Skin color of Indians (and others)

Indians more than any other ethnic group are aware of the dramatic variation in their skin color. This may be seen right in the same family (especially in brahmin households from the southern part of the peninsula) with all the shades ranging from fair as Western Eurasians to dark as sub-Saharan Africans. What plays out in a family also plays out over the entire country. There are some well-known trends: 1) The gradient from north to south of the sub-continent with the extreme Northerners close to European complexions and extreme Southerners more closer to sub-Saharan Africans. 2) A less visible trend that is still in need of more detailed study is the gradation between the varNa and avarNa populations, with the latter tending to have more dark individuals and possibly being darker on an average and than the former.

World skin color distribution of native populations (I believe this is not entirely accurate but a an approximation that serves for illustrative purposes).

In a sense India reflects the variation that is seen throughout the world all within a subcontinent. Now on the global scale the populations closer to the equator tend to be dark, while those from higher latitudes are fair. In Eurasia, two otherwise only distantly related populations are characterized by fair skin: the Europeans, and the Northern East Asians like Chinese, Japanese and Koreans. All these have for long suggested a major selective role played by solar radiation damage and vitamin D production in shaping the biogeography of skin. The quest to understand the mysteries of Indian skin color genetics began with Haldane’s shot at the dark in the form of a 4 gene model to explain the majority of skin color diversity in India. But we have now come close to a real understanding with several new papers supplying nuggets of useful data on genetics of skin color. I paste some of those references below and provide a synthetic account with some related observations of mine.

The mouse below is the Underwhite mutant (SLC45A2 gene). The fish on top is the zebrafish golden mutant. The human on the left is the corresponding human natural variant (both in SLC24A5 gene).

There are only two major biological purveyors of skin color shared by all humans– melanin the skin pigment and hemoglobin the blood pigment. A combination of the reflectivities of these two compounds colors skin. In East Asians, additionally the layer of fat below the skin may provide an additional yellowish tinge. Melanin is an ancient pigment in vertebrates produced by melanocytes from the amino acid tyrosine (and cysteine in the case of the brown pigments) in the derived lysosomes termed melanosomes. Melanosomes are transported along the dendritic projections of the melanocytes (which in someways resemble neural cells being derived as sister cells in the neural crest) to neighboring keratinocytes in skin, which are directly responsible for skin and hair color. Earlier studies on mouse pigmentation identified over 100 genes several orthologs of which are also mutated in human pigmentation disorders. More recently the zebrafish golden gene was sequenced and its human ortholog was found to show an allele that in humans shows a mutation that correlates with lighter skin color. Thus a common pathway of melanocyte function and skin pigmentation appears to function throughout vertebrates.

The main results of the studies on Indian skin color genetics are thus:

-3 genes account for a major component of color variation amongst Indians:
SLC24A5 (ortholog of zebrafish golden), TYR (ortholog of mouse Albino gene) and SLC45A2 (ortholog of mouse Underwhite).

-These genes are also been implicated in color variation between Europeans and Africans. Light skin in Europeans and dark skin in Africans is due to alternate alleles in the same genes. At least some of the same alleles responsible for this color difference between Europeans and Africans seem to be involved in Indians. The SLC45A2 shows an allelic gradient from north to south in Europe. It might similarly be different in distribution between north and south Indians, with decrease in frequency in South India.

-While the above allelic variation responsible for skin color variation in Indians is as seen in Europeans and Central Asians it does not match the situation in Asians (people from East like Chinese Japanese Koreans etc). Here different set of genes appear to have convergently given rise to light skin color.

The region around the SLC24A5 gene shows a considerable reduction of heterozygosity suggesting a major selective sweep of this light skinned allele through the population. Interestingly, it appears that the light-skin in Indians and Central/Western Eurasians had a common origin independent of those in Asians, especially in terms of the contribution of the SLC24A5 allele. In a general sense this is not very surprising given that the Indian varNa populations show a closer relationship and Central/Western Eurasian populations to the exclusion of Asians. But what does this mean? After all several Indians who have the SLC24A5 A111T “light allele” (SNP: rs1426654) are not exactly in sun-starved climes where low vitamin D could trigger a major selective sweep. In fact it is found even in Shri-Lankans in fairly significant frequency. This might suggest that indeed it is ancestry rather than adaptation that has resulted in deep South Indian possessing this SLC24A5 allele in the observed frequencies. It means that Indians, including many from South India and Shri Lanka might have had ancestors from Northern latitudes. The results from these studies also suggest that in South India and Lanka there might actually be selection by UV-damage against the “light alleles” like that of SLC45A2 keeping its frequency low (Given that it is much lower than expected when compared to SLC24A5. In fact Soejima et al propose that it might be a useful forensic marker for Lankans and Tamils). Yet the retention of certain percentage of light alleles in tropical India might also suggest a sexual selection for such alleles.

So what do these genes do?

TYR encodes the tyrosinase which catalyze the oxidation of phenols to quinones in the first step of melanin biosynthesis. So evident the allele in Indian populations directly influence the amount or efficiency of the enzyme at first step. SLC24A5 and the SLC45A2 genes are more mysterious transporter proteins. SLC24A5 protein is a cation antiporter that is likely to exchange 4 Na+ for Ca+ and K+. The SLC45A2 protein is a member of the major facilitator superfamily of transporters that performs symport of sugar and sugar derivatives with Na+. This suggests that the SLC24A5 and SLC45A2 are functionally linked via the transport of Na+. Workers have suggested that the ion transport by these proteins may be related to efficiency of melanosome function by affecting pH (which is different in white and black skin) and providing calcium for proteolysis of the SILV protein.

However, I suspect that they are ignoring the issue of sugar transport by SLC45A2. The SLC24A5 A111T mutation maps to a loop between the TM regions of the transporter and the ancestral A in this position is highly conserved, and is seen in orthologous proteins in plants. Thus this state was present common ancestor of plants and animals, suggesting a key functional role in formation of a structure in the soluble part of the transporter. Hence, the A111T mutation probably seriously compromises function of this promoter. Likewise the SLC45A2 L374F mutation is in a soluble loop and the position is alway occupied by an aliphatic side chain residue in orthologous proteins, but never aromatic. Thus, again the L374F probably seriously affect the effective functioning of the transporter. We wonder, given that the melanosome is a derived lysosome, if this transport system ultimately affects the transport of a sugar. The transport of this sugar might be critical for processing of glycosylated proteins in the melanosome and there by their appropriate localization. Future investigation of this angle might be appropriate.

The key papers:
Norton HL, Kittles RA, Parra E, McKeigue P, Mao X, Cheng K, Canfield VA,
Bradley DG, McEvoy B, Shriver MD.
Genetic evidence for the convergent evolution of light skin in Europeans and East Asians.
Mol Biol Evol. 2007 Mar;24(3):710-22.

Soejima M, Koda Y.
Population differences of two coding SNPs in pigmentation-related genes SLC24A5 and SLC45A2.
Int J Legal Med. 2007 Jan;121(1):36-9

A genome-wide association study of skin pigmentation in a South Asian population
Renee P. Stokowski, P.V. Krishna Pant, Tony Dadd, Amelia Fereday, David A. Hinds, Carl Jarman, Wendy Filsell, Rebecca S. Ginger, Martin R. Green, Frans J. van der Ouderaa, David R. Cox
The American Journal of Human Genetics in press

This entry was posted in Scientific ramblings. Bookmark the permalink.