|
Post by Admin on Apr 11, 2018 18:51:06 GMT
Figure 3. The structure of the human SLC24A5 gene (Chromosome 15q21.1, 48409019 to 48434692). Exons of the gene are shown in yellow, introns in blue and 5′ flanking region in pink. The black lines underneath the gene show the regions resequenced in this study (total of 11741 bp) spanning 25674 bp. rs1426654 is the functional SNP located in the third exon. Fine-scale genetic variation of SLC24A5 We resequenced 11.74 kb of SLC24A5 (Figure 3), covering all the nine exons (1617 bp), introns (5797 bp), 5′ flanking (4150 bp), and 3′ flanking (177 bp) regions (Figure 3) in a global sample set of 95 individuals (see Materials and Methods) grouped into 8 broad geographic regions. A total of 60 variable sites (including 23 singletons), one insertion, and one tetranucleotide repeat were identified with derived allele frequencies ranging from 0.005 to 0.39. Results of the resequencing study for these variable sites are presented in Table S9. According to dbSNP ( www.ncbi.nlm.nih.gov/projects/SNP/) build 137 (June 2012), 21 of these 62 identified variants were novel. The insertion present in the 5′ flanking region (position 48411803) was confined to two San individuals (San 15 and San 17). Comparison of polymorphic sites across different regions revealed that the exons of SLC24A5 are highly conserved in humans. We detected only two variable positions within exons, with rs1426654 being the only non-synonymous SNP. The other variant, a synonymous (Ser-Ser) mutation identified at exon 7 at position 48431227, was shared by four Africans. In contrast to low variation in the exonic region, a highly polymorphic tetranucleotide repeat (GAAA) was observed in the 5′ flanking region (GAAA-GA-GAAA-GAAAAA-(GAAA)n-GAAAAA-GAAAA) at position 48412029. These repeats varied from 3 to 12 copies. A detailed analysis of the repeats did not reveal any correlation with the geographical origin of the samples or the haplogroups studied, in general (Table S10). However, chromosomes belonging to haplogroup H (Figure S3), defined by the rs1426654-A allele, were associated with larger repeat lengths (7–13), albeit this association was not restricted only to them (Table S10). Figure 4. Heat map showing the intra- and inter-population variation measured by average pairwise sequence differences of the SLC24A5 gene. The upper triangle of the matrix (green) shows average pairwise differences between populations (PiXY). The average number of pairwise differences (PiX) within each population is shown along the diagonal (orange). The lower triangle of the matrix (blue) shows differences between populations based on Nei's distance, i.e., corrected average pairwise differences (PiXY−(PiX+PiY)/2). The nucleotide diversity estimated for the consensus resequenced region (11741 bp) was observed to be 0.00042±0.00004 (with Jukes-Cantor correction), which is low compared to the average of 0.00071±0.00042 for 647 genes resequenced in the NIEHS SNP database ( egp.gs.washington.edu/). A sliding window approach based on similar measures (window size = 100 bp, step size = 25 bp) for the 5′ flanking region (4150 bp) sequenced revealed that the 2726–2875 region demonstrates the highest nucleotide diversity of 0.00651 (Figure S4). Various molecular diversity indices studied for the eight geographical groups are presented in Table S11 and Figure S5. Average pairwise differences observed among and within 8 different geographical regions using 11741 bp sequence data are summarized in Figure 4. Populations from regions previously reported to exhibit a high frequency of the rs1426654-A allele (North Africa and Middle East, Central Asia, South Asia and Europe; see Figure 2) show low levels of intra- and inter-population diversity in the resequenced region (Figure 4, Table S11). Evidence for positive selection We tested if our sequence data supports the well-documented evidence of positive selection for SLC24A5 in previous studies [4], [13], [20], [21], [29], [30] and whether it provides any additional evidence of selection. None of the populations tested showed significant departure from neutrality, except for Europeans, who had negative Tajima's D (p = 0.02) and Fu and Li's F* (p = 0.04) as estimated from calibrated population genetic models using COSI (Table S12). Hence, these observations confirm that SLC24A5 has been under strong selective pressure in Europeans. In addition to this, we also performed haplotype-based selection tests based on genome-wide data (see Materials and Methods) of 1035 individuals including 145 Indians. XP-EHH scores demonstrated that SLC24A5 ranks among the top 10 candidate genes for positive selection in Europe, Middle East and Pakistan, and among the top 1% in Central Asia, Iran and North India (Table S13). Likewise, scores from our iHS analysis had significant empirical p-values for Central Asia and North India (Table S13). It is interesting to note that both of our haplotype-based selection tests demonstrated evidence of positive selection in North Indians, but no such evidence of positive selection was found in South Indians (Table S13). The difference in detecting selection signals from genotype and sequence data has also been pinpointed in a previous study [31]. Basu Mallick C, Iliescu FM, Möls M, Hill S, Tamang R, Chaubey G, et al. (2013) The Light Skin Allele of SLC24A5 in South Asians and Europeans Shares Identity by Descent. PLoS Genet9(11): e1003912.
|
|
|
Post by Admin on Sept 6, 2018 19:11:46 GMT
Cape Verde is an archipelago of ten islands located 300 miles off the coast of Western Africa. The previously uninhabited islands were discovered and colonized by the Portuguese in the 15th Century, and subsequently prospered during the transatlantic slave trade. In this population, extensive genetic admixture between the African and European ancestral populations during the last several centuries [16], [17] has facilitated assortment of pigmentary alleles. Blond or red hair color is very rare in Cape Verde, but there is a wide spectrum of variation in both eye and skin color, and individuals with dark skin and blue eyes are not infrequent. Pigmentary variation in Cape Verde also provides an opportunity to investigate and disentangle the effects of locus-specific vs. genome-wide ancestry effects, a problem that is especially relevant to admixed populations, which are poorly represented in existing GWAS efforts, and for whom health disparities often correlate with genome-wide admixture proportions. For example, in previous studies of African-American cohorts, darker skin color (used as a proxy for genome-wide ancestry) correlates with higher blood pressure and with lower socioeconomic status, but the extent to which genetic factors contribute to these correlations is unclear [18]. Previous studies of African-European skin color variation have focused mostly on candidate genes that exhibit large allele frequency differences between ancestral populations, and have led to the view that a small number of loci account for most of the phenotypic variation [13], [15], in which case skin color should correlate poorly with genome-wide ancestry. Availability of dense genotyping information and new analytical tools allow a more rigorous approach in which the effects of individual loci and genome-wide ancestry can be disentangled and comprehensively investigated. In an admixed population, the effects of individual loci on a quantitative trait can be detected either by a correlation between genotype and phenotype, or by a correlation between local ancestry and phenotype. Genotype-based approaches are expected to be more powerful for traits where the causative allele exists at similar frequencies in ancestral populations, while ancestry-based approaches should be more powerful for traits where the causative allele exhibits large frequency differences in ancestral populations. Here we apply and compare genotype-based and ancestry-based association approaches for skin and eye color in 699 Cape Verdean individuals; both approaches identify two major loci for eye color, and four major loci for skin color. Surprisingly, the genetic component with the greatest effect on skin color is not a single locus but average genomic ancestry, which, together with these four major loci, accounts for most of the estimated heritable variation. Our results indicate that Cape Verdean pigmentary variation is the result of variation in a different set of genes from those determining variation within Europe, suggest that long-range regulatory effects help to explain the relationship between skin and eye color, and highlight the potential and the pitfalls of using allele distribution patterns and signatures of selection as indicators of phenotypic differences. We first investigated the pattern and distribution of continental ancestry in 699 Cape Verdeans for whom detailed pigmentary phenotype and high density genotype information was available, the latter obtained with the Illumina 1 M platform as described below. Data from CEU and YRI HapMap individuals [19] was used to partition each Cape Verdean individual's ancestry into “European” and “African” components. We note that there is population substructure and pigmentary heterogeneity within Europe and, of course, within Africa. We also note that the true ancestral populations for Cape Verde are likely to include contributions from several areas of Southern Europe, and several areas of West Africa. In what follows, we use the terms “European” and “African” to refer to different continental origins, and emphasize that our use of these terms should not be taken to infer homogeneity of genotype or phenotype for the true ancestral populations of Cape Verde. The program frappe [20] implements a maximum likelihood clustering approach to infer individual genomic ancestry proportions. Overall, African genomic ancestry in Cape Verdeans ranges from 23.5% to 87.9%, with a median of 58%. Across different islands, the distributions of African genomic ancestry exhibit substantial overlap in range but vary in their median values, from 50.5% in Fogo to 74.4% in the capital island of Santiago (Figure 1a), which suggests a population history of extensive intercontinental admixture accompanied by reduced gene flow between islands. Figure 1. Relationship of geography and ancestry to skin and eye color. Individual ancestry proportions for Cape Verdeans displayed on all four panels were obtained from a supervised analysis in frappe with K = 2 and HapMap's CEU and YRI fixed as European and African parental populations. (a) Bar plots of individual ancestry proportions for Cape Verdeans across the islands. The width of the plots is proportional to sample size (Santiago, n = 172; Fogo, n = 129; NW cluster, n = 192; Boa Vista, n = 27). The proportion of African vs. European ancestry of the individuals is indicated by the proportion of blue vs. red color in each plot. (b) Individual African ancestry distribution in the total cohort of 685 Cape Verdeans (histogram) and in 802 African Americans (kernel density curve) from the Family Blood Pressure Program (FBPP) [21]. (c) Scatter-plot of skin color vs. Individual African ancestry proportions. Skin color is measured by the MM index described in Material and Methods. (d) Scatter-plot of eye color vs. Individual African ancestry proportions. Eye color is measured by the T-index, described in Figure 2 and Material and Methods. Points in scatter-plots are color coded according to the island of origin of the individuals. Differences in human skin color have been shaped principally and dramatically by selection [1], [2], and therefore the underlying genetic architecture may inform disease-related traits such as obesity or hypertension, for which common susceptibility alleles may have been advantageous to early humans that came to inhabit different environments [3]. But our knowledge of skin and eye color genetics is incomplete, based mostly on the limited range of phenotypic diversity represented by individuals of European ancestry [4]–[10], or on candidate gene studies in African Americans [11]–[15]. Besides serving as a model system for gene action and genetic architecture, more knowledge about human pigmentary variation has the potential to provide important insights into human diversity, disentangling biology from both scientific and social stereotypes.
|
|
|
Post by Admin on Sept 8, 2018 20:07:58 GMT
Quantitative assessment of pigmentary phenotypes For skin color, we used reflectance spectroscopy on the upper inner arm to calculate a modified melanin (MM) index, which appears normally distributed with a mean of 7.39 and standard deviation of 0.85 for the Cape Verde cohort. By comparison, MM in individuals of European ancestry is smaller and much more narrowly distributed, with a mean of 5.38 and standard deviation of 0.27 [23]. For eye color, we developed a new measure based on automated analysis of digital photographs that captures the full range of African-European variation. We observed that RGB reflectance values project onto an empirical curve that begins and ends, respectively, with very light blue, and very dark brown eyes (Figure 2). We describe the distance along this curve as the “T index”, a quantitative measure in which the categorical descriptions of “hazel”, “light brown”, and “medium brown” progressively increase in value (Figure 2). Figure 2. Quantitative assessment of eye color. Plotted are the normalized median values of green (x-axis) and blue (y-axis) levels of each individual's irises. We fitted a principal curve that explains most of the variation in the data (red dashed curve). The T-index is defined by the arc-length from the projection of each point on the curve to the end of the curve that corresponds to the lightest eye color. In the figure are examples of eye photos at their respective position in the T-index curve. The correlation between skin color and African genomic ancestry (R∧2 = 0.44) is apparent both across and within (Table S1) islands clusters, and is higher than anticipated for a trait determined by the action of a small number of genes. (For example, in a model with 3 major skin color genes that act additively and equally, we predict an R∧2160 kb in length for HERC2 (OCA2) and >90 kb in length for APBA2 [OCA2]), whereas the ancestral alleles are found on multiple, shorter, haplotypes. However, measures of linkage disequilibrium between the two loci are not significant (R∧2 = 0.036, D′ = 0.071 with Χ2 = 0.517), extended haplotypes for the derived alleles are separated by ∼800 kb, and approximately half of the European chromosomes we examined carry an ancestral HERC2 (OCA2) haplotype together with a derivative APBA2 (OCA2) haplotype, or vice versa (Figure 6c). These observations are consistent with the presence of discrete CMS peaks under the HERC2 (OCA2) and APBA2 (OCA2) regions (Figure 4d) and suggest that the loci may have undergone selection independently in European populations.
|
|
|
Post by Admin on Sept 9, 2018 18:56:52 GMT
Genetic architecture of pigmentary variation in Cape Verde The four skin color loci we identified by association analysis act in an additive fashion: we found no evidence of dominance at any of the loci, nor evidence of opposite-sign epistasis between loci. For eye color, the ancestral HERC2 (OCA2) allele is mostly dominant over the derived allele, consistent with the near recessive mode of inheritance of blue eye phenotype in Europeans (Figure S3). By contrast, the effects of SLC24A5 on eye color are semi-dominant, and no interaction was found between this gene and HERC2 (Figure S3). Quantitative assessment of the skin color loci can be considered from two perspectives: effect size in an individual, and contribution to total phenotypic variance in a population. The former depends only on genotype, and measures the strength of allelic substitution, whereas the latter is influenced by the distributions and potential correlations of allele frequencies in the specific population being studied. From the first perspective, alleles of the four major loci shift the MM index 0.16–0.33 units (in Cape Verde, skin color ranges from ∼5.5–10.5 units), and the sum of effect sizes for homozygous substitution at all four loci would shift, in an individual, an amount about the same as ∼1/3 of the total range of skin color we observed (Figure 6a). From the second perspective, the proportions of phenotypic variance attributed solely to genotype at each of the four major skin color loci are quite small, about 2% each for GRM5-TYR, APBA2 (OCA2), and SLC45A2, and about 7% for SLC24A5. However, these estimates, based on conventional regression analyses in which individual genomic ancestry is considered as a covariate, fail to consider admixture stratification, wherein admixture proportions vary widely due to recent admixing, and genotypes at unlinked loci remain correlated. For example, even though SLC24A5 and SLC45A2 lie on different chromosomes, their genotypes in the Cape Verde population are correlated with genomic ancestry and, therefore, with each other. Thus, in any particular individual, genomic ancestry has predictive value for genotype at each of the four major skin color loci, and vice versa. Figure 7. Genetic architecture of skin color variation. (a) Effect sizes of the loci associated with skin color. Effect values represent the beta values obtained from a regression model containing the four associated loci plus ancestry. (b) The pie chart represents the proportion of phenotypic variation accounted for by the different components, including non-heritable factors (∼20%), the four major loci (∼35%, color-coded as in a), and average genomic ancestry (44%). The heritable contributions were estimated by regression and variance decomposition as described in Material and Methods, and are also represented below the pie chart separately as grey (genomic ancestry) or open (four major loci) areas. However, because of admixture stratification, the heritable contributions overlap as described in the text. To capture this predictive value, we applied the epidemiological concept of population-attributable risk—the extent that an environmental risk factor contributes to phenotypic variance in a population—to determine the quantitative impact of allelic substitution at each of the four major loci. This quantity, to which we refer as “population-attributable variance”, is calculated by determining the fractional reduction in phenotypic variance that would occur if genotypes at a locus of interest were set to a common baseline in all individuals. For skin color, population-attributable variances of SLC24A5, GRM5-TYR, APBA2 (OCA2), and SLC45A2 are 18%, 7%, 9% and 7%, respectively, considerably less than the proportion of variance attributed to individual genomic ancestry, 44% (Figure 7b, also Figure 1c).
|
|
|
Post by Admin on Sept 11, 2018 19:44:12 GMT
Discussion Previous genome-wide studies of human skin and eye color have focused on populations of European ancestry and have been based primarily on categorical and subjective assessment of pigmentary phenotype [4]–[7]. The work reported here—in which the choice of population, phenotype assessment, and statistical analysis strategies were developed to provide robust information across a broad range of genetic and phenotypic backgrounds—reveals a likely new regulatory locus and mechanism of action, provides a more refined view of the underlying genetic architecture than previously appreciated, and has important implications for studying quantitative phenotypes in admixed populations. OCA2 was originally considered a strong candidate gene for Afro-European pigmentary variation based on allele distribution patterns of closely linked SNPs [11], [14], [34], [35], and because of a strong association with blue vs. brown eye color in populations of European ancestry [4], [41]. This eye color locus was later refined to a small region 50 kb upstream of OCA2 in an intron of the HERC2 gene [7], [8], [10]. Very recently, Visser et al. [33] showed that the critical SNP in this region, rs12913832, is part of an enhancer that forms a chromatin loop with the OCA2 promoter, and that the derivative allele causes reduced recruitment of the chromatin remodeler HLTF, leading to reduced binding of additional transcription factors and impairment of loop formation and OCA2 expression in cultured human melanocytes. Our work confirms the very strong effect of rs12913832 on eye color but provides little evidence for an effect on skin color, since neither the transcribed region of OCA2 nor HERC2 has a major influence on skin color in Cape Verde. Instead, the 15q13.1 skin color locus maps to an intronic region of APBA2 that lies ∼1 Mb away from OCA2, and has a haplotype set, linkage disequilibrium architecture, and signature of selection that is clearly distinct from the HERC2 (OCA2) eye color locus. Neither functional nor expression studies suggest a role for the APBA2 protein in pigmentation [42]–[44]; instead, we favor the hypothesis that the action of ABPA2 (OCA2) is analogous to that of HERC2 (OCA2), with distinct cis-acting regulatory regions that can affect eye and skin color independently. Critical evaluation of this hypothesis will require functional studies similar to those described by Visser et al. [33], but it is interesting to note that the ∼1 Mb interval between HERC2 and APBA2 contains a cluster of segmental duplications that serves as a frequent deletion breakpoint (BP3) involved in Prader-Willi and Angelman syndromes [45]. Hypopigmentation associated with these conditions is thought to be caused by hemizygosity for OCA2 [46] and is therefore independent of any potential cis-acting regulatory region in APBA2. If, however, a chromatin loop does form between APBA2 and OCA2, the BP3 region must be contained within that loop. Furthermore, the notion that expression of OCA2 can be modulated and selected separately in specific pigment cell compartments could help to explain the correlation and evolutionary history of skin and eye color. In particular, loss-of-function for OCA2 can cause blindness due to loss of melanin from retinal pigment epithelial cells [31], and we speculate that natural selection for the APBA2 region in Europeans represents cell type-specific regulation that promotes fair skin while preserving visual function. Finally, our population genetic analysis of HERC2 and APBA2, together with similar work on the HERC2 - OCA2 interval from Donnelly et al. [37], is consistent with the possibility that blue eyes were selected independently from fair skin for reasons having to do with reproductive success rather than survival. Our analysis of skin color genetic architecture generally assigns smaller contributions to individual genes than in previous work. In particular, SLC24A5 and KITLG have been reported to have effect sizes that correspond to 25%–38% and 20%–25% of the European-West African difference in skin color, respectively [13], [15]. In our work, the effect size of SLC24A5 corresponds to 13% of the range of skin color variation in Cape Verde, and we did not detect KITLG as a skin color locus. For SLC24A5, this apparent discrepancy stems from whether effect sizes are described as a fraction of the “average pigmentation difference between European-Americans and African Americans of about 30 melanin units” [13], or the entire dynamic range within a population, which is 77 melanin units in Cape Verde. Indeed, we estimate an effect size for a single SLC24A5 allele in Cape Verde that corresponds to 4.9 melanin units, compared to 4.8 melanin units estimated by Lamason et al. [13] in an African-American sample. This somewhat trivial explanation underscores the importance of how quantitative genetic effects are described, reported, and interpreted, especially when the results have potential explanatory and predictive value for morphologic traits that distinguish different biogeographic ancestries. Published: March 21, 2013https://doi.org/10.1371/journal.pgen.1003372
|
|