A genome- and phenome-wide association study to identify genetic variants influencing platelet count and volume and their pleiotropic effects

Abstract

Platelets are enucleated cell fragments derived from megakaryocytes that play key roles in hemostasis and in the pathogenesis of atherothrombosis and cancer. Platelet traits are highly heritable and identification of genetic variants associated with platelet traits and assessing their pleiotropic effects may help to understand the role of underlying biological pathways. We conducted an electronic medical record (EMR)-based study to identify common variants that influence inter-individual variation in the number of circulating platelets (PLT) and mean platelet volume (MPV), by performing a genome-wide association study (GWAS). We characterized genetic variants associated with MPV and PLT using functional, pathway and disease enrichment analyses; we assessed pleiotropic effects of such variants by performing a phenome-wide association study (PheWAS) with a wide range of EMR-derived phenotypes. A total of 13,582 participants in the electronic MEdical Records and GEnomic network had data for PLT and 6,291 participants had data for MPV. We identified five chromosomal regions associated with PLT and eight associated with MPV at genome-wide significance (P < 5E-8). In addition, we replicated 20 SNPs [out of 56 SNPs (α: 0.05/56 = 9E-4)] influencing PLT and 22 SNPs [out of 29 SNPs (α: 0.05/29 = 2E-3)] influencing MPV in a published meta-analysis of GWAS of PLT and MPV. While our GWAS did not find any new associations, our functional analyses revealed that genes in these regions influence thrombopoiesis and encode kinases, membrane proteins, proteins involved in cellular trafficking, transcription factors, proteasome complex subunits, proteins of signal transduction pathways, proteins involved in megakaryocyte development, and platelet production and hemostasis. PheWAS using a single-SNP Bonferroni correction for 1,368 diagnoses (0.05/1368 = 3.6E-5) revealed that several variants in these genes have pleiotropic associations with myocardial infarction, autoimmune, and hematologic disorders. We conclude that multiple genetic loci influence interindividual variation in platelet traits and also have significant pleiotropic effects; the related genes are in multiple functional pathways including those relevant to thrombopoiesis.

Figure 1. Manhattan and Quantile-Quantile Plots for Genome-wide Association Analysis of a) PLT (n=13, 852) and b) MPV(n=6, 291) in the eMERGE cohort

The vertical axis indicates (−log10 transformed) observed P values, and the horizontal line indicates the genome-wide significance level of p = 5E-8. In the quantile-quantile (QQ) plot, the horizontal axis shows (−log10 transformed) expected p values, and the vertical axis indicates (−log10 transformed) observed p values.

Figure 2. Regional plots of 5 loci associated with platelet count (PLT) on chromosomal regions: 3p14.3, 6q23.3, 9p24.1, 11p15.5 and 12q24.12

Single-nucleotide polymorphisms (SNPs) are plotted by position on the chromosome (x-axis) and the association with PLT (−log₁₀ P value) on the y-axis. The rs number for the most significant SNP in the joint analysis is shown on the plot. Estimated recombination rates (from HapMap) are plotted in cyan to reflect the local linkage disequilibrium (LD) structure. The SNPs near the most significant SNP are color coded to depict their LD with the lead SNP (derived using pairwise r² values from the HapMap CEU data). Genes, the position of exons and the direction of transcription from the University of California, Santa Cruz genome browser are also plotted. ▾ = nonsynonymous; ○ = no annotation; ⊠ = conserved in mammals; * = conserved TFBS (transcription factor binding site). cM/Mb = centimorgan/megabase; Mb = megabase.

Figure 3. Regional plots of 8 loci associated with mean platelet volume (MPV) in chromosomal regions: 1q24.3, 1q32.1, 3p14.3, 7q22.3, 10q21.2, 12q13.3, 12q24.1 and 17q11.2

Single-nucleotide polymorphisms (SNPs) are plotted by position on the chromosome (x-axis) and the association with MPV (−log₁₀ P value) on the y-axis. The rs number for the most significant SNP in the joint analysis is shown on the plot. Estimated recombination rates (from HapMap) are plotted in cyan to reflect the local linkage disequilibrium (LD) structure. The SNPs near the most significant SNP are color coded to depict their LD with the lead SNP (derived using pairwise r² values from the HapMap CEU data). Genes, the position of exons and the direction of transcription from the University of California, Santa Cruz genome browser are also plotted. ▾ = nonsynonymous; ○ = no annotation; ⊠ = conserved in mammals; * = conserved TFBS (transcription factor binding site). cM/Mb = centimorgan/megabase; Mb = megabase.

Figure 4. Manhattan plots summarizing results of a phenome-wide association analyses of A) platelet count and B) mean platelet volume

Results are from logistic regression analysis for 1368 EMR-defined phenotypes in 13,688 individuals, adjusted for age and gender. Broad phenotypic categories (X-axis) and −log10 (P-values) (Y-axis) are provided. Horizontal line indicates threshold for statistical significance (P-value 3.6E-5).

Figure 5. Functional repertoire of genes associated with PLT and MPV mapped to the platelet life cycle

Genes associated with variants influencing PLT and MPV are mapped to various stages of platelet biogenesis.