sign in sign up
<<
Home , Gender, Marc Breedlove, Ring finger

and Behavior 111 (2019) 23–30

Contents lists available at ScienceDirect

Hormones and Behavior

journal homepage: www.elsevier.com/locate/yhbeh

Review article Individual differences in the biological basis of androphilia in mice and men T ⁎ Ashlyn Swift-Gallant

Neuroscience Program, Michigan State University, 293 Farm Lane, East Lansing, MI 48824, USA Department of Psychology, Memorial University of Newfoundland, St. John's, NL A1B 3X9,

ARTICLE INFO ABSTRACT

Keywords: For nearly 60 years since the seminal paper from W.C Young and colleagues (Phoenix et al., 1959), the principles Androphilia of sexual differentiation of the and behavior have maintained that -typical sexual behaviors (e.g., Transgenic mice lordosis) and sexual preferences (e.g., attraction to males) are the result of low levels during devel- Androgen opment, whereas higher androgen levels promote male-typical sexual behaviors (e.g., mounting and thrusting) Sexual behavior and preferences (e.g., attraction to ). However, recent reports suggest that the relationship between Sexual preferences and male-typical behaviors is not always linear – when androgen signaling is increased in male rodents, via exogenous androgen exposure or androgen receptor overexpression, males continue to exhibit male- typical sexual behaviors, but their sexual preferences are altered such that their interest in same- partners is increased. Analogous to this rodent literature, recent findings indicate that high level androgen exposure may contribute to the sexual orientation of a subset of men who prefer insertive anal sex and report more male- typical gender traits, whereas who prefer receptive anal sex, and who on average report more gender nonconformity, present with biomarkers suggestive of low androgen exposure. Together, the evidence indicates that for both mice and men there is an inverted-U curvilinear relationship between androgens and sexual pre- ferences, such that low and high androgen exposure increases androphilic sexual attraction, whereas relative mid-range androgen exposure leads to gynephilic attraction. Future directions for studying how individual differences in biological development mediate sexual behavior and sexual preferences in both mice and humans are discussed.

1. Introduction down to more androgens results in more masculine behavior; however, recent evidence points to a more complex nonlinear relationship be- The neuroendocrine basis of sexual behavior and sexual preferences tween androgens and male-typical sexual preferences, and here I will is well studied in rodents. The research indicates that gonadal hor- review the literature suggesting that high androgen signaling promotes mones act during critical periods in development to organize neural androphilia (i.e., sexual preference for male partners) in male rodents. I in a sex-typical fashion (e.g., Morris et al., 2004). In adult- will then discuss how these findings may apply to humans, and review hood, activational or transient effects of hormones further act on these the recent literature suggesting multiple biological pathways (i.e., in- sexually differentiated brain structures to promote sex-typical beha- cluding high androgen signaling) underlie androphilia in men. viors, including sexual behavior towards opposite-sex partners. The landmark paper by Phoenix et al. (1959) demonstrated that male-ty- 2. Androphilia and androgens in mice pical sexual behaviors like mounting are increased in female guinea pigs androgenized during embryonic development, whereas female 2.1. From manipulation to transgenic mouse models: androgens, sexual behaviors, like lordosis, are decreased in these females even sexual behavior and sexual preferences in rodents when primed with estrus-inducing hormone regimens in adulthood. Since this formative study, a plethora of evidence has emerged sup- Hormone manipulation studies were the first line of evidence to porting the of androgen action during early critical periods in de- suggest that androgen action during early development organize the velopment as well as in (reviewed in Schulz and Sisk, 2016), for brain and behavior in a sex-typical fashion. Females exposed to an- masculinizing and defeminizing neural circuits and adult sex-typed drogens during early critical periods display male-typical sexual beha- behaviors. By and large, the story of sexual differentiation seems to boil viors in adulthood; Phoenix et al. (1959), first reported these findings in

⁎ Department of Psychology, Memorial University of Newfoundland, 232 Elizabeth Ave, St. John's, NL A1B 3X9, Canada. E-mail address: [email protected]. https://doi.org/10.1016/j.yhbeh.2018.12.006 Received 31 July 2018; Received in revised form 21 November 2018; Accepted 11 December 2018 Available online 27 December 2018 0018-506X/ © 2018 Elsevier Inc. All rights reserved. A. Swift-Gallant Hormones and Behavior 111 (2019) 23–30 guinea pigs, and many others have replicated these findings in mice and the enzyme aromatase to assess estrogenic effects on behavior; indeed, rats (for review, see Cooke et al., 1998). In male rodents, removal of such studies have indicated that androgen action via ERs contributes to gonadal androgens during early development (e.g., Gerall et al., 1967), the display of male-typical sexual behavior and sexual preferences (e.g., or prenatal and postnatal treatment with anti-androgens (e.g., fluta- Brand et al., 1991; Clemens and Gladue, 1978; for review in rats, see mide: Clemens et al., 1978; Casto et al., 2003; cyproterone acetate: Bakker et al., 1996; in mice, see Bodo, 2008; Brock and Bakker, 2011; Ward and Renz, 1972), results in dramatic decreases in male-typical for review comparing mice and rats, see Bonthuis et al., 2010). How- sexual behaviors (i.e., decreased mounting, intromissions and ejacula- ever, a second limitation of androgen or estrogen manipulation studies tion), and increased female-typical behaviors such as lordosis (e.g., is that simply manipulating circulating hormones may not alter hor- Gladue and Clemens, 1978; Ward and Renz, 1972) when treated with mone levels in the brain; for example, removal of both gonadal and hormone regimens that typically induce receptivity in females (i.e., adrenal hormones at birth does not seem to alter the neural endocrine estradiol and progesterone). Similarly, sexual preferences, as measured environment, even 3 days following gonad and adrenal removal by simultaneous exposure to male and female partner or sexual (Konkle and McCarthy, 2011); these findings could be accounted for by stimuli, are mediated by androgens: male-typical levels of androgen de novo steroid synthesis in the brain (Robel and Baulieu, 1995; re- exposure during early development increases preference for female viewed in Diotel et al., 2018 and Forger et al., 2016). Neural hormone sexual stimuli, whereas the absence of or low-level androgen exposure implantations were first to address the role of hormones directly in the leads to a preference for male sexual stimuli (e.g., Domınguez-Salazar brain for behavior (e.g., Davidson, 1966; for review see McEwen et al., et al., 2002; Stern, 1970). Together, such studies suggest that low level 1979; Frye, 2001), and modern transgenic technology (and sponta- androgen exposure during development results in female-typical sexual neous mutations of the androgen receptor ) has allowed for the behaviors and preferences, while higher male-typical androgen levels refined testing of the effects of androgenic signaling compared to es- promote male-typical sexual behaviors and preferences in adult mice trogenic signaling on sexual behaviors by targeting hormone receptors. and rats (see Table 1). With transgenic mouse models, we can ensure hormone sensitivity is There are a couple caveats to consider when evaluating these early altered specifically in neural tissue, even if hormone synthesis con- androgen manipulation studies; for one, androgens can act via both tinues in the brain following gonadectomy. androgenic and estrogenic signaling pathways. Testosterone, the pri- Prior to modern transgenic technology, Lyon and Hawkes (1970) mary androgen produced by the gonads, can be metabolized into either described the condition of testicular feminization mutation (Tfm) in the more potent androgen dihydrotestosterone (DHT) or, with the help mice, in which XY chromosomal male mice showed a female phenotype of the enzyme aromatase, converted to estrogens (Naftolin, 1994; re- at birth. It was later discovered that these rodents have a single nu- viewed in Roselli et al., 2009). Thus, it is difficult to parse out the ef- cleotide deletion in the androgen receptor gene, causing a frameshift fects of androgen action via the androgen receptor (AR) versus the es- mutation that renders AR nonfunctional (Charest et al., 1991). Males trogen receptors (ERs). Many studies have manipulated estrogens and with this mutation present as females in their somatic features, having a

Table 1 Endocrine mediation of male-typical and female-typical sexual behaviors and preferences.

Endocrine manipulation Male-typical sexual Female-typical sexual Gynephilia Androphilia References behavior behavior

Neonatal decrease in T in males For review, see Cooke et al. (1998)

Neonatal increase in T in males a e.g., Henley et al. (2010) and Cruz and Pereira (2012)

Estrogenic manipulations Global ERα KO in males Ogawa et al. (1998) and Wersinger and Rissman (2000)

Global ERβ KO in males Kudwa et al. (2005)

Global AFP KO in females Bakker et al. (2006, 2007)

Global Arom KO in males a Honda et al. (1998) and Bakker et al. (2002)

Androgenic manipulation Global AR KO in males Reviewed in Zuloaga et al. (2008), Bodo and Rissman (2007) and Sato et al. (2004)

Neural AR KO in males a Juntti et al. (2010) and Raskin et al. (2009)

Global AR overexpression in males a Swift-Gallant et al. (2016a, 2016b)

Neural AR overexpression in males a Swift-Gallant et al. (2016a, 2016b)

Abbreviations: T = testosterone; ERα = estrogen receptor alpha; ERβ = estrogen receptor beta; KO = knockout; AFP = alpha-feto protein; Arom = aromatase; AR = androgen receptor. Note: Comparison group is wildtype same-sex counterparts. Male-typical sexual behaviors refer to behaviors such as mounting and thrusting, while female-typical behavior refers to lordosis; these behaviors are typically measured in response to a receptive female partner or a male partner presented to experimental mice. Gynephilia (attraction to females) and androphilia (attraction to males) has generally been tested in the above experiments via a choice preference test where male and female are presented simultaneously to experimental mice. a Not tested with estrus priming hormones, thus it is possible that female-typical behaviors could be displayed if primed.

24 A. Swift-Gallant Hormones and Behavior 111 (2019) 23–30 smaller anogenital distance, blind vaginal canal, nipples and a reduced female-typical sexual behaviors in adulthood, suggesting maternal es- body weight, falling within the female-typical range. Their behavior is trogens are sufficient to masculinize and defeminize the female CNS also more female-typical, such that are demasculinized in sexual and behavior in the absence of AFP (Bakker et al., 2006). However, behaviors such as mounting (reviewed in Zuloaga et al., 2008). Tfm sexual preference for female sexual stimuli is not masculinized in AFP- male mice also exhibit a female-typical preference for male sexual sti- KO females (Bakker et al., 2007), supporting a role for androgen action muli and show a female-typical pattern of neural activity in response to via AR for male-typical sexual preferences (reviewed in Forger et al., same-sex odors, even when treated with exogenous estrogens (Bodo and 2016). Rissman, 2007). Together these findings suggest AR is necessary for Without aromatase, androgens can only act directly via AR because male-typical behaviors and sexual preferences, and the lack of andro- they cannot be metabolized to estradiol to act via ERs. Interestingly, genic signaling increases female-typical sexual preferences. Subsequent male aromatase KO mice display dramatic decreases in male-typical studies indicated that the lack of functioning AR affects estrogenic sexual behaviors (i.e., mounting of receptive females; Honda et al., signaling (Bodo and Rissman, 2007), and there is some evidence for 1998) and they show a decreased preference for female odor stimuli non-classical effects of AR or a novel/yet-to-be identified AR that may (Bakker et al., 2002), suggesting that aromatization of testosterone to be functional in Tfm male mice (Tejada and Rissman, 2012). Never- estradiol is necessary for these behaviors. However, aromatase KO theless, the Tfm mouse model has been valuable in studying the role of males continue to show similar patterns of neural activation in response classical AR function, and provides evidence for a role of AR in male to opposite and same-sex odors to wildtype males (Pierman et al., sexual behaviors and preferences (reviewed in Zuloaga et al., 2008; 2008), suggesting that the male-typical neural response to opposite-sex Swift-Gallant and Monks, 2017). odor stimuli depends on androgen action via AR. With modern transgenic technology, mouse models are now avail- Together, the evidence suggests that androgens act via both AR and able with AR knocked-out either globally or selectively in neural tissue, ERs to promote male-typical sexual behavior and inhibit female-typical allowing for a more refined understanding of the role of AR for male- sexual behavior. All knockout models interrupting the ability for an- typical sexual behaviors and preferences. Results from global AR drogens to act via AR or ERα drastically alter male-typical behaviors knockout (KO) models largely recapitulate the finding from the Tfm such as mounting, intromissions and ejaculation. ERβ, on the mouse literature. For example, Sato et al. (2004) found that male mice , is required for the defeminization of behaviors such as lordosis, with complete AR-KO did not display male-typical sexual behaviors and AR is uniquely required for male sexual odor preferences and including mounting, intromissions and ejaculation; estradiol adminis- corresponding neural activity in response to sexual odor stimuli. tration was partially successful in restoring mounting and intromissions in these mice, but not ejaculation. Similarly, neural-only AR-KO re- 2.2. High androgen signaling promotes androphilia in male mice sulted in decreased male sexual behaviors (Juntti et al., 2010; Raskin et al., 2009), although not to the extent found with complete loss of AR Sexual behavior and preferences have been extensively studied in function. Furthermore, sexual odor preferences were reported to be relation to little or no androgen exposure. Far fewer studies have undisturbed in neural-only AR-KO males (Raskin et al., 2009), whereas evaluated the effects of androgen levels at the high end of the male- Tfm male mice with complete AR insensitivity displayed a sex-reversal typical range and/or supraphysiological levels, and only a couple stu- in this behavior. These results support a role for androgens via AR for dies have evaluated the effects of enhanced sensitivity to androgens on male sexual behaviors, although KO of neural AR alone is insufficient to behavior. The available research suggests that while greater androgen affect sexual preferences, suggesting non-neural AR plays a role in signaling is associated with enhanced male-typical behaviors, sup- sexual preferences in mice (reviewed in Swift-Gallant and Monks, porting a linear relationship between androgens and male-typical 2017). Although, Chen et al. (2015) raise the concern that AR is not phenotype, other behaviors such as sexual preferences are more female- completely knocked-out in the brain of the Nestin-AR KO mouse model, typical in response to high androgen signaling, indicative of a nonlinear and thus it is possible that residual AR expression could continue to relationship for these behaviors. masculinize some aspects of behavior. At least 6 studies have evaluated the effects of elevated androgen The influences of both estrogen receptor alpha (ERα) and estrogen exposure during early critical periods in development on the display of receptor beta (ERβ) have also been tested using mouse KO models. With adult male sexual behavior and/or preferences in rodents (Diamond ERα-KO, male mice exhibit decreases in male-typical sexual behaviors, et al., 1973; Piacsek and Hostetter, 1984; Zadina et al., 1979; Henley although testosterone and DHT are both able to induce mounts and et al., 2010; Cruz and Pereira, 2012). Overall, these studies suggest that intromissions in ERα-KO males, but not ejaculation (Ogawa et al., exogenous androgen treatment during early development increases 1998). Interest in investigating female partners is also decreased in preference for same-sex partners in male rodents. For example, Henley ERα-KO male mice, although they continue to show increased neural et al. (2010) found that sexual behaviors were reduced in response to activation in response to female odors, similar to wildtype males, and female conspecifics among male rats exposed to supraphysiological they exhibit the expected surge in luteinizing hormone in response to levels of testosterone during early postnatal development (i.e., day of opposite-sex odors (Wersinger and Rissman, 2000). On the other hand, birth-postnatal day 21). Instead, these hyperandrogenized male rats while ERβ-KO males continue to show male-typical sexual behaviors exhibited an increase interest in investigating male partners compared and a preference for female sexual stimuli, they exhibit a higher pro- to unmanipulated male rats that preferred female partners. Similarly, pensity for female-typical sexual behaviors such as lordosis when Cruz and Pereira (2012) found that male rats administered testosterone primed with estrus-inducing hormones when compared with wildtype between embryonic days 17–19 (i.e., administered to pregnant dams; male mice (Kudwa et al., 2005). Together, these results suggest ERα is coinciding with the endogenous spike in androgens in unmanipulated required for masculinization of male sexual behaviors, while ERβ is male rats), showed a decrease preference for female partners and in- responsible for the defeminization of sexual behavior in male mice creased preference for male partners compared to control male rats. (reviewed in Kudwa et al., 2006). Collectively, androgen manipulation studies indicate that at the high Similar conclusions have been drawn from studying alpha-feto extent of androgen signaling male-typical sexual behaviors persist, but protein (AFP) and aromatase KO mouse models. AFP in embryonic fe- male-typical preferences are altered, such that male rodents show in- males restricts maternal estrogens from entering the central nervous creased androphilic sexual preference. system (CNS) and masculinizing neural (McEwen et al., The role of higher androgen signaling via AR was recently tested 1975). Thus, sans AFP, maternal estrogens can enter the developing with modern transgenic technology. When AR is overexpressed in male brain in female offspring and masculinize neural development. Indeed, mice to levels 3–4× higher than in wildtype counterparts, male mice AFP-KO females exhibited male-typical sexual behaviors and not continued to exhibit male-typical sexual behaviors while androphilic

25 A. Swift-Gallant Hormones and Behavior 111 (2019) 23–30 sexual preferences were increased (Swift-Gallant et al., 2016a, 2016b). doubt continue to evolve. Next, I will examine how the principles of Specifically, using cre-loxP system, two models of AR overexpression sexual differentiation derived from this rodent literature may apply to were created, one with global overexpression and a second with neural- humans and propose future steps for understanding neuroendocrine specific overexpression of AR. Swift-Gallant et al. (2016a) found that influences on human sexual orientation. male mice with global AR overexpression exhibited an increase in male- typical sexual behaviors (e.g., number of thrusts/mount) in response to 3. Androgens and androphilia in men a receptive female partner. While these males continued to show sexual interest in female partners, males with global overexpression showed a Given the role of androgens on sexual preferences reported in the greater preference for the anogenital investigation of male partners and non-human animal literature, one of the first hypotheses for the me- a decrease in aggressive behaviors towards male intruders (Swift- chanisms behind same-sex sexual orientation in humans was the neu- Gallant et al., 2016a). Moreover, male mice with global AR over- rohormonal theory of sexual orientation (Ellis and Ames, 1987). In expression show a greater preference for male odors when presented essence, this theory states that opposite-sex attraction in humans is si- simultaneously with male and female odor stimuli (Swift-Gallant et al., milarly dependent on androgen exposure, as it is in rodents: gynephilia 2016b). Neural activity in response to female odors along the accessory (i.e., sexual attraction to females) results from high androgen titers olfactory pathway was also decreased in male mice with global AR during early critical periods, whereas relatively low levels of androgen overexpression compared to control males, corresponding with the during these same critical periods results in androphilia (i.e., sexual behavioral results (Swift-Gallant et al., 2016b). Conversely, male mice attraction to males). Thus, it has been predicted that same-sex sexual with neural-only AR overexpression (Nestin-AR; Swift-Gallant et al., orientation in men would result from lower androgen exposure during 2016a, 2016b) did not exhibit differences in sexual behavior or sexual development, whereas higher androgen exposure in women was pre- preferences compared to wildtype males, though these mice did display dicted to result in same-sex sexual orientation in women. However, the a dramatic decrease in inter-male aggression. Together, these results human literature investigating this theory has been fraught with con- suggest that global increases in androgenic signaling promote andro- flicting findings, and there are no clear indications that sexual or- philic sexual preferences and male-typical sexual behaviors in male ientation in men is mediated by androgens (e.g., Breedlove, 2017; al- mice, whereas neural-only increases in AR were insufficient to alter though gynephilia in women does seem related to exposure to prenatal these behaviors. androgens). It is possible that the mixed findings in the literature are The literature to date suggests that at the high end of androgen due to the methods of research (i.e., use of noisy biomarkers, as we signaling androphilic sexual preferences are increased, while male-ty- cannot manipulate and experiment on humans as we do in rodents), pical sexual behaviors (i.e., mounting, thrusting) remain intact. In other and/or there may be subgroups of gay men who owe their sexual or- words, male rodents treated with exogenous androgens or with global ientation to distinct bio-developmental factors. If the latter were the increases in AR signaling continue to perform the same sexual beha- case, then studying all gay men as a single group would mask any ef- viors as wildtype males (i.e., mounting/thrusting), but these hyperan- fects that pertain to particular subsets of same-sex oriented men. For drogenized males exhibit an increased preference for same-sex partners example, given the findings in the rodent literature, it is possible that and sexual stimuli. However, there remain questions for future research both low and high androgen levels during early development result in to address: namely, in some studies hyperandrogenization decreased androphilic attraction; if both subgroups of gay men are studied as a preference for female partners while increasing androphilia (e.g., Cruz single group, then conflicting findings would be expected. Next, I will and Pereira, 2012), while in other studies preference for female part- briefly review the literature evaluating retrospective markers of pre- ners remained intact but androphilia was increased (e.g., Swift-Gallant natal androgens for male sexual orientation and end by discussing a few et al., 2016b). One possibility is that there are species differences given recent studies that support the possibility that subgroups of gay men that the first set of experiments reported were conducted in rats may owe their sexual orientation to distinct bio-developmental me- whereas the latter were in mice. Alternatively, these inconsistencies chanisms. could be related to androgen dose timing: in the first set of studies supraphysiological levels of androgen were administered only during 3.1. Retrospective markers of the prenatal endocrine environment early critical periods whereas the latter studies, using transgenic mouse models, AR signaling was increased throughout the entire lifespan of It is admittedly difficult to study the prenatal neuroendocrine pro- the mouse. Future studies will be required to delineate the nature of file in humans. Amniocentesis would be required to assess hormones these differences. during prenatal neural development, when hormones would be pre- dicted to affect sexual orientation; given the risks associated with such a 2.3. Conclusions procedure, this is rarely done. Plus, this method provides only a snap shot of hormone levels on a single day and time, when androgen ex- The rodent literature suggests that low androgen exposure results in posure and its effects happen over weeks. Thus, putative biological both female-typical sexual behaviors and female-typical sexual pre- markers are often used as proxies to indicate the prenatal hormonal ferences, whereas higher male-typical levels of androgen exposure and milieu. sensitivity to androgens via both AR and ERs is required for the display One well studied biomarker of prenatal development is the second- of male-typical sexual behaviors and preferences - of or to-fourth finger ratio (2D:4D). is determined prenatally insensitivity to the male-typical androgen surge during early critical (Galis et al., 2010; Malas et al., 2006), and for reasons not well un- periods results in sex-reversals in behavior. Paradoxically, when an- derstood, androgens play a large part in establishing this ratio. Speci- drogen signaling is abnormally high, male rodents also exhibit a sex- fically, males on average have a lower ratio (i.e., index finger [2D] is reversal (i.e., increased androphilia) in sexual preferences, while con- shorter than the ring finger [4D]), whereas females tend to have a ratio tinuing to display male-typical sexual behaviors. Many in the field closer to 1 (i.e., equal or longer index to ring finger length). Convincing continue to explore the intricacies of sex hormones, and their precise evidence that androgens influence this ratio comes from studies of in- role in promoting sex differences (e.g., reviewed in de Vries and Forger, dividuals that have altered androgen exposure or sensitivity. For ex- 2015; Swift-Gallant and Monks, 2017) besides the many other factors ample, chromosomal XY males with complete androgen insensitivity that influence sexual differentiation (e.g., reviewed in McCarthy and syndrome (CAIS) have a mutation in the androgen receptor gene that Arnold, 2011; Forger, 2016); however, it is clear that the processes renders the receptor dysfunctional (Hughes et al., 2012; Gottlieb and underlying sexual behavior and sexual preferences are multifaceted and Trifiro, 1999). Thus, these XY individuals are insensitive to androgens, the principles of sexual differentiation of the brain and behavior will no presenting at birth with female-typical genitalia. Interestingly, their

26 A. Swift-Gallant Hormones and Behavior 111 (2019) 23–30

finger digit ratio is also female-typical, indicating that androgens cri- 3.2. Retrospective markers of prenatal androgen and male sexual tically mediate the digit length ratio, and that sex chromosomes and orientation anti-Mullerian hormone do not play a role in masculinizing digit ratio (Berenbaum et al., 2009). Similar conclusions were drawn from Same-sex sexual orientation in men has been associated with all of studying exposed to high levels of androgens during early devel- the above-described biomarkers, however the literature has been in- opment; girls with congenital adrenal hyperplasia (CAH) are exposed to consistent. For example, a number of studies have indicated that 2D:4D high androgen levels prenatally due to an enzyme deficiency that sti- is more female-typical among gay men (e.g., Lippa, 2003; McFadden mulates the adrenals to produce excess androgens (Yau et al., 2015); and Shubel, 2002; Hall and Schaeff, 2008), whereas others have found the finger digit ratios on the right hand are more like males in girls with that 2D:4D is more male-typical (e.g., Rahman and Wilson, 2003; CAH, further implicating androgens (Brown et al., 2002; Ökten et al., Robinson and Manning, 2000; Rahman, 2005) when compared to het- 2002). Though the reasons remain unknown, there is a lateral asym- erosexual men. Other studies found no differences in 2D:4D ratios be- metry, such that the 2D:4D ratio is associated with androgen exposure tween gay and heterosexual men (e.g., Williams et al., 2000; Kraemer on the right but not the left hand (e.g., Brown et al., 2002; Buck et al., et al., 2006; Voracek et al., 2005), and two meta-analyses concluded 2003). Evidence from the rodent literature also supports a role for an- that 2D:4D is not associated with sexual orientation in men across drogens for 2D:4D; mice with AR knocked-out either globally or locally geographical locations (McFadden et al., 2005; Grimbos et al., 2010). in the paws display a more female-typical digit ratio (Zheng and Cohn, Similarly, handedness and physical development measures have 2011). Together, these data make a strong case for androgens mediating also received mixed findings when studied in relation to male sexual 2D:4D, however, of course other biological processes can affect 2D:4D, orientation. Non-right-handedness has been reported to be higher including estrogens (Zheng and Cohn, 2011) and genetic factors among gay men compared to heterosexual men (e.g., see meta-analysis, (Gobrogge et al., 2008). Nonetheless, 2D:4D is one of the most well- Lalumiere et al., 2000; Blanchard and Lippa, 2007), although other established biomarkers of prenatal androgens (reviewed in Breedlove, studies did not replicate this effect (e.g., Bogaert and Blanchard, 2017). 1996a). Gay men have also been reported to be shorter, weigh less and Another biomarker linked to the processes of sexual differentiation report an earlier pubertal onset when compared to heterosexual men and androgens is handedness. Preference for the use of the right or left ( Bogaert and Blanchard, 1996b; Bogaert et al., 2002; Bogaert, 2010; hand for various tasks is determined prenatally (Hepper et al., 1991; Skorska and Bogaert, 2017) although other reports did not replicate Hepper, 2013) and is sexually differentiated (for meta-analysis, see these findings (Bogaert and Friesen, 2002; Savin-Williams and Ream, Papadatou-Pastou et al., 2008); males tend to have a greater left-hand 2006). preference than females. There are reports that girls with CAH and girls The fraternal birth order effect (FBOE), or the higher preponderance that are gender nonconforming have an increased left-hand preference of older brothers among gay men compared to heterosexual men, is the (CAH: Nass et al., 1987; Kelso et al., 1999, but see Helleday et al., 1994; most well-established biomarker of male sexual orientation to date (see gender nonconformity: Casey and Nuttall, 1990; Nicholls and Forbes, meta-analysis, Blanchard, 2018). Although no direct evidence im- 1996), implicating androgens as the factor driving sex differences in plicates androgens with this biomarker (although see James and Grech, handedness. There have also been reports of an association between 2017), it is related to handedness, and this literature suggests that there normal polymorphisms in the androgen receptor gene (i.e., CAG re- may be subgroups of gay men who may owe their sexual orientation to peats) and mixed handedness. The number of CAG repeats in the first distinct biological mechanisms. Briefly, the FBOE is a biomarker at- exon of the AR gene varies normally in the human population (i.e., tributed to prenatal biological processes rather than a social influence 11–31 repeats; Edwards et al., 1992). Fewer CAG repeats is associated of having more older brothers; for example, having more non-biological with greater AR potency (e.g., increased risk of prostate cancer; older brothers does not affect male sexual orientation (Bogaert, 2006), Giovannucci et al., 1997) whereas a larger number of CAGs is asso- whereas evidence supports the hypothesis that an immunological me- ciated with reduced AR potency (e.g., hypogonadism; Tirabassi et al., chanism is behind this biomarker (i.e., the maternal immune hypoth- 2015). Thus, the association of AR CAG repeats and handedness im- esis; e.g., Bogaert et al., 2018). Interestingly, FBOE is more consistently plicates androgens in determining this trait (Arning et al., 2015; found among gay men who are right-handed, suggesting that the in- Hampson and Sankar, 2012; Medland et al., 2005). Although other creased incidence of left-handedness among gay men is present in a influences likely also contribute to hand preference, including genetic, separate subgroup of gay men in which same-sex sexual orientation epigenetic factors and immunological mechanisms (reviewed in likely depends on mechanisms distinct from right-handed gay men (for Lalumiere et al., 2000; Schmitz et al., 2017). review, see Blanchard, 2008). One of the largest sex differences between men and women is their Together, the inconsistencies in the literature suggest that prenatal physical development. Males tend to be taller, heavier, hairier and go androgens are not associated with androphilia in human males (re- through puberty later than females (reviewed in Underwood and Van viewed in Breedlove, 2017). Some studies suggest that gay men were Wyk, 1992; Barber, 1995). These sex differences are thought to be as- exposed to higher levels of androgens, while other studies suggest low sociated with climbing titers of circulating androgens in adolescent levels of androgen lead to same-sex sexual orientation. Many others males. Strong support for the link between physical development and have not found evidence that these biomarkers are consistently present androgens comes from studies of XY individuals with CAIS; these across samples of gay men. However, as found in mice, the relationship chromosomal males who are insensitive to androgens develop female- between androgens and sexual orientation may be more complex, typical somatic measures, being shorter, lighter and less hairy com- namely, not necessarily linear. For example, both low and high an- pared to control males (e.g., Danilovic et al., 2007). Another example is drogen signaling in male mice can lead to same-sex preferences – if this the timing of puberty. Even though males typically go through puberty were also the case in humans, then it could explain the mixed findings later than females, an earlier pubertal onset in males is associated with in the literature. The literature on the FBOE and handedness suggests higher androgen levels and more masculine somatic measures (Yousefi that at least two separate subgroups of gay men exist who owe their et al., 2013; Rey et al., 2016). Together, given the role of androgens on sexual orientation to separate bio-developmental processes. Thus, it is these measures, they are also used as biomarkers of neuroendocrine indeed possible that the conflicting findings in the literature reflect influences on sexual orientation and gender. multiple distinct bio-developmental pathways underlying male sexual orientation, including separate subgroups influenced by the FBOE and both low and high androgen signaling.

27 A. Swift-Gallant Hormones and Behavior 111 (2019) 23–30

3.3. Retrospective biomarkers and anal sex role among gay men investigate other biomarkers, such as 2D:4D, in relation to anal sex role. Other important questions include whether anal sex role serves as the Recently, we have tested the hypothesis that subgroups of gay men best proxy for biological subgroups; previous research has found that exist who differ in their bio-developmental trajectories that contribute left-handedness and FBOE are associated with distinct subgroups of gay to their sexual orientation development. We asked whether anal sex men, whereas the literature on these two biomarkers and anal sex role role could serve as a valid proxy for subgroups of gay men, because anal suggest that both handedness and FBOE are present among gay men sex role groups tend to differ on measures affected by the processes of with a bottom anal sex role. One possibility is that the biological pro- sexual differentiation, such as gender-typicality (Zheng et al., 2012; cesses underlying handedness and FBOE contribute to the development Swift-Gallant et al., 2017; Swift-Gallant et al., 2018a, 2018b; of sexual orientation for two distinct subgroups of gay men that both Moskowitz and Hart, 2011, and cognitive style: Zheng et al., 2015). have a bottom anal sex role preference. The findings on anal sex role Specifically, since gender nonconformity has been associated with early and biomarkers also indicate that gay men with a bottom anal sex role androgen exposure (Auyeung et al., 2009; Rieger et al., 2008; Csathó tend to report more gender nonconformity, raising the possibility that et al., 2003; McIntyre, 2003; Wallien et al., 2008; Hisasue et al., 2012), the processes underlying the FBOE and handedness (i.e., biomarkers it was predicted that gay men who prefer a receptive (i.e., bottom) anal with higher prevalence among gay men with a bottom anal sex role sex role, and who report higher gender nonconforming, would differ in preference) affect both the sexual differentiation of gender traits and retrospective biomarkers of androgens from gay men with an insertive sexual orientation, whereas the research suggests that the processes (i.e., top) anal sex role who tend to be more gender conforming. To underlying the development of physical development and puberty date, anal sex role has been studied in relation to handedness, physical promote increased in gender and somatic measures while traits and the FBOE (Swift-Gallant et al., 2017, 2018a, 2018b; also increasing androphilia. These results are in some ways analogous Wampold, 2017). to the mouse literature; in rodents, low androgens result in female-ty- Consistent with previous literature, Swift-Gallant et al. (2017) found pical sexual behaviors and preferences, whereas with high androgen that gay men delineated by anal sex role preferences differed in recalled signaling, male-typical sexual behaviors persist while androphilia is childhood gender nonconformity, such that gay men with a bottom anal increased. It is noteworthy that the possible influence of prenatal and/ sex role preference reported more gender nonconformity compared to or pubertal hormones on anal sex role could be the result of a direct gay men with a top anal sex role preference. We also found that gay influence of these hormones on sex role preferences, or the relationship men with a bottom anal sex role, but not gay men with a top anal sex could be indirect in that hormone action on masculine gender traits role, have an increased left-hand preference when compared to het- and/or body type could influence anal sex role identity. For example, erosexual men. Interestingly, we also found that handedness was a with the development of more masculine body type and gender traits, significant mediator of the relationship between anal sex role and re- there may be an increased social expectation to take on sex that called childhood gender nonconformity. Together, these findings sug- are perceived as more masculine (Moskowitz and Hart, 2011). gest that the developmental processes underlying handedness and In sum, this research on bio-developmental subgroups among gay gender nonconformity are relevant to a subgroup of gay men with a men is young and requires further investigation and replication; how- bottom anal sex role. ever, to date the evidence supports the notion that there are multiple In Wampold (2017) and Swift-Gallant et al. (2018a) it was predicted distinct biological pathways (e.g., low androgen, high androgen, im- that the FBOE would be found among gay men with a bottom anal sex munological) underlying same-sex sexual orientation in men. role, given the separate lines of literature linking both FBOE and re- ceptive anal sex with increased gender nonconformity (e.g., reviewed in 4. Conclusions Wampold, 2013). Indeed, both Wampold (2017) and Swift-Gallant et al. (2018a) found that gay men with a bottom anal sex role have a higher The field of behavioral neuroendocrinology has established princi- preponderance of older brothers, whereas other gay men did not show ples of sexual differentiation dating back to the 1950's that continue to evidence of the FBOE. These results suggest that the processes under- guide research: androgens act during developmental critical periods to lying the FBOE contribute to the development of sexual orientation in a shape the brain and behavior. However, as the field progresses, we are subgroup of gay men who also develop a bottom anal sex role pre- unveiling complexities in the processes by which the brain undergoes ference in adulthood. sexual differentiation to lead to a myriad, rather than just the two long Physical development measures also differ by anal sex role. Swift- recognized sexual phenotypes. There is still a lot to learn about how Gallant et al. (2018b) report that gay men who typically take on a top sexual differentiation of the brain and behavior is affected by the in- anal sex role were more male-typical on measures such as height and teraction of androgenic and estrogenic pathways, site of hormone ac- body , whereas gay men with a bottom anal sex role were shorter tion, dose effects and environmental factors, among others. Fortunately, and reported less body hair compared to both heterosexual men and the field is sufficiently well established that we can begin to test whe- other gay men. Furthermore, gay men with a top anal sex role reported ther the principles emerging from the rodent literature generalize to the earliest pubertal onset compared to all other groups, while gay men other species, including humans. There are sure to be differences in the with a bottom anal sex role were intermediate between heterosexual details of how hormones affect the brain and behavior between species and top gay men. These results suggest that the biological processes but given the many similarities in our , it is likely that the underlying the development of height, body hair and pubertal onset overarching themes will translate. Indeed, recent findings reviewed may contribute to the development of sexual orientation of gay men here support the hypothesis that there are multiple biological pathways with a top anal sex role. Given these measures were more male-typical that lead to same-sex sexual orientation, including both low and high in tops compared to all other groups, these results suggest that this androgen signaling in humans, as it does in mice. Although this review subgroup of gay men may be exposed to higher androgen levels. largely focused on the effects of prenatal androgens on androphilia in Together, this line of research supports the hypothesis subgroups of males, these principles likely also hold true for female sexual orienta- gay men delineated by anal sex role differ in their bio-development. tion. For example, the literature has consistently linked increased an- Specifically, these results indicate that gay men who prefer a bottom drogens to same-sex sexual attraction in women, but multiple factors anal sex role are influenced by the biological processes underlying are likely at play, given that biomarkers of androgen exposure and handedness and the FBOE, whereas the biological processes underlying circulating testosterone are higher among more gender nonconforming physical development and pubertal onset contribute to the sexual or- (e.g., self-identified “butch”) compared to gender conforming ientation of a subgroup of gay men with a top anal sex role. It will be of (e.g., self-identified “femmes”; reviewed in Breedlove, 2017; Singh interest to replicate these findings in future research, as well as et al., 1999). Thus, it is likely that for both men and women, there exist

28 A. Swift-Gallant Hormones and Behavior 111 (2019) 23–30 more than a single bio-developmental pathway to same-sex sexual or- frameshift mutation destabilizes androgen receptor messenger RNA in the Tfm ientation. Further exploring such questions into the sexual differentia- mouse. Mol. Endocrinol. 5 (4), 573–581. Chen, C.V., Brummet, J.L., Jordan, C.L., Breedlove, S.M., 2015. Down, but not out: partial tion of rodents and humans will contribute to our understanding of how elimination of androgen receptors in the male mouse brain does not affect androgenic individual differences develop in the brain and behavior. regulation of anxiety or HPA activity. Endocrinology 157 (2), 764–773. Clemens, L.G., Gladue, B.A., 1978. Feminine sexual behavior in rats enhanced by prenatal inhibition of androgen aromatization. Horm. Behav. 11 (2), 190–201. Acknowledgements Clemens, L.G., Gladue, B.A., Coniglio, L.P., 1978. Prenatal endogenous androgenic in- fl uences on masculine sexual behavior and genital morphology in male and female rats. Horm. Behav. 10 (1), 40–53. I thank Cynthia L. Jordan and S. Marc Breedlove for their thoughtful Cooke, B., Hegstrom, C.D., Villeneuve, L.S., Breedlove, S.M., 1998. Sexual differentiation comments on an earlier draft of this manuscript. This work was sup- of the vertebrate brain: principles and mechanisms. Front. Neuroendocrinol. 19 (4), – ported by a Natural Sciences and Engineering Research Council of 323 362. Cruz, C.D., Pereira, O.C., 2012. Prenatal testosterone supplementation alters puberty Canada Postdoctoral Fellowship to ASG. onset, aggressive behavior, and partner preference in adult male rats. J. Physiol. Sci. 62 (2), 123–131. Csathó, Á., Osváth, A., Bicsák, É., Karádi, K., Manning, J., Kállai, J., 2003. Sex role References identity related to the ratio of second to fourth digit length in women. Biol. Psychol. 62 (2), 147–156. Arning, L., Ocklenburg, S., Schulz, S., Ness, V., Gerding, W.M., Hengstler, J.G., ... Beste, Danilovic, D.L.S., Correa, P.H.S., Costa, E.M.F., Melo, K.F.S., Mendonca, B.B., Arnhold, C., 2015. Handedness and the X chromosome: the role of androgen receptor CAG- I.J.P., 2007. Height and bonemineral density in androgen insensitivity syndrome with – repeat length. Sci. Rep. 5, 8325. mutations in the androgen receptor gene. Osteoporos. Int. 18, 369 374. Auyeung, B., Baron-Cohen, S., Ashwin, E., Knickmeyer, R., Taylor, K., Hackett, G., et al., Davidson, J.M., 1966. Activation of the male rat's sexual behavior by intracerebral im- – 2009. Fetal testosterone predicts sexually differentiated childhood behavior in girls plantation of androgen. Endocrinology 79 (4), 783 794. ff and in boys. Psychol. Sci. 20 (2), 144–148 (Feb 1). de Vries, G.J., Forger, N.G., 2015. Sex di erences in the brain: a whole body perspective. ff Bakker, J., Van Ophemert, J., Slob, A.K., 1996. Sexual differentiation of odor and partner Biol. Sex Di er. 6 (1), 15. preference in the rat. Physiol. Behav. 60 (2), 489–494. Diamond, M., Llacuna, A., Wong, C.L., 1973. Sex behavior after neonatal progesterone, – – Bakker, J., Honda, S., Harada, N., Balthazart, J., 2002. Sexual partner preference requires testosterone, estrogen, or antiandrogens. Horm. Behav. 4 (1 2), 73 88. a functional aromatase (cyp19) gene in male mice. Horm. Behav. 42 (2), 158–171. Diotel, N., Charlier, T.D., Lefebvre d'Hellencourt, C., Couret, D., Trudeau, V.L., Nicolau, Bakker, J., De Mees, C., Douhard, Q., Balthazart, J., Gabant, P., Szpirer, J., Szpirer, C., J.C., ... Pellegrini, E., 2018. Steroid transport, local synthesis, and signaling within 2006. Alpha-fetoprotein protects the developing female mouse brain from masculi- the brain: roles in neurogenesis, neuroprotection, and sexual behaviors. Front. nization and defeminization by estrogens. Nat. Neurosci. 9 (2), 220. Neurosci. 12, 84. ı ff Bakker, J., De Mees, C., Szpirer, J., Szpirer, C., Balthazart, J., 2007. Exposure to oestrogen Dom nguez-Salazar, E., Portillo, W., Baum, M.J., Bakker, J., Paredes, R.G., 2002. E ect of prenatally does not interfere with the normal female-typical development of odour prenatal androgen receptor antagonist or aromatase inhibitor on sexual behavior, preferences. J. Neuroendocrinol. 19 (5), 329–334. partner preference and neuronal Fos responses to estrous female odors in the rat – Barber, N., 1995. The evolutionary psychology of physical attractiveness: sexual selection accessory olfactory system. Physiol. Behav. 75 (3), 337 346. and human morphology. Ethol. Sociobiol. 16, 395–424. Edwards, A.L., Hammond, H.A., Jin, L., Caskey, C.T., Chakraborty, R., 1992. Genetic fi Berenbaum, S.A., Bryk, K.K., Nowak, N., Quigley, C.A., Moffat, S., 2009. as a variation at ve trimeric and tetrameric tandem repeat loci in four human population – marker of prenatal androgen exposure. Endocrinology 150 (11), 5119–5124. groups. Genomics 12 (2), 241 253. Blanchard, R., 2008. Review and theory of handedness, birth order, and in Ellis, L., Ames, M.A., 1987. Neurohormonal functioning and sexual orientation: a theory – men. Laterality 13 (1), 51–70. of homosexuality-. Psychol. Bull. 101, 233 258. https://doi.org/10. Blanchard, R., 2018. Fraternal birth order, family size, and male homosexuality: meta- 1037/0033-2909.101.2.233. ff analysis of studies spanning 25 years. Arch. Sex. Behav. 47 (1), 1–15. Forger, N.G., 2016. Epigenetic mechanisms in sexual di erentiation of the brain and Blanchard, R., Lippa, R.A., 2007. Birth order, sibling , handedness, and sexual behaviour. Philos. Trans. R. Soc. B 371 (1688), 20150114. orientation of male and female participants in a BBC internet research project. Arch. Forger, N.G., Strahan, J.A., Castillo-Ruiz, A., 2016. Cellular and molecular mechanisms of ff Sex. Behav. 36 (2), 163–176. sexual di erentiation in the mammalian nervous system. Front. Neuroendocrinol. 40, – Bodo, C., 2008. A role for the androgen receptor in the sexual differentiation of the ol- 67 86. ff factory system in mice. Brain Res. Rev. 57 (2), 321–331. Frye, C.A., 2001. The role of neurosteroids and non-genomic e ects of progestins and – Bodo, C., Rissman, E.F., 2007. Androgen receptor is essential for sexual differentiation of androgens in mediating sexual receptivity of rodents. Brain Res. Rev. 37 (1 3), – responses to olfactory cues in mice. Eur. J. Neurosci. 25 (7), 2182–2190. 201 222. Bogaert, A.F., 2006. Biological versus nonbiological older brothers and men's sexual or- Galis, F., Ten Broek, C.M., Van Dongen, S., Wijnaendts, L.C., 2010. in – ientation. Proc. Natl. Acad. Sci. 103 (28), 10771–10774. the prenatal digit ratio (2D:4D). Arch. Sex. Behav. 39 (1), 57 62. ff Bogaert, A.F., 2010. Physical development and sexual orientation in men and women: an Gerall, A.A., Hendricks, S.E., Johnson, L.L., 1967. E ects of early castration in male rats analysis of NATSAL-2000. Arch. Sex. Behav. 39 (1), 110–116. on adult sexual behavior. J. Comp. Physiol. Psychol. 64 (2), 206. Bogaert, A.F., Blanchard, R., 1996a. Handedness in homosexual and heterosexual men in Giovannucci, E., Stampfer, M.J., Krithivas, K., Brown, M., Brufsky, A., Talcott, J., ... ff the Kinsey interview data. Arch. Sex. Behav. 25 (4), 373. Kanto , P.W., 1997. The CAG repeat within the androgen receptor gene and its re- – Bogaert, A.F., Blanchard, R., 1996b. Physical development and sexual orientation in men: lationship to prostate cancer. Proc. Natl. Acad. Sci. 94 (7), 3320 3323. fl height, weight and age of puberty differences. Personal. Individ. Differ. 21 (1), Gladue, B.A., Clemens, L.G., 1978. Androgenic in uences on feminine sexual behavior in 77–84. male and female rats: defeminization blocked by prenatal antiandrogen treatment. – Bogaert, A.F., Friesen, C., 2002. Sexual orientation and height, weight, and age of pub- Endocrinology 103 (5), 1702 1709. fl erty: new tests from a British national probability sample. Biol. Psychol. 59 (2), Gobrogge, K.L., Breedlove, S.M., Klump, K.L., 2008. Genetic and environmental in u- fi 135–145. ences on 2D:4D nger length ratios: a study of monozygotic and dizygotic male and – Bogaert, A.F., Friesen, C., Klentrou, P., 2002. Age of puberty and sexual orientation in a female twins. Arch. Sex. Behav. 37 (1), 112 118. fi national probability sample. Arch. Sex. Behav. 31 (1), 73–81. Gottlieb, B., Tri ro, M.A., 1999. Androgen insensitivity syndrome. In: Adam, M.P., ® Bogaert, A.F., Skorska, M.N., Wang, C., Gabrie, J., MacNeil, A.J., Hoffarth, M.R., ... Ardinger, H.H., Pagon, R.A. (Eds.), GeneReviews . University of Washington, Seattle Blanchard, R., 2018. Male homosexuality and maternal immune responsivity to the [updated May 2017]. Y-linked protein NLGN4Y. Proc. Natl. Acad. Sci. 115 (2), 302–306. Grimbos, T., Dawood, K., Burriss, R.P., Zucker, K.J., Puts, D.A., 2010. Sexual orientation fi Bonthuis, P.J., Cox, K.H., Searcy, B.T., Kumar, P., Tobet, S., Rissman, E.F., 2010. Of mice and the second to fourth nger length ratio: a meta-analysis in men and women. and rats: key species variations in the sexual differentiation of brain and behavior. Behav. Neurosci. 124 (2), 278. ff fl Front. Neuroendocrinol. 31 (3), 341–358. Hall, P.A., Schae , C.M., 2008. Sexual orientation and uctuating asymmetry in men and – Brand, T., Kroonen, J., Mos, J., Slob, A.K., 1991. Adult partner preference and sexual women. Arch. Sex. Behav. 37 (1), 158 165. behavior of male rats affected by perinatal endocrine manipulations. Horm. Behav. Hampson, E., Sankar, J.S., 2012. Hand preference in humans is associated with testos- 25 (3), 323–341. terone levels and androgen receptor gene polymorphism. Neuropsychologia 50 (8), – Breedlove, S.M., 2017. Prenatal influences on human sexual orientation: expectations 2018 2025. versus data. Arch. Sex. Behav. 46 (6), 1583–1592. Helleday, J., Siwers, B., Ritzén, E.M., Hugdahl, K., 1994. Normal lateralization for Brock, O., Bakker, J., 2011. Potential contribution of prenatal estrogens to the sexual handedness and advantage in a verbal dichotic listening task in women with – differentiation of mate preferences in mice. Horm. Behav. 59 (1), 83–89. congenital adrenal hyperplasia (CAH). Neuropsychologia 32 (7), 875 880. Brown, W.M., Hines, M., Fane, B.A., Breedlove, S.M., 2002. Masculinized finger length Henley, C.L., Nunez, A.A., Clemens, L.G., 2010. Exogenous androgen during development patterns in human males and females with congenital adrenal hyperplasia. Horm. alters adult partner preference and mating behavior in gonadally intact male rats. – – Behav. 42 (4), 380–386. Horm. Behav. 57 (4 5), 488 495. Buck, J.J., Williams, R.M., Hughes, I.A., Acerini, C.L., 2003. In-utero androgen exposure Hepper, P.G., 2013. The developmental origins of laterality: fetal handedness. Dev. – and 2nd to 4th digit length ratio—comparisons between healthy controls and females Psychobiol. 55 (6), 588 595. with classical congenital adrenal hyperplasia. Hum. Reprod. 18 (5), 976–979. Hepper, P.G., Shahidullah, S., White, R., 1991. Handedness in the human fetus. – Casey, M.B., Nuttall, R.L., 1990. Differences in feminine and masculine characteristics in Neuropsychologia 29 (11), 1107 1111. women as a function of handedness: support for the Geschwind/Galaburda theory of Hisasue, S.I., Sasaki, S., Tsukamoto, T., Horie, S., 2012. The relationship between second- – brain organization. Neuropsychologia 28 (7), 749–754. to-fourth digit ratio and female . J. Sex. Med. 9 (11), 2903 2910. Casto, J.M., Ward, O.B., Bartke, A., 2003. Play, copulation, , and testosterone in Honda, S.I., Harada, N., Ito, S., Takagi, Y., Maeda, S., 1998. Disruption of sexual behavior fi gonadally intact male rats prenatally exposed to flutamide. Physiol. Behav. 79 (4–5), in male aromatase-de cient mice lacking exons 1 and 2 of the cyp19 gene. Biochem. – 633–641. Biophys. Res. Commun. 252 (2), 445 449. Charest, N.J., Zhou, Z.X., Lubahn, D.B., Olsen, K.L., Wilson, E.M., French, F.S., 1991. A Hughes, I.A., Davies, J.D., Bunch, T.I., Pasterski, V., Mastroyannopoulou, K., MacDougall,

29 A. Swift-Gallant Hormones and Behavior 111 (2019) 23–30

J., 2012. Androgen insensitivity syndrome. Lancet 380 (9851), 1419–1428. (1), 46. James, W.H., Grech, V., 2017. Homosexual sons and the fraternal birth order effect: the Robel, P., Baulieu, E.E., 1995. Neurosteroids: biosynthesis and function. Crit. Rev. role of elevated maternal intrauterine testosterone. Arch. Sex. Behav. 1–4. Neurobiol. 9 (4), 383–394. Juntti, S.A., Tollkuhn, J., Wu, M.V., Fraser, E.J., Soderborg, T., Tan, S., ... Shah, N.M., Robinson, S.J., Manning, J.T., 2000. The ratio of 2nd to 4th digit length and male 2010. The androgen receptor governs the execution, but not programming, of male homosexuality. Evol. Hum. Behav. 21, 333–345. sexual and territorial behaviors. Neuron 66 (2), 260–272. Roselli, C.E., Liu, M., Hurn, P.D., 2009. Brain aromatization: classical roles and new Kelso, W.M., Nicholls, M.E., Warne, G.L., 1999. Effects of prenatal androgen exposure on perspectives. In: Seminars in Reproductive Medicine. vol. 27. NIH Public Access, pp. cerebral lateralization in patients with congenital adrenal hyperplasia (CAH). Brain 207 (No. 3). Cogn. 40 (1), 153–156. Sato, T., Matsumoto, T., Kawano, H., Watanabe, T., Uematsu, Y., Sekine, K., ... Konkle, A.T., McCarthy, M.M., 2011. Developmental time course of estradiol, testos- Nakamichi, Y., 2004. Brain masculinization requires androgen receptor function. terone, and dihydrotestosterone levels in discrete regions of male and female rat Proc. Natl. Acad. Sci. 101 (6), 1673–1678. brain. Endocrinology 152 (1), 223–235. Savin-Williams, R.C., Ream, G.L., 2006. Pubertal onset and sexual orientation in an Kraemer, B., Noll, T., Delsignore, A., Milos, G., Schnyder, U., Hepp, U., 2006. adolescent national probability sample. Arch. Sex. Behav. 35 (3), 279. length ratio (2D:4D) and dimensions of sexual orientation. Neuropsychobiology 53 Schmitz, J., Kumsta, R., Moser, D., Güntürkün, O., Ocklenburg, S., 2017. DNA methyla- (4), 210–214. tion in candidate for handedness predicts handedness direction. Laterality Kudwa, A.E., Bodo, C., Gustafsson, J.Å., Rissman, E.F., 2005. A previously un- 1–21. characterized role for estrogen receptor β: defeminization of male brain and beha- Schulz, K.M., Sisk, C.L., 2016. The organizing actions of adolescent gonadal steroid vior. Proc. Natl. Acad. Sci. 102 (12), 4608–4612. hormones on brain and behavioral development. Neurosci. Biobehav. Rev. 70, Kudwa, A.E., Michopoulos, V., Gatewood, J.D., Rissman, E.F., 2006. Roles of estrogen 148–158. receptors α and β in differentiation of mouse sexual behavior. 138 (3), Singh, D., Vidaurri, M., Zambarano, R.J., Dabbs Jr., J.M., 1999. erotic role 921–928. identification: behavioral, morphological, and hormonal correlates. J. Pers. Soc. Lalumiere, M.L., Blanchard, R., Zucker, K.J., 2000. Sexual orientation and handedness in Psychol. 76 (6), 1035. men and women: a meta-analysis. Psychol. Bull. 126 (4), 575. Skorska, M.N., Bogaert, A.F., 2017. Sexual orientation, objective height, and self-reported Lippa, R.A., 2003. Are 2D:4D finger-length ratios related to sexual orientation? Yes for height. J. Sex Res. 54 (1), 19–32. men, no for women. J. Pers. Soc. Psychol. 85, 179–188. Stern, J.J., 1970. Responses of male rats to sex odors. Physiol. Behav. 5 (4), 519–524. Lyon, M.F., Hawkes, S.G., 1970. X-linked gene for testicular feminization in the mouse. Swift-Gallant, A., Monks, D.A., 2017. Androgenic mechanisms of sexual differentiation of Nature 227 (5264), 1217. the nervous system and behavior. Front. Neuroendocrinol. 46, 32–45. Malas, M.A., Dogan, S., Evcil, E.H., Desdicioglu, K., 2006. Fetal development of the hand, Swift-Gallant, A., Coome, L.A., Ramzan, F., Monks, D.A., 2016a. Nonneural androgen digits and digit ratio (2D:4D). Early Hum. Dev. 82 (7), 469–475. receptors affect sexual differentiation of brain and behavior. Endocrinology 157 (2), McCarthy, M.M., Arnold, A.P., 2011. Reframing sexual differentiation of the brain. Nat. 788–798. Neurosci. 14 (6), 677. Swift-Gallant, A., Coome, L., Srinivasan, S., Monks, D.A., 2016b. Non-neural androgen McEwen, B.S., Plapinger, L., Chaptal, C., Gerlach, J., Wallach, G., 1975. Role of feto- receptor promotes androphilic odor preference in mice. Horm. Behav. 83, 14–22. neonatal estrogen binding proteins in the associations of estrogen with neonatal brain Swift-Gallant, A., Coome, L.A., Monks, D.A., VanderLaan, D.P., 2017. Handedness is a cell nuclear receptors. Brain Res. 96 (2), 400–406. biomarker of variation in anal sex role behavior and recalled childhood gender McEwen, B.S., Davis, P.G., Parsons, B., Pfaff, D.W., 1979. The brain as a target for steroid nonconformity among gay men. PLoS One 12 (2), e0170241. hormone action. Annu. Rev. Neurosci. 2 (1), 65–112. Swift-Gallant, A., Coome, L.A., Monks, D.A., VanderLaan, D.P., 2018a. Gender non- McFadden, D., Shubel, E., 2002. Relative lengths of fingers and in human males and conformity and birth order in relation to anal sex role among gay men. Arch. Sex. females. Horm. Behav. 42, 492–500. Behav. 1–12. McFadden, D., Loehlin, J.C., Breedlove, S.M., Lippa, R.A., Manning, J.T., Rahman, Q., Swift-Gallant, A., Coome, L.A., Monks, D.A., VanderLaan, D.P., 2018b. Pubertal onset and 2005. A reanalysis of five studies on sexual orientation and the relative length of the anal sex role among gay men. In: Psychology of Sexual Orientation and Gender 2nd and 4th fingers (the 2D:4D ratio). Arch. Sex. Behav. 34 (3), 341–356. , (E Pub Ahead of Print). McIntyre, M.H., 2003. Letter to the editor: digit ratios, childhood behavior, Tejada, L.D., Rissman, E.F., 2012. Sex differences in social investigation: effects of an- and erotic role preferences of gay men. Arch. Sex. Behav. 32 (6), 495–497. drogen receptors, hormones and test partner. J. Neuroendocrinol. 24 (8), 1144–1153. Medland, S.E., Duffy, D.L., Spurdle, A.B., Wright, M.J., Geffen, G.M., Montgomery, G.W., Tirabassi, G., Corona, G., Biagioli, A., Buldreghini, E., delli Muti, N., Maggi, M., Balercia, Martin, N.G., 2005. Opposite effects of androgen receptor CAG repeat length on in- G., 2015. Influence of androgen receptor cag polymorphism on sexual function re- creased risk of left-handedness in males and females. Behav. Genet. 35 (6), 735–744. covery after testosterone therapy in late-onset hypogonadism. J. Sex. Med. 12 (2), Morris, J.A., Jordan, C.L., Breedlove, S.M., 2004. Sexual differentiation of the vertebrate 381–388. nervous system. Nat. Neurosci. 7 (10), 1034. Underwood, L.E., Van Wyk, J.J., 1992. Normal and aberrant growth. In: Wilson, J.D., Moskowitz, D.A., Hart, T.A., 2011. The influence of physical body traits and masculinity Foster, D.W. (Eds.), Williams Textbook of Endocrinology, 8th ed. W. B. Saunders, on anal sex roles in gay and bisexual men. Arch. Sex. Behav. 40 (4), 835–841. Philadelphia, pp. 1079–1138. Naftolin, F., 1994. Brain aromatization of androgens. J. Reprod. Med. 39 (4), 257–261. Voracek, M., Manning, J.T., Ponocny, I., 2005. Digit ratio (2D:4D) in homosexual and Nass, R., Baker, S., Speiser, P., Virdis, R., Balsamo, A., Cacciari, E., ... New, M., 1987. heterosexual men from Austria. Arch. Sex. Behav. 34 (3), 335–340. Hormones and handedness: Left-hand in female congenital adrenal hyperplasia Wallien, M.S., Zucker, K.J., Steensma, T.D., Cohen-Kettenis, P.T., 2008. 2D:4D finger- patients. Neurology 37 (4), 711–715. length ratios in children and adults with gender identity disorder. Horm. Behav. 54 Nicholls, M.E., Forbes, S., 1996. Handedness and its association with gender-related (3), 450–454. psychological and physiological characteristics. J. Clin. Exp. Neuropsychol. 18 (6), Wampold, C.H., 2013. Redesigning fraternal birth order studies from top to bottom. Arch. 905–910. Sex. Behav. 42 (8), 1387–1389. Ogawa, S., Washburn, T.F., Taylor, J., Lubahn, D.B., Korach, K.S., Pfaff, D.W., 1998. Wampold, C., 2017. Fraternal birth order and anal-erotic roles of men who have sex with Modifications of testosterone-dependent behaviors by estrogen receptor-α gene dis- men. J. Sex. Med. 14 (5), e295–e296. ruption in male mice. Endocrinology 139 (12), 5058–5069. Ward, I.L., Renz, F.J., 1972. Consequences of perinatal hormone manipulation on the Ökten, A., Kalyoncu, M., Yariş, N., 2002. The ratio of second-and fourth-digit lengths and adult sexual behavior of female rats. J. Comp. Physiol. Psychol. 78 (3), 349–355. congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Early Hum. Dev. 70 Wersinger, S.R., Rissman, E.F., 2000. Oestrogen receptor α is essential for female-directed (1–2), 47–54. chemo-investigatory behaviour but is not required for the pheromone-induced lu- Papadatou-Pastou, M., Martin, M., Munafo, M.R., Jones, G.V., 2008. Sex differences in teinizing hormone surge in male mice. J. Neuroendocrinol. 12 (2), 103–110. left-handedness: a meta-analysis of 144 studies. Psychol. Bull. 134 (5), 677. Williams, T.J., Pepitone, M.E., Christensen, S.E., Cooke, B.M., Huberman, A.D., Phoenix, C.H., Goy, R.W., Gerall, A.A., Young, W.C., 1959. Organizing action of pre- Breedlove, N.J., et al., 2000. Finger-length ratios and sexual orientation. Nature 404, natally administered testosterone propionate on the tissues mediating mating beha- 455–456. vior in the female guinea pig. Endocrinology 65 (3), 369–382. Yau, M., Nimkarn, S., New, M.I., 2015. Congenital adrenal hyperplasia. In: Genetic Piacsek, B.E., Hostetter, M.W., 1984. Neonatal androgenization in the male rat: evidence Diagnosis of Endocrine Disorders, Second edition. pp. 203–210. for central and peripheral defects. Biol. Reprod. 30 (2), 344–351. Yousefi, M., Karmaus, W., Zhang, H., Roberts, G., Matthews, S., Clayton, B., Arshad, S.H., Pierman, S., Douhard, Q., Bakker, J., 2008. Evidence for a role of early oestrogens in the 2013. Relationships between age of puberty onset and height at age 18 years in girls central processing of sexually relevant olfactory cues in female mice. Eur. J. Neurosci. and boys. World J. Pediatr. 9 (3), 230–238. 27 (2), 423–431. Zadina, J.E., Dunlap, J.L., Gerall, A.A., 1979. Modifications induced by neonatal steroids Rahman, Q., 2005. Fluctuating asymmetry, second to fourth finger length ratios and in reproductive organs and behavior of male rats. J. Comp. Physiol. Psychol. 93 (2), human sexual orientation. Psychoneuroendocrinology 30 (4), 382–391. 314. Rahman, Q., Wilson, G.D., 2003. Sexual orientation and the 2nd to 4th finger length ratio: Zheng, Z., Cohn, M.J., 2011. Developmental basis of sexually dimorphic digit ratios. Proc. evidence for organizing effects of sex hormones or developmental instability? Natl. Acad. Sci. 108 (39), 16289–16294. Psychoneuroendocrinology 28, 288–303. Zheng, L., Hart, T.A., Zheng, Y., 2012. The relationship between intercourse preference Raskin, K., De Gendt, K., Duittoz, A., Liere, P., Verhoeven, G., Tronche, F., Mhaouty- positions and personality traits among gay men in China. Arch. Sex. Behav. 41 (3), Kodja, S., 2009. Conditional inactivation of androgen receptor gene in the nervous 683–689. system: effects on male behavioral and neuroendocrine responses. J. Neurosci. 29 Zheng, L., Hart, T.A., Zheng, Y., 2015. Top/bottom sexual self-labels and empathi- (14), 4461–4470. zing–systemizing cognitive styles among gay men in China. Arch. Sex. Behav. 44 (5), Rey, R.A., Campo, S.M., Ropelato, M.G., Bergadá, I., 2016. Hormonal changes in child- 1431–1438. hood and puberty. In: Puberty. Springer, Cham, pp. 23–37. Zuloaga, D.G., Puts, D.A., Jordan, C.L., Breedlove, S.M., 2008. The role of androgen re- Rieger, G., Linsenmeier, J.A., Gygax, L., Bailey, J.M., 2008. Sexual orientation and ceptors in the masculinization of brain and behavior: what we've learned from the childhood gender nonconformity: evidence from home videos. Dev. Psychol. 44 testicular feminization mutation. Horm. Behav. 53 (5), 613–626.

30