Testosterone (T) affects myriad central and peripheral physiological systems in both men and women. These include muscle growth, endocrine function, maintenance of external and internal gonadal tissues, maintenance of autonomic function, energy balance, bone, hair growth, augmentation of mood and the stimulation of brain systems for sexual arousal, desire and orgasm. ¹ The effects of T on the brain and behavior occur via binding to androgen receptors (ARs) as T or its 5α reduced metabolite, dihydrotestosterone (DHT), or by aromatization to estradiol (E2) and binding to estrogen receptors (ERs). Although aromatase exists in high quantity in granulosa cells of the ovaries (to convert T into E2), it exists in neurons within regions of the male brain associated with reproduction, such as the medial preoptic area (mPOA). In fact, the conversion of T to E2 is critical for sexual differentiation of the brain into a “male phenotype.” ² In adult females and males, ARs and ERs exist throughout the brain, but especially in limbic and hypothalamic regions that act as processing centers for olfactory and genital stimulation, arousal, mood, reward and bonding.

Figure 1. Genomic mechanism of T binding to nuclear receptors.

Average blood plasma levels of free T range between 300 and 1,100 ng/dl for men. However, these levels fluctuate daily, being generally higher in the morning than the evening. Although overall, women have one-tenth to one-twentieth the plasma levels of T as men, these levels fluctuate monthly for women, with a steep phasic rise and fall on the day of ovulation, coming on the heels of the earlier rise and fall in E2. ³ This can raise plasma levels of free T to those observed in the medium range of men (ie 500
ng/dl). Although hormonal replacement therapy for postmenopausal women with E2 restores and maintains gonadal tissues, bone, muscle, neuronal integrity and cardiovascular flow (all components necessary for genital sexual arousal), the addition of T over E2 augments sexual desire. Indeed, peak sexual desire occurs naturally in premenopausal women during ovulation. Periovulatory increases in sexual solicitations indicative of desire are observed in all mammalian females and appear to be driven by similar hormonal effects. In males, T is also an important regulator of sexual arousal, desire and copulatory behavior. T replacement in hypogonadal men increases the proportion who acquire and sustain erection, and restores sexual desire and orgasm.

Figure 2. Nongenomic mechanism of T and other androgens binding to cell surface receptors and associated second messenger effects. Adapted with permission from Lucas-Herald AK et al: Genomic and non-genomic effects of androgens in the cardiovascular system: clinical implications. Clin Sci (Lond) 2017; 131: 1405.

How does T accomplish this? There are 2 modes of steroid hormone binding to receptors, “genomic” and “nongenomic.” The genomic mechanism involves nuclear hormone receptors (fig. 1). The steroid binds to these receptors and alters their shape so that they make contact with hormone response elements on different genes that stimulate transcription of those genes into functional proteins. The proteins can act as receptors for different neurotransmitters, enzymes that synthesize or metabolize different neurotransmitters or that act as substrates for second messengers (eg protein kinases), or that perform other regulatory functions in cells. Some proteins activated by steroid hormone binding are themselves neurotransmitters, such as GnRH, oxytocin, vasopressin and ß-endorphin. In the case of T, an important and ubiquitous protein it facilitates is nitric oxide (NO) synthase. NO is critical for clitoral and penile erection and for neurotransmitter release throughout the brain. One hallmark of protein synthesis is that it takes time. Thus, there is a delay of hours between peak free T levels in plasma and the constellation of proteins it makes in the brain that ultimately alter behavior. In contrast, the nongenomic mechanism is faster, working on the order of seconds to minutes. Steroids can bind to receptors on the cell membrane that are linked to second messengers, such as phosphatidyl inositol, cyclic AMP and MAP kinases (fig 2). In the case of T, these linkages lead to increased calcium mobilization within neurons, thus depolarizing them (which makes them easer to “fire”) and facilitating neurotransmitter release through the action of calcium binding proteins.

Figure 3. Synergistic actions of ER and AR binding on brain dopamine systems and dopamine-driven response in mPOA to erotic cues. Adapted with permission from Arnow BA et al: Women with hypoactive sexual desire disorder compared to normal females: a functional magnetic resonance imaging study. Neuroscience 2009; 158: 484, and Kingsburg SA et al: The female sexual response: current models, neurobiological underpinnings and agents currently approved or under investigation for the treatment of hypoactive sexual desire disorder. CNS Drugs 2015; 29: 915.

Does T work alone or in concert with its aromatized metabolite, E2? This is still an area of debate in the neuroendocrine literature, especially regarding the role of T in women’s sexual desire. In some critical parts of the brain (eg mPOA), neurons contain nuclear ARs, aromatase and nuclear ERs. This makes it possible for T to induce specific proteins by binding to ARs, and then be aromatized into E2 to produce other proteins via ERs in those or neighboring neurons. Nongenomic effects of T could occur in virtually all limbic and hypothalamic brain regions associated with sexual behavior and reward. One synergic action related to sexual desire in both women and men involves the facilitation of dopamine synthesis by E2, and a subsequent action of T to facilitate calcium mobilization and synthesize NO. In sequence, these 2 mechanisms would lead to greater dopamine release in the presence of competent sexual cues and partner actions that stimulate sexual desire (fig. 3). Indeed, we have shown recently that E2 and T stimulate sexual solicitations in female rats better than either alone but can do so dramatically in the presence of the aromatase inhibitor fadrozole. ⁴ Similarly, administration of E2 and DHT dramatically increases the number of sexual solicitations and lordosis in female rats relative to either hormone alone. ⁵

A prescient idea was raised in the 1930s by Korenchevsky, a pioneer in gerontology. He showed that E2 and T restored gonadal tissue shape, vascularization and function following ovariectomy or orchiectomy in aged rats. The ability of both steroids to do this suggested to him that they were “bisexual” hormones, able to act on both female and male tissues. This idea was overshadowed in the late 1950s with the finding that exposure to T during perinatal and pubertal development organizes the brain, body and behavior into a male phenotype, whereas no T generates a female phenotype that is activated by E2 during puberty. Rather than T being “blue” and E2 “pink,” both work in the brains and bodies of males, females and trans individuals, on their own and synergistically. What T might be stimulating in the brain is fast becoming a new vista in neuroendocrinology.