A 'Long Tradition'
The biological basis of sexual orientation is a research area that is coming out of the closet
By Robert Finn
For the last few years each new result has been covered extensively in the popular press. And last May, Minot State University in North Dakota hosted an international conference on the topic, in which 89 scientists participated. Organized by Lee Ellis, a professor of sociology at Minot State, the conference was sponsored by the Eugene Garfield Foundation, and was the first to assemble virtually all the top researchers in the field.
Each new finding, however, seems to generate its own mini-controversy over its scientific aspects, over and above the social and political turmoil such research engenders. In fact, there are even allegations that one of the leaders in the field engaged in scientific misconduct to enhance his results.
Despite the field's high level of interest and scientific productivity, "I think a lot of people are having trouble getting funding for this kind of work," says J. Michael Bailey, an associate professor of psychology at Northwestern University. Government funding is very hard to come by, he notes, and "when I approach foundations, they pretty much smile and say, 'This is something we can't get into.' There are a surprising number of [scientists] doing related stuff, but I think a lot of people just do it on the side."
As the first suggestion that there was a neuroanatomic difference between gay and heterosexual men, LeVay's finding garnered a great deal of public attention, and a great deal of controversy about both its scientific and its social implications. For example, many in the gay community argued that LeVay's findings bolster their contention that people are born gay, and do not become gay as a matter of choice. And if this is the case, the argument goes, homosexuals should be granted protection against discrimination. Some scientists have also assailed LeVay's methodology and the conclusions of his work.
LeVay, a neuroanatomist who was at the La Jolla, Calif.-based Salk Institute at the time he performed this research, now resides in West Hollywood, Calif., where he cofounded the Institute of Gay and Lesbian Education. He published his INAH3 study in Science (S. LeVay, 253:1034-7, 1991) and discussed it further in his book The Sexual Brain (MIT Press, 1993).
"You could say that my research was in a long tradition of looking for a basis of sexual orientation in a sort of sex reversal," LeVay comments. "This goes back to the last century, when people thought of homosexuality as 'congenital inversion.' . . . Gay people were [thought of as] a kind of chimera, where their brains were of one sex and their bodies another. It took decades of activism for people to realize that wasn't what was going on."
Gay people have long objected to the "chimera" idea, pointing out that, except for the direction of their sexual attraction, they seemed more like other members of their own sex. "To defend myself," continues LeVay, "I would say that first of all I reject that whole picture of congenital inversion. But if you're talking about one specific issue, which is the direction of sexual attraction, then you have to admit that is one thing gay men have in common with heterosexual women, and vice versa, lesbians with straight men. So if there is some brain basis for the direction of sexual attraction, then you would imagine you'd have a sort of sex atypicality in those circuits. In fact, it sort of has to be that way."
Another neuroanatomical difference between gay and heterosexual men was reported in 1992 by University of California, Los Angeles, investigators Laura S. Allen, a research scientist, and Roger Gorski, a professor of neurobiology (L.S. Allen, R. Gorski, Proceedings of the National Academy of Sciences, 89:7199-202). The anterior commissure, a relatively small bundle of axons connecting the two brain hemispheres, is larger in women than in men, and larger in gay men than in heterosexual men. And in related work, Dick F. Swaab and coworkers at the Netherlands Institute for Brain Research recently reported that a hypothalamic structure called the central subdivision of the bed nucleus of the stria terminalis is larger in men than women, and also larger in men than in male-to-female transsexuals (J.-N. Zhou et al., Nature, 378:68-70, 1995).
But the anatomical work has come under heavy criticism by William Byne, director of the neuroanatomy laboratory of neuropsychiatric disease at New York's Mount Sinai Medical Center. "A general problem with this work is that there have been dozens and dozens of reports of sex differences in the human brain since the middle of the last century. But not a single one of these has been corroborated, except for the one that men tend to have slightly larger brains than women.
"The reason for that is that it's tremendously difficult to do morphometric studies in the human brain. I would be surprised if there weren't sex differences in the human brain, since there are sex differences in just about every organ system in humans. But to date we can't say with any confidence where the sex differences are."
And Byne thinks that even if the sex differences are real, LeVay's findings could have been confounded by the fact that all his original gay subjects died of AIDS. LeVay maintains that he controlled for this by examining the brains of heterosexual men who died of AIDS, as well as one gay man who died of other causes. "I'm pretty damn confident that it's not a disease effect," he says. "But it's something that still gives me sleepless nights."
LeVay's controls are not adequate, counters Byne. For one thing, many people with AIDS suffer testicular atrophy before death, and since gonadal hormones are known to regulate the size of several hypothalamic nuclei in animals, disease effects can't be excluded.
Byne himself is currently engaged in an attempt to replicate LeVay's study. He's using extremely stringent criteria that exclude subjects outside a narrow age range and those with even a hint of endocrine pathology. Preliminary results have so far confirmed the sex difference in INAH3 and excluded the possibility that HIV status might affect the size of the nucleus. The data have not yet been examined for findings related to sexual orientation, but Byne says he expects to have results in six months to a year.
A second line of evidence concerning the biology of sexual orientation comes from genetic studies. In studies on male and female homosexuals who have twins, Northwestern's Bailey and his associates found clear evidence for genetic transmission. In a study on gay men, for example, 52 percent of their identical twins, 22 percent of their fraternal twins of the same sex, and only 11 percent of their adopted brothers were also gay (J.M. Bailey, R.C. Pillard, Archives of General Psychiatry, 48:1089-1096, 1991).
Among lesbian women, 48 percent of their identical twins, 16 percent of their fraternal twins of the same sex, and only 6 percent of their adopted sisters were also lesbian (J.M. Bailey et al., Arch. Gen. Psych., 50:217-223, 1993). Aside from these twins studies, there have been few, if any, firm results showing neuroanatomical or genetic correlates for female homosexuality.
Nonetheless, Bailey says that his data in men lead to an estimate that the overall heritability of sexual orientation ranges from 25 percent to 75 percent, depending on a number of assumptions. Byne, however, criticizes Bailey's study also, noting that subjects were recruited by ads placed in gay publications. "Some people have suggested that identical twins who are concordant on a variety of measures are more likely to respond to these ads than ones who are different," which would lead to an increased estimate of heritability, notes Byne.
Further supporting a genetic influence on sexual orientation are studies by Dean H. Hamer, chief of the gene structure and regulation section of the National Cancer Institute's laboratory of biochemistry, and his coworkers. In studying the pedigrees of gay men who also had gay brothers, Hamer found that such people had an excess of gay relatives on the mother's, but not the father's, side of the family.
Reasoning that this might indicate that sexual orientation might be linked to the X chromosome, Hamer conducted a linkage analysis to determine if any DNA markers on the X chromosome would be inherited along with the putative gene for sexual orientation. In 33 of 40 pairs of gay brothers he found such a marker near the tip of the long arm of the X chromosome, in a location called Xq28, an area that contains several hundred genes (D.H. Hamer et al., Science, 261:321-7, 1993). Hamer recently replicated this finding in a new set of families (S. Hu et al., Nature Genetics, 11:248, 1995).
CASTING DOUBT: a study by George Ebers contradicts the X linkage theory. However, a group led by George C. Ebers, a professor of neurology at the University of Western Ontario in Canada, has failed to replicate Hamer's findings in a study that is currently being submitted for publication. "We've been collecting families that have more than one gay person for five years, and we've gone through something like 400 pedigrees," explains Ebers. "In those [families] there is really no support for the idea that male homosexuality is X-linked. The DNA tests that were done didn't even support Dean's idea a bit. There wasn't even a trend toward increased sharing of haplotypes down there at Xq28."
Ebers speculates that there may be a simple explanation for Hamer's finding of maternal transmission. "There may be an excess of all kinds of things on the maternal side because mothers know more about their family history than fathers. Something of a personal nature like this is perhaps even more likely to be something that you would [learn] from the maternal side."
Hamer responds: "At this point, no one has seen [Ebers's] data in its entirety. It hasn't been peer reviewed, and it hasn't been published, so it's difficult to say. I have seen some of the data myself, and what's clear is that it's a very different study than ours. They made no effort to focus on families that are X-linked, and in fact their aim really has been to find non-X-linked loci, and they seem to have focused largely on families that are most likely to show autosomal linkages."
But Byne offers another possible confounding factor in Hamer's work: "The hallmark of X-linked transmission is the absence of father-to-son transmission. . . . It's possible that just the relative absence of father-to-son transmission-because gay men tend not to have children-could have given Dean the impression of X-linked transmission in his first pedigree study."
Support for Hamer's findings of maternal transmission, X-linkage, and even the idea that a gene related to sexual orientation may be found at Xq28 comes from a study by William J. Turner, a clinical professor of psychiatry at the University of New Mexico (W.J. Turner, Archives of Sexual Behavior, 24:109-34, 1995). Turner studied a total of 259 families of male homosexuals and found that "the mothers of homosexuals have more sisters than brothers at a rate suggesting that 50 percent of the males conceived in the mother's generation never came to term. On the mother's side there are many more miscarriages than on the father's side, many more suicides, much more infertility, and many more people who remained single past the age of 30." Turner argues that these are all clues that indicate genetic involvement. Moreover, says Turner, similar proportions of "fetal wastage" of the mother's male siblings are evident in nine "semi-lethal," Xq28-linked disorders, including adrenoleucodystrophy, color blindness, and fragile X syndrome. Such proportions of fetal wastage are unique to Xq28 disorders, claims Turner, and this argues for an Xq28 linkage for sexual orientation, as well.
Despite this experimental support, Hamer's research has been cast into doubt not only by arguments over his interpretation of the data, but also by allegations of scientific misconduct. According to a front-page article by John Crewdson in the Chicago Tribune (June 25, 1995), an anonymous former member of Hamer's lab has alleged that Hamer engaged in selective presentation of data in his 1993 Science paper. Crewdson reported that an investigation has been launched by the Office of Research Integrity (ORI) of the United States Department of Health and Human Services. ORI director Lyle Bivens tells The Scientist, "We don't confirm or deny the existence of any investigation, so I can't make any comments on the topic at all."
ORGANIZER: Lee Ellis put together a recent international conference on the biology of sexual orientation. The suggestion that Hamer may have engaged in impropriety worries Byne. "If the allegations were coming from the religious or political right, it would be one thing," notes Byne. "But it's disturbing to me that the allegations come from a member of his research team." Hamer declines comment, except to say, "The only thing I can say about the Crewdson article is that it was seriously in error."
In addition to the neuroanatomical and genetic work, several other widely disparate types of studies argue for a biological basis for sexual orientation. For example, Lee Ellis published a study that indicated that "if the mother experiences a lot of stress-and it's got to be severe-during the second trimester [of pregnancy], there will be a significantly higher chance that her male offspring would be homosexual when they became sexually mature" (L. Ellis et al., Journal of Sex Research, 25:152-7, 1988). Another study suggests that, compared with heterosexual men, gay men show a leftward asymmetry in the number of fingerprint ridges on their thumbs and little fingers (J.A.Y. Hall, D. Kimura, Behavioral Neuroscience, 108:1203-6, 1994).
And Bailey says that a common behavioral finding about male homosexuals provides additional support for a biological hypothesis. "The most consistent finding about male homosexuality is that as children gay men were feminine boys," as judged by such factors as lack of interest in sports or rough play, reputation as a "sissy," or a desire to be a girl. "Perhaps 75 percent of feminine boys grow up to be gay men, which is a huge increase over expected rates. That's generally consistent with a biological hypothesis because you have these boys playing atypically at a very early age-three to five-in a way they haven't been socialized to behave. In fact, they're often punished for behaving that way" (J.M. Bailey, K.J. Zucker, Developmental Psychology, 31:43-55, 1995).
But whatever the biological basis of sexual orientation, the research-and the results-can be politically charged. Explains Byne, "My work has been taken up by the political and religious right, . . . claiming that by debunking the biology I was showing that homosexuality was a matter of choice, and a poor choice that could be changed." In fact, Byne has learned that one group, the Washington, D.C.-based Family Research Council, went so far as to set up a toll-free telephone number to distribute reprints of one of Byne's articles, something Byne says they did without his permission.
CONTROVERSIAL RESEARCH: Dean Hamer, left, and J. Michael Bailey converse at the May conference. "The fact of the matter is that this biological issue should not figure in gay rights," maintains Byne. "Gays have to premise their rights on principles of equality, justice, and liberty, just as any minority does. Any set of rights that was premised on the assumption of genetic or biological immutability would be impoverished at best." Robert Finn, a freelance science writer based in Long Beach Calif., is online at firstname.lastname@example.org.