Malic enzyme 2 may underlie susceptibility to adolescent-onset idiopathic generalized epilepsy

Abstract

Idiopathic generalized epilepsy (IGE) is a class of genetically determined, phenotypically related epilepsy syndromes. Linkage analysis identified a chromosome 18 locus predisposing to a number of adolescent-onset IGEs. We report a single-nucleotide polymorphism (SNP) association analysis of the region around the marker locus with the high LOD score. This analysis, which used both case-control and family-based association methods, yielded strong evidence that malic enzyme 2 (ME2) is the gene predisposing to IGE. We also observed association among subgroups of IGE syndromes. An ME2-centered nine-SNP haplotype, when present homozygously, increases the risk for IGE (odds ratio 6.1; 95% confidence interval 2.9-12.7) compared with any other genotype. Both the linkage analysis and the association analysis support recessive inheritance for the locus, which is compatible with the fact that ME2 is an enzyme. ME2 is a genome-coded mitochondrial enzyme that converts malate to pyruvate and is involved in neuronal synthesis of the neurotransmitter gamma-aminobutyric acid (GABA). The results suggest that GABA synthesis disruption predisposes to common IGE and that clinical seizures are triggered when mutations at other genes, or perhaps other insults, are present.

Figure 1

Evidence for association of adolescent-onset IGE on chromosome 18. The graph shows the χ² based on the case-control analysis for each SNP tested in this region, which is just centromeric to marker D18S474. There appear to be two peaks, the first of which occurs directly over the promoter and locus for ME2. The second occurs in a region of no known genes that exhibits evidence of strong marker-marker disequilibrium. ME2 is located between SNP locations 200 and 300. The origin of the X-axis in figures 1–3 is the first SNP (rs3752087), which is located at position 46442.427 kb. SNP numbers are given next to each data point.

Figure 2

Evidence for association in IGE, from family-based association analysis rather than case-control analysis. SNP numbers are given next to each data point.

Figure 3

Evidence for association in two separate diagnostic groups with adolescent-onset IGE: JME (left) and non-JME (right) IGE. Non-JME IGE consists of the two groups: JAE and EGTCS. There were 88 JME cases and 68 JAE + EGTCS cases. The results shown are for case-control analysis. SNP numbers are given next to each data point. Results (not shown) for family-based association were similar for JME, but the sample size for non-JME was not sufficiently large to reach significance.

Figure 4

Graph of marker-marker LD in the studied region. The colors correspond to the LD measure D′, with red showing the strongest evidence for LD and blue the weakest. The promoter region of ME2 begins just after SNP rs1822459; ME2 ends at about SNP rs654136. Note that LD appears to become most pronounced starting just after the ME2 locus. We suggest that the second peak in association evidence seen in figures 1–3 is due to the LD with ME2 and not to an independent disease association. The plot uses all SNPs listed in table 1.