Long-Lasting Consequences of Testosterone Exposure

Evidence of reciprocal regulation of androgen signaling and obesity has been well documented (1, 2); it is clear that late gestational and early postnatal periods are most vulnerable to perturbation of these signaling pathways (3). Exposure to high caloric nutrition early in life causes significant changes in reproductive physiology which cannot be reversed by resumption of the normal diet. A maternal high-fat diet produces prostatic hyperplasia in adult male and ovarian failure in adult female offspring (4, 5). Importantly, the embryo exposure to high caloric diet does not immediately activate androgen signaling and postnatal androgen surge in females does not immediately affect the weight; the effects of these stimuli are delayed until the organism is mature. Similarly, extensive experimental data indicate that the perturbations of hormonal balance during this developmental period have long-lasting effects not only on reproductive characteristics, including ovarian and neuroendocrine changes (6), but also changes in metabolic functions (7). The phenotype resembling human polycystic ovarian syndrome often associated with impaired insulin sensitivity and increased weight is observed in rodent females subjected to transient neonatal or prepubertal androgen treatment (8).

The study published in this issue of Endocrinology by Jang et al (9) narrows down the critical period of sensitivity to exogenous androgen treatment in mice (9). Three temporal windows were used: within 24 hours after birth and at 3 or 6 weeks of age. A single injection of testosterone enanthate (TE) was used at every time point. Reproductive abnormalities were detected in adult females treated with androgens 24 hours after birth. No abnormalities were detected in females treated with TE at 3 or 6 weeks, suggesting that the developmental program at that stage was fully determined. An increase in blood testosterone concentration in newborn female pups partially mimicked the surge of testosterone in late neonatal/early postnatal period in male rodents (10). In males, the increase of testosterone at the end of pregnancy is required for completion of the full masculinization program (11), including scrotum development, testicular descent, and the substitution of fetal with adult Leydig cells (12). In females, the oocyte nest breakdown occurs, and the primordial follicle pool is established the early postnatal period (13). In this report, neonatal TE treatment caused reproductive abnormalities in adult females such as irregular cycling, anovulatory phenotype and increased ovarian weight. Surprisingly, despite clearly abnormal cystic ovaries and the absence of corpora lutea in treated females, no changes in estradiol, LH, or FSH levels and expression of estrogen receptors α and β and androgen receptor genes were detected. This distinguishes the results of these experiments from previous reports on androgen treatment in the prenatal or neonatal rodent females (14). It is possible that the much lower dose and significantly more stable derivative of testosterone used in the Jang et al study was responsible for these differences and that the signaling pathways in the target cells downstream from steroid receptors were affected. Additionally, metabolic parameters were changed in TE-treated females when compared with the vehicle-treated controls such as a greater body weight, including lean body and fat mass. When insulin resistance was compared in 30-week-old females injected with oil or TE 24 hours after birth, TE-treated females displayed significantly lower blood glucose levels in response to insulin injections as compared with controls. Increases in lipoprotein lipase and resistin mRNA levels, the key genes in lipogenic and lipolytic pathways in adipose tissues, were detected in inguinal adipose tissue. Changes in morphology of brown adipose tissue (BAT) in 8-month-old females that received early TE administration were also reported. BAT tissue was increased in weight due in part to infiltration of white adipose tissue with the corresponding decrease in expression of BAT markers.

Although the data established a critical period of response to androgen treatment, the mechanism underlying developmental abnormalities in adulthood remains to be determined. It is possible that endocrine changes may affect the selection of many cell types within gonads and other organs, including differential survival of various stem or precursor cells. Alternatively, changes in the differentiation program of precursor cells may be affected. Importantly, it was shown that some of the changes caused by prenatal testosterone exposure may be ameliorated by postnatal administration of antiandrogens and insulin sensitizers (15). Taking into account the recent epidemic of obesity and widespread contamination of our environment with endocrine disruptors, it is important to determine similar periods of increased sensitivity during human development. Long-lasting consequences of such short-term or 1-time exposures, even after their cessation, make these studies relevant to human health. Findings in animal models are not easily extrapolated to humans, as hormonal control and timing of various developmental stages differ in rodents and humans. However, by discovering the developmental targets and understanding the mechanisms of early endocrine and metabolic changes, we might be able to design interventions to mitigate effects of abnormal nutrition and acute or chronic exposure to endocrine disruptors in early childhood.

Funding Statement

This work was supported by National Cancer Institute Grants U01CA177711 (to A.I.A.) and R15CA179287 (to I.U.A.).