Genome-wide association study of Lp-PLA(2) activity and mass in the Framingham Heart Study

Abstract

Lipoprotein-associated phospholipase A(2) (Lp-PLA(2)) is an emerging risk factor and therapeutic target for cardiovascular disease. The activity and mass of this enzyme are heritable traits, but major genetic determinants have not been explored in a systematic, genome-wide fashion. We carried out a genome-wide association study of Lp-PLA(2) activity and mass in 6,668 Caucasian subjects from the population-based Framingham Heart Study. Clinical data and genotypes from the Affymetrix 550K SNP array were obtained from the open-access Framingham SHARe project. Each polymorphism that passed quality control was tested for associations with Lp-PLA(2) activity and mass using linear mixed models implemented in the R statistical package, accounting for familial correlations, and controlling for age, sex, smoking, lipid-lowering-medication use, and cohort. For Lp-PLA(2) activity, polymorphisms at four independent loci reached genome-wide significance, including the APOE/APOC1 region on chromosome 19 (p = 6 x 10(-24)); CELSR2/PSRC1 on chromosome 1 (p = 3 x 10(-15)); SCARB1 on chromosome 12 (p = 1x10(-8)) and ZNF259/BUD13 in the APOA5/APOA1 gene region on chromosome 11 (p = 4 x 10(-8)). All of these remained significant after accounting for associations with LDL cholesterol, HDL cholesterol, or triglycerides. For Lp-PLA(2) mass, 12 SNPs achieved genome-wide significance, all clustering in a region on chromosome 6p12.3 near the PLA2G7 gene. Our analyses demonstrate that genetic polymorphisms may contribute to inter-individual variation in Lp-PLA(2) activity and mass.

Figure 1. Q-Q plot of genome-wide P values for association with Lp-PLA₂ activity.

Regression analysis of Lp-PLA₂ activity in Caucasians from the Framingham Heart Study. Model controlled for age, age², sex, lipid-lowering medication use, cohort, and smoking status.

Figure 2. Manhattan plot of genome-wide P values for association with Lp-PLA₂ activity.

Polymorphisms in four regions reached genome-wide significance and are labeled with candidate genes in the associated region.

Figure 3. Q-Q plot of genome-wide P values for association with Lp-PLA₂ mass.

Regression analysis of Lp-PLA₂ activity in Caucasians from the Framingham Heart Study. Model controlled for age, age², sex, lipid lowering medication use, cohort, and smoking status.

Figure 4. Manhattan plot of genome-wide P values for association with Lp-PLA₂ mass.

Only the region around the PLA2G7 gene reached genome-wide significance.

Figure 5. Genomic region surrounding the top hit near APOE/APOC1.

Location of rs4420638 is shown relative to genes in the region of chromosome 19q13.

Figure 6. PLA2G7 gene and surrounding region.

Linkage disequilibrium (r²) plot across the PLA2G7 gene region, including all SNPs significantly associated with Lp-PLA₂ mass (listed). Figure was generated from HapMap Caucasian data using Haploview software. Star indicates location of the top SNP, rs1805017, in exon 4 of the PLA2G7 gene. Table to the right of the figure shows linkage disequilibrium (r²) between all of the SNPs associated with Lp-PLA₂ mass and the top hit, rs1805017, generated from Framingham data.

Figure 7. Scatter plot of Lp-PLA₂ activity against LDL cholesterol, HDL cholesterol, and triglycerides.

Units are mg/dL for all lipids/lipoproteins and nmol/ml/min for Lp-PLA₂ activity. Triglycerides are log-transformed.