Subjects

Abstract

We present results from a genome-wide association study for variants associated with human pigmentation characteristics among 5,130 Icelanders, with follow-up analyses in 2,116 Icelanders and 1,214 Dutch individuals. Two coding variants in TPCN2 are associated with hair color, and a variant at the ASIP locus shows strong association with skin sensitivity to sun, freckling and red hair, phenotypic characteristics similar to those affected by well-known mutations in MC1R.

Main

We recently reported a genome-wide study identifying six loci associated with variation in hair, eye and skin pigmentation in Europeans1. To search for additional sequence variants correlating with human pigmentation, we expanded the genome-wide scan from 2,986 individuals to 5,130. The findings of this discovery phase were followed up in 2,116 Icelanders and 1,214 Dutch individuals. The Icelandic and Dutch studies were both approved by their respective ethical review boards, and informed consent was obtained from all study participants (Supplementary Methods online). We examined the association of sequence variants with pigmentation traits in eight genome-wide association analyses: three analyses for eye color (blue versus green, blue versus brown and blue versus nonblue), two for hair color (red versus nonred and blond versus brown) and three for skin pigmentation traits (skin sensitivity to sun, the presence of freckles and a combination of skin sensitivity to sun and presence of freckles, herein referred to as 'burning and freckling'). These analyses identified 99 distinct SNPs (Supplementary Table 1 online) with genome-wide significant associations (P < 1.5 × 10−7) in at least one of the eight pigmentation scans, 88 of which are located in areas described in our initial report1. The 11 remaining SNPs are located at three loci on chromosomes 9p23, 11q13 and 20q12. The findings of the discovery phase were confirmed in an analysis of an additional 2,116 Icelanders and 1,214 Dutch individuals (see Supplementary Table 2 online for quality control information).

A total of six SNPs within a region of strong linkage disequilibrium (LD) on 20q11.22 showed association with burning and freckling that reached genome-wide significance (max OR = 1.60, P = 3.9 × 10−9; Supplementary Table 1). Multipoint analysis of the area revealed an extended haplotype tagged by a two-SNP haplotype (rs1015362[G], rs4911414[T]) that we will refer to as the ASIP haplotype (Table 1, Supplementary Methods and Supplementary Fig. 1 online). The ASIP haplotype accounts for the association of the other SNPs in the region (Supplementary Table 3 online) and was significantly associated in both the Icelandic and Dutch replication samples (Table 1). In the combined analysis of the discovery and replication samples, the ASIP haplotype reached genome-wide significance for red hair color, freckling and skin sensitivity to sun, in addition to burning and freckling (Table 1). This pattern of association is very similar to that observed for the mutations associated with red hair color in MC1R (encoding melanocortin-1 receptor)2 (Supplementary Tables 3, 4 and 5 online). The region covered by the extended haplotype contains a large number of genes, including ASIP (encoding agouti signaling protein), a gene with a well-documented role in pigmentation. In melanocytes, the agouti signaling protein antagonizes the activation of MC1R by α-MSH (alpha melanocyte-stimulating hormone), resulting in a switch to the production of red or yellow phaeomelanin. Sequence variants at the agouti locus are responsible for animal coat colors such as yellow and dark color3,4. A polymorphism in the 3′ untranslated region of ASIP, rs6058017 (8818A>G), has been studied for its association with pigmentation characteristics within populations of European ancestry5,6,7 and has also been related to differences in skin pigmentation among populations of mixed African and European ancestry8. The ASIP haplotype, rs1015362[G] and rs4911414[T], occurs on the background of the major allele of rs6058017, but the correlation between the two is very weak (D′ = 1; r2 = 0.008). The strength of association of rs6058017 with the pigmentation traits is much less than that of the ASIP haplotype, and after adjustment for rs6058017, the ASIP haplotype remains highly significant for burning and freckling (P = 1.3 × 10−46). On the other hand, after adjustment for the haplotype, rs6058017 is only marginally associated with pigmentation characteristics (P = 0.057 for burning and freckling; Fig. 1a and Supplementary Table 3). We sequenced the exons and promoter of ASIP in 368 individuals without detecting any sequence variant likely to account for the observed association. A stronger association of the ASIP haplotype with skin sensitivity to sun was observed for males than females (P = 0.0033; Supplementary Table 4), but the difference was not significant after correction for the number of variants tested for sex-specific differences.

Table 1: Association of previously unreported SNPs with pigmentation characteristics in Icelandic and Dutch individuals
Figure 1: Estimates of odds ratio for haplotypes at ASIP and TPCN2.
Figure 1

(a) At ASIP, the previously reported variant 8818A is compared to the ASIP haplotype (AH) in individuals who burn and freckle and those who tan and do not freckle. Chromosomes not carrying the ASIP haplotype are denoted by notAH. (b) At TPCN2, the two variants encoding G734E and M484L are compared to the wild-type haplotype and to each other. Frequencies in the two pigmentation groups are displayed in brackets (fairer pigmentation, darker pigmentation). Estimated ORs and P values, from the pair-wise comparison of the haplotype at the end of arrow versus haplotype at the beginning of the arrow adjusted for all other haplotypes, are shown beside each arrow.

Four SNPs on 11q13.2 showed association with blond versus brown hair color in the Icelandic discovery sample that reached genome-wide significance (Supplementary Table 1). These SNPs are located within a single LD block that overlaps with only one gene, TPCN2 (encoding two-pore segment channel 2). We identified three common nonsynonymous variants in exons of TPCN2 (rs3829241, rs35264875, rs3750965) that, on the basis of the HapMap data, correlate with the four SNPs on the 300K chip giving significant association. These SNPs were typed in the Icelandic discovery samples as well as the two replication samples. The replication samples were also typed with rs1011176, which showed the strongest association in the initial discovery scan. All of the observed association with blond versus brown hair could be explained by two of the coding SNPs: rs35264875 (encoding M484L) and rs3829241 (encoding G734E) (Supplementary Table 3), which replicated with similar effects (Table 1 and Fig. 1b). We did not observe strong association of these two variants with the other pigmentation traits (Supplementary Tables 5, 6, 7 online), similar to what had been observed for the KITLG (encoding the ligand for KIT receptor tyrosine kinase) variant that also associates with blond versus brown hair. A link between TPCN2 and pigmentation has not been previously suspected. The protein encoded by TPCN2 participates in calcium transport, similar to the proteins encoded by known pigmentation genes SLC24A4 (ref. 1) and SLC24A5 (ref. 9).

A single SNP, rs1408799, on 9p23 showed genome-wide significant association with blue versus nonblue eyes (OR = 1.41, P = 1.5 × 10−9). This association was confirmed in both the Icelandic and Dutch replication samples with a similar effect (Table 1). A suggestive association with blond versus brown hair was also observed for this SNP. The SNP belongs to an LD block that encompasses only one gene, TYRP1 (encoding the tyrosinase-related protein 1)10. TYRP1 encodes a melanosomal enzyme with a role in the eumelanin pathway. In humans, rare mutations in TYRP1 are responsible for oculocutaneous albinism type 3 (ref. 11). Previous studies on the genetics of eye color in Europeans have associated polymorphisms at TYRP1 with eye color12. The SNP reported here, rs1408799, is in strong LD with one of the previously reported SNPs, rs2733832, in HAPMAP CEU13 (D′ = 0.96; r2 = 0.67).

The larger sample size studied here allows us to make some additional claims for the loci we reported previously. The TYR (encoding tyrosinase) coding variant rs1126809 (encoding R402Q) reaches genome-wide significance for skin sensitivity to sun in addition to its previously reported association with eye color (Supplementary Tables 5,6, 7). Compound heterozygosity for a mutant allele of TYR and the R402Q polymorphism can result in ocular albinism14.

This report adds sequence variants at two loci to the ever growing list of variants affecting normal variation in pigmentation among Europeans. The strength of the association of the newly identified ASIP haplotype described here is close to that of variants in MC1R and much stronger than that of the previously reported variants near ASIP. This haplotype thus seems to tag more effectively the actual causal variant(s). It is of note that genes encoding calcium ion transporters are emerging as a family of pigmentation genes, as three have now been associated with pigmentation: SLC24A4, SLC24A5 and now TPCN2.

Note: Supplementary information is available on the Nature Genetics website.

References

  1. 1.

    et al. Nat. Genet. 39, 1443–1452 (2007).

  2. 2.

    Am. J. Hum. Genet. 75, 739–751 (2004).

  3. 3.

    & Pigment Cell Res. 15, 10–18 (2002).

  4. 4.

    , , , & J. Hered. 98, 351–355 (2007).

  5. 5.

    et al. Am. J. Hum. Genet. 70, 770–775 (2002).

  6. 6.

    et al. J. Natl. Cancer Inst. 97, 998–1007 (2005).

  7. 7.

    et al. J. Dermatol. Sci. 40, 133–136 (2005).

  8. 8.

    et al. Mol. Biol. Evol. 24, 710–722 (2007).

  9. 9.

    et al. Science 310, 1782–1786 (2005).

  10. 10.

    et al. Mamm. Genome 9, 50–53 (1998).

  11. 11.

    et al. Am. J. Hum. Genet. 61, 1095–1101 (1997).

  12. 12.

    et al. Genetics 165, 2071–2083 (2003).

  13. 13.

    International HapMap Consortium. Nature 437, 1299–1320 (2005).

  14. 14.

    et al. Nat. Genet. 9, 92–95 (1995).

Download references

Acknowledgements

This research was supported in part by the Intramural Research Program of the US National Institutes of Health National Cancer Institute.

Author information

Affiliations

  1. deCODE genetics, Sturlugata 8, 101 Reykjavik, Iceland.

    • Patrick Sulem
    • , Daniel F Gudbjartsson
    • , Simon N Stacey
    • , Agnar Helgason
    • , Thorunn Rafnar
    • , Margret Jakobsdottir
    • , Stacy Steinberg
    • , Sigurjon A Gudjonsson
    • , Arnar Palsson
    • , Gudmar Thorleifsson
    • , Snæbjörn Pálsson
    • , Jeffrey Gulcher
    • , Augustine Kong
    • , Unnur Thorsteinsdottir
    •  & Kari Stefansson
  2. Department of Dermatology, Landspitali-University Hospital, 101 Reykjavik, Iceland.

    • Bardur Sigurgeirsson
    • , Kristin Thorisdottir
    • , Kristrun R Benediktsdottir
    •  & Jon H Olafsson
  3. Department of Pathology, Landspitali-University Hospital, 101 Reykjavik, Iceland.

    • Bardur Sigurgeirsson
    • , Kristin Thorisdottir
    • , Kristrun R Benediktsdottir
    •  & Jon H Olafsson
  4. Department of Plastic Surgery, Landspitali-University Hospital, 101 Reykjavik, Iceland.

    • Kristin Thorisdottir
    •  & Rafn Ragnarsson
  5. Comprehensive Cancer Center, 6501 BG Nijmegen, The Netherlands.

    • Katja K Aben
    •  & Lambertus A Kiemeney
  6. Department of Epidemiology and Biostatistics, Radboud University, Nijmegen Medical Center, 6500 HB Nijmegen, The Netherlands.

    • Katja K Aben
    • , Sita H Vermeulen
    •  & Lambertus A Kiemeney
  7. Genetic Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland 20892, USA.

    • Alisa M Goldstein
    •  & Margaret A Tucker
  8. Department of Urology, Radboud University Nijmegen Medical Center, 6500 HB Nijmegen, The Netherlands.

    • Lambertus A Kiemeney

Authors

  1. Search for Patrick Sulem in:

  2. Search for Daniel F Gudbjartsson in:

  3. Search for Simon N Stacey in:

  4. Search for Agnar Helgason in:

  5. Search for Thorunn Rafnar in:

  6. Search for Margret Jakobsdottir in:

  7. Search for Stacy Steinberg in:

  8. Search for Sigurjon A Gudjonsson in:

  9. Search for Arnar Palsson in:

  10. Search for Gudmar Thorleifsson in:

  11. Search for Snæbjörn Pálsson in:

  12. Search for Bardur Sigurgeirsson in:

  13. Search for Kristin Thorisdottir in:

  14. Search for Rafn Ragnarsson in:

  15. Search for Kristrun R Benediktsdottir in:

  16. Search for Katja K Aben in:

  17. Search for Sita H Vermeulen in:

  18. Search for Alisa M Goldstein in:

  19. Search for Margaret A Tucker in:

  20. Search for Lambertus A Kiemeney in:

  21. Search for Jon H Olafsson in:

  22. Search for Jeffrey Gulcher in:

  23. Search for Augustine Kong in:

  24. Search for Unnur Thorsteinsdottir in:

  25. Search for Kari Stefansson in:

Contributions

P.S., D.F.G., S.N.S. and K.S. wrote the first draft of the paper. S.N.S., B.S., K.T., R.R., K.R.B. and J.H.O. collected the Icelandic samples and phenotypes. S.N.S., M.J. and U.T. performed the genotyping. K.K.A., S.H.V. and L.A.K. collected the Dutch samples and phenotypes. P.S., D.F.G., A.H., S.S., S.A.G., A.P., G.T., and S.P. analyzed the data. P.S., D.F.G., S.N.S., T.R., A.M.G., M.A.T., L.A.K., J.H.O., J.G., A.K., U.T., K.S. planned, coordinated and supervised the work. All authors contributed to the final version of the paper.

Competing interests

The authors from deCODE own stock and stock options in the company.

Corresponding authors

Correspondence to Daniel F Gudbjartsson or Kari Stefansson.

Supplementary information

PDF files

  1. 1.

    Supplementary Text and Figures

    Supplementary Methods, Supplementary Figure 1 and Supplementary Tables 1–7

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/ng.160

Further reading Further reading

  • 1.

    Genome-wide association study in 176,678 Europeans reveals genetic loci for tanning response to sun exposure

    • Alessia Visconti
    • , David L. Duffy
    • , Fan Liu
    • , Gu Zhu
    • , Wenting Wu
    • , Yan Chen
    • , Pirro G. Hysi
    • , Changqing Zeng
    • , Marianna Sanna
    • , Mark M. Iles
    • , Peter A. Kanetsky
    • , Florence Demenais
    • , Merel A. Hamer
    • , Andre G. Uitterlinden
    • , M. Arfan Ikram
    • , Tamar Nijsten
    • , Nicholas G. Martin
    • , Manfred Kayser
    • , Tim D. Spector
    • , Jiali Han
    • , Veronique Bataille
    •  & Mario Falchi

    Nature Communications (2018)

  • 2.

    Genome-wide association meta-analysis of individuals of European ancestry identifies new loci explaining a substantial fraction of hair color variation and heritability

    • Pirro G. Hysi
    • , Ana M. Valdes
    • , Fan Liu
    • , Nicholas A. Furlotte
    • , David M. Evans
    • , Veronique Bataille
    • , Alessia Visconti
    • , Gibran Hemani
    • , George McMahon
    • , Susan M. Ring
    • , George Davey Smith
    • , David L. Duffy
    • , Gu Zhu
    • , Scott D. Gordon
    • , Sarah E. Medland
    • , Bochao D. Lin
    • , Gonneke Willemsen
    • , Jouke Jan Hottenga
    • , Dragana Vuckovic
    • , Giorgia Girotto
    • , Ilaria Gandin
    • , Cinzia Sala
    • , Maria Pina Concas
    • , Marco Brumat
    • , Paolo Gasparini
    • , Daniela Toniolo
    • , Massimiliano Cocca
    • , Antonietta Robino
    • , Seyhan Yazar
    • , Alex W. Hewitt
    • , Yan Chen
    • , Changqing Zeng
    • , Andre G. Uitterlinden
    • , M. Arfan Ikram
    • , Merel A. Hamer
    • , Cornelia M. van Duijn
    • , Tamar Nijsten
    • , David A. Mackey
    • , Mario Falchi
    • , Dorret I. Boomsma
    • , Nicholas G. Martin
    • , David A. Hinds
    • , Manfred Kayser
    •  & Timothy D. Spector

    Nature Genetics (2018)

  • 3.

    The origin of chow chows in the light of the East Asian breeds

    • Hechuan Yang
    • , Guodong Wang
    • , Meng Wang
    • , Yaping Ma
    • , Tingting Yin
    • , Ruoxi Fan
    • , Hong Wu
    • , Li Zhong
    • , David M. Irwin
    • , Weiwei Zhai
    •  & Yaping Zhang

    BMC Genomics (2017)

  • 4.

    Meta-analysis identifies novel risk loci and yields systematic insights into the biology of male-pattern baldness

    • Stefanie Heilmann-Heimbach
    • , Christine Herold
    • , Lara M. Hochfeld
    • , Axel M. Hillmer
    • , Dale R. Nyholt
    • , Julian Hecker
    • , Asif Javed
    • , Elaine G. Y. Chew
    • , Sonali Pechlivanis
    • , Dmitriy Drichel
    • , Xiu Ting Heng
    • , Ricardo C. -H. del Rosario
    • , Heide L. Fier
    • , Ralf Paus
    • , Rico Rueedi
    • , Tessel E. Galesloot
    • , Susanne Moebus
    • , Thomas Anhalt
    • , Shyam Prabhakar
    • , Rui Li
    • , Stavroula Kanoni
    • , George Papanikolaou
    • , Zoltán Kutalik
    • , Panos Deloukas
    • , Michael P. Philpott
    • , Gérard Waeber
    • , Tim D. Spector
    • , Peter Vollenweider
    • , Lambertus A. L. M. Kiemeney
    • , George Dedoussis
    • , J. Brent Richards
    • , Michael Nothnagel
    • , Nicholas G. Martin
    • , Tim Becker
    • , David A. Hinds
    •  & Markus M. Nöthen

    Nature Communications (2017)

  • 5.

    Global skin colour prediction from DNA

    • Susan Walsh
    • , Lakshmi Chaitanya
    • , Krystal Breslin
    • , Charanya Muralidharan
    • , Agnieszka Bronikowska
    • , Ewelina Pospiech
    • , Julia Koller
    • , Leda Kovatsi
    • , Andreas Wollstein
    • , Wojciech Branicki
    • , Fan Liu
    •  & Manfred Kayser

    Human Genetics (2017)