No evidence for an association between genetic variation at the MMP2 and MMP9 loci and aneurysmal subarachnoid haemorrhage

Abstract

Subarachnoid haemorrhage (SAH) is a serious condition that has a young age onset and poor outcome [1]. Approximately 75% of the cases of SAH are due to the rupture of an intracranial aneurysm (IA) [1]. The most important risk factors are smoking, hypertension, and excessive alcohol intake [1]. In addition, genetic factors have been shown to play an important role [2]. The matrix metalloproteinases (MMP) can degrade extracellular matrix (ECM) components, and alteration of the ECM leads to changes of the vessel wall. There is evidence that matrix remodelling plays a role in the formation and rupture of IA. MMP9 levels has been found to be increased in the aneurysmal wall [3], and both MMP2 and MMP9 was found to be overexpressed in cerebral ruptured aneurysms compared with unruptured aneurysms [4]. To evaluate the role of genetic variants in MMP2 and MMP9 in SAH due to ruptured IA (aSAH), the genes were tagged using HapMap CEU data (Rel 23a) and Haploview (r = 0.8 and MAF = 0.1), resulting in 11 tagSNPs. SNPs in the promoter region of MMP2 (rs243865) and in the 30UTR of MMP9 (rs20544) were also genotyped. Consecutive patients (n = 183) presenting with aSAH admitted to the Neurointensive Care Unit at the Sahlgrenska University Hospital in Gothenburg, Sweden and 366 matched controls were recruited. A detailed description of the inclusion criteria has been published [5, 6]. All participants gave written informed consent, and the Medical Ethics Committee at the University of Gothenburg approved the study. Genotyping was performed by TaqMan SNP Genotyping Assays. The assay for rs243865 failed and was replaced by an assay for rs243864. All genotype frequencies were in Hardy–Weinberg equilibrium, and genotyping success rates were above 98%. Haplotype analyses were conducted using HelixTree 6.4.3 (Golden Helix). Association between single SNPs and case–control status was investigated with logistic regression using an additive model. The study has 80% power to detect an OR of 1.45 at the 5% level for SNPs with a MAF of 0.3 for the risk allele. Mean age of patients and controls was 55 years and 74% were females. As expected, smokers were more frequent in cases compared to controls. Baseline characteristics have been presented previously [6]. Genotype frequencies are presented in Table 1. No SNP showed association to aSAH, and haplotype analysis did not add any further information (data not shown). As SAH is more common in women [1] we conducted a sex-stratified subanalysis. Two strongly linked SNPs in MMP9 (rs3918256 and rs20544, r = 0.99 in our material) showed significant associations with aSAH in men, both with OR 1.6 (95% CI 1.00–2.60). These associations did not remain after adjustment for smoking and hypertension. No association was found in the female subgroup. However, the effect of the SNPs did not differ significantly between men and women. In line with previous studies on aSAH, hemorrhagic stroke, and IA [7–10], we did not detect any association with genetic variants in MMP9, although conflicting results has been reported for rs20544 and IA [11]. Furthermore, in concordance with the present study, a lack of association between genetic variants in MMP2 and IA has been reported [11]. S. Olsson (&) L. Z. Csajbok K. Jood K. Nylen B. Nellgard C. Jern Department of Clinical Neuroscience and Rehabilitation, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at University of Gothenburg, Box 403, SE-405 30 Gothenburg, Sweden e-mail: sandra.olsson@neuro.gu.se