Red hair color: The myth
Some people have red hair, and some have hair that is various shades of blond or brown. The myth is that red hair is determined by a single gene, with the allele for red being recessive to alleles for other colors.
Hair color as a character
Hair color is determined by the amount of eumelanin (which is dark brown) and pheomelanin (which is reddish). The amount of eumelanin ranges continuously from very little, producing light-blonde hair, to large amounts, producing black hair. People with large amounts of pheomelanin have red hair, which can range from pale red ("strawberry blond") to bright red to reddish brown.
Most studies divide hair color into a small number of categories, such as blond, red, and brown. Reed (1952) criticized the subjectiveness of this approach and used a reflectance spectrophotometer to measure the amount of light reflected by hair at different wavelengths. He found that there was no clear separation of hair into two categories; instead there were some individuals with intermediate hair that could not easily be classified as red or non-red. Reed (1952) also examined hair under a microscope and found the reddish pheomelanin granules that are common in red hair were also in some individuals with brown hair. This suggests that when the red pheomelanin pigment is present, whether a person has red hair is determined by the amount of brown eumelanin pigment that the person also has.
The variety of hair colors makes it difficult to summarize the results of family studies in detail. Davenport and Davenport (1909) found numerous examples of two brown-haired parents having red-haired offspring, which would suggest that it is determined by a recessive allele, but Neel (1943) found 13 out of 114 offspring of two red-haired parents to have non-red hair. Reed (1952) reviewed the various hypotheses that had been proposed, including that red was recessive; that red was dominant; that red was dominant, but could be masked by brown; or that red was usually recessive but could sometimes be dominant.
Valverde et al. (1995) surveyed DNA sequence variation in the melanocortin 1 receptor (MC1R) gene. They found several amino acid variants that were found in red-haired people but rare in non-red people. Box et al. (1997) identified the three most common amino acid polymorphisms that are associated with red hair: R151C, R160W, and D294H. This shorthand means that the common amino acids at positions 151, 160 and 294 in the protein are arginine (R), arginine, and aspartic acid (D), while the amino acids cysteine (C), tryptophan (W), and histidine (H) are found in redheads. Most alleles have only one of these three red-associated amino acids; for example, some alleles have cysteine at position 151 but arginine and aspartic acid at positions 160 and 294.
There are a large number of rare amino acid polymorphisms in the MC1R gene, some of which may also be associated with red hair (Beaumont et al. 2007). Sulem et al. (2007) surveyed genetic variation throughout the genome of a large sample of Icelanders and found that MC1R is the only gene with a strong association with red hair. However, knowing the genotype of an individual at the MC1R locus is not enough to predict whether they have red hair. Beaumont et al. (2007) found that only 74% of individuals who were homozygous for tryptophan at position 160 have red hair, while 4% of individuals who were heterozygous for this amino acid had red hair. Box et al. (1997) found five pairs of dizygotic twins which had identical genotypes for the MC1R gene, yet one twin had red hair and the other didn't. Sulem et al. (2007) used the variation at the MC1R gene to try to predict hair color, and about a third of the individuals who were predicted to have red hair actually had blond or brown hair.
The alleles associated with red hair are fairly common in northern European populations; in Britain and Ireland, the R151C allele has a frequency of about 10 percent, while R160W and D294H are at 9 and 2 percent (Gerstenblith et al. 2007). Red hair is rare in other populations, which has led to speculation that the alleles for red hair were favored by selection by differing amounts of ultraviolet radiation, since red hair is associated with pale skin (Jablonski and Chaplin 2010) and is most common in areas with gloomy winters. Harding et al. (2000) applied several statistical tests for DNA sequence data and found no evidence that the large amount of amino acid variation at MC1R resulted from positive selection; in particular, the large number of amino acid sites that vary within human populations are comparable to the large number of amino acid differences between human and chimp MC1R. Harding et al. (2000) did not find any amino acid variation within African populations, so they concluded that there is strong negative selection there (new amino acid mutations are selected against). They concluded that the variation outside of Africa reflects a relaxation of negative selection allowing new alleles to drift in frequency, rather than new alleles being favored. In contrast, Savage et al. (2008) concluded that MC1R was affected by positive selection, based on the greater geographic variation in allele frequency than most human genes, greater levels of polymorphism, and an unusually large number of low frequency polymorphisms.
Red hair color is not an example of a simple genetic trait. While the amount of red pigment may be mainly determined by one gene (MC1R), there are a large number of different MC1R alleles, and other genes affecting the amount of brown pigment that plays a major role in determining hair color. The complicated genetics means that it is possible for two red-haired parents to have non-red-haired children.
Beaumont, K. A., S. L. Shekar, R. A. Newton, M. R. James, J. L. Stow, D. L. Duffy, and R. A. Sturm. 2007. Receptor function, dominant negative activity and phenotype correlations for MC1R variant alleles. Human Molecular Genetics 16: 2249-60.
Box, N., J. Wyeth, L. O’Gorman, N. Martin, and R. Sturm. 1997. Characterization of melanaocyte stimulating hormone receptor variant alleles in twins with red hair. Human Molecular Genetics 6:1891–1897.
Davenport, G. C., and C. B. Davenport. 1909. Heredity of hair color in man. American Naturalist 43: 193-211.
Gerstenblith, M. R., A. M. Goldstein, M. C. Fargnoll, K. Peris, and M. T. Landi. 2007. Comprehensive evaluation of allele frequency differences of MC1R variants across populations. Human Mutation 28: 495-505.
Harding, R. M., et al. (11 co-authors). 2000. Evidence for variable selective pressures at MC1R. American Journal of Human Genetics 66: 1351-1361.
Jablonski, N.G., and G. Chaplin. 2010. Human skin pigmentation as an adaptation to UV radiation. Proceedings of the National Academy of Sciences 107: 8962-8968.
Neel, J. V. 1943. Concerning the inheritance of red hair. Journal of Heredity 34: 93-96.
Reed, T. E. 1952. Red hair colour as a genetical character. Annals of Eugenics 17: 115-139.
Savage, S. A., M. R. Gerstenblith, A. M. Goldstein, L. Mirabello, M. C. Fargnoll, K. Peris, and M. T. Landi. 2008. Nucleotide diversity and population differentiation of the melanocortin 1 receptor gene, MC1R. BMC Genetics 9: 31.
Sulem, P., et al. (25 co-authors). 2007. Genetic determinants of hair, eye and skin pigmentation in Europeans. Nature Genetics 39: 1443-1452.
Valverde, P., E. Healy, I. Jackson, J. L. Rees, and A. J. Thody. 1995. Variants of the melanocyte-stimulating hormone receptor gene are associated with red hair and fair skin in humans. Nature Genetics 11: 328-330.
This page was last revised December 8, 2011. Its address is http://udel.edu/~mcdonald/mythredhair.html. It may be cited as pp. 37-39 in: McDonald, J.H. 2011. Myths of Human Genetics. Sparky House Publishing, Baltimore, Maryland.
©2011 by John H. McDonald. You can probably do what you want with this content; see the permissions page for details.