« Nature "vs." Nurture (again!?!?!) | Gene Expression Front Page | Bowel Disease Gene »
December 23, 2003

Genetics of Hair Color (again)

We talk about a lot of complex genetic traits. Many of them are only partially genetic of course, but because of their social importance (intelligence, anti-sociality, etc.) we feel it important to approach all angles, including the genetic one. But one thing seems to show up in the emails that I receive, and the conversations I have with friends who know of my interest in genetics.

Here is a recent email:

I have a question about color of hair , would you like to help me to answer ? Thanks you anyway.

I have seen a family which has 4 members. Father is a American . He's blond hair.

Mother is a Asian. She 's black hair. They has 2 children. They have blue eyes. But they have different colors of hair. The first is a man. He has black hair. The second is woman. She has blond hair. May you expain for me why they have different hair colors, please !

Another (from a friend):

Someone told me that since I had dark hair/eyes that I would be able to get really tan or brown if in the sunlight. So I believed that for years until I tried last summer and i only got burnt and peeled.

Well, I told my hairdresser this and she told me the reason why I cannot tan is because my mother is a red-head. Is that true? Even though someone has really dark hair and eyes, he cannot tan if he has redheaded family? Most blonds tan better than I do. Thanks for your time

What am I to say? When we were kids in high school they showed us Punnett Squares which illustrate simple crosses between parents, like so:

 
Mother
    A A
Father a Aa Aa
a Aa Aa

Assume that "A" is the gene for dark hair, while "a" is the gene for blonde hair. All the children should then have dark hair, right (thinking back to the capital usually being "dominant")? You got the test question correct if you said "Yes!" "A" could be converted to many things, but it is "dominant," while "a" is recessive, end of story....

There are a few problems with this model.

1) More than one gene might influence the phenotype.
2) They might not be on-off, but additive.
3) The very definition of the phenotype might slant our perspective a bit.

(note that Mendellian genetics does serve as the building block for modelling polygenic traits, but the Punnett Square can get really out of hand when you have many genes)

There isn't one "blonde" color, it is a range. Similarly, there isn't one "dark" color (in hair). Rather, in the fashion of a continuous trait, like height or intelligence, where there is a spectrum, you have light blonde and black hair as normal ranges of the phenotype (though getting a aproximate normal distribution might be problematic if you sampled the world's population and classified their phenotype from visual inspection, assigning a integer value to "how dark" the hair is-but perhaps not as difficult if you used a machine to sample the concentration of pigment). The phenotype is caused by the presence of melanin. To add to the confusion you need to include another dimension with hair color: red. So you have auburn, strawberry blond, etc. thrown into the mix. I have posted several times on coloration in human beings:

I think in the near future we'll get a handle on these genes. There is obviously great interest in this topic, the "skin color" article was the top-viewed on PLOS a few weeks ago.

Back to the emails. I feel uncomfortable about addressing them because I feel they want cut & dried answers, but this isn't like asking the chance of having a Cystic Fibrosis child if both parents carry the CF recessive gene. They're asking questions about more complex genetics, filtered through their own perception of the phenotype. What is blonde? What is light? And so on.

But here is what I would say....

1) Children have less melanin than adults
2) Women have less melanin than men
3) Environment also affects hair color

Bruce Lee one once said his son Brandon was the only "blonde Chinaman" in the world. If you know what Brandon looked like as an adult, he had dark brown hair (which could be perceived as black). Most European blonde children become non-blonde adults. It stands to reason that a half-Asian "blonde" (again, "blonde" can be relative, I have met many self-described blondes who I would not describe as such) would have dark hair as an adult. A similar pattern can be seen with blue eyes (note that skin, hair and eye color are not tightly linked). The saying that "all babies have blue eyes" is dependent on parental phenotype. In my family, the saying would more appropriately be that "all babies are born white." Though we have brown eyes at birth, our skin is white until we are 6 months to 1 year old (I once talked to a friend of Korean origin who thought that all babies were born with the Mongoloid blue spot, and I have been told that black babies do not develop kinky hair for a few months). Additionally, there are also other eye colors besides blue and brown. Hazel and green for instance can be confusingly classified sometimes. I have known a few people who are half-Asian (of various kinds, east and south) who have hazel and green eyes. If that is classified as "non-blue," then blue can be confirmed to be recessive. If it is interpreted as a more continuous trait, it gets complicated. If it is interpreted as a color caused by a different gene, well....

You get the point. People look different at different ages. Men and women are affected by the different levels of testerone and estrogen. The coloration of South Asian field-workers is often black, that of the Brahmin caste is lighter, some of this is genetic, but some of it is environment. Gay men often have blonde tips (thank you fashion!). Surfers are often very blonde (thank you sun & chemicals).

As to my friend with the red-haired mother, the abstract linked in the red hair post indicated that heterozygotes for the "red hair" locii are more prone to sun-burns. Score one for genetics + experience.

To sum, we have a case here where people want cut & dried answers, like physics. All I can present are the odds. I can tell you how the whole population should behave, but individual cases are ... individual cases. The term educated guess is very appropriate.

Finally, the abiding interesting in this topic is pretty interesting. Free time in the consumer class and preoccupation with aesthetics is surely part of it. But theories about sexual selection for blondism are really interesting in light of contemporary concerns. I think it is safe to say that "the substance of style" has always been around (in biology it might be re-termed the function of style)....

Posted by razib at 04:08 PM