Citation: Barsh GS (2003) What Controls Variation in Human Skin Color? PLoS Biol 1(1): e27. https://doi.org/10.1371/journal.pbio.0000027
Published: October 13, 2003
Copyright: © 2003 Public Library of Science. This is an open-access article distributed under the terms of the Public Library of Science Open-Access License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Diversity of human appearance and form has intrigued biologists for centuries, but nearly 100 years after the term “genetics” was coined by William Bateson in 1906, the genes that underlie this diversity are an unsolved mystery. One of the most obvious phenotypes that distinguish members of our species, differences in skin pigmentation, is also one of the most enigmatic. There is a tremendous range of human skin color in which variation can be correlated with climates, continents, and/or cultures, yet we know very little about the underlying genetic architecture. Is the number of common skin color genes closer to five, 50, or 500? Do gain- and loss-of-function alleles for a small set of genes give rise to phenotypes at opposite ends of the pigmentary spectrum? Has the effect of natural selection on similar pigmentation phenotypes proceeded independently via similar pathways? And, finally, should we care about the genetics of human pigmentation if it is only skin-deep?
Why Should We Care?
From a clinical perspective, inadequate protection from sunlight has a major impact on human health (Armstrong et al. 1997; Diepgen and Mahler 2002). In Australia, the lifetime cumulative incidence of skin cancer approaches 50%, yet the oxymoronic “smart tanning” industry continues to grow, and there is controversy over the extent to which different types of melanin can influence susceptibility to ultraviolet (UV) radiation (Schmitz et al. 1995; Wenczl et al. 1998). At the other end of the spectrum, inadequate exposure to sunlight, leading to vitamin D deficiency and rickets, has been mostly cured by nutritional advances made in the early 1900s. In both cases, understanding the genetic architecture of human skin color is likely to provide a greater appreciation of underlying biological mechanisms, much in the same way that mutational hotspots in the gene TP53 have helped to educate society about the risks of tobacco (Takahashi et al. 1989; Toyooka et al. 2003).
From a basic science perspective, variation in human skin color represents an unparalleled opportunity for cell biologists, geneticists, and anthropologists to learn more about the biogenesis and movement of subcellular organelles, to better characterize the relationship between genotypic and phenotypic diversity, to further investigate human origins, and to understand how recent human evolution may have been shaped by natural selection.
The Color Variation Toolbox
Historically, measurement of human skin color is often based on subjective categories, e.g., “moderate brown, rarely burns, tans very easily.” More recently, quantitative methods based on reflectance spectrophotometry have been applied, which allow reddening caused by inflammation and increased hemoglobin to be distinguished from darkening caused by increased melanin (Alaluf et al. 2002b; Shriver and Parra 2000; Wagner et al. 2002). Melanin itself is an organic polymer built from oxidative tyrosine derivatives and comes in two types, a cysteine-rich red–yellow form known as pheomelanin and a less-soluble black--brown form known as eumelanin (Figure 1A). Discriminating among pigment types in biological samples requires chemical extraction, but is worth the effort, since the little we do know about common variation in human pigmentation involves pigment type-switching. The characteristic phenotype of fair skin, freckling, and carrot-red hair is associated with large amounts of pheomelanin and small amounts of eumelanin and is caused by loss-of-function alleles in a single gene, the melanocortin 1 receptor (MC1R) (Sturm et al. 1998; Rees 2000) However, MC1R variation has a significant effect on pigmentation only in populations where red hair and fair skin are common (Rana et al. 1999; Harding et al. 2000), and its primary effects—to promote eumelanin synthesis at the expense of pheomelanin synthesis, or vice versa— contribute little to variation of skin reflectance among or between major ethnic groups (Alaluf et al. 2002a).
(A) Activation of the melanocortin 1 receptor (MC1R) promotes the synthesis of eumelanin at the expense of pheomelanin, although oxidation of tyrosine by tyrosinase (TYR) is required for synthesis of both pigment types. The membrane-associated transport protein (MATP) and the pink-eyed dilution protein (P) are melanosomal membrane components that contribute to the extent of pigment synthesis within melanosomes. (B) There is a gradient of melanosome size and number in dark, intermediate, and light skin; in addition, melanosomes of dark skin are more widely dispersed. This diagram is based on one published by Sturm et al. (1998) and summarizes data from Szabo et al. (1969), Toda et al. (1972), and Konrad and Wolff (1973) based on individuals whose recent ancestors were from Africa, Asia, or Europe.
More important than the ratio of melanin types is the total amount of melanin produced. In addition, histological characteristics of different-colored skin provide some clues as to cellular mechanisms that are likely to drive pigmentary variation (Figure 1B). For the same body region, light- and dark-skinned individuals have similar numbers of melanocytes (there is considerable variation between different body regions), but pigment-containing organelles, called melanosomes, are larger, more numerous, and more pigmented in dark compared to intermediate compared to light skin, corresponding to individuals whose recent ancestors were from Africa, Asia, or Europe, respectively (Szabo et al. 1969; Toda et al. 1972; Konrad and Wolff 1973). From these perspectives, oxidative enzymes like tyrosinase (TYR), which catalyzes the formation of dopaquinone from tyrosine, or melanosomal membrane components like the pink-eyed dilution protein (P) or the membrane-associated transporter protein (MATP), which affect substrate availability and activity of TYR (Orlow and Brilliant 1999; Brilliant and Gardner 2001; Newton et al. 2001; Costin et al. 2003), are logical candidates upon which genetic variation could contribute to the diversity of human skin color.
Of equal importance to what happens inside melanocytes is what happens outside. Each pigment cell actively transfers its melanosomes to about 40 basal keratinocytes; ultimately, skin reflectance is determined by the amount and distribution of pigment granules within keratinocytes rather than melanocytes. In general, melanosomes of African skin are larger and dispersed more widely than in Asian or European skin (Figure 1). Remarkably, keratinocytes from dark skin cocultured with melanocytes from light skin give rise to a melanosome distribution pattern characteristic of dark skin, and vice versa (Minwalla et al. 2001). Thus, at least one component of skin color variation represents a gene or genes whose expression and action affect the pigment cell environment rather than the pigment cell itself.
Genetics of Skin Color
For any quantitative trait with multiple contributing factors, the most important questions are the overall heritability, the number of genes likely to be involved, and the best strategies for identifying those genes. For skin color, the broad sense heritability (defined as the overall effect of genetic vs. nongenetic factors) is very high (Clark et al. 1981), provided one is able to control for the most important nongenetic factor, exposure to sunlight.
Statements regarding the number of human skin color genes are attributed to several studies; one of the most complete is by Harrison and Owen (1964). In that study, skin reflectance measurements were obtained from 70 residents of Liverpool whose parents, grandparents, or both were of European (“with a large Irish component”) or West African (“mostly from coastal regions of Ghana and Nigeria”) descent and who were roughly classified into “hybrid” and “backcross” groups on this basis. An attempt to partition and analyze the variance of the backcross groups led to minimal estimates of three to four “effective factors,” in this case, independently segregating genes. Aside from the key word minimal (Harrison and Owen's data could also be explained by 30–40 genes), one of the more interesting findings was that skin reflectance appeared to be mainly additive. In other words, mean skin reflectance of “F1 hybrid” or “backcross hybrid” groups is intermediate between their respective parental groups.
An alternative approach for considering the number of potential human pigmentation genes is based on mouse coat color genetics, one of the original models to define and study gene action and interaction, for which nearly 100 different genes have been recognized (Bennett and Lamoreux 2003; Jackson 1994). Setting aside mouse mutations that cause white spotting or predominant effects outside the pigmentary system, no more than 15 or 20 mutations remain, many of which have been identified and characterized, and most of which have human homologs in which null mutations cause albinism.
This brings us to the question of candidate genes for skin color, since, like any quantitative trait, a reasonable place to start is with rare mutations known to cause an extreme phenotype, in this case Mendelian forms of albinism. The underlying assumption is that if a rare null allele causes a complete loss of pigment, then a set of polymorphic, i.e., more frequent, alleles with subtle effects on gene expression will contribute to a spectrum of skin colors. The TYR, P, and MATP genes discussed earlier are well-known causes of albinism whose primary effects are limited to pigment cells (Oetting and King 1999); among these, the P gene is highly polymorphic but the phenotypic consequences of P gene polymorphisms are not yet known.
Independent of phenotype, a gene responsible for selection of different skin colors should exhibit a population signature with a large number of alleles and rates of sequence substitution that are greater for nonsynonymous (which change an amino acid in the protein) than synonymous (which do not change any amino acid) alterations. Data have been collected only for MC1R, in which the most notable finding is a dearth of allelic diversity in African samples, which is remarkable given that polymorphism for most genes is greater in Africa than in other geographic regions (Rana et al. 1999; Harding et al. 2000). Thus, while MC1R sequence variation does not contribute significantly to variation in human skin color around the world, a functional MC1R is probably important for dark skin.
Selection for Skin Color?
Credit for describing the relationship between latitude and skin color in modern humans is usually ascribed to an Italian geographer, Renato Basutti, whose widely reproduced “skin color maps” illustrate the correlation of darker skin with equatorial proximity (Figure 2). More recent studies by physical anthropologists have substantiated and extended these observations; a recent review and analysis of data from more than 100 populations (Relethford 1997) found that skin reflectance is lowest at the equator, then gradually increases, about 8% per 10° of latitude in the Northern Hemisphere and about 4% per 10° of latitude in the Southern Hemisphere. This pattern is inversely correlated with levels of UV irradiation, which are greater in the Southern than in the Northern Hemisphere. An important caveat is that we do not know how patterns of UV irradiation have changed over time; more importantly, we do not know when skin color is likely to have evolved, with multiple migrations out of Africa and extensive genetic interchange over the last 500,000 years (Templeton 2002).
(A) A traditional skin color map based on the data of Biasutti. Reproduced from http://anthro.palomar.edu/vary/ with permission from Dennis O'Neil. Erratum note: The source of this image was incorrectly acknowledged. Corrected 12/19/03. (B) Summary of 102 skin reflectance samples for males as a function of latitude, redrawn from Relethford (1997).
Regardless, most anthropologists accept the notion that differences in UV irradiation have driven selection for dark human skin at the equator and for light human skin at greater latitudes. What remains controversial are the exact mechanisms of selection. The most popular theory posits that protection offered by dark skin from UV irradiation becomes a liability in more polar latitudes due to vitamin D deficiency (Murray 1934). UVB (short-wavelength UV) converts 7-dehydrocholesterol into an essential precursor of cholecaliferol (vitamin D3); when not otherwise provided by dietary supplements, deficiency for vitamin D causes rickets, a characteristic pattern of growth abnormalities and bony deformities. An oft-cited anecdote in support of the vitamin D hypothesis is that Arctic populations whose skin is relatively dark given their latitude, such as the Inuit and the Lapp, have had a diet that is historically rich in vitamin D. Sensitivity of modern humans to vitamin D deficiency is evident from the widespread occurrence of rickets in 19th-century industrial Europe, but whether dark-skinned humans migrating to polar latitudes tens or hundreds of thousands of years ago experienced similar problems is open to question. In any case, a risk for vitamin D deficiency can only explain selection for light skin. Among several mechanisms suggested to provide a selective advantage for dark skin in conditions of high UV irradiation (Loomis 1967; Robins 1991; Jablonski and Chaplin 2000), the most tenable are protection from sunburn and skin cancer due to the physical barrier imposed by epidermal melanin.
Solving the Mystery
Recent developments in several areas provide a tremendous opportunity to better understand the diversity of human pigmentation. Improved spectrophotometric tools, advances in epidemiology and statistics, a wealth of genome sequences, and efficient techniques for assaying sequence variation offer the chance to replace misunderstanding and myths about skin color with education and scientific insight. The same approaches used to investigate traits such as hypertension and obesity—genetic linkage and association studies—can be applied in a more powerful way to study human pigmentation, since the sources of environmental variation can be controlled and we have a deeper knowledge of the underlying biochemistry and cell biology.
This approach is especially appealing given the dismal success rate in molecular identification of complex genetic diseases. In fact, understanding more about the genetic architecture of skin color may prove helpful in designing studies to investigate other quantitative traits. Current debates in the human genetics community involve strategies for selecting populations and candidate genes to study, the characteristics of sequence polymorphisms worth pursuing as potential disease mutations, and the extent to which common diseases are caused by common (and presumably ancient) alleles. While specific answers will be different for every phenotype, there may be common themes, and some answers are better than none.
Harrison and Owen concluded their 1964 study of human skin color by stating, “The deficiencies in the data in this study are keenly appreciated by the writers, but since there appear at present to be no opportunities for improving the data, it seems justifiable to take the analysis as far as possible.” Nearly 40 years later, opportunities abound, and the mystery of human skin color is ready to be solved.
I am grateful to members of my laboratory and colleagues who study pigment cells in a variety of different experimental organisms for useful discussions and to Sophie Candille for helpful comments on the manuscript. Many of the ideas presented here emerged during a discussion series on Unsolved Mysteries in Biomedical Research that was initiated by Mark Krasnow and the Medical Scientist Training Program at Stanford University.
- 1. Alaluf S, Atkins D, Barrett K, Blount M, Carter N, et al. (2002a) Ethnic variation in melanin content and composition in photoexposed and photoprotected human skin. Pigment Cell Res 15: 112–118.S. AlalufD. AtkinsK. BarrettM. BlountN. CarterEthnic variation in melanin content and composition in photoexposed and photoprotected human skin.Pigment Cell Res2002a15112118
- 2. Alaluf S, Atkins D, Barrett K, Blount M, Carter N, et al. (2002b) The impact of epidermal melanin on objective measurements of human skin colour. Pigment Cell Res 15: 119–126.S. AlalufD. AtkinsK. BarrettM. BlountN. CarterThe impact of epidermal melanin on objective measurements of human skin colour.Pigment Cell Res2002b15119126
- 3. Armstrong BK, Kricker A, English DR (1997) Sun exposure and skin cancer. Australas J Dermatol 38: Suppl 1S1–S6.BK ArmstrongA. KrickerDR EnglishSun exposure and skin cancer.Australas J Dermatol199738Suppl 1S1S6
- 4. Bennett DC, Lamoreux ML (2003) The color loci of mice—A genetic century. Pigment Cell Res 16: 333–344.DC BennettML LamoreuxThe color loci of mice—A genetic century.Pigment Cell Res200316333344
- 5. Brilliant M, Gardner LJ (2001) Melanosomal pH, pink locus protein and their roles in melanogenesis. J Invest Dermatol 117: 386–387.M. BrilliantLJ GardnerMelanosomal pH, pink locus protein and their roles in melanogenesis.J Invest Dermatol2001117386387
- 6. Clark P, Stark AE, Walsh RJ, Jardine R, Martin NG (1981) A twin study of skin reflectance. Ann Hum Biol 8: 529–541.P. ClarkAE StarkRJ WalshR. JardineNG MartinA twin study of skin reflectance.Ann Hum Biol19818529541
- 7. Costin GE, Valencia JC, Vieira WD, Lamoreux ML, Hearing VJ (2003) Tyrosinase processing and intracellular trafficking is disrupted in mouse primary melanocytes carrying the underwhite (uw) mutation: A model for oculocutaneous albinism (OCA) type 4. J Cell Sci 116: 3203–3212.GE CostinJC ValenciaWD VieiraML LamoreuxVJ HearingTyrosinase processing and intracellular trafficking is disrupted in mouse primary melanocytes carrying the underwhite (uw) mutation: A model for oculocutaneous albinism (OCA) type 4.J Cell Sci200311632033212
- 8. Diepgen TL, Mahler V (2002) The epidemiology of skin cancer. Br J Dermatol 146: Suppl 611–6.TL DiepgenV. MahlerThe epidemiology of skin cancer.Br J Dermatol2002146Suppl 6116
- 9. Harding RM, Healy E, Ray AJ, Ellis NS, Flanagan N, et al. (2000) Evidence for variable selective pressures at MC1R. Am J Hum Genet 66: 1351–1361.RM HardingE. HealyAJ RayNS EllisN. FlanaganEvidence for variable selective pressures at MC1R.Am J Hum Genet20006613511361
- 10. Harrison GA, Owen JJT (1964) Studies on the inheritance of human skin colour. Ann Hum Genet 28: 27–37.GA HarrisonJJT OwenStudies on the inheritance of human skin colour.Ann Hum Genet1964282737
- 11. Jablonski NG, Chaplin G (2000) The evolution of human skin coloration. J Hum Evol 39: 57–106.NG JablonskiG. ChaplinThe evolution of human skin coloration.J Hum Evol20003957106
- 12. Jackson IJ (1994) Molecular and developmental genetics of mouse coat color. Annu Rev Genet 28: 189–217.IJ JacksonMolecular and developmental genetics of mouse coat color.Annu Rev Genet199428189217
- 13. Konrad K, Wolff K (1973) Hyperpigmentation, melanosome size, and distribution patterns of melanosomes. Arch Dermatol 107: 853–860.K. KonradK. WolffHyperpigmentation, melanosome size, and distribution patterns of melanosomes.Arch Dermatol1973107853860
- 14. Loomis WF (1967) Skin-pigment regulation of vitamin-D biosynthesis in man. Science 157: 501–506.WF LoomisSkin-pigment regulation of vitamin-D biosynthesis in man.Science1967157501506
- 15. Minwalla L, Zhao Y, Le Poole IC, Wickett RR, Boissy RE (2001) Keratinocytes play a role in regulating distribution patterns of recipient melanosomes in vitro. J Invest Dermatol 117: 341–347.L. MinwallaY. ZhaoIC Le PooleRR WickettRE BoissyKeratinocytes play a role in regulating distribution patterns of recipient melanosomes in vitro.J Invest Dermatol2001117341347
- 16. Murray FG (1934) Pigmentation, sunlight, and nutritional disease. Am Anthropologist 36: 438–445.FG MurrayPigmentation, sunlight, and nutritional disease.Am Anthropologist193436438445
- 17. Newton JM, Cohen-Barak O, Hagiwara N, Gardner JM, Davisson MT, et al. (2001) Mutations in the human orthologue of the mouse underwhite gene (uw) underlie a new form of oculocutaneous albinism, OCA4. Am J Hum Genet 69: 981–988.JM NewtonO. Cohen-BarakN. HagiwaraJM GardnerMT DavissonMutations in the human orthologue of the mouse underwhite gene (uw) underlie a new form of oculocutaneous albinism, OCA4.Am J Hum Genet200169981988
- 18. Oetting WS, King RA (1999) Molecular basis of albinism: Mutations and polymorphisms of pigmentation genes associated with albinism. Hum Mutat 13: 99–115.WS OettingRA KingMolecular basis of albinism: Mutations and polymorphisms of pigmentation genes associated with albinism.Hum Mutat19991399115
- 19. Orlow SJ, Brilliant MH (1999) The pink-eyed dilution locus controls the biogenesis of melanosomes and levels of melanosomal proteins in the eye. Exp Eye Res 68: 147–154.SJ OrlowMH BrilliantThe pink-eyed dilution locus controls the biogenesis of melanosomes and levels of melanosomal proteins in the eye.Exp Eye Res199968147154
- 20. Rana BK, Hewett-Emmett D, Jin L, Chang BH, Sambuughin N, et al. (1999) High polymorphism at the human melanocortin 1 receptor locus. Genetics 151: 1547–1557.BK RanaD. Hewett-EmmettL. JinBH ChangN. SambuughinHigh polymorphism at the human melanocortin 1 receptor locus.Genetics199915115471557
- 21. Rees JL (2000) The melanocortin 1 receptor (MC1R): More than just red hair. Pigment Cell Res 13: 135–140.JL ReesThe melanocortin 1 receptor (MC1R): More than just red hair.Pigment Cell Res200013135140
- 22. Relethford JH (1997) Hemispheric difference in human skin color. Am J Phys Anthropol 104: 449–457.JH RelethfordHemispheric difference in human skin color.Am J Phys Anthropol1997104449457
- 23. Robins AH, editor. (1991) Biological perspectives on human pigmentation. Cambridge: Cambridge University Press. 253 p.AH RobinsBiological perspectives on human pigmentation1991CambridgeCambridge University Press
- 24. Schmitz S, Thomas PD, Allen TM, Poznansky MJ, Jimbow K (1995) Dual role of melanins and melanin precursors as photoprotective and phototoxic agents: Inhibition of ultraviolet radiation-induced lipid peroxidation. Photochem Photobiol 61: 650–655.S. SchmitzPD ThomasTM AllenMJ PoznanskyK. JimbowDual role of melanins and melanin precursors as photoprotective and phototoxic agents: Inhibition of ultraviolet radiation-induced lipid peroxidation.Photochem Photobiol199561650655
- 25. Shriver MD, Parra EJ (2000) Comparison of narrow-band reflectance spectroscopy and tristimulus colorimetry for measurements of skin and hair color in persons of different biological ancestry. Am J Phys Anthropol 112: 17–27.MD ShriverEJ ParraComparison of narrow-band reflectance spectroscopy and tristimulus colorimetry for measurements of skin and hair color in persons of different biological ancestry.Am J Phys Anthropol20001121727
- 26. Sturm RA, Box NF, Ramsay M (1998) Human pigmentation genetics: The difference is only skin deep. Bioessays 20: 712–721.RA SturmNF BoxM. RamsayHuman pigmentation genetics: The difference is only skin deep.Bioessays199820712721
- 27. Szabo G, Gerald AB, Pathak MA, Fitzpatrick TB (1969) Racial differences in the fate of melanosomes in human epidermis. Nature 222: 1081–1082.G. SzaboAB GeraldMA PathakTB FitzpatrickRacial differences in the fate of melanosomes in human epidermis.Nature196922210811082
- 28. Takahashi T, Nau MM, Chiba I, Birrer MJ, Rosenberg RK, et al. (1989) p53: A frequent target for genetic abnormalities in lung cancer. Science 246: 491–494.T. TakahashiMM NauI. ChibaMJ BirrerRK Rosenbergp53: A frequent target for genetic abnormalities in lung cancer.Science1989246491494
- 29. Templeton A (2002) Out of Africa again and again. Nature 416: 45–51.A. TempletonOut of Africa again and again.Nature20024164551
- 30. Toda K, Pathak MA, Parrish JA, Fitzpatrick TB, Quevedo WC (1972) Alteration of racial differences in melanosome distribution in human epidermis after exposure to ultraviolet light. Nat New Biol 236: 143–145.K. TodaMA PathakJA ParrishTB FitzpatrickWC QuevedoAlteration of racial differences in melanosome distribution in human epidermis after exposure to ultraviolet light.Nat New Biol1972236143145
- 31. Toyooka S, Tsuda T, Gazdar AF (2003) The TP53 gene, tobacco exposure, and lung cancer. Hum Mutat 21: 229–239.S. ToyookaT. TsudaAF GazdarThe TP53 gene, tobacco exposure, and lung cancer.Hum Mutat200321229239
- 32. Wagner JK, Jovel C, Norton HL, Parra EJ, Shriver MD (2002) Comparing quantitative measures of erythema, pigmentation and skin response using reflectometry. Pigment Cell Res 15: 379–384.JK WagnerC. JovelHL NortonEJ ParraMD ShriverComparing quantitative measures of erythema, pigmentation and skin response using reflectometry.Pigment Cell Res200215379384
- 33. Wenczl E, van der Schans GP, Roza L, Kolb RM, Timmerman AJ, et al. (1998) (Pheo)melanin photosensitizes UVA-induced DNA damage in cultured human melanocytes. J Invest Dermatol 111: 678–682.E. WenczlGP van der SchansL. RozaRM KolbAJ Timmerman(Pheo)melanin photosensitizes UVA-induced DNA damage in cultured human melanocytes.J Invest Dermatol1998111678682