| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Program in Neuroscience (J.A.M., C.L.J., S.M.B.), Departments of Psychology (K.L.G., C.L.J., S.M.B.) and Zoology (S.M.B.), Michigan State University, East Lansing, Michigan 48824-1101
Address all correspondence and requests for reprints to: S. Marc Breedlove, Neuroscience Program, 108 Giltner Hall, Michigan State University, East Lansing, Michigan 48824-1101. E-mail: breedsm{at}msu.edu.
"Tell me where is fancy bred, or in the heart or in the head?"—William Shakespeare, The Merchant of Venice, Act III, Scene 2
Today’s neuroscientist has no hesitation answering Shakespeare’s question taken literally: of course love and desire are generated in the brain, that heady organ in charge of all our thoughts and feelings. Therefore, it must be the brain that is responsible for most men being sexually attracted to women and for most women being sexually attracted to men. Starting in 1985, several structural sex differences were described in the human brain (1), including the third interstitial nucleus of the anterior hypothalamus (INAH3), which has a greater volume in men than in women (2). But there was no way to know which, if any, of these neural sexual dimorphisms were responsible for the robust human sex difference in sexual partner preference. Then in 1991, Simon LeVay (3) not only replicated the sex difference in human INAH3 but reported that the nucleus in homosexual men was indistinguishable from that of women, i.e. significantly smaller than in heterosexual men. Suddenly it became plausible to think that possessing a large INAH3 might cause a person to be attracted to women, whereas having a small INAH3 might cause a person to be attracted to men. In this issue, Charles Roselli et al. (4) describe a sexually dimorphic hypothalamic nucleus in sheep, another mammalian species in which a minority of males display a lasting preference for mating with males rather than with females. Furthermore, they find that those rams that prefer mounting males have a smaller nucleus in this brain region than do rams that prefer mounting ewes. These observations lend strong support to the idea that the structure of the anterior hypothalamus in mammals influences sexual orientation in males.
To understand the context of this finding, we first have to deal with some confusing issues of terminology. The region of the anterior hypothalamus known as the preoptic area (POA) has long been implicated in male sexual behavior. Lesions of this region in males, across a wide variety of vertebrate species, cause the animals to lose interest in mating with females (5). So it was perfectly reasonable for Roger Gorski and colleagues (6) to look for sexual dimorphisms in the rat POA, and they struck gold. Within the POA was a subregion that in Nissl stain is five to six times greater in volume in male rats than in females. Impressed by the findings in rats, Dick Swaab and Eric Fliers (1) of The Netherlands Institute for Brain Research found a nucleus in the POA of humans that was larger in men than in women, so they named it the sexually dimorphic nucleus (SDN)-POA. Unfortunately, when Gorski’s group examined the human POA, they were unable to decide which of the several nuclei there were homologous to the rat SDN-POA. So they named the four neuron groups the INAH and numbered them, starting with the group Swaab and Fliers termed the SDN-POA (=INAH1). Gorski’s group found that INAH3 was larger in men than in women (2), a finding that LeVay would later replicate. Unfortunately, neither Gorski’s group nor LeVay found a sexual dimorphism in INAH1, failing to replicate Swaab and Fliers’ findings, which may represent a matter of statistical power, because the Dutch group have much larger sample sizes (7). Given this complicated nomenclature, the Roselli group decided to call the nucleus in the sheep POA the ovine SDN (oSDN). This title represents a strictly descriptive name (it’s larger in males than in females), without taking a stand on whether it is homologous with the rat SDN-POA and/or the human INAH3. That’s a wise decision because there is no sure way to determine whether any of these nuclei are homologous across species. So we cannot be sure that the oSDN is homologous with the human INAH3, but they are definitely in roughly the same part of the brain. Plus in both species, the nuclei are smaller in those minority of males that are sexually attracted to other males, which certainly suggests that the two nuclei (homologous or not) are involved in sexual attraction in male mammals.
Of course, Shakespeare was also posing a deeper question. Do we hit upon the object of our affection through carefully considered, rational analysis of the sort we might use in selecting tires for an automobile, or through emotional reactions, which are often so difficult to understand, far less to explain to others? Most people, even scientists, would agree that sexual attraction is more a matter of emotion than intellectual reflection if for no other reason than because it often seems so arbitrary. Before you object that surely there is a continuum of attractiveness, remember that the vast majority of people are attracted exclusively to people of the opposite sex, and find same-sex individuals completely unacceptable as potential mates. This dichotomy of desire is so pervasive in our culture, so automatic in our response to other people, that it’s easy to overlook how truly strange it is. How can half the population regard Brad Pitt as dreamy, whereas the other half feels absolutely no attraction to him at all? How did each of us come to be so finicky that we find billions of people totally unworthy as sexual partners? And does this selectivity result from a choice made during childhood? "Let’s see, I’ll have gym at school today, so I’ll wear white socks and tennis shoes. Gosh, as long as I’m making decisions I guess I better be attracted to girls for the rest of my life, too."
If we don’t use our head to decide which gender to lust after, maybe the decision is made somewhere even further down than our heart. Neuroendocrinologists’ work on nonhuman animals offers a theory that the gonads play a role in sexual orientation. In mammalian models, a testicular hormonal signal—androgen—masculinizes the developing genitalia and also masculinizes the developing brain. For a wide variety of behaviors, we can arrange for an animal to display either typically male-like or typically female-like behaviors, or something in between, just by manipulating androgen levels at the right time in development. For most behaviors, a single exposure to androgen early in life will masculinize the animal’s brain and behavior forever (8). The rat SDN-POA conformed to this notion beautifully: males deprived of androgen early in life display a small SDN-POA in adulthood, whereas females exposed to androgen during the perinatal period display a large SDN-POA. So might there be a similar process at work in humans? Do the fetal androgens that provide a baby boy with his penis also masculinize his brain, perhaps even the POA, so that 10 yr later he starts fantasizing about Britney Spears, kicking off a lifelong fascination with women?
The question isn’t easy to answer. For one thing, the fetal androgens that provide a baby boy with a penis also solicit an entire program of experience. People around the baby boy take careful note of his androgen-augmented genitalia and bring to bear a lifetime of social pressure to behave like the other boys. Conversely, girls developing with relatively little androgen are born with a clitoris and are therefore instructed to behave like other girls. So there’s no way to dissociate androgen’s influence on the genitalia vs. the brain when you consider most people. But homosexuals, that minority of people who do not follow society’s rigid instruction about sexual attraction, offer an opportunity. Were homosexual men exposed to reduced levels of fetal androgen compared with heterosexual men? Were lesbians exposed to higher levels of fetal androgen than heterosexual women?
Interestingly, so far there has been more evidence that early androgens might play a role in homosexuality in women than in men. There are recent reports of three previously unsuspected body markers that seem to reflect fetal androgen in humans. The markers are quite varied, involving the ears (the production of tiny sounds by the cochlea), the eyes (eye-blink reflexes), and the fingers (the pattern of relative finger lengths). Yet in each case, there is a sex difference in function or structure, and in each case lesbians display characteristics that are more male-like compared with heterosexual women (9, 10, 11). These same putative somatic markers of early androgen have provided conflicting results when comparing homosexual and heterosexual men (11, 12, 13). The conflicting results in men suggest that some boys may turn out homosexual as a result of lower-than-normal fetal androgen, some may result from higher-than-normal levels, and some may turn out homosexual for reasons having nothing to do with androgens. For sheep, Roselli et al. (4) have some evidence that male-oriented rams receive lower-than-normal androgen stimulation of the brain.
The effects of steroids on the SDN-POA of rats suggest that estrogens might affect the volume of the oSDN. Testing this angle, Roselli et al. (4) examined aromatase, the enzyme that converts androgens such as testosterone into estrogens such as estradiol. This enzyme plays a crucial role in the masculine development of the rat SDN-POA because testicular androgens are aromatized to estrogens that act through estrogen receptors (not androgen receptors) to masculinize the rat POA and (not so coincidentally) also to masculinize many rat behaviors. Roselli et al. (4) found that mRNA levels for aromatase in the oSDN were higher in males than in females and were higher in female-oriented rams than in male-oriented rams. These results certainly suggest that there is a relationship between steroid hormones and oSDN morphology and, by extension, with sexual orientation in rams. These provocative results in sheep are sure to stoke the flames of debate about the origin of human sexual orientation, where groups condemning homosexuality as choosing a sinful lifestyle conflict with those gay activists who claim that homosexuality is a normal, if not widespread, component of a behavioral continuum found in nature (14).
Naturally, the list of things we do not know about the sheep POA and sexual orientation is a long one. For one, which is established first: oSDN size or mate preference? One might assume that the neural structure is determined first and that this guides the development of sexual preference but, as with the human INAH3, it’s always possible that one or more other factors, including social influences, might shape sexual preference first. Then either the establishment of sexual preference and/or the experience of implementing sexual preference might later affect the size of the POA nuclei. For comparison, in humans the sex difference in the SDN-POA (also known as INAH1) does not arise until 6–10 yr of age (7), so that it is entirely possible that differential social stimulation of boys and girls is responsible for the sexual dimorphism in INAH1 in adulthood. We know nothing about the ontogeny of human INAH3.
Likewise, does the level of aromatase activity determine oSDN size, or does oSDN size determine the level of aromatase? Because present methods of assaying preoptic aromatase require removing the brain, it’s not possible to track the ontogeny of aromatase expression in female-oriented vs. male-oriented rams (brains would have to be analyzed before males have had a chance to reveal their preference). But it should be possible to compare the ontogeny of aromatase expression in males vs. females to get some hints about the sex difference in enzyme activity. For that matter, it’s possible that the higher levels of aromatase in males vs. females simply reflects differing levels of circulating androgen in adulthood. But circulating levels of androgen probably cannot explain the difference between the two types of rams, because androgen levels seem equivalent in the two types of males, despite the lower levels of aromatase expression in the oSDN of male-oriented rams (15).
If aromatase activity in the POA is responsible for the size of the oSDN, and if oSDN size in turn is responsible for sexual preference, then it might be possible to interfere with hypothalamic aromatase in developing rams to reliably shrink the oSDN and produce male-oriented rams. Would it be possible to shift a female-oriented ram’s preference, even in adulthood, by reducing aromatase activity in the POA? Either of these demonstrations would be powerful proof of a causal relationship between steroid action, hypothalamic morphology, and sexual preference in sheep. Either demonstration would also, by extension, strengthen the notion that steroids affect the human hypothalamus to influence sexual orientation in adulthood. If the results pan out this way, it’s going to be increasingly difficult to condemn homosexuality as a lifestyle choice rather than an ingrained property of the human heart.
 |
Footnotes |
|---|
Received October 29, 2003.
Accepted for publication November 5, 2003.
|
References |
|---|
 Top References |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
