Aktuelle Neurologie 2007; 34 - P518
DOI: 10.1055/s-2007-987789

Genetic variation in the lymphotoxin alpha pathway and the risk of ischaemic stroke

M Dichgans 1, A Gschwendtner 1, S Bevan 1, P Lichtner 1, S Ripke 1, B Müller-Myhsok 1, E Wichmann 1, H Markus 1, T Meitinger 1, T Freilinger 1
  • 1München; London, UK; Neuherberg, München

Background and purpose: Recent evidence suggests a role of the lymphotoxin alpha pathway in myocardial infarction (MI). Genetic variation in the LTA, LGALS2 and PSMA6 genes is associated with MI in Japanese (Ozaki et al., Nature Genetics 2002 and 2006, Nature 2004). Given the close relationship between MI and ischaemic stroke (IS) and the involvement of inflammatory cytokines in atherosclerosis, the LTA, LGALS2 and PSMA6 genes represent candidate genes also for IS. We therefore set out to perform a comprehensive analysis of all three genes using SNP and haplotype association tests in a large European IS sample.

Patients and methods: 601 German patients with IS were consecutively recruited from our stroke unit. Patients were phenotyped in detail and IS was further classified into subtypes (TOAST classification). The control group comprised 736 unrelated age- and sex-matched individuals recruited from the KORA study, a community-based project near Munich. Blood samples were obtained from all participants, and DNA was extracted according to standard techniques. Selection of tagging SNPs was performed based on data of the International HapMap Project (CEU population); the LTA, LGALS2 and PSMA6 genes (including 10kb of 5' and 3'-sequence; NCBI B35, dbSNP b125) were analysed with the tagger algorithm (pairwise only; r2 >0,8). 24 SNPs (including one variant to control for sex) were genotyped using the iPlex-method on a MassArray System (Sequenom). The mean call-rate was 97%. No deviations from Hardy-Weinberg equilibrium were observed. Single-marker analyses were done using a chi square test with correction for multiple testing according to Westphal and Young. SNPs which were nominally significant in the German sample were genotyped in a large independent sample from the UK for replication purposes.

Results: Several SNPs, including a functional SNP in PSMA6 previously reported by Ozaki et al., showed nominally significant associations with overall stroke or stroke subtypes (p<0,05; allelic model). However, following correction for multiple testing, none of them remained significant. Replication of nominally significant SNPs in the UK sample was negative.

Conclusion: Our findings provide no evidence that genetic variation in the genes LTA, LGALS2 and PSMA6 contributes to IS.