doi: 10.1111/j.1469-1809.2006.00341.x Signatures of Positive Selection in Genes Associated with Human Skin Pigmentation as Revealed from Analyses of Single Nucleotide Polymorphisms ∗ O. Lao1,2, J. M. de Gruijter1,2, K. van Duijn1,2, A. Navarro3 and M. Kayser1 1Department of Forensic Molecular Biology, Erasmus University Medical Centre Rotterdam, The Netherlands 2Departments of Biology, Netherlands Forensic Institute, The Hague, The Netherlands 3Institucio Catalana de Reserca i Estudis Avancats (ICREA), and Unitat de Biologia Evolutiva, Departament de Ciencies de la vida i de la salut, Universitat Pompeu Fabra, Barcelona, Catalonia, Spain Summary Phenotypic variation between human populations in skin pigmentation correlates with latitude at the continental level. A large number of hypotheses involving genetic adaptation have been proposed to explain human variation in skin colour, but only limited genetic evidence for positive selection has been presented. To shed light on the evolutionary genetic history of human variation in skin colour we inspected 118 genes associated with skin pigmentation in the Perlegen dataset, studying single nucleotide polymorphisms (SNPs), and analyzed 55 genes in detail. We identified eight genes that are associated with the melanin pathway (SLC45A2, OCA2, TYRP1, DCT, KITLG, EGFR, DRD2 and PPARD) and presented significant differences in genetic variation between Europeans, Africans and Asians. In six of these genes we detected, by means of the EHH test, variability patterns that are compatible with the hypothesis of local positive selection in Europeans (OCA2, TYRP1 and KITLG) and in Asians (OCA2, DCT, KITLG, EGFR and DRD2), whereas signals were scarce in Africans (DCT, EGFR and DRD2). Furthermore, a statistically significant correlation between genotypic variation in four pigmentation candidate genes and phenotypic variation of skin colour in 51 worldwide human populations was revealed. Overall, our data also suggest that light skin colour is the derived state and is of independent origin in Europeans and Asians, whereas dark skin color seems of unique origin, reflecting the ancestral state in humans. Keywords: human pigmentation, skin color, positive selection, genetic adaptation, Perlegen database, SNP, EHH test Introduction protecting against the harmful effects of UVB radiation, is determined both by the interaction of genetic factors The skin is the largest organ in the human body and including pigmentation genes and hormones, as well as serves as a barrier between the organism and the envi- by environmental factors such as the individual’s age, ronment (Jablonski, 2004). It is involved in a wide range the region of skin (Jablonski, 2004) and the amount of of critical roles in maintaining body integrity, including UV exposure (Tadokoro et al. 2005). The pattern of defense against pathogens, homeostasis, thermoregula- melanosome distribution within the epidermis and the tion, and protection against harmful effects of UVB ra- quantity, as well as the type, of melanin comprise the diation. Skin pigmentation, which is the key factor for primary determinants of colour (Thong et al. 2003). Differences in skin pigmentation are observed within ∗ Address for correspondence and reprints: Prof. Dr. Manfred and between human populations (Jablonski, 2004). The Kayser, Department of Forensic Molecular Biology, Erasmus MC geographic distribution of phenotypic variation in skin - University Medical Centre Rotterdam, Medical-Genetic Clus- ter, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands, pigmentation tends to show sharp gradients between E-mail: [email protected] populations of different continents (Parra et al. 2004). 354 Annals of Human Genetics (2007) 71,354–369 C 2007 The Authors Journal compilation C 2007 University College London Signatures of Positive Selection at Human Pigmentation Genes Almost 85% of the total variance of skin colour is ex- Sforza & Feldman, 2003). On the other hand, it plained when human populations are grouped at the could be imagined that light skin had already arisen continental level (Relethford, 2002). Therefore, skin in Africa, for instance in the Khoisan who appear colour has been traditionally used to classify human in- in the most basal branch of a tree of worldwide dividuals into groups (Romualdi et al. 2002), despite Y chromosome diversity (Underhill et al. 2000), and the fact that the evolutionary and functional mechanism who have somewhat lighter skin colour than other shaping pigmentation differences are not sufficiently un- African groups (Jablonski & Chaplin, 2000). What also derstood. remains unclear is whether light skin pigmentation arose A large number of hypotheses involving genetic adap- independently and more than once in different popu- tation have been suggested to explain the phenotypic lations (i.e. Europeans and Asians), as well as whether variation of human skin pigmentation, including pro- some dark skinned populations (e.g. New Guineans) tection against the harmful effects of UVB radiation, derived secondarily from already lightly pigmented pop- heat load, concealment, resistance against pathogens and ulations and acquired dark pigmentation as a secondary resistance against cold injury (for a review see Robins, trait (Diamond, 2005). 1991). In addition, sexual selection has been proposed to To shed light on the evolutionary genetic history of explain the lighter constitutive pigmentation of females human skin pigmentation we performed a survey of relative to males (Aoki, 2002). Many genes were previ- genes putatively associated with pigmentation in hu- ously suggested as candidates for human skin pigmenta- mans to search for evidence of positive selection. We tion due to their involvement in human pigmentation inspected a set of 118 putative pigmentation genes in disorders such as oculocutaneous albinism (OCA), or the Perlegen SNP dataset (Hinds et al. 2005), com- in the pigmentation of model organisms, such as the prising Europeans, Africans and Asians, from which OCA2/‘p’ gene, KITLG and SLC45A2 (for a review 55 genes contained suitable SNP information for see Slominski et al. 2004). However, only in a few cases detailed analysis. have the patterns of genetic variation in genes associated with skin pigmentation been correlated with changes in melanin content (i.e. Akey et al. 2001; Shriver et al. Material and Methods 2003) and, to the best of our knowledge, only a few Gene Ascertainment genes associated with human skin pigmentation [MC1R (Rana et al. 1999; Harding et al. 2000), SLC45A2 We ascertained putative mammalian genes involved in (Soejima et al. 2006) and SLC24A5 (Lamason et al. the skin pigmentation pathway from the literature prior 2005)] have been specifically tested for evidence of posi- to June 2005 (Slominski et al. 2004; Imokawa, 2004) tive selection. A recent study based on a full genome scan and from gene expression databases (Hill et al. 2004; for signatures of positive selection using the database of UniGene Build #188), while focusing on genes re- the International HapMap project database found ev- lated to skin pathologies with a modifying effect on idence for only four genes being associated with skin skin pigmentation. The recently described SLC24A5 pigmentation OCA2, MYO5A, DTNBP1 and TYRP1 gene (Lamason et al. 2005) was not included, since the (Voight et al. 2006). gene ascertainment was carried out prior to Lamason Limited knowledge is available on the evolution- et al’s publication. We then used the Perlegen database ary history of human skin pigmentation. Dark skin (Hinds et al. 2005), comprising Africans, Europeans, and pigmentation has been suggested as the ancestral trait Asians, to study the genetic variability of SNPs within in humans (Jablonski & Chaplin, 2000). If this is each gene, as well as in the respective surrounding re- true, light skin pigmented populations could have gions of the human genome. This database was chosen arisen after humans spread from Africa into the rest over the HapMap database as the SNP ascertainment of the world, according to the “Out of Africa” hy- bias is higher in the HapMap dataset than in the Perlegen pothesis about 100,000-150,000 years ago (Cavalli- dataset (Clark et al. 2005). C 2007 The Authors Annals of Human Genetics (2007) 71,354–369 355 Journal compilation C 2007 University College London O. Lao et al. lent to Fst when this statistic is computed between pairs Testing for Interpopulation Differentiation of populations (Rosenberg et al. 2003). The mean I n We applied a sliding window approach in order to de- was computed considering all SNPs within each win- tect regions of unusual patterns of genetic variation dow, and this value was compared with an empirical between populations within the putative skin pigmen- distribution of mean I n obtained in windows of the tation genes. same size and with the same number of markers from This approach is based on comparing the observed 10,145 genes (>10 SNPs) from the entire genome and value of a particular statistic in one of the windows with described in the Perlegen database (genes on the X chro- the observed value in windows of the same size in the rest mosome were excluded because of the different effective of the genome. Regions showing an excess of statistically population size for this chromosome (Schaffner, 2004)). significant empirical p values suggest interesting candi- We computed how often the number of statistically dates for further statistical analyses (Marques-Bonet et al. significant sliding windows found for the candidate 2005). Since the average SNP density in the Perlegen genes could be found for other genes in the Perlegen dataset is one SNP in each two kb (Hinds et al. 2005), database. This allowed us to obtain estimates of the sta- each window was centered on each SNP with 2.5 kb tistical significance of the number of significant sliding on both sides of the SNP, thus including (on average) windows for gene. We defined two stringent cutoffs three SNPs per window. An alternative approach, based for considering a sliding window statistically significant: on the number of markers within the window, was not α = 0.02 and α = 0.01.