A genome-wide association study of red-blood cell fatty acids and ratios incorporating dietary covariates: Framingham Heart Study Offspring Cohort.

Anya Kalsbeek,Greg C Shearer,Karen Fischer,Jenna Veenstra,Nathan L Tintle,Kristin Koch,Jason Vander Woude,William S Harris,Jason Westra,Craig Disselkoen,Jacob O’Bott,Katelyn A Mckenzie,Andre Scherag

doi:10.1371/journal.pone.0194882

Abstract

Recent analyses have suggested a strong heritable component to circulating fatty acid (FA) levels; however, only a limited number of genes have been identified which associate with FA levels. In order to expand upon a previous genome wide association study done on participants in the Framingham Heart Study Offspring Cohort and FA levels, we used data from 2,400 of these individuals for whom red blood cell FA profiles, dietary information and genotypes are available, and then conducted a genome-wide evaluation of potential genetic variants associated with 22 FAs and 15 FA ratios, after adjusting for relevant dietary covariates. Our analysis found nine previously identified loci associated with FA levels (FADS, ELOVL2, PCOLCE2, LPCAT3, AGPAT4, NTAN1/PDXDC1, PKD2L1, HBS1L/MYB and RAB3GAP1/MCM6), while identifying four novel loci. The latter include an association between variants in CALN1 (Chromosome 7) and eicosapentaenoic acid (EPA), DHRS4L2 (Chromosome 14) and a FA ratio measuring delta-9-desaturase activity, as well as two loci associated with less well understood proteins. Thus, the inclusion of dietary covariates had a modest impact, helping to uncover four additional loci. While genome-wide association studies continue to uncover additional genes associated with circulating FA levels, much of the heritable risk is yet to be explained, suggesting the potential role of rare genetic variation, epistasis and gene-environment interactions on FA levels as well. Further studies are needed to continue to understand the complex genetic picture of FA metabolism and synthesis.

Highlights

Genome-wide association studies (GWAS) continue to be used in order to attempt to understand potential genetic contributions to phenotypes
In our previous GWAS, we identified five loci associated with fatty acid (FA) levels which reached genome-wide significance (TRIM58, PCOLCE2, ELOVL2, FADS genes and LPCAT3), three of which were new observations
Across the 15 FA ratios tested in models without dietary covariates, ten yielded significant associations (D5D_C20, D6D_C18, D9D_16_18, D9D_C18, ELONG5_N6, ELONG2_N6, D6D+ELONG5(N6;C18), OXD_N3, OXD_N3_N6 and OXD_N6 –see S1 Table for FA involved in each ratio), while five did not (D9D_C16, ELONG2_N3, ELONG2_N3_N6, ELONG6_SAT and ELONG6_MONO)

Summary

Introduction

Genome-wide association studies (GWAS) continue to be used in order to attempt to understand potential genetic contributions to phenotypes. Numerous studies have conducted genome-wide association analyses testing for associations with FA levels [2,3,4,5] identifying numerous loci. In our previous GWAS, we identified five loci associated with FA levels which reached genome-wide significance (TRIM58, PCOLCE2, ELOVL2, FADS genes and LPCAT3), three of which were new observations. This analysis was limited by only analyzing 14 FAs and no FA ratios, not considering dietary intake of FAs or fish oil supplements, considering only 2.5 million SNPs and only conducting single-marker tests of variant-FA association

Methods

Results

Discussion

Conclusion