Genetically regulated gene expression underlies lipid traits in Hispanic cohorts

Abstract

Plasma lipid levels are risk factors for cardiovascular disease, a leading cause of death worldwide. While many studies have been conducted in genetic variation underlying lipid levels, they mainly comprise individuals of European ancestry and thus their transferability to non-European populations is unclear. We performed genome-wide (GWAS) and imputed transcriptome-wide association studies of four lipid traits in the Hispanic Community Health Study/Study of Latinos cohort (HCHS/SoL, n = 11,103), replicated top hits in the Multi-Ethnic Study of Atherosclerosis (MESA, n = 3,855), and compared the results to the larger, predominantly European ancestry meta-analysis by the Global Lipids Genetics Consortium (GLGC, n = 196,475). In our GWAS, we found significant SNP associations in regions within or near known lipid genes, but in our admixture mapping analysis, we did not find significant associations between local ancestry and lipid phenotypes. In the imputed transcriptome-wide association study in multiple tissues and in different ethnicities, we found 59 significant gene-tissue-phenotype associations (P < 3.61×10-8) with 14 unique significant genes, many of which occurred across multiple phenotypes, tissues, and ethnicities and replicated in MESA (45/59) and in GLGC (44/59). These include well-studied lipid genes such as SORT1, CETP, and PSRC1, as well as genes that have been implicated in cardiovascular phenotypes, such as CCL22 and ICAM1. The majority (40/59) of significant associations colocalized with expression quantitative trait loci (eQTLs), indicating a possible mechanism of gene regulation in lipid level variation. To fully characterize the genetic architecture of lipid traits in diverse populations, larger studies in non-European ancestry populations are needed.

Fig 1. Genotypic principal component eigenscores of HCHS/SoL participants by self-identified region.

Each line represents an individual in HCHS/SoL connected by their eigenscores calculated in KING and colored by self-identified region. Hispanic populations are mainly admixed between Native American, West African, and European populations, resulting in a genetically diverse and structured cohort under the umbrella term “Hispanic”.

Fig 2. Conditional association of rs117961479 with CHOL in HCHS/SoL.

Conditional and joint analysis (COJO) [21] of the HCHS/SoL CHOL GWAS results revealed a genome-wide significant signal for rs117961479 (pJ = 1.5×10⁻⁹). Plotted p-values for other SNPs in the region are from a conditional GWAS where the rs117961479 genotype was used as a covariate. The color of each dot represents the SNP’s linkage disequilibrium (LD) r² with the labeled SNP in the 1000 Genomes American populations. Many nearby SNPs within or near SUGP1 and MAU2 are in high LD with rs117961479. These conditional results indicate one signal in this region, which was plotted with LocusZoom [24].

Fig 3. Predicted expression vs. observed phenotype for CETP.

Using the GTEx Artery Coronary expression prediction model, increased predicted gene expression of CETP is significantly associated with decreased observed HDL in both HCHS/SoL (A) and MESA (B) (Table 2), which is consistent with previous studies of CETP [35].

Fig 4. Colocalization of HCHS/SoL HDL GWAS and MESA eQTL signals at the CETP locus.

(A) Zoomed in Manhattan plot of the key SNPs driving the CETP association in the MESA HIS and MESA CAU PrediXcan models. Filled circles represent HDL associations in HCHS/SoL and open triangles represent eQTLs in MESA (up- or down-triangles indicate the effect allele is associated with increased or decreased CETP expression, respectively). Blue symbols represent SNPs in the MESA HIS PrediXcan model and black symbols represent SNPs in the MESA CAU PrediXcan model. Linkage disequilibrium (LD) plots are labeled with the population genotypes used to calculate r². Note several SNPs present in the Hispanic populations were monomorphic in MESA CAU and thus not included on the plot. Comparison between eQTL and HCHS/SoL HDL GWAS p-values for SNPs in either the MESA HIS (B) or MESA CAU (C) PrediXcan model. The most significant HCHS/SoL HDL GWAS SNP is the index SNP (purple diamond) in each plot. LD is calculated from the 1000 Genomes American (AMR) or European (EUR) populations. Note the index SNP is linked to the strongest eQTL signals in the MESA HIS model, but not in the MESA CAU model. Plots were generated with snp.plotter (A) and LocusCompare (B,C) [38, 39].