No Replication of Association Between Estrogen Receptor Gene Polymorphisms and Susceptibility to Myocardial Infarction in a Large Sample of
http://www.100md.com
《循环学杂志》
the Deutsches Herzzentrum München and 1. Medizinische Klinik, Klinikum rechts der Isar (W.K., P.H., A.S., A.K.)
Institut für Humangenetik, Klinikum rechts der Isar, Technische Universitt München, Munich (A.P.)
Institut für Humangenetik, GSF Forschungszentrum, Neuherberg (A.P.)
Institut für medizinische Statistik und Epidemiologie, Klinikum rechts der Isar, Technische Universitt München, Munich (J.C.M.), Germany.
Abstract
Background— The effects of estrogen on blood vessels are partly due to changes in vascular cell gene expression and protein synthesis that are mediated by estrogen receptors. In previous association studies, the –397T/C (rs2234693) and –351A/G (rs9340799) single nucleotide polymorphisms in the estrogen receptor gene (ESR1) have been implicated in the risk of coronary atherosclerosis and myocardial infarction. To test these findings, we examined the relationship of the polymorphisms to myocardial infarction in a large sample of white patients and control individuals of predominantly European descent.
Methods and Results— The case group included 3657 patients with myocardial infarction, and the control group comprised 1211 individuals with angiographically normal coronary arteries and without signs or symptoms of myocardial infarction. TaqMan assays were used for the determination of genotypes. Genotype distributions of the –397T/C and –351A/G polymorphisms were not significantly different between the control and patient groups (P0.85). The frequencies of haplotypes defined by the –397T/C and –351A/G polymorphisms were similar in the control group and the patient group (P=0.42). In addition, the distributions of haplotype-defined genotypes (diplotypes) were not significantly different between the control group and the patient group (P=0.81). Separate analyses in women and men did not reveal sex-related associations of specific genotypes or haplotypes of the polymorphisms with myocardial infarction (P0.25).
Conclusions— The results indicate that the –397T/C and –351A/G polymorphisms of ESR1 or haplotypes based on these polymorphisms are not associated with myocardial infarction in a white population.
Key Words: estrogen receptor alpha genetics myocardial infarction polymorphism receptors, estrogen
Introduction
Estrogen receptor is a ligand-activated transcription factor involved in the regulation of cellular pathways that are of potential importance in vascular physiology and disorders.1 Human vascular endothelial and smooth muscle cells contain estrogen receptor protein, and experiments with transgenic mice showed that it is able to cause stimulation of endothelial nitric oxide production, acceleration of reendothelialization, and inhibition of the vascular injury response.1–4 A cardioprotective function of estrogen receptor in humans was indicated by the finding that a disruptive mutation in both alleles of the estrogen receptor gene (ESR1; gene map locus 6q25.1) was associated with endothelial dysfunction and coronary artery disease in a 31-year-old man.5,6
Editorial p 2081
To examine the potential role of estrogen receptor in coronary artery disease and myocardial infarction (MI), common sequence variations in ESR1, especially the –397T/C (rs2234693) and –351A/G (rs9340799) single nucleotide polymorphisms (SNPs), have been employed in both prospective and retrospective association studies.7–11 Although several previous studies have found associations of the –397T/C and –351A/G SNPs with cardiovascular disease phenotypes,8–11 a comparison of their results reveals inconsistencies, including the identity of the risk genotype or haplotype. The diversity of findings between the prior studies may result, at least in part, from the low prevalence of events observed in the prospective trials10,11 and the relatively small number of cases enrolled in the retrospective examinations.7–9 To avoid these potential limitations, we investigated the association between the –397T/C and –351A/G SNPs of ESR1 and MI in a case-control study that included a large number of participants.
Methods
Patients and Controls
The study population comprised a group of 3657 white patients with MI and a group of 1211 white control individuals who were recruited consecutively from 1993 to 2002. All patients and control persons were examined with coronary angiography at Deutsches Herzzentrum München or 1 Medizinische Klinik rechts der Isar der Technischen Universitt München. Written informed consent was obtained from all study participants. The study protocol was approved by the institutional ethics committee, and the reported investigations were in accordance with the principles of the current version of the Declaration of Helsinki.12
Definitions
Individuals were considered disease free and therefore eligible as controls when their coronary arteries were angiographically normal and when they had no history of MI, no symptoms suggestive of MI, no electrocardiographic signs of MI, and no regional wall motion abnormalities. Coronary angiography in the control individuals was performed for the evaluation of chest pain. The diagnosis of MI was established in the presence of chest pain lasting >20 minutes combined with ST-segment elevation or pathological Q waves on a surface ECG. Patients with MI had to show either an angiographically occluded infarct-related artery or regional wall motion abnormalities corresponding to the electrocardiographic infarct localization or both. Systemic arterial hypertension was defined as a systolic blood pressure of 140 mm Hg and/or a diastolic blood pressure of 90 mm Hg,13 on at least 2 separate occasions, or antihypertensive treatment. Hypercholesterolemia was defined as a documented total cholesterol value 240 mg/dL (6.2 mmol/L) or current treatment with cholesterol-lowering medication. Persons reporting regular smoking in the previous 6 months were considered current smokers. Diabetes mellitus was defined as the presence of an active treatment with insulin or an oral antidiabetic agent; for patients on dietary treatment, documentation of an abnormal fasting blood glucose or glucose tolerance test based on the World Health Organization criteria14 was required for establishing this diagnosis.
Determination of ESR1 Genotypes
Genomic DNA was extracted from peripheral blood leukocytes with the QIAamp DNA Blood Kit (Qiagen) or the High Pure PCR Template Preparation Kit (Roche Applied Science). We designed and used TaqMan allelic discrimination assays for genotype analysis.15 The primers 5 TCCATCAGTTCATCTGAGTTCCAA 3 and 5 TTCAGAACCATTAGAGACCAATGCT 3 were used to initiate the amplification of a 117-bp portion of intron 1 of ESR1, which contains the polymorphic sites –397T/C (rs2234693) and –351A/G (rs9340799). The –397T/C and –351A/G SNPs are located 397 and 351 bp, respectively, upstream from the first nucleotide of exon 2 of ESR1. To accomplish allele-specific signaling, the probes contained the fluorogenic dyes 6-carboxy-fluorescein (FAM) or VIC (proprietary dye of Applied Biosystems). The structures of the probes were as follows (allele-specific nucleotides are underlined): FAM-5 TGTCCCAGCTGTTTT 3 (for –397T), VIC-5 CCCAGCCGTTTTA 3 (for –397C), FAM-5 CCCAACTCTAGACCAC 3 (for –351A), and VIC-5 TCCCAACTCCAGACCA 3 (for –351G). Minor groove binder groups were conjugated with the 3 ends of the oligonucleotides to facilitate formation of stable duplexes between the probes and their single-stranded DNA targets.16 TaqMan reactions were performed on 96-well plates, each of which included, at the outset of the thermocycling procedure, 90 assays that contained DNA samples of unknown genotype, 2 assays that contained no DNA, and 2 assays each that contained DNA samples of the homozygous genotypes as standards. As a control, genotyping was repeated for 20% of the samples with the use of DNA prepared separately from the original blood sample.
Restriction enzyme analyses involving PvuII and XbaI are established methods for genotyping of the –397T/C and –351A/G SNPs of ESR1, respectively.7–10 Examinations with these allele-discriminating restriction enzymes (Roche Applied Science) served as measures to verify the accuracy of genotyping with the use of the new TaqMan systems for the –397T/C and –351A/G SNPs. Digestion of the –397T-specific polymerase chain reaction (PCR) product with PvuII resulted in 2 fragments of 84 and 33 bp, and digestion of the –351A-specific PCR product with XbaI resulted in 2 fragments of 78 and 39 bp. PCR products specific for the –397C and –351G alleles were not cleaved with the use of PvuII and XbaI, respectively. In the test, randomly selected samples (n=153) were subjected to the TaqMan assays and, in parallel, separate digestions with PvuII and XbaI. The results obtained with the different methods were fully corresponding, which demonstrated the reliability of the TaqMan systems that were used for genotyping of the –397T/C and –351A/G SNPs in the entire study population. The genotypes of the DNA samples that served as standards in the TaqMan reactions were determined with the use of restriction enzyme analysis and verified with the use of sequence analysis of both DNA strands. Genotype determination was done by workers who had no knowledge of clinical, laboratory, or angiographic data of the individuals enrolled in the study.
Statistical Analysis
The analysis consisted of comparing separately genotype distributions, haplotype frequencies, and haplotype-based diplotype distributions between the control group and the group of patients with MI. Discrete variables are expressed as counts (percentage) and compared with the use of the 2 or Fisher exact test, as appropriate. Continuous variables are expressed as mean±SD and compared by means of the unpaired, two-sided t test. The measures of linkage disequilibrium (D and r2) between the –397T/C and –351A/G SNPs were calculated from primary genotype data with the use of the software package Haploview.17 We tested for the independent association effect of SNP-related genotypes and haplotypes in multiple logistic regression models of MI that included as covariates age, gender, history of arterial hypertension, history of hypercholesterolemia, current cigarette smoking, and diabetes mellitus. Adjusted odds ratios (ORs) and 95% Wald CIs were calculated on the basis of these models. Statistical significance was accepted for probability values <0.05.
Results
The main baseline characteristics and genotype distributions of the –397T/C and –351A/G SNPs in the control group (n=1211) and the group of patients with MI (n=3657) are shown in Table 1. Mean age of the patients was higher than that of the control group, the proportion of women was lower in the patient group than in the control group, and history of arterial hypertension and hypercholesterolemia, current cigarette smoking, and diabetes mellitus were more prevalent in the patient group than in the control group (P<0.0001 for all comparisons; Table 1). The genotype distributions of the SNPs were not significantly different between the control group and the patient group (P0.85; Table 1). Genotype distributions of the SNPs in the control and patient groups were consistent with those expected for samples in Hardy-Weinberg equilibrium (P0.21). The sample sizes of the control group and the patient group provided the analysis with 84% power (–397T/C SNP) and 89% power (–351A/G SNP) to detect a 25% higher risk of MI associated with carriage of the –397C and –351G alleles, respectively (2-sided -level of 0.05).
The possibility existed that a specific combination of the alleles of the –397T/C and –351A/G SNPs was associated with MI. We observed a high degree of linkage disequilibrium between the –397T/C and –351A/G SNPs (D=0.999; 95% CI, 0.98 to 1.00; r2=0.64). Table 2 shows the calculated frequencies of the 4 haplotypes, –397T/–351A (TA), CG, CA, and TG, in the control group and in the patient group. Haplotype phases were assigned unambiguously in the individuals who were homozygous at least at 1 of the 2 polymorphic positions (n=3032). Among these individuals, the TG haplotype was only present on 2 of the total of 6064 chromosomes. In these 2 cases, correct genotyping was verified in separate analyses with the use of the restriction enzymes PvuII and XbaI. Because of the rarity of the TG haplotype among the unambiguously typed individuals, the haplotypes TA and CG were assigned to the 1836 carriers of the double heterozygous genotype combination –397TC/–351AG. The probability is 0.99987 that this inference is correct, indicating that the assignment was incorrect theoretically in <1 of the 1836 double heterozygous individuals. In essence, haplotype frequencies were not substantially different between the control and patient groups (overall P=0.42; Table 2). Table 3 shows the distributions of haplotype-related genotypes (diplotypes) in the study groups. Similar to the haplotype frequencies, the diplotype distributions were not significantly different between the control group and the patient group (overall P=0.81).
We addressed the question of whether sex-dependent relationships existed between the –397T/C and –351A/G SNPs and MI. The genotype distributions of the SNPs were not essentially different between the women and men of the entire study population (P0.41). Comparisons of the genotype distributions between the women in the control group and the patient group and between the men in the 2 groups did not reveal significant differences (P0.41; Table 4). Haplotype frequencies were not substantially different between the women in the control and patient groups (P=0.25) and between the men in the 2 groups (P=0.75). Similarly, diplotype distributions were not associated with MI in women (P=0.66) and men (P=0.94).
The possibility of independent associations of genotypes or haplotypes of the SNPs with MI was tested, after adjustment for other possible risk factors, in multiple logistic regression models of MI that included the baseline clinical characteristics as covariates. We observed that the occurrence of MI was not different between the carriers of the –397TT genotype and the carriers of the –397CC genotype (OR, 1.17; 95% CI, 0.82 to 1.69) and between the carriers of the –351AA genotype and the carriers of the –351GG genotype (OR, 0.84; 95% CI, 0.57 to 1.25). The MI risk of the homozygous carriers of the TA, CG, or CA haplotype was not substantially different from that of the noncarriers of the respective haplotype (TA haplotype: OR, 0.95; 95% CI, 0.77 to 1.17; CG haplotype: OR, 0.99; 95% CI, 0.80 to 1.24; CA haplotype: OR, 1.17; 95% CI, 0.82 to 1.66).
Discussion
The results of the present study show that the –397T/C and –351A/G SNPs of ESR1 are not associated with MI in a large sample of white patients. Lack of association relates to genotypes, haplotypes, and diplotypes and applies to women and men. Recruitment bias may be considered a possible reason for the negative results of this study. Because all control subjects had some indication for coronary angiography, they did not represent a typical sample of healthy controls. However, we consider this group to be especially suitable as a control group in the setting of this study because the absence of MI was rigorously established on the basis of history, electrocardiography, left ventricular angiography, and coronary angiography. The control group used in this study appears not to be fundamentally different from typical control groups because the genotype distributions were in Hardy-Weinberg equilibrium and not significantly different from the genotype distributions observed in another control group that consisted of white individuals.18 Two reasons argue against the possibility of false-negative results due to a systematic genotyping error: (1) Genotype and haplotype frequencies in the control and patient groups were in good correlation with findings of independent investigations that included populations of European origin,8,10,18,19 and (2) repetition of genotyping with the use of restriction enzymes provided reassurance that the TaqMan assays produced correct data.
Associations of ESR1 SNPs with cardiovascular disease were found in prospective studies,10,11 yet outcomes were discordant, and essential findings were not in agreement with our negative results. It was observed in the group of men of one study population (subset of individuals from the offspring cohort of the Framingham Heart Study; n=1739) that recognized MI events (n=54) occurred more often among the carriers of the –397CC genotype than among the carriers of the –397TT or –397TC genotype.10 In another population (subset of individuals from The Rotterdam Study; n=6408), the women homozygous or heterozygous for the TA haplotype encountered MI (115 cases) and ischemic heart disease (168 cases) events more often than the noncarriers of this haplotype.11 No association of the TA haplotype with MI (170 cases) or ischemic heart disease (272 cases) was observed in the group of men.11 The utility of the –397T/C and –351A/G SNPs as cardiovascular disease markers was also tested in retrospective settings, and heterogeneous results were obtained.7–9 A relationship was found between the –397TC and –397CC genotypes and the complexity of atherosclerotic lesions and the presence of thrombosis in coronary arteries in an autopsy study of Finnish men (262 cases).8 In a study that included 119 patients with coronary artery disease and 176 individuals without coronary artery disease from Japan, the –351GG genotype and homozygosity of the CG haplotype, but not the –397T/C SNP on its own, were found to be associated with angina pectoris and MI.9 No association between the –397T/C or –351A/G SNP with the severity of coronary stenosis was observed in another study from Japan that included 87 cases with MI or angina pectoris and 94 control individuals.7
Population stratification is a potential problem for case-control studies and may lead to false-positive results or mask an existing association.21–23 In this study the probability of a false-negative association result due to population stratification is relatively low because the study participants were consecutively recruited from a defined geographic area of southern Germany with limited recent immigration. Family-based studies, which use unaffected family members as control individuals, may help to clarify the importance of the –397T/C and –351A/G SNPs for MI because they are less prone to errors stemming from admixture than population-based studies.
The inability to identify an association in this relatively large and well-defined population may reflect the possibility that the –397T/C and –351A/G SNPs of ESR1 have at best a modest impact on the process leading to MI, a disease phenotype that may be more strongly affected by other genetic factors24–26 and environmental and lifestyle influences. The negative results of this report suggest that polymorphic markers tightly linked to the –397T/C and –351A/G SNPs, including a (TA)n dinucleotide repeat polymorphism located upstream from exon 1 of ESR1,18,27 may not be relevant as indicators of MI risk in this study population.
In conclusion, the results of this study do not provide evidence of an association of specific genotypes or haplotypes of the –397T/C and –351A/G SNPs of ESR1 with the occurrence of MI in whites.
Acknowledgments
The authors thank Marianne Eichinger, Claudia Ganser, and Wolfgang Latz for excellent technical assistance.
References
Mendelsohn ME, Karas RH. The protective effects of estrogen on the cardiovascular system. N Engl J Med. 1999; 340: 1801–1811.
Darblade B, Pendaries C, Krust A, Dupont S, Fouque M-J, Rami J, Chambon P, Bayard F, Arnal J-F. Estradiol alters nitric oxide production in the mouse aorta through the -, but not -, estrogen receptor. Circ Res. 2002; 90: 413–419.
Brouchet L, Krust A, Dupont S, Chambon P, Bayard F, Arnal JF. Estradiol accelerates reendothelialization in mouse carotid artery through estrogen receptor- but not estrogen receptor-. Circulation. 2001; 103: 423–428.
Pare G, Krust A, Karas RH, Dupont S, Aronovitz M, Chambon P, Mendelsohn ME. Estrogen receptor- mediates the protective effects of estrogen against vascular injury. Circ Res. 2002; 90: 1087–1092.
Sudhir K, Chou TM, Messina LM, Hutchison SJ, Korach KS, Chatterjee K, Rubanyi GM. Endothelial dysfunction in a man with disruptive mutation in oestrogen-receptor gene. Lancet. 1997; 349: 1146–1147.
Sudhir K, Chou TM, Chatterjee K, Smith EP, Williams TC, Kane JP, Malloy MJ, Korach KS, Rubanyi GM. Premature coronary artery disease associated with a disruptive mutation in the estrogen receptor gene in a man. Circulation. 1997; 96: 3774–3777.
Matsubara Y, Murata M, Kawano K, Zama T, Aoki N, Yoshino H, Watanabe G, Ishikawa K, Ikeda Y. Genotype distribution of estrogen receptor polymorphisms in men and postmenopausal women from healthy and coronary populations and its relation to serum lipid levels. Arterioscler Thromb Vasc Biol. 1997; 17: 3006–3012.
Lehtimki T, Kunnas TA, Mattila KM, Perola M, Penttil A, Koivula T, Karhunen PJ. Coronary artery wall atherosclerosis in relation to the estrogen receptor 1 gene polymorphism: an autopsy study. J Mol Med. 2002; 80: 176–180.
Lu H, Higashikata T, Inazu A, Nohara A, Yu W, Shimizu M, Mabuchi H. Association of estrogen receptor- gene polymorphisms with coronary artery disease in patients with familial hypercholesterolemia. Arterioscler Thromb Vasc Biol. 2002; 22: 817–823.
Shearman AM, Cupples LA, Demissie S, Peter I, Schmid CH, Karas RH, Mendelsohn ME, Housman DE, Levy D. Association between estrogen receptor gene variation and cardiovascular disease. JAMA. 2003; 290: 2263–2270.
Schuit SCE, Oei H-HS, Witteman JCM, Geurts van Kessel CH, van Meurs JBJ, Nijhuis RL, van Leeuwen JPTM, de Jong FH, Zillikens MC, Hofman A, Pols HAP, Uitterlinden AG. Estrogen receptor gene polymorphisms and risk of myocardial infarction. JAMA. 2004; 291: 2969–2977.
World Medical Association Declaration of Helsinki: recommendations guiding physicians in biomedical research involving human subjects. JAMA. 1997; 277: 925–926.
Guidelines Subcommittee. 1999 World Health Organization–International Society of Hypertension Guidelines for the Management of Hypertension. J Hypertens. 1999; 17: 151–183.
World Health Organization Study Group. Diabetes mellitus. WHO Tech Rep Ser. 1985; 727: 1–104.
Livak KJ. Allelic discrimination using fluorogenic probes and the 5 nuclease assay. Genet Anal. 1999; 14: 143–149.
Kutyavin IV, Afonina IA, Mills A, Gorn VV, Lukhtanov EA, Belousov ES, Singer MJ, Walburger DK, Lokhov SG, Gall AA, Dempcy R, Reed MW, Meyer RB, Hedgpeth J. 3-Minor groove binder-DNA probes increase sequence specificity at PCR extension temperatures. Nucleic Acids Res. 2000; 28: 655–661.
Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005; 21: 263–265.
Albagha OME, McGuigan FEA, Reid DM, Ralston SH. Estrogen receptor gene polymorphisms and bone mineral density: haplotype analysis in women from the United Kingdom. J Bone Miner Res. 2001; 16: 128–134.
Deng H-W, Li J, Li J-L, Johnson M, Gong G, Recker RR. Association of VDR and estrogen receptor genotypes with bone mass in postmenopausal Caucasian women: different conclusions with different analyses and the implications. Osteoporos Int. 1999; 9: 499–507.
Kobayashi S, Inoue S, Hosoi T, Ouchi Y, Shiraki M, Orimo H. Association of bone mineral density with polymorphism of the estrogen receptor gene. J Bone Miner Res. 1996; 11: 306–311.
Freely associating [editorial]. Nat Genet. 1999; 22: 1–2.
Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN. Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet. 2003; 33: 177–181.
Hegele RA. SNP judgments and freedom of association. Arterioscler Thromb Vasc Biol. 2002; 22: 1058–1061.
Ozaki K, Ohnishi Y, Iida A, Sekine A, Yamada R, Tsunoda T, Sato H, Sato H, Hori M, Nakamura Y, Tanaka T. Functional SNPs in the lymphotoxin- gene that are associated with susceptibility to myocardial infarction. Nat Genet. 2002; 32: 650–654.
Yamada Y, Izawa H, Ichihara S, Takatsu F, Ishihara H, Hirayama H, Sone T, Tanaka M, Yokota M. Prediction of the risk of myocardial infarction from polymorphisms in candidate genes. N Engl J Med. 2002; 347: 1916–1923.
Helgadottir A, Manolescu A, Thorleifsson G, Gretarsdottir S, Jonsdottir H, Thorsteinsdottir U, Samani NJ, Gudmundsson G, Grant SFA, Thorgeirsson G, Sveinbjornsdottir S, Valdimarsson EM, Matthiasson SE, Johannsson H, Gudmundsdottir O, Gurney ME, Sainz J, Thorhallsdottir M, Andresdottir M, Frigge ML, Topol EJ, Kong A, Gudnason V, Hakonarson H, Gulcher JR, Stefansson K. The gene encoding 5-lipoxygenase activating protein confers risk of myocardial infarction and stroke. Nat Genet. 2004; 36: 233–239.
Becherini L, Gennari L, Masi L, Mansani R, Massart F, Morelli A, Falchetti A, Gonnelli S, Fiorelli G, Tanini A, Brandi ML. Evidence of a linkage disequilibrium between polymorphisms in the human estrogen receptor gene and their relationship to bone mass variation in postmenopausal Italian women. Hum Mol Genet. 2000; 9: 2043–2050.(Werner Koch, PhD; Petra H)
Institut für Humangenetik, Klinikum rechts der Isar, Technische Universitt München, Munich (A.P.)
Institut für Humangenetik, GSF Forschungszentrum, Neuherberg (A.P.)
Institut für medizinische Statistik und Epidemiologie, Klinikum rechts der Isar, Technische Universitt München, Munich (J.C.M.), Germany.
Abstract
Background— The effects of estrogen on blood vessels are partly due to changes in vascular cell gene expression and protein synthesis that are mediated by estrogen receptors. In previous association studies, the –397T/C (rs2234693) and –351A/G (rs9340799) single nucleotide polymorphisms in the estrogen receptor gene (ESR1) have been implicated in the risk of coronary atherosclerosis and myocardial infarction. To test these findings, we examined the relationship of the polymorphisms to myocardial infarction in a large sample of white patients and control individuals of predominantly European descent.
Methods and Results— The case group included 3657 patients with myocardial infarction, and the control group comprised 1211 individuals with angiographically normal coronary arteries and without signs or symptoms of myocardial infarction. TaqMan assays were used for the determination of genotypes. Genotype distributions of the –397T/C and –351A/G polymorphisms were not significantly different between the control and patient groups (P0.85). The frequencies of haplotypes defined by the –397T/C and –351A/G polymorphisms were similar in the control group and the patient group (P=0.42). In addition, the distributions of haplotype-defined genotypes (diplotypes) were not significantly different between the control group and the patient group (P=0.81). Separate analyses in women and men did not reveal sex-related associations of specific genotypes or haplotypes of the polymorphisms with myocardial infarction (P0.25).
Conclusions— The results indicate that the –397T/C and –351A/G polymorphisms of ESR1 or haplotypes based on these polymorphisms are not associated with myocardial infarction in a white population.
Key Words: estrogen receptor alpha genetics myocardial infarction polymorphism receptors, estrogen
Introduction
Estrogen receptor is a ligand-activated transcription factor involved in the regulation of cellular pathways that are of potential importance in vascular physiology and disorders.1 Human vascular endothelial and smooth muscle cells contain estrogen receptor protein, and experiments with transgenic mice showed that it is able to cause stimulation of endothelial nitric oxide production, acceleration of reendothelialization, and inhibition of the vascular injury response.1–4 A cardioprotective function of estrogen receptor in humans was indicated by the finding that a disruptive mutation in both alleles of the estrogen receptor gene (ESR1; gene map locus 6q25.1) was associated with endothelial dysfunction and coronary artery disease in a 31-year-old man.5,6
Editorial p 2081
To examine the potential role of estrogen receptor in coronary artery disease and myocardial infarction (MI), common sequence variations in ESR1, especially the –397T/C (rs2234693) and –351A/G (rs9340799) single nucleotide polymorphisms (SNPs), have been employed in both prospective and retrospective association studies.7–11 Although several previous studies have found associations of the –397T/C and –351A/G SNPs with cardiovascular disease phenotypes,8–11 a comparison of their results reveals inconsistencies, including the identity of the risk genotype or haplotype. The diversity of findings between the prior studies may result, at least in part, from the low prevalence of events observed in the prospective trials10,11 and the relatively small number of cases enrolled in the retrospective examinations.7–9 To avoid these potential limitations, we investigated the association between the –397T/C and –351A/G SNPs of ESR1 and MI in a case-control study that included a large number of participants.
Methods
Patients and Controls
The study population comprised a group of 3657 white patients with MI and a group of 1211 white control individuals who were recruited consecutively from 1993 to 2002. All patients and control persons were examined with coronary angiography at Deutsches Herzzentrum München or 1 Medizinische Klinik rechts der Isar der Technischen Universitt München. Written informed consent was obtained from all study participants. The study protocol was approved by the institutional ethics committee, and the reported investigations were in accordance with the principles of the current version of the Declaration of Helsinki.12
Definitions
Individuals were considered disease free and therefore eligible as controls when their coronary arteries were angiographically normal and when they had no history of MI, no symptoms suggestive of MI, no electrocardiographic signs of MI, and no regional wall motion abnormalities. Coronary angiography in the control individuals was performed for the evaluation of chest pain. The diagnosis of MI was established in the presence of chest pain lasting >20 minutes combined with ST-segment elevation or pathological Q waves on a surface ECG. Patients with MI had to show either an angiographically occluded infarct-related artery or regional wall motion abnormalities corresponding to the electrocardiographic infarct localization or both. Systemic arterial hypertension was defined as a systolic blood pressure of 140 mm Hg and/or a diastolic blood pressure of 90 mm Hg,13 on at least 2 separate occasions, or antihypertensive treatment. Hypercholesterolemia was defined as a documented total cholesterol value 240 mg/dL (6.2 mmol/L) or current treatment with cholesterol-lowering medication. Persons reporting regular smoking in the previous 6 months were considered current smokers. Diabetes mellitus was defined as the presence of an active treatment with insulin or an oral antidiabetic agent; for patients on dietary treatment, documentation of an abnormal fasting blood glucose or glucose tolerance test based on the World Health Organization criteria14 was required for establishing this diagnosis.
Determination of ESR1 Genotypes
Genomic DNA was extracted from peripheral blood leukocytes with the QIAamp DNA Blood Kit (Qiagen) or the High Pure PCR Template Preparation Kit (Roche Applied Science). We designed and used TaqMan allelic discrimination assays for genotype analysis.15 The primers 5 TCCATCAGTTCATCTGAGTTCCAA 3 and 5 TTCAGAACCATTAGAGACCAATGCT 3 were used to initiate the amplification of a 117-bp portion of intron 1 of ESR1, which contains the polymorphic sites –397T/C (rs2234693) and –351A/G (rs9340799). The –397T/C and –351A/G SNPs are located 397 and 351 bp, respectively, upstream from the first nucleotide of exon 2 of ESR1. To accomplish allele-specific signaling, the probes contained the fluorogenic dyes 6-carboxy-fluorescein (FAM) or VIC (proprietary dye of Applied Biosystems). The structures of the probes were as follows (allele-specific nucleotides are underlined): FAM-5 TGTCCCAGCTGTTTT 3 (for –397T), VIC-5 CCCAGCCGTTTTA 3 (for –397C), FAM-5 CCCAACTCTAGACCAC 3 (for –351A), and VIC-5 TCCCAACTCCAGACCA 3 (for –351G). Minor groove binder groups were conjugated with the 3 ends of the oligonucleotides to facilitate formation of stable duplexes between the probes and their single-stranded DNA targets.16 TaqMan reactions were performed on 96-well plates, each of which included, at the outset of the thermocycling procedure, 90 assays that contained DNA samples of unknown genotype, 2 assays that contained no DNA, and 2 assays each that contained DNA samples of the homozygous genotypes as standards. As a control, genotyping was repeated for 20% of the samples with the use of DNA prepared separately from the original blood sample.
Restriction enzyme analyses involving PvuII and XbaI are established methods for genotyping of the –397T/C and –351A/G SNPs of ESR1, respectively.7–10 Examinations with these allele-discriminating restriction enzymes (Roche Applied Science) served as measures to verify the accuracy of genotyping with the use of the new TaqMan systems for the –397T/C and –351A/G SNPs. Digestion of the –397T-specific polymerase chain reaction (PCR) product with PvuII resulted in 2 fragments of 84 and 33 bp, and digestion of the –351A-specific PCR product with XbaI resulted in 2 fragments of 78 and 39 bp. PCR products specific for the –397C and –351G alleles were not cleaved with the use of PvuII and XbaI, respectively. In the test, randomly selected samples (n=153) were subjected to the TaqMan assays and, in parallel, separate digestions with PvuII and XbaI. The results obtained with the different methods were fully corresponding, which demonstrated the reliability of the TaqMan systems that were used for genotyping of the –397T/C and –351A/G SNPs in the entire study population. The genotypes of the DNA samples that served as standards in the TaqMan reactions were determined with the use of restriction enzyme analysis and verified with the use of sequence analysis of both DNA strands. Genotype determination was done by workers who had no knowledge of clinical, laboratory, or angiographic data of the individuals enrolled in the study.
Statistical Analysis
The analysis consisted of comparing separately genotype distributions, haplotype frequencies, and haplotype-based diplotype distributions between the control group and the group of patients with MI. Discrete variables are expressed as counts (percentage) and compared with the use of the 2 or Fisher exact test, as appropriate. Continuous variables are expressed as mean±SD and compared by means of the unpaired, two-sided t test. The measures of linkage disequilibrium (D and r2) between the –397T/C and –351A/G SNPs were calculated from primary genotype data with the use of the software package Haploview.17 We tested for the independent association effect of SNP-related genotypes and haplotypes in multiple logistic regression models of MI that included as covariates age, gender, history of arterial hypertension, history of hypercholesterolemia, current cigarette smoking, and diabetes mellitus. Adjusted odds ratios (ORs) and 95% Wald CIs were calculated on the basis of these models. Statistical significance was accepted for probability values <0.05.
Results
The main baseline characteristics and genotype distributions of the –397T/C and –351A/G SNPs in the control group (n=1211) and the group of patients with MI (n=3657) are shown in Table 1. Mean age of the patients was higher than that of the control group, the proportion of women was lower in the patient group than in the control group, and history of arterial hypertension and hypercholesterolemia, current cigarette smoking, and diabetes mellitus were more prevalent in the patient group than in the control group (P<0.0001 for all comparisons; Table 1). The genotype distributions of the SNPs were not significantly different between the control group and the patient group (P0.85; Table 1). Genotype distributions of the SNPs in the control and patient groups were consistent with those expected for samples in Hardy-Weinberg equilibrium (P0.21). The sample sizes of the control group and the patient group provided the analysis with 84% power (–397T/C SNP) and 89% power (–351A/G SNP) to detect a 25% higher risk of MI associated with carriage of the –397C and –351G alleles, respectively (2-sided -level of 0.05).
The possibility existed that a specific combination of the alleles of the –397T/C and –351A/G SNPs was associated with MI. We observed a high degree of linkage disequilibrium between the –397T/C and –351A/G SNPs (D=0.999; 95% CI, 0.98 to 1.00; r2=0.64). Table 2 shows the calculated frequencies of the 4 haplotypes, –397T/–351A (TA), CG, CA, and TG, in the control group and in the patient group. Haplotype phases were assigned unambiguously in the individuals who were homozygous at least at 1 of the 2 polymorphic positions (n=3032). Among these individuals, the TG haplotype was only present on 2 of the total of 6064 chromosomes. In these 2 cases, correct genotyping was verified in separate analyses with the use of the restriction enzymes PvuII and XbaI. Because of the rarity of the TG haplotype among the unambiguously typed individuals, the haplotypes TA and CG were assigned to the 1836 carriers of the double heterozygous genotype combination –397TC/–351AG. The probability is 0.99987 that this inference is correct, indicating that the assignment was incorrect theoretically in <1 of the 1836 double heterozygous individuals. In essence, haplotype frequencies were not substantially different between the control and patient groups (overall P=0.42; Table 2). Table 3 shows the distributions of haplotype-related genotypes (diplotypes) in the study groups. Similar to the haplotype frequencies, the diplotype distributions were not significantly different between the control group and the patient group (overall P=0.81).
We addressed the question of whether sex-dependent relationships existed between the –397T/C and –351A/G SNPs and MI. The genotype distributions of the SNPs were not essentially different between the women and men of the entire study population (P0.41). Comparisons of the genotype distributions between the women in the control group and the patient group and between the men in the 2 groups did not reveal significant differences (P0.41; Table 4). Haplotype frequencies were not substantially different between the women in the control and patient groups (P=0.25) and between the men in the 2 groups (P=0.75). Similarly, diplotype distributions were not associated with MI in women (P=0.66) and men (P=0.94).
The possibility of independent associations of genotypes or haplotypes of the SNPs with MI was tested, after adjustment for other possible risk factors, in multiple logistic regression models of MI that included the baseline clinical characteristics as covariates. We observed that the occurrence of MI was not different between the carriers of the –397TT genotype and the carriers of the –397CC genotype (OR, 1.17; 95% CI, 0.82 to 1.69) and between the carriers of the –351AA genotype and the carriers of the –351GG genotype (OR, 0.84; 95% CI, 0.57 to 1.25). The MI risk of the homozygous carriers of the TA, CG, or CA haplotype was not substantially different from that of the noncarriers of the respective haplotype (TA haplotype: OR, 0.95; 95% CI, 0.77 to 1.17; CG haplotype: OR, 0.99; 95% CI, 0.80 to 1.24; CA haplotype: OR, 1.17; 95% CI, 0.82 to 1.66).
Discussion
The results of the present study show that the –397T/C and –351A/G SNPs of ESR1 are not associated with MI in a large sample of white patients. Lack of association relates to genotypes, haplotypes, and diplotypes and applies to women and men. Recruitment bias may be considered a possible reason for the negative results of this study. Because all control subjects had some indication for coronary angiography, they did not represent a typical sample of healthy controls. However, we consider this group to be especially suitable as a control group in the setting of this study because the absence of MI was rigorously established on the basis of history, electrocardiography, left ventricular angiography, and coronary angiography. The control group used in this study appears not to be fundamentally different from typical control groups because the genotype distributions were in Hardy-Weinberg equilibrium and not significantly different from the genotype distributions observed in another control group that consisted of white individuals.18 Two reasons argue against the possibility of false-negative results due to a systematic genotyping error: (1) Genotype and haplotype frequencies in the control and patient groups were in good correlation with findings of independent investigations that included populations of European origin,8,10,18,19 and (2) repetition of genotyping with the use of restriction enzymes provided reassurance that the TaqMan assays produced correct data.
Associations of ESR1 SNPs with cardiovascular disease were found in prospective studies,10,11 yet outcomes were discordant, and essential findings were not in agreement with our negative results. It was observed in the group of men of one study population (subset of individuals from the offspring cohort of the Framingham Heart Study; n=1739) that recognized MI events (n=54) occurred more often among the carriers of the –397CC genotype than among the carriers of the –397TT or –397TC genotype.10 In another population (subset of individuals from The Rotterdam Study; n=6408), the women homozygous or heterozygous for the TA haplotype encountered MI (115 cases) and ischemic heart disease (168 cases) events more often than the noncarriers of this haplotype.11 No association of the TA haplotype with MI (170 cases) or ischemic heart disease (272 cases) was observed in the group of men.11 The utility of the –397T/C and –351A/G SNPs as cardiovascular disease markers was also tested in retrospective settings, and heterogeneous results were obtained.7–9 A relationship was found between the –397TC and –397CC genotypes and the complexity of atherosclerotic lesions and the presence of thrombosis in coronary arteries in an autopsy study of Finnish men (262 cases).8 In a study that included 119 patients with coronary artery disease and 176 individuals without coronary artery disease from Japan, the –351GG genotype and homozygosity of the CG haplotype, but not the –397T/C SNP on its own, were found to be associated with angina pectoris and MI.9 No association between the –397T/C or –351A/G SNP with the severity of coronary stenosis was observed in another study from Japan that included 87 cases with MI or angina pectoris and 94 control individuals.7
Population stratification is a potential problem for case-control studies and may lead to false-positive results or mask an existing association.21–23 In this study the probability of a false-negative association result due to population stratification is relatively low because the study participants were consecutively recruited from a defined geographic area of southern Germany with limited recent immigration. Family-based studies, which use unaffected family members as control individuals, may help to clarify the importance of the –397T/C and –351A/G SNPs for MI because they are less prone to errors stemming from admixture than population-based studies.
The inability to identify an association in this relatively large and well-defined population may reflect the possibility that the –397T/C and –351A/G SNPs of ESR1 have at best a modest impact on the process leading to MI, a disease phenotype that may be more strongly affected by other genetic factors24–26 and environmental and lifestyle influences. The negative results of this report suggest that polymorphic markers tightly linked to the –397T/C and –351A/G SNPs, including a (TA)n dinucleotide repeat polymorphism located upstream from exon 1 of ESR1,18,27 may not be relevant as indicators of MI risk in this study population.
In conclusion, the results of this study do not provide evidence of an association of specific genotypes or haplotypes of the –397T/C and –351A/G SNPs of ESR1 with the occurrence of MI in whites.
Acknowledgments
The authors thank Marianne Eichinger, Claudia Ganser, and Wolfgang Latz for excellent technical assistance.
References
Mendelsohn ME, Karas RH. The protective effects of estrogen on the cardiovascular system. N Engl J Med. 1999; 340: 1801–1811.
Darblade B, Pendaries C, Krust A, Dupont S, Fouque M-J, Rami J, Chambon P, Bayard F, Arnal J-F. Estradiol alters nitric oxide production in the mouse aorta through the -, but not -, estrogen receptor. Circ Res. 2002; 90: 413–419.
Brouchet L, Krust A, Dupont S, Chambon P, Bayard F, Arnal JF. Estradiol accelerates reendothelialization in mouse carotid artery through estrogen receptor- but not estrogen receptor-. Circulation. 2001; 103: 423–428.
Pare G, Krust A, Karas RH, Dupont S, Aronovitz M, Chambon P, Mendelsohn ME. Estrogen receptor- mediates the protective effects of estrogen against vascular injury. Circ Res. 2002; 90: 1087–1092.
Sudhir K, Chou TM, Messina LM, Hutchison SJ, Korach KS, Chatterjee K, Rubanyi GM. Endothelial dysfunction in a man with disruptive mutation in oestrogen-receptor gene. Lancet. 1997; 349: 1146–1147.
Sudhir K, Chou TM, Chatterjee K, Smith EP, Williams TC, Kane JP, Malloy MJ, Korach KS, Rubanyi GM. Premature coronary artery disease associated with a disruptive mutation in the estrogen receptor gene in a man. Circulation. 1997; 96: 3774–3777.
Matsubara Y, Murata M, Kawano K, Zama T, Aoki N, Yoshino H, Watanabe G, Ishikawa K, Ikeda Y. Genotype distribution of estrogen receptor polymorphisms in men and postmenopausal women from healthy and coronary populations and its relation to serum lipid levels. Arterioscler Thromb Vasc Biol. 1997; 17: 3006–3012.
Lehtimki T, Kunnas TA, Mattila KM, Perola M, Penttil A, Koivula T, Karhunen PJ. Coronary artery wall atherosclerosis in relation to the estrogen receptor 1 gene polymorphism: an autopsy study. J Mol Med. 2002; 80: 176–180.
Lu H, Higashikata T, Inazu A, Nohara A, Yu W, Shimizu M, Mabuchi H. Association of estrogen receptor- gene polymorphisms with coronary artery disease in patients with familial hypercholesterolemia. Arterioscler Thromb Vasc Biol. 2002; 22: 817–823.
Shearman AM, Cupples LA, Demissie S, Peter I, Schmid CH, Karas RH, Mendelsohn ME, Housman DE, Levy D. Association between estrogen receptor gene variation and cardiovascular disease. JAMA. 2003; 290: 2263–2270.
Schuit SCE, Oei H-HS, Witteman JCM, Geurts van Kessel CH, van Meurs JBJ, Nijhuis RL, van Leeuwen JPTM, de Jong FH, Zillikens MC, Hofman A, Pols HAP, Uitterlinden AG. Estrogen receptor gene polymorphisms and risk of myocardial infarction. JAMA. 2004; 291: 2969–2977.
World Medical Association Declaration of Helsinki: recommendations guiding physicians in biomedical research involving human subjects. JAMA. 1997; 277: 925–926.
Guidelines Subcommittee. 1999 World Health Organization–International Society of Hypertension Guidelines for the Management of Hypertension. J Hypertens. 1999; 17: 151–183.
World Health Organization Study Group. Diabetes mellitus. WHO Tech Rep Ser. 1985; 727: 1–104.
Livak KJ. Allelic discrimination using fluorogenic probes and the 5 nuclease assay. Genet Anal. 1999; 14: 143–149.
Kutyavin IV, Afonina IA, Mills A, Gorn VV, Lukhtanov EA, Belousov ES, Singer MJ, Walburger DK, Lokhov SG, Gall AA, Dempcy R, Reed MW, Meyer RB, Hedgpeth J. 3-Minor groove binder-DNA probes increase sequence specificity at PCR extension temperatures. Nucleic Acids Res. 2000; 28: 655–661.
Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005; 21: 263–265.
Albagha OME, McGuigan FEA, Reid DM, Ralston SH. Estrogen receptor gene polymorphisms and bone mineral density: haplotype analysis in women from the United Kingdom. J Bone Miner Res. 2001; 16: 128–134.
Deng H-W, Li J, Li J-L, Johnson M, Gong G, Recker RR. Association of VDR and estrogen receptor genotypes with bone mass in postmenopausal Caucasian women: different conclusions with different analyses and the implications. Osteoporos Int. 1999; 9: 499–507.
Kobayashi S, Inoue S, Hosoi T, Ouchi Y, Shiraki M, Orimo H. Association of bone mineral density with polymorphism of the estrogen receptor gene. J Bone Miner Res. 1996; 11: 306–311.
Freely associating [editorial]. Nat Genet. 1999; 22: 1–2.
Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN. Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet. 2003; 33: 177–181.
Hegele RA. SNP judgments and freedom of association. Arterioscler Thromb Vasc Biol. 2002; 22: 1058–1061.
Ozaki K, Ohnishi Y, Iida A, Sekine A, Yamada R, Tsunoda T, Sato H, Sato H, Hori M, Nakamura Y, Tanaka T. Functional SNPs in the lymphotoxin- gene that are associated with susceptibility to myocardial infarction. Nat Genet. 2002; 32: 650–654.
Yamada Y, Izawa H, Ichihara S, Takatsu F, Ishihara H, Hirayama H, Sone T, Tanaka M, Yokota M. Prediction of the risk of myocardial infarction from polymorphisms in candidate genes. N Engl J Med. 2002; 347: 1916–1923.
Helgadottir A, Manolescu A, Thorleifsson G, Gretarsdottir S, Jonsdottir H, Thorsteinsdottir U, Samani NJ, Gudmundsson G, Grant SFA, Thorgeirsson G, Sveinbjornsdottir S, Valdimarsson EM, Matthiasson SE, Johannsson H, Gudmundsdottir O, Gurney ME, Sainz J, Thorhallsdottir M, Andresdottir M, Frigge ML, Topol EJ, Kong A, Gudnason V, Hakonarson H, Gulcher JR, Stefansson K. The gene encoding 5-lipoxygenase activating protein confers risk of myocardial infarction and stroke. Nat Genet. 2004; 36: 233–239.
Becherini L, Gennari L, Masi L, Mansani R, Massart F, Morelli A, Falchetti A, Gonnelli S, Fiorelli G, Tanini A, Brandi ML. Evidence of a linkage disequilibrium between polymorphisms in the human estrogen receptor gene and their relationship to bone mass variation in postmenopausal Italian women. Hum Mol Genet. 2000; 9: 2043–2050.(Werner Koch, PhD; Petra H)