNA) polymorphism of the alpha 1-antitrypsin gene in chronic lung disease. BMJ 1987; 294: 1511–14. Morgan K, Scobie G, Kalsheker NA. The characterisation of a mutation of 3' flanking sequence of the ␣1-antitrypsin gene commonly associated with chronic obstuctive airways disease. Eur J Clin Invest 1992; 22: 134–37. Hoffman D, Hecht SS. Advances in tobacco carcinogenesis. Handbook of Exp Pharmacol 1990; 94: 63–102. Smith CAD, Smith G, Wolf CR. Genetic predisposition to cancer: polymorphisms in xenobiotic metabolising enzymes. Eur J Cancer 1994; 30A: 1921–36. Tingle MD, Pirmohamed M, Templeton E, et al. An investigation of the formation of cytotoxic, genotoxic, protein-reactive and stable metabolites from naphthalene by human liver microsomes. Biochem Pharm 1993; 46: 1529–38. Bartsch H, Petruzzelli S, De Flora S, et al. Carcinogen metabolism in human lung tissues and the effect of tobacco smoking: results from a case-control multicenter study on lung cancer patients. Environ Health Perspect 1992; 98: 119–24. McGlynn KA, Rosvold EA, Lustbader EA, et al. Susceptibility to hepatocellular carcinoma is associated with genetic variation in the enzymatic detoxification of Aflatoxin B1. Proc Natl Acad Sci 1995; 92: 2384–87.
633