Humphries_1998_Arterioscler.Thromb.Vasc.Biol_18_526

The H-allele of the intron 8 HindIII polymorphism in the lipoprotein lipase (LPL) gene has been associated with a lower risk of myocardial infarction (MI) and plasma levels of triglycerides (TG). To test whether the HindIII site was in linkage disequilibrium with the functional variant LPL Serine447Stop (S447X), subjects from the European Atherosclerosis Research Study (EARS I) were genotyped for both polymorphic sites. This study included 515 offspring of fathers with a premature (<55 years old) MI, who were designated cases, and 930 age- and sex-matched control subjects from five different regions of Europe. Linkage disequilibrium between the two sites was very strong (>.99), with only three of the four possible haplotypes identified: H+S447, H-S447, and H-X447. The frequency of the H-X447 but not of the H-S447 haplotype was significantly lower in cases than in control subjects (.090 versus .117, P<.01) suggesting a protective effect for MI, with this difference being consistent in all five regions of Europe. Compared with individuals homozygous for the H+S447 haplotype, the odds ratio of having a paternal history of premature MI for H-X447 heterozygotes (approximately 20% of the population) was 0.71 (95% confidence interval, 0.55 to 0.92). In addition, there was an increase of the H-X447 haplotype frequency from north to south in control subjects (0.119 in Finland to 0.143 in the Mediterranean region, P<.01). Compared with the H+S447 haplotype, the H-X447 haplotype was associated with significantly lower concentrations of plasma TGs (5.4% lower, P=.01), with this effect being consistent over the regions of Europe. There was no significant evidence for a heterogeneity of effect between males and females or between cases and control subjects, although the effect on TG levels appeared to be the greatest in male cases (11% lower, P=.05). In a second study (EARS II), of 332 cases and 342 control subjects, postprandial clearance of TGs after a standard fat meal was examined. The H-X447 haplotype was associated with significantly lower postprandial triglyceride levels than was the H+S447 haplotype (9.4% smaller area under the curve, P<.05). Thus, the effects on MI risk and plasma lipids associated with the H allele appeared to be mainly mediated by the X447 mutation, and although the lowering effects associated with the H-X447 haplotype on fasting and postprandial TGs are not large, they are consistent with the lowering effect observed on MI risk throughout Europe.