Where does gender come from?
I doubt that this is a question many non-trans people would pose. I mean, really, there must be a bunch of people out in the world who have a gender that is congruent with their body and who will more often ask themselves, “Where did I leave the remote?” than, “What caused me to have this gender?”
For trans people, on the other hand, gender can be a topic that receives a lot of thought and contemplation. I can say that I have spent quite a bit of time thinking about my own gender. One of the main questions I asked myself in the months following the realization that I was transsexual was “How did this happen?”
Well, I was more wondering how I ended up with a male gender and a female body, but the origin of gender is interesting, if you think about it. What indeed does cause a person to feel like a man? Or a woman? Or both? Or neither?
There are some people who think they might know the answer.
Let’s start this discussion by talking about anatomy and physiology. We know that in vertebrates, males and females have sexual dimorphism of many different anatomical structures and physiological characteristics. In other words, there are physical differences between males and females.
Of course, Nature rarely adheres in a strict way to a binary — there is biological variation all around us — so I am speaking here in general, “average” terms.
Included on the list of sexual dimorphisms is the brain and peripheral nervous system (Morris et al., 2004). The thought is (by people who think about these things) that gender- “typical” behavior is dependent upon the action of sex steroids on the brain.
Scientists believe that mammalian brains are intrinsically female and masculinization is required during fetal development to induce male-typical behavior later in life.
Originally, they thought that the masculinizing factor in the brain was estradiol that had been produced by aromatization (conversion) of testosterone. Then with more recent data, they concluded that both testosterone and estradiol were required to masculinize the brain.
These conclusions came predominantly from studies with animals, mainly rodents (Sato et al., 2004; reviewed in Zuloaga et al., 2008). Well, that’s fine and good for rodents, but there are reasons why it can’t hold for primates, including humans.
Take, for example, the issue with estrogen produced from the ovary. In rodents, alpha-fetoprotein binds estrogen and keeps it from entering and masculinizing the female brain (Keller et al., 2010). In humans, on the other hand, alpha-fetoprotein doesn’t bind estrogen (Swartz & Soloff, 1974), so if estrogens were the masculinizing factor in humans, then most females would have male genders and behavior.
Then there are the case reports of men with mutations in the gene for aromatase (which I wrote about in a previous post). These men would not be able to produce estrogens from testosterone because of the lack of the aromatase enzyme that makes the conversion, and yet, they were living as men and so presumably had masculinized brains.
Complete Androgen Insensitivity Syndrome
Finally, there are the case studies of individuals with Complete Androgen Insensitivity Syndrome, known as CAIS. A disorder of sex development (DSD, or intersex condition), CAIS is due to genetic mutations in the androgen receptor (AR) gene that either render the AR non-functional or eradicate it altogether. (See previous posts for information on AIS and the AR and its mutations.)
The AR gene is located on the X-chromosome. If a 46,XX female (meaning a person with the standard 23 pairs of human chromosomes with two “X” sex chromosomes) inherits a mutated copy of the AR gene from her mother, she’ll still have a good copy of the gene from her father and so there will be no phenotype (i.e. physical manifestation) of the mutated copy. That means that women can be carriers of a mutated AR gene and not even know it.
On the other hand, 46,XY “males” who inherit a mutated AR gene from their mother won’t have another wild-type (i.e. non-mutated) copy from their father, so they’re left without a functional AR. Because testosterone and other androgens, like dihydrotestosterone (DHT), can only exert their actions through the AR, which these individuals don’t have, genetic males with CAIS are born with female external genitalia, a short, blind-ending vagina and internalized testes (testicles).
Because they usually appear as unremarkable females at birth, 46,XY CAIS children are raised as females and often are not diagnosed until they fail to menstruate at puberty (due to a lack of female reproductive organs), but with the exception of axillary and pubic hair which is absent or sparse, they do develop secondary female characteristics because the testosterone produced by their internal testes is aromatized to estrogen.
Individuals with CAIS are sometimes identified before puberty because of an inguinal hernia due to the testes being partially descended, or because the fetus/child was karyotyped and found to have XY chromosomes that didn’t match their female appearance.
If the condition is discovered before puberty, the testes are sometimes removed and the children are treated with estrogen to induce puberty at the correct age, but more recent guidelines suggest leaving the testes intact so that the estrogen aromatized from testicular-derived testosterone can induce female secondary characteristics at puberty. Eventually, the testes must be removed because of the increased risk of testicular cancer after puberty (see review by Galani et al., 2008).
No matter when the condition is discovered, these children are raised as females, but what is their gender identity?
In a study involving 22 women with CAIS (46,XY), there was no evidence (based on their answers to questionnaires) that they were less masculine or more feminine in their gender role behavior or identity than the 33 non-CAIS (46,XX) women who were included in the study (Hines et al., 2003). Later, a review of the world literature by Mazur (2005) found 156 cases of individuals with CAIS — all were raised as females, all identified as female and there were no documented reports of gender dysphoria.
So, then, these data support the belief that the action of androgen in the brain causes masculinization, male-typical behavior and male gender identity because in the presence of XY sex chromosomes but the absence of a functional AR, individuals have a female gender identity.
We trans people, however, know that Nature is not so cut and dried.
Male Gender Identity in an Individual with CAIS
A recent report about an individual with CAIS and a male gender identity calls into question this entire theory about testosterone being the factor that causes masculinization of the human brain and gender.
T’Sjoen and colleagues (2010) provide a case study of an individual who was unambiguously female at birth, was raised female, and who was diagnosed (because of lack of menstruation) with CAIS at 17 years of age. Karyotyping revealed 46,XY chromosomes and sequencing of the individual’s androgen receptor gene found a mutation that has been reported in other cases of CAIS. The testes were removed and a hernia was repaired the year after diagnosis.
Estrogen replacement therapy was begun but caused a significant emotional response and treatment was stopped because of this person’s displeasure with feminization of their body. A few years later, he reported a male gender identity that he’d had since childhood.
He began therapy and his male gender was subsequently confirmed. He began living as a man, took a male name and had surgery to masculinize his chest. A course of testosterone administration and later treatment with DHT had no virilizing effects because of his lack of androgen receptors. With no option for metoidioplasty because his genitalia were not masculinized, he eventually chose to undergo phalloplasty and scrotoplasty. He continues to take testosterone and so has no issues with osteoporosis. He lives, according to the authors, “a well adjusted life.”
What About the Original Question?
So we’re back to square one, right? Where does gender come from? What does this mean for the theory of brain masculinization and male gender identity?
Well, it wasn’t all that good of a theory to begin with, at least in my opinion. Let’s just look at the statistics of the situation. If we pull 156 female-bodied individuals off the street, this number coming from Mazur’s review of reports involving people with CAIS that I mentioned above, and we put them in a room (and ply them with cookies and coffee so they’ll hang around for a little while), what would the chances be of finding one person in the group with a male gender identity?
Granted, it’s difficult to count transsexuals and know our frequency in the general population, but I think we can be confident that the chances would be pretty slim that a random group of 156 female-bodied individuals would yield one with a male gender identity, and that’s basically what you have with the documented cases of individuals with CAIS.
In other words, if testosterone were not the sole factor that determines male gender identity, we might not know it because our sample size of CAIS individuals isn’t big enough to prove it.
In addition, Nature is rarely so cooperative with simple theories. Take the situation of biological sex determination. When I was a just a baby scientist in graduate school, we thought that a Y chromosome, a functional androgen receptor and testosterone was all a person would need to have typical male biological sex, but now we know that sex determination is much more complicated than that (Piprek, 2009). Then I ask, “Why would gender determination be any different?”
So now, having said all that, I have a confession to make. I have simplified the situation a bit.
Although T’Sjoen and co-authors of this case study stated, “To our knowledge, there is no documented case of persistent gender dysphoria leading to cross-sex hormonal treatment and sex reassignment surgery in an individual with CAIS,” there actually were hints in the literature along these lines before their report was published.
In a letter to the editor of Archives of Sexual Behavior regarding the case study by T’Sjoen et al., Meyer-Bahlburg (2010) points out:
…the absence of documented similar cases in the world literature (see the review by Mazur, 2005) had led to a virtual textbook dogma that CAIS was associated with female psychological development. If this dogma were valid, one would expect that psychological studies of gender-related behavior in girls and women with 46,XY CAIS show means and variances comparable to those of non-DSD girls or women, if not even lower, given the normal variability in androgen exposure of girls and women and the effects of such exposure on gender-behavior variability… However, a careful examination of the limited psychological studies of CAIS that provide details on gender-related behavior and identity reveals a more variable gender outcome.
After listing the pertinent points from these psychological studies that he mentioned from the literature, Meyer-Bahlburg adds:
Although these diverse studies suffered from small sample sizes and possible selection biases, in combination they suggest more heterogeneity in the development of gender-related behavior and identity than one would expect on the basis of the stereotypic notion of CAIS femininity. From that perspective, the occasional development of a male gender identity and behavior in 46,XY CAIS appears less surprising.
There is much more to this story than what I have presented here. In order to get a more complete picture of brain masculinization and gender, I will devote Part 3 of this post series to discussing what we know about fetal brain and genital development, the timing of sex steroid production in the fetus, genetic and physiologic diversity of androgen receptor mutations, and a look at gender identity and behavior in other disorders of sexual development.
Acknowledgment: I would like to thank Jamison Green for kindly procuring and providing pdf files of many of the references cited in this post series.
References (some with free pdf downloads)
Holterhus, Sinnecker, Hiort, 2000. Phenotypic Diversity and Testosterone-Induced Normalization of Mutant L712F Androgen Receptor Function in a Kindred with Androgen Insensitivity. J Clin Endocrinol Metab 85:3245-3250 (free article)
Keller, Pawluski, Brock et al., 2010. The alpha-fetoprotein knock-out mouse model suggests that parental behavior is sexually differentiated under the influence of prenatal estradiol. Horm Behav 57:434-440
Sato, Matsumoto, Kawano et al. 2004. Brain masculinization requires androgen receptor function. PNAS 101:1673-1678 (free article)
Swartz & Soloff, 1974. The Lack of Estrogen Binding by Human -Fetoprotein. J Clin Endocrinol Metab 39:589-591
Zuloaga, Puts, Jordan, Breedlove, 2008. The role of androgen receptors in the masculinization of brain and behavior: what we’ve learned from the testicular feminization mutation. Horm Behav 53:613-626 (free article)