October 06, 2008

SLC45A2 and human hair color

Yann Klimentidis points me to this new article which shows a link between variations in the SLC45A2 gene and hair color.

Journal of Human Genetics doi: 10.1007/s10038-008-0338-3

Association of the SLC45A2 gene with physiological human hair colour variation

Wojciech Branicki et al.


Pigmentation is a complex physical trait with multiple genes involved. Several genes have already been associated with natural differences in human pigmentation. The SLC45A2 gene encoding a transporter protein involved in melanin synthesis is considered to be one of the most important genes affecting human pigmentation. Here we present results of an association study conducted on a population of European origin, where the relationship between two non-synonymous polymorphisms in the SLC45A2 gene — rs26722 (E272K) and rs16891982 (L374F) — and different pigmentation traits was examined. The study revealed a significant association between both variable sites and normal variation in hair colour. Only L374F remained significantly associated with hair colour when both SNPs were included in a logistic regression model. No association with other pigmentation traits was detected in this population sample. Our results indicate that the rare allele L374 significantly increases the possibility of having black hair colour (OR = 7.05) and thus may be considered as a future marker for black hair colour prediction.



Kosmo said...

The study called the L374 allele rare and then indicated it could be a marker for black hair. I'm curious which population was used for the study and hence which population, exactly, this allele is rare in.

L374 could be the wild-type allele, and NW Europeans simply have a majority of the derived allele-- meaning that L374 isn't rare at all outside of NW Europe.

The vast majority of humans all have the exact same hair and eye color. (black and black) So I am curious about this study that identifies an allele for black hair and then calls it rare. I need more info.

Kosmo said...

Ah, I now see from reading Yann's post that the study was done on a Polish population. I know poland clines from light hair near the Baltic to black hair to the south. Very intersting, also, that having the black hair gene didn't prevent you from having red hair if you had two copies for red hair at the MC1R loci. This is pretty much what you'd expect, I think, if there are two different pigment systems in place-- one which controls type (red or black) and the other which controls amount (dilute or full pigment). The MC1R loci is a qualitative control loci, and the SLC45A2 is a quantitative control loci. The L374F allele simply codes for full pigment. (I suspect this will be shown to be the wild-type for the rest of the world outside Europe)

And a further prediction: those red-heads who have the 374F allele will be phenotypically distinct from red-heads without that allele by virtue of having extremely intense, "fire-engine" red hair, instead of more strawberry, or auburn red hair.

cacio said...


good points. So if you check snp entrez:

it shows that in Hapmap, the european sample (Utah, ie NW Europe) have 97% GG and 3% CG - ie as you were saying, G is almost fixed in Europe. In contrast, Africa and Asia have 100% CC.

rs26722 (same source) gives Europe 100% C, Africa predominantly (but not uniquely) C, and Asia about 50-50. So this second allele is hard to relate to color or pigmentation (unless it codes for pigmentation in Asians).

Given the percentages in Europe, I wonder what the precision would be. How many observations could he have gotten?