Skip to main content
Premium Trial:

Request an Annual Quote

Candidate Hair Color Variants Identified in Meta-Analysis of Canadian Cohort

Hair color

NEW YORK — Researchers have homed in on a number of candidate causal variants influencing blond, red, and brown hair in a new genetic study of Canadians.

Hair color and other pigmentation traits in people are polygenic, and hundreds of regions in the human genome have been tied to hair color. In a new analysis appearing in Communications Biology on Thursday, a team led by researchers at the University of Toronto at Mississauga conducted a meta-analysis of genome-wide association studies of hair color using data on more than 12,000 individuals from the Canadian Partnership for Tomorrow's Health, or CanPath, dataset. Through fine-mapping and additional analyses, they uncovered candidate causal variants, including several associated with expression and methylation quantitative trait loci that highlight possible pigmentation mechanisms. A transcriptome-wide association study further underscored the association of two genes, EDRB and CDK10, with hair color.

"Our main outcomes include the identification of multiple candidate causal variants through fine-mapping of significant loci across distinct regions of the genome, including putative causal variants not previously reported for hair color," senior author Esteban Parra, a professor at UT-Mississauga, and colleagues wrote in their paper.

As part of the CanPath project, nearly 13,000 individuals of European ancestry underwent genotyping on one of five genotyping arrays and self-reported their natural hair color as either black, dark brown, light brown, blond, or red. The researchers then conducted GWAS for blond versus brown or black hair color; brown versus black hair color; and red versus brown or black hair color for each of the five genotyping arrays and then bundled them into three hair color meta-analyses.

Through this, they identified a number of significant loci overlapping with or near genes known to affect variation in pigmentation. For instance, the most significant locus affecting blond hair was OCA2/HERC2 on chromosome 15, though others overlapping with SLC45A2, IRF4, MC1R, and other genes were also significant. A locus in HERC2 — in linkage disequilibrium with the one influencing blond hair — was also significantly linked to brown hair, as were others overlapping with SLC45A2 and IRF4. A SNP in HERC2, the researchers noted, has also been tied to eye color. For red hair, the only significant locus was in MC1R.

After investigating the number of independent SNPs driving the associations at those loci, the researchers fine-mapped them to zoom in on candidate causal SNPs. This, they reported, highlighted known functional pigmentation SNPs, such as rs12203592 in IRF4 for blond and brown hair. Other candidate causal SNPs included rs12913832 in the HERC2/OCA2 region and rs1805005 in MC1R for blond hair.

Some GWAS signals colocalized with eQTLs or meQTLs. For instance, five GWAS loci colocalized with meQTLs that were associated with the methylation of CpGs near MC1R, OCA2, IRF4, SLC45A2, and SLC24A4. At the same time, GWAS signals also colocalized with eQTLs of SLC45A2 and OCA2. SLC24A5, SLC45A2, and SLC24A4 are transmembrane proteins involved in ion transport, and SNPs in them have been linked to light skin pigmentation among individuals of European ancestry, the researchers noted.

Additionally, transcriptome-wide association studies using cultured melanocytes found that decreased expression of OCA2 and SCLA24A4 was associated with blond hair, as was the increased expression of EDNRB. At the same time, the decreased expression of CDK10, which is located near MC1R, was associated with red hair.

A gene set analysis further noted that most of the candidate causal SNPs are in genes involved in pigmentation processes or other signaling or synthesis pathways that affect pigmentation-related processes.

"Our results provide insights on the general mechanisms regulating pigmentation biology in humans," the researchers wrote.

They added that they are particularly interested in further studying the HERC2/OCA2 gene region as it appears to harbor a number of independent candidate variants.