Estrogen Receptor Alpha Gene Polymorphisms and

Original Paper Cerebrovasc Dis 2007;24:500–508 DOI: 10.1159/000110419

Received: June 14, 2007 Accepted: June 21, 2007 Published online: October 29, 2007

Estrogen Receptor Alpha Gene Polymorphisms and First-Ever Intracerebral Hemorrhage Magnus Strand a Ingegerd Söderström a Per-Gunnar Wiklund a Göran Hallmans b Lars Weinehall c Stefan Söderberg a Tommy Olsson a Department of Public Health and Clinical Medicine, a Medicine, b Nutritional Research and c Epidemiology, Umeå University, Umeå, Sweden

Abstract Background: Signaling through estrogen receptor alpha (ER␣) regulates vasodilatation and atherogenesis. Since hypertension and atherosclerosis are major mechanisms in stroke development, we hypothesized that genetic variants of the ER␣ gene (ESR1) are determinants of future ischemic stroke or intracerebral hemorrhage (ICH). Methods: In a population-based prospective nested case-control study, the relationships between ESR1 polymorphisms (c.454–397T1C and c.454–351A1 G) and ischemic stroke and ICH were examined in univariate and multivariate models using conditional logistic regression, which included established risk factors. Definitive first-ever stroke events (n = 388), including ischemic stroke (n = 320), ICH (n = 61), and unspecified stroke (n = 7) cases, and controls without cardiovascular disease (n = 773), matched for age, sex, and geographical region were included. Results: Carriers of the c.454–397T/T genotype had a significantly (p = 0.017) increased risk of ICH (OR 2.31, 95% CI 1.16–4.60) in a univariate analysis. This association persisted (OR 3.94, 95% CI 1.54–10.03), after adjustment for stroke risk determinants. Carriers of c.454–397T/T or

© 2007 S. Karger AG, Basel 1015–9770/07/0246–0500$23.50/0 Fax +41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/ced

c.454–397T/C genotypes had significantly (p = 0.002 and p = 0.004, respectively) higher mean systolic blood pressure (SBP), than carriers of c.454–397C/C, and a similar relationship was observed for diastolic blood pressure (DBP). The combinations of c.454–397T/T genotype and hypertension (OR 21.46, 95% CI 5.20–88.51), or high SBP (OR 18.17, 95% CI 4.91– 67.31) or DBP (OR 11.94, 95% CI 3.75–38.03), were strongly associated with increased risk of ICH. Conclusions: In this population, the c.454–397T/T genotype associates with first-ever ICH, particularly in combination with hypertension. This implies that alterations in ER␣-mediated signaling may be involved in the pathophysiology of this disease, with a putative impact on primary prevention. Copyright © 2007 S. Karger AG, Basel

Introduction

Stroke is a multifactorial disease influenced by genetic and environmental factors. Notably, the incidence of stroke is approximately 50% higher in northern compared to southern Sweden [1], suggesting genetic influences, with putative gene-environment interactions, in this population. The age-specific stroke incidence is higher for men than women, indicating that hormonal factors have protective properties. If so, endogenous estrogens may be Prof. Tommy Olsson Department of Public Health and Clinical Medicine, Medicine Umeå University Hospital, SE–901 85 Umeå (Sweden) Tel. +46 90 785 1845, Fax +46 90 137 633 E-Mail [email protected]

Downloaded by: Umea University 198.143.54.65 - 10/12/2015 9:28:37 AM

Key Words Estrogen receptor ␣ ⴢ Genetics ⴢ Hypertension ⴢ Intracerebral hemorrhage ⴢ Ischemic stroke ⴢ Polymorphism

Materials and Methods Study Populations This population-based study included subjects living in the two northernmost counties of Sweden, Norrbotten and Västerbotten. It was a substudy within the Northern Sweden Monitoring of Trends and Determinants in Cardiovascular Diseases (MONICA) project and the Västerbotten Intervention Program (VIP). In the MONICA project, randomly selected samples of 2,000– 2,500, 25- to 74-year-old inhabitants of Norrbotten and Västerbotten, stratified by age and sex, were invited to participate in health surveys in 1986, 1990, 1994, and 1999 [13]. The mean participation rate in these surveys was 77.2%, resulting in a total of 6,942 screened individuals. The VIP is an ongoing community intervention program targeting cardiovascular disease and diabetes prevention in Västerbotten. When subjects turned 30, 40, 50, and 60 years of age they were invited to participate in a health survey, with a design similar to the MONICA surveys, at their primary healthcare center. The overall participation rate has been 60% and possible selection bias has been investigated [14]. Between January 1, 1985 and Sep-

ESR1 Polymorphisms and Intracerebral Hemorrhage

tember 20, 2000, approximately 66,300 individuals took part in the VIP. As part of both the MONICA and the VIP surveys, participants were invited to donate fasting blood samples to be stored at the Northern Sweden Medical Research Bank. Case Ascertainment Since 1985, all acute stroke events in the age group 25–74 years living in the study area have been included in the Northern Sweden MONICA stroke registries, using WHO criteria and MONICA methodology [13]. Between January 1, 1985 and September 20, 2000, all acute definitive first-ever stroke events in Västerbotten and Norrbotten occurring in those aged 25–74 years were recorded. Detailed descriptions of the criteria for stroke diagnosis and subtype classification have been given elsewhere [13]. Briefly, the WHO definition ‘rapidly developing clinical signs of focal (or global) disturbance of cerebral function lasting more than 24 h (unless interrupted by surgery or death), with no apparent cause other than a vascular origin’ was applied and putative first-ever stroke events were identified through screening hospital discharge records, general practitioners’ reports, and death certificates. The subtypes of acute stroke were diagnosed according to International Classification of Diagnosis (ICD)-9 and 10 and based on the following examinations: ICH (ICD-9, 431; ICD-10, I61) – positive finding on computed tomography (CT) or magnetic resonance scan (MR) or autopsy; ischemic stroke (ICD-9, 434; ICD10, I63) – no signs of hemorrhage on CT/MR scan or autopsy; unspecified stroke (ICD-9, 436; ICD-10, I64) – not investigated by CT/MR scan or autopsy. Additional data collected included information on medical history, symptoms, and examinations. In this study, only cases classified as ‘definitive stroke’ were included as non-fatal events. In the case of fatal events, the ‘unclassified stroke’ category was also included [13]. Blood samples and DNA have been stored in the Northern Sweden Medical Research Bank. Subjects with prior stroke, myocardial infarction, or cancer were excluded, as were subjects with subarachnoid hemorrhage and traumatic ICH. The number of subjects with confirmed or suspected stroke included in the registry but who were unwilling to participate in further studies after information gathering has averaged 6 per year (0.6%). For each case, 2 matched controls without known cardiovascular disease or cancer were selected from the MONICA and VIP cohorts. Matching was by age (82 years), cohort (MONICA or VIP), date (81 year) of health survey and geographical locale. For 3 cases of ischemic stroke, only 1 matched control was available. This approach resulted in a cohort comprising 388 cases of firstever stroke (160 women and 228 men) and 773 controls (319 women and 454 men). In total, 320 (82.5%) ischemic stroke, 61 (15.7%) ICH, and 7 (1.8%) unspecified stroke cases were identified who, prior to the stroke event, had participated in a MONICA (44 cases) or VIP (344 cases) survey. Informed consent was obtained. This study was approved by the Research Ethics Committee of Umeå University, Umeå, and the National Computer Data Inspection Board, Stockholm, Sweden. Biomedical Factors and Biochemical Analyses Methods for measurements of blood pressure, height and weight, and blood sampling including estimation of blood lipids and an oral glucose tolerance test have been reported elsewhere

Cerebrovasc Dis 2007;24:500–508

501


key. The effects of estrogens are mediated through the estrogen receptors alpha (ER␣) and beta (ER␤). These function as transcription factors, targeting numerous processes in the body, including lipid metabolism and vessel wall function. Importantly, estrogens increase arterial vasodilatation and prevent atherosclerosis [2]. The regulatory effects of estrogens on blood pressure are at least partly mediated through ER␣-dependent genomic and non-genomic stimulation of endothelial nitric oxide synthase [3, 4]. A disruptive mutation in the ER␣ gene (ESR1) causes endothelial dysfunction, dyslipidemia, and premature coronary artery disease [5]. Two single nucleotide polymorphisms in the first intron of ESR1 (c.454–397T 1C and c.454–351A1G) have been implicated in cardiovascular diseases. The c.454 –397T allele has been linked to lower ESR1 expression [6, 7]. Homozygosity for this allele associates with increased systolic blood pressure (SBP) [8], increased waist circumference [9], increased low-density lipoprotein and total cholesterol [10], and increased risk of aortic valvular sclerosis [10], and in-stent restenosis [11]. An ESR1 haplotype (c.454–397T and c.454–351A alleles) associates with lower plasma estradiol levels [12], higher diastolic blood pressure (DBP), myocardial infarction, and ischemic heart disease [7]. These findings imply that ESR1 expression and ER␣ function have strong modulatory effects on the cardiovascular system. Our hypothesis was that specific ESR1 polymorphisms are related to stroke subtypes ischemic stroke or intracerebral hemorrhage (ICH).

Intron 1 46 bp 351 bp

Exon 1

Exon 2

Pvu II Xba I c.454–397T>C c.454–351A>G

Fig. 1. Genomic localizations and RFLP GTCCCAGCTGTTTTATGCTTTGTCTCTGTTTCCCAGAGACCCTGAGTGTGG TCTAGAGTT

TC

CC

GG

TT

AA

TC

AA

1,374 bp 936 bp 438 bp

[13]. Hypertension was defined according to WHO criteria as SBP 6140 mm Hg and/or DBP 690 mm Hg and/or use of antihypertensive medication. The subjects were classified as diabetic or non-diabetic based on self-reported data, information regarding pharmacological treatment, and/or fasting plasma glucose levels 67.0 mM and/or post-load glucose levels 611.0 mM (612.2 mM in VIP, measured in capillary plasma). Smoking was defined as daily tobacco smoking of cigarettes, cigars, or pipes. Subjects were sorted into categories (current, previous, or never) depending on frequency of smoking. Occasional smokers, without a previous history of smoking, were included in the ‘never’ category. Genotyping DNA was extracted from peripheral blood leukocytes using standard methods. In the first part of the cohort (sampled from January 1, 1985 to August 31, 1996; 113 cases and 226 controls), genotyping was performed by RFLP-PCR using the same primer design for both polymorphisms. Genomic DNA (50 ng) was amplified in 100 ␮l reaction mix containing: 0.4 ␮ M 5ⴕ-CTGCCACCCTATCTGTATCTTTTCCTATTCTCC-3ⴕ (forward), 0.4 ␮ M 5ⴕ-TCTTTCTCTGCCACCCTGGCGTCGATTATCTGA-3ⴕ (reverse), 0.2 mM each of dNTPs, 50 mM MgCl2, and 3.5 U Taq polymerase (Gibco-BRL Life Technologies). The PCR was run for 30 cycles at 94 ° C for 30 s, 61 ° C for 40 s, and 72 ° C for 90 s (Peltier Thermocycler, MJ Research Inc.). The resulting PCR fragment (1,374 bp) contained both polymorphic sites. PCR products were digested at 37 ° C by an excess of PvuII or XbaI (Gibco-BRL Life

502

AG

Cerebrovasc Dis 2007;24:500–508

981 bp

393 bp

Technologies), and fragment lengths were analyzed using electrophoresis (fig. 1). In the second part of the cohort (sampled from September 1, 1996 to September 20, 2000; 275 cases and 547 controls), the TaqMan 5ⴕ endonuclease PCR allelic discrimination technique was used (Applied Biosystems), according to the manufacturer’s instructions with the following designs: c.454–397T1C: 5ⴕ-CTGTGTTGTCCATCAGTTCATCT-3ⴕ (forward) 5ⴕ-ACTCAGGGTCTCTGGGAAACA-3ⴕ (reverse) 5ⴕ-VIC-CCAGCTGTTTTAT-3ⴕ (MGB probe) 5ⴕ-FAM-CCAGCCGTTTTAT-3ⴕ (MGB probe) c.454–351A1G: 5ⴕ-CATCAGTTCATCTGAGTTCCAAAT-3ⴕ (forward) 5ⴕ-TTTCAGAACCATTAGAGACCAATGC-3ⴕ (reverse) 5ⴕ-VIC-TGTGGTCTAGAGTT-3ⴕ (MGB probe) 5ⴕ-FAM-TGGTCTGGAGTT-3ⴕ (MGB probe) The genotyping methods were validated against each other by running DNA from 24 control subjects in duplicate using both RFLP and TaqMan, resulting in complete match. Of these controls with validated genotypes, two in each allelic cluster were used as internal standards on each TaqMan plate. In cases of unsuccessful genotyping a second attempt was performed before subjects were excluded due to lack of DNA or poor DNA quality. In total, 1,045 and 1,046 subjects were genotyped for c.454 –397T 1C and c.454–351A1G, respectively. Of these, 241 subjects (164 controls and 77 cases) were genotyped for both polymorphisms using RFLP in the first part of the cohort, and 804 subjects

Strand /Söderström /Wiklund /Hallmans / Weinehall /Söderberg /Olsson


patterns of ESR1 polymorphisms. RFLP = Restriction fragment length polymorphism. The c.454–397T 1C polymorphism (also denoted IVS1–397T/C or IVS1 –401T/C) and the c.454–351A1G polymorphism (also denoted IVS1–351A/G or IVS1–354A/G), have identification numbers rs2234693 and rs9340799, respectively, at http://www.ncbi.nlm.nih.gov/SNP, according to the reference sequence NM_000125. For c.454–397T 1C, the presence of C eliminates a PvuII restriction site, while for c.454–351A1G the presence of G eliminates an XbaI restriction site.

Table 1. Baseline characteristics

Continuous variables Age at health survey (SD), year SBP (SD), mm Hg DBP (SD), mm Hg BMI (SD), kg m–2 Triglycerides (SD), mM Total cholesterol (SD), mM Fasting glucose (SD), mM Post-load glucose (SD), mM Categorical variables Sex, n (%) Women Men Diabetes, n (%) No Yes Hypertension, n (%) No Yes Smoking, n (%) Current Previous Never

Controls (n = 773)

Cases (n = 388)

p value

55.088.1 136.5818.1 85.389.3 26.083.6 1.580.8 6.281.3 5.381.1 6.781.9

55.088.1 144.2820.3 89.2810.3 27.184.3 1.781.0 6.481.4 5.681.5 7.082.4

Matched