Extrahypothalamic ER Alpha Are Required for Testosterone Effects on Physical Activity and Fat Mass in Mice

Testosterone replacement therapy increases lean mass and decreases fat mass in hypogonadal men. These changes in the body composition are dependent on the dose of testosterone in short-term studies of GnRH analogue induced male hypogonadism in younger and older men. Both testosterone and the nonaromatizable androgen DHT increase muscle mass. Aromatizable and nonaromatizable androgens promote myogenesis and inhibit adipose tissue differentiation and adipogenesis (1, 2). Finkelstein et al demonstrated that when testosterone gel was coadministered with an aromatase inhibitor, the effect of testosterone on lean mass was maintained but decreases in both subcutaneous and intra-abdominal fat mass were abrogated (3). The results suggest that testosterone decreases fat mass through aromatization to estradiol.

Male and female mice with aromatase deficiency are obese with elevated insulin levels. Men with aromatase deficiency have truncal obesity and insulin resistance. Both male and female mice with estrogen receptor α (ERα) deletions develop progressive increases in adipose tissues with aging. In a man with nonfunctional ERα, obesity and insulin resistance were the presenting clinical features. ERα decreases proliferation and size of adipocytes, whereas ERβ increases adipocyte size, suggesting that ERα and ERβ may have opposing effect on adipocyte function. ERβ knock-out mice do not have the obesity phenotype of ERɑ knock-out mice, indicating that the effects of ERɑ signaling predominate in metabolic networks. These studies in mice and men provide sufficient evidence to support that estrogens acting on ERα drive adipose tissue differentiation, function, and accumulation (4, 5). In humans, the gene Cyp19a1 encoding aromatase, and its tissue-specific promoters, are expressed widely in placenta, adipose tissue, skin, brain, bone, endothelial cells, gonads, and fetal liver (6). In contrast, in male mice serum estradiol is not measurable in the blood and Cyp 19a1 expression is limited to the hypothalamus and testis with little expression in adipose tissue (7). The newly published investigations of Kim et al (8) in this issue of Endocrinology have addressed the question of the mechanisms of action of androgens in obese hypogonadal mice to decrease fat mass, given that estrogens are not detected in the circulation and aromatase is not expressed in adipocytes in mice.

To address this issue, Kim et al (8) demonstrated in mice with GnRH induced hypogonadism fed a high-fat diet, both testosterone and DHT increased lean mass, seminal vesicle and levator ani muscle weight, and food intake. Both androgens increased oxygen consumption and heat production that were related to increase in lean mass. The authors further demonstrated that testosterone but not the nonaromatizable DHT decreased total, gonadal, subcutaneous, and perirenal fat that was associated with increased nocturnal ambulatory movement in mice. This indicates that the decrease in fat mass is mediated by estrogens, as was demonstrated in men (3). Because circulating estrogens in mice are not detectable even by sensitive methods, the action of testosterone must be mediated by conversion to estradiol by the aromatase enzyme acting locally at the target tissue. The authors then confirmed that Cyp19a1 transcripts were only expressed in gonads and brain mainly in the hypothalamus and the extrahypothalamic regions including the striatum, hippocampus, and amygdala. Cyp19a1 transcripts were not expressed in gonadal, subcutaneous, and perirenal fat. The authors then repeated the experiments in extrahypothalamic neuron-specific ERα knock-out mice treated with GnRH antagonist to induce hypogonadism and then replaced the animals with testosterone or DHT. Although both androgens increased lean mass, the decreases in total and tissue fat mass and nocturnal increase in physical activity were no longer observed in the neuron-specific ERα knock-out mice. The authors concluded that in the model of obese mice with induced hypogonadism, testosterone increases lean mass and indirectly energy expenditure acting through androgen receptor signaling. However, the decrease in fat depends on aromatization of testosterone into estrogens that is at least partially mediated by increased physical activity mediated by ERα in the extrahypothalamic regions of the brain. This is a new proposed action of testosterone though local aromatization to estradiol at the extrahypothalamic neurons to increase ambulatory activity to reduce body fat.

Although this new paper clearly proves these actions for estrogens in the murine model, it is important to note that the action of testosterone on adipose tissues in hypogonadal men with obesity may be different from the mouse model. This is because, in men, circulating estrogens are measurable and aromatase activity and ERs are present in adipose tissues, allowing estrogens to have a direct effect on adipocyte function. Whether testosterone has additional effects on brain regions to increase physical activity that required aromatization to estradiol acting on ERα has yet to be demonstrated. Interestingly, in a randomized, controlled placebo trial in mildly overweight older men with partial androgen deficiency, administration of the nonaromatizable androgen DHT for 3 months at doses mimicking that of testosterone in adult men was associated with suppression of serum testosterone resulting from negative feedback on the hypothalamus-pituitary axis. Lean mass was not increased but fat mass was significantly decreased in the men receiving DHT compared with placebo gel (9). However, despite randomization, the baseline characteristics of the 2 groups of men were different. The lower baseline body mass index and fat mass and higher serum testosterone concentrations in the DHT treated compared with the placebo group may be responsible for the observed decrease in fat mass consequent to DHT treatment. The implications of these clinical studies suggest that a well-designed and controlled study in obese hypogonadal men treated for 3 to 4 months with testosterone with or without concomitant aromatase inhibitor may potentially address the requirement of estrogens for testosterone action on adipose mass. In addition, careful monitoring of physical activity in these men with modern activity monitors may also address whether estrogens act by increasing physical activity in men, as suggested by Kim et al in mice (8).

Additional Information

Disclosures: The authors have no conflict of interest with the study reviewed in this commentary.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.