The sexiest part of a man, a Swiss zoologist has found, may be his armpits: They’re an odoriferous window on his genes.
A woman enters a room alone and seats herself before a table covered with seven small boxes. She picks up each box, sniffs it carefully, and jots something on a pad of paper. Shortly after she leaves, another woman enters and reenacts the performance. This is not market research at a perfume company; it’s the lab of Claus Wedekind, a zoologist at Bern University in Switzerland. Wedekind is testing women’s responses to sweaty T-shirts--men’s sweaty T-shirts. He has found that women have strong preferences in that department. They’re attracted to the scent of men who are most unlike them in a very particular way--in the array of immune system genes known as MHC, for major histocompatibility complex.
MHC genes are the most diverse of all genes. In fact, they differ so widely from person to person that they constitute a molecular John Hancock, one that helps an organism recognize its own healthy cells, identify pathogens, and reject foreign tissue. The idea that they may also confer a distinctive odor, and thereby influence behavior, was first suggested in 1974 by biologist Lewis Thomas. Laboratory studies soon proved Thomas right in the case of mice. Inbred mice--who were alike in all genes but MHC--could detect a difference in the scent of a relative that harbored an ever-so-slightly different MHC gene. Moreover, their odor preferences were not innate but learned. Young mice tend to prefer the odor of their nest mates, but when they hit puberty, it’s vive la différence: they prefer to mate with mice whose MHC genes are unlike their own.
Since the mouse studies were done, researchers have found that humans, too, can detect MHC-dependent odors. But no one had shown that we prefer some such odors to others. Wedekind and his colleagues didn’t really set out to do that; they’re zoologists, and their professional interest is not in how humans choose mates but in how fish do. Their work had led them to the hypothesis that MHC genes are somehow involved in producing signals that influence mate choice--but since humans can talk, and our MHC genes are better known than those of fish, for once we were the better laboratory animals. We know a lot about the immune system in humans, and we can do experiments in which we ask women or men what their preferences are, says Wedekind. We used humans just as a model species to get an idea about a very basic question in evolutionary biology.
For his study, Wedekind recruited a group of 49 women and 44 men who harbored a wide range of MHC genes. Wedekind gave each man a clean T- shirt on a Sunday morning and asked him to wear it for two nights. He decided to gather male scent rather than female scent simply because unshaved armpits collect more odor. In fact, to ensure a strong body odor, he gave the men supplies of odor-free soap and aftershave and asked them to remain as odor neutral as possible.
On Tuesday morning, the men returned, sweaty T-shirts in hand. Wedekind put each shirt in a plastic-lined cardboard box with a sniffing hole on top. Then he brought in the women. Each was scheduled for the experiment at the midpoint of her menstrual cycle, when women’s noses are reputedly the keenest, and each was presented with a different set of seven boxes. Three of the seven boxes contained T-shirts from men harboring MHC similar to the woman’s own; three contained T-shirts from MHC-dissimilar men; and one contained an unworn T-shirt as a control. The women were asked to rate each of the seven T-shirts as pleasant or unpleasant.
As is so often the case in biology, what is true of mice is true of men--or in this case, women. Overall, says Wedekind, the women he tested were more likely to prefer the scent of men with dissimilar MHC. In fact, that scent tended to remind them of their boyfriends, both past and present. Says Wedekind, This is the first indication that MHC still plays a role in mate choice today.
But the extent of the similarity between mice and humans surprised even Wedekind. When mice are pregnant, their preferences revert, and they prefer the familiar odor of MHC-similar males, says Wedekind. By nesting with relatives, the mothers get help nursing the young pups as well as protection from strange and murderous males. Surprisingly enough, the women in Wedekind’s study who were on the pill--which raises estrogen levels in the body, as in pregnancy--also preferred the odor of MHC-similar males.
Does that mean that women who are on the pill tend to choose mates they would otherwise avoid, with untold consequences for Western society? Wedekind can’t really say; presumably smell is not the only thing a woman looks for in a man. But in some respects, Wedekind speculates, a woman on the pill may behave like a pregnant woman, and maybe pregnant women have these opposite preferences. They prefer the odor of relatives. It may well be that their odor perception has a different function. It’s not for mate choice, it’s for choice of cooperative partners, and that would be relatives.
The benefits of raising young among relatives are easy to see. But the benefit of preferring MHC-dissimilar mates is still far from clear. One possibility is that it reduces the risk of inbreeding and thus of producing children with genetic diseases. Wedekind thinks that MHC-based mating does indeed serve that function among mice, who live in small, genetically homogeneous communities. He doubts, though, that preventing inbreeding is the main purpose of MHC-based preferences in humans. We have cultural systems--such as incest taboos--to do that, he explains.
Instead, Wedekind believes that preferring MHC-dissimilar mates helps ensure offspring with more effective immune systems. Although MHC genes somehow influence a person’s body odor, their main function is to encode cell-surface proteins that offer a window display of the proteins being made inside a cell. That helps the immune system identify cells that have been invaded by a virus or some other pathogen. The more genetically dissimilar a child’s parents are, Wedekind thinks, the more diverse his complement of MHC genes will be, and the more capable his cell-surface proteins will be of presenting for immune attack any pathogen that happens to beset him. Wedekind is now working on experiments to test this hypothesis.
Meanwhile there is yet another reason for humans to avoid a mate with similar MHC: MHC-similar couples, according to several recent studies, are less fertile. For example, geneticist Carole Ober of the University of Chicago studied a South Dakota community of Hutterites, a religious group who do not believe in contraception and tend to marry within the community- -and are thus more likely to marry partners with similar MHC. She found that Hutterite couples with similar MHC tended to have longer periods between pregnancies and higher rates of miscarriage than other Hutterite couples. Fetuses who receive the same MHC genes from each parent, Ober suggests, may be more likely to abort--perhaps because of some unknown immunologic mechanism.
Choosing MHC-dissimilar mates may, therefore, serve three ends: increasing fertility, producing hardier offspring, and reducing the risk of genetic disease. Whatever its ultimate aim, Wedekind believes the mechanism is still strong in both sexes (although he hasn’t yet tested how men respond to the scent of a woman). Some women were very surprised, Wedekind says. Some told me that a certain odor was absolutely beautiful or extremely dreadful. I sniffed a T-shirt and I was astonished by how different odors can be. I had the impression from our data, though, that there’s no man who smells very good for everyone--it depends on who is sniffing him. It is a result I very much like. Wedekind’s wife says she likes his odor. They have two children.
