Significance: Men are defined by androgens (testosterone), which drive fetal masculinization (male development) and puberty and maintain masculinity in adulthood, including sex drive, erectile function, and fertility. Moreover, Western cardiometabolic diseases are all associated with lowered testosterone levels in men. Therefore, influences on testosterone levels in adulthood have pervasive importance for masculinity and health. Our study shows, for the first time, to our knowledge, that testosterone levels during fetal masculinization can (re)program adult testosterone levels through effects on stem cells, which develop into adult Leydig cells (the source of testosterone) after puberty. These stem cells are present in fetal testes of humans and animals, and using the latter, we show how these cells are reprogrammed to affect adult testosterone levels.

Most common male reproductive disorders are linked to lower testosterone exposure in fetal life, although the factors responsible for suppressing fetal testosterone remain largely unknown. Protracted use of acetaminophen during pregnancy is associated with increased risk of cryptorchidism in sons, but effects on fetal testosterone production have not been demonstrated. We used a validated xenograft model to expose human fetal testes to clinically relevant doses and regimens of acetaminophen. Exposure to a therapeutic dose of acetaminophen for 7 days significantly reduced plasma testosterone (45% reduction; P = 0.025) and seminal vesicle weight (a biomarker of androgen exposure; 18% reduction; P = 0.005) in castrate host mice bearing human fetal testis xenografts, whereas acetaminophen exposure for just 1 day did not alter either parameter. Plasma acetaminophen concentrations (at 1 hour after the final dose) in exposed host mice were substantially below those reported in humans after a therapeutic oral dose. Subsequent in utero exposure studies in rats indicated that the acetaminophen-induced reduction in testosterone likely results from reduced expression of key steroidogenic enzymes (Cyp11a1, Cyp17a1). Our results suggest that protracted use of acetaminophen (1 week) may suppress fetal testosterone production, which could have adverse consequences. Further studies are required to establish the dose-response and treatment-duration relationships to delineate the maximum dose and treatment period without this adverse effect.

The testicular dysgenesis syndrome (TDS) hypothesis, which proposes that common reproductive disorders of newborn and adult human males may have a common fetal origin, is largely untested. We tested this hypothesis using a rat model involving gestational exposure to dibutyl phthalate (DBP), which suppresses testosterone production by the fetal testis. We evaluated if induction of TDS via testosterone suppression is restricted to the “masculinization programming window” (MPW), as indicated by reduction in anogenital distance (AGD). We show that DBP suppresses fetal testosterone equally during and after the MPW, but only DBP exposure in the MPW causes reduced AGD, focal testicular dysgenesis, and TDS disorders (cryptorchidism, hypospadias, reduced adult testis size, and compensated adult Leydig cell failure). Focal testicular dysgenesis, reduced size of adult male reproductive organs, and TDS disorders and their severity were all strongly associated with reduced AGD. We related our findings to human TDS cases by demonstrating similar focal dysgenetic changes in testes of men with preinvasive germ cell neoplasia (GCNIS) and in testes of DBP-MPW animals. If our results are translatable to humans, they suggest that identification of potential causes of human TDS disorders should focus on exposures during a human MPW equivalent, especially if negatively associated with offspring AGD.

Common male reproductive abnormalities including cryptorchidism, hypospadias, and low sperm counts may comprise a testicular dysgenesis syndrome (TDS), resulting from fetal testis dysfunction during a critical developmental period involving reduced androgen production/action. The recent increase in TDS prevalence suggests environmental/lifestyle factors may be etiologically important. The developing fetus is exposed to multimodal challenges, and we hypothesized that exposure to a combination of factors rather than single agents may be important in the pathogenesis of TDS. We experimentally induced fetal testis dysfunction in rats via treatment of pregnant females daily from embryonic day (e) 13.5 to e21.5 with vehicle, 100 or 500 mg/kg · d dibutyl phthalate (DBP), 0.1 mg/kg · d dexamethasone (Dex), or a combination of DBP + Dex. In adulthood, penile length/normality, testis weight/descent, prostate weight, and plasma testosterone levels were measured plus anogenital distance (AGD) as a measure of androgen action within the masculinization programming window. Intratesticular testosterone and steroidogenic enzyme gene expression were measured in fetal testes at e17.5. High-dose DBP reduced fetal intratesticular testosterone and steroidogenic gene expression; induced mild hypospadias (31%) and cryptorchidism (53%); and reduced penile length, AGD, and testis and prostate weight in adulthood. Dex alone had no effect except to reduce birth weight but amplified the adverse effects of 500 mg/kg · d DBP and exacerbated the effects of 100 mg/kg · d DBP. All adverse effects were highly correlated to AGD, emphasizing the etiological importance of the masculinization programming window. These findings suggest that exposure to common environmental chemicals in combination with, for example, maternal stress, may increase the risk of common male reproductive abnormalities, with implications for human populations.

Reproductive disorders that are common/increasing in prevalence in human males may arise because of deficient androgen production/action during a fetal ‘masculinization programming window’. We identify a potentially important role for Chicken Ovalbumin Upstream Promoter-Transcription Factor II (COUP-TFII) in Leydig cell (LC) steroidogenesis that may partly explain this. In rats, fetal LC size and intratesticular testosterone (ITT) increased ∼3-fold between e15.5-e21.5 which associated with a progressive decrease in the percentage of LC expressing COUP-TFII. Exposure of fetuses to dibutyl phthalate (DBP), which induces masculinization disorders, dose-dependently prevented the age-related decrease in LC COUP-TFII expression and the normal increases in LC size and ITT. We show that nuclear COUP-TFII expression in fetal rat LC relates inversely to LC expression of steroidogenic factor-1 (SF-1)-dependent genes (StAR, Cyp11a1, Cyp17a1) with overlapping binding sites for SF-1 and COUP-TFII in their promoter regions, but does not affect an SF-1 dependent LC gene (3β-HSD) without overlapping sites. We also show that once COUP-TFII expression in LC has switched off, it is re-induced by DBP exposure, coincident with suppression of ITT. Furthermore, other treatments that reduce fetal ITT in rats (dexamethasone, diethylstilbestrol (DES)) also maintain/induce LC nuclear expression of COUP-TFII. In contrast to rats, in mice DBP neither causes persistence of fetal LC COUP-TFII nor reduces ITT, whereas DES-exposure of mice maintains COUP-TFII expression in fetal LC and decreases ITT, as in rats. These findings suggest that lifting of repression by COUP-TFII may be an important mechanism that promotes increased testosterone production by fetal LC to drive masculinization. As we also show an age-related decline in expression of COUP-TFII in human fetal LC, this mechanism may also be functional in humans, and its susceptibility to disruption by environmental chemicals, stress and pregnancy hormones could explain the origin of some human male reproductive disorders.

Vascular endothelial growth factor (VEGF)-dependent angiogenesis is essential for normal luteal development. Although it is believed that hypoxia is the primary inducer of VEGF, in the corpus luteum it is up-regulated by human chorionic gonadotrophin (hCG). As hypoxia-inducible factor (HIF)1A has been shown to regulate VEGFA under ligand-stimulated conditions, we hypothesized that the effect of hCG on luteal VEGFA was mediated through HIF1A. We studied the effect of hCG on VEGFA and HIF1A expression in human luteinized granulosa cells in vitro and in human corpora lutea in vivo. HCG up-regulated VEGFA (P < 0.05) and HIF1A (P < 0.001) in vitro and VEGFA (P < 0.05) and HIF1A (P < 0.05) in vivo. There was a correlation between HIF1A and VEGFA in vivo (P < 0.005) and in vitro (P < 0.05). Nuclear HIF1A in granulosa-lutein cells was highest during luteal formation and absent from the fully functional corpus luteum (P < 0.05). Both VEGFA (P < 0.001) and HIF1A (P < 0.01) were up-regulated by dibutyryl-cAMP, through a PKA pathway. Hypoxia increased VEGFA (P < 0.001) and HIF1A (P < 0.05) expression and hCG further augmented VEGFA (P < 0.001) and HIF1A (P < 0.01) under hypoxic conditions. However, progesterone increased hCG-stimulated VEGFA but had no effect on HIF1A expression. The expression of HIF1A is therefore hormonally regulated in luteal cells in vitro and in vivo and may regulate VEGFA expression under normoxic and hypoxic conditions. However, the differential effects of progesterone suggest that not all regulation of VEGFA is associated with an up-regulation of HIF1A.