Battle of sex hormones in genitalia anomalies

To cope with their transition to a terrestrial lifestyle, vertebrates had to extensively modify their reproductive organs to facilitate reproduction on land (1). The mammalian penis represents such a pinnacle in mammalian evolution, which enabled internal fertilization and successful land invasion. There are two phases of external genitalia development in mammals: (i) an ambisexual stage, in which male and female embryos undergo the same genital outgrowth program; and (ii) a sexually dimorphic stage, in which further growth and differentiation of the male penis depend on androgen. Consequently, it is the second phase of genitalia development that is sensitive to perturbations of the hormonal milieu during development. Congenital penile anomalies (CPAs), including ambiguous genitalia, hypospadias, chordee, and micropenis, represent one of the most common birth defects, second only to congenital cardiac defects. Hypospadias is an arrest in urethra development, such that the urethra opening is located anywhere along the ventral side of the penile shaft instead of at the glans penis. Chordee is the abnormal bending of the penis, resulting from tethering of urethral epithelium to the skin. Over the past few decades, the reported CPA incidents have increased significantly, raising the speculation that in utero exposure to endocrine disrupting chemicals (EDCs) may be responsible for this increase. Published studies using either the mouse or rat model showed that developmental exposures to antiandrogen or estrogenic compounds can lead to a range of penile anomalies similar to human CPAs (2). However, how and when EDCs may influence normal external genitalia development is not very clear. In PNAS, Zheng et al. (3) use a state-of-the-art conditional androgen receptor (AR) knockout mouse model to show that disruption of androgen signaling at different developmental stages can produce the full spectrum of CPAs. The authors go on to show that androgen and estrogen signaling antagonize each other during neonatal mouse genital development and identify a signaling molecule, Indian hedgehog (Ihh), as a novel AR target required for penile masculinization.

Mammalian external genitalia development begins with the outgrowth of a genital tubercle (GT), which is the anlage of penis in males and clitoris in females. In mice, the GT develops as a pair of genital swellings around the cloaca and slightly caudal to the hindlimbs on embryonic day (E) 10.5 (Fig. 1A). By E11.5, the two swellings have merged and formed a single GT (4, 5). The GT and limb bud use a common genetic cassette to promote their outgrowth, which involves a signaling center containing Wnt and Fgf activities (6). However, the signaling center in the GT is the endodermally derived distal urethral epithelium, in contrast to the ectodermally derived apical ectodermal ridge in limbs (4). By E13.5, a pair of preputial swellings emerge on either side of the GT, which eventually will engulf the GT, while inside the GT tubulogenesis commence in a proximal-distal direction involving canalization of urethral epithelium from its bilaminar plate precursor (Fig. 1A).

Fig. 1.

(A) A schematic diagram of mouse external genitalia development from the sexual indifferent stage (E10.5–E15) to adult sexual dimorphic stage. Reprinted from ref. 6. (B) A developmental time window when penile development is sensitive to endocrine disruption. (C) Signaling pathways implicated in genital masculinization. b, baculum (os penis); EB, estradiol benzoate; Flut, flutamide; gp, glans penis; p, prepuce; u, urethra.

Work from the Yamada laboratory demonstrated that before E15, male and female mouse GTs are morphologically indistinguishable (7). However, after E15.5, GT development enters the sexually dimorphic stage, where they are sensitive to androgen and antiandrogen treatments. Consistent with this finding, Zheng et al. (3) examine the sexual dimorphic expression of AR and estrogen receptor (ERα) in both male and female genitalia throughout development and show that sexually dimorphic AR and ER staining patterns were first observed at E15.5 (3). More importantly, exposure to exogenous estrogen or androgen can up-regulate its own receptor expression, while simultaneously repressing expression of the other receptor. These findings lend support to the idea that sexual differentiation of the genitalia is a balance between the two sex hormones, because for a long time it was thought that female genitalia represent the default state until it was found that ERα knockout females showed masculinized external genitalia (8). During normal male development, androgen produced by the Leydig cells promotes further genital growth, as well as mesenchymal condensation within the glans penis, to give rise to various corporal tissues and penile bones that are either missing or rudimentary in females (Fig. 1A). Consequently it is the mesenchymal AR but not ectodermal or endodermal AR that is required in genital masculinization (3, 7). In the Zheng et al. study (3), the authors varied the timing of AR disruption either pharmacologically or genetically. Surprisingly, early (E13.5) deletion of AR led to feminization of male penis similar to that found in tfm mice, whereas late (E17.5) deletion resulted in micropenis but did not affect urethral tube closure (3). On the other hand, disruption of androgen signaling between E12.5 to E16.5 with an AR antagonist, flutamide, caused urethral tube closure defects, similar to human genital defect-hypospadias with chordee (Fig. 1B). These results show that disruption of a single genetic pathway at different times during development can generate the wide spectrum of CPA. In addition to defining the androgen-sensitive period, Zheng et al. (3) also discovered that postnatal day 0–6 represents an estrogen-sensitive period for male penis development. Males develop micropenis when exposed to estradiol benzoate during the neonatal period (Fig. 1B). Taken together, these data highlight the importance of the temporal window of endocrine disruption during genitalia development and may serve as a guide to endocrine disruption studies in humans. Because of the long gestation period for humans (40 wk), in utero exposure to EDCs can span the whole length of human pregnancy, including both hormone-sensitive and -insensitive periods. The sensitive period to androgen disruption in mice corresponds to the first trimester of human pregnancy, whereas the estrogen-sensitive period corresponds to mostly the second trimester (2, 9). Thus, if the EDC exposure window can be more precisely defined and separated into hormone-sensitive and -insensitive periods, epidemiological studies may be able to increase their predicative power as to whether in utero exposure to certain chemicals is associated with genital defects later in the adult.

Although it has been known for quite some time that sex hormones play important roles in genital differentiation, the downstream targets—especially those paracrine factors mediating local tissue interactions—remain mostly illusive. The Cohn (10) and Yamada (7) groups have previously identified Fgf and Wnt pathways as targets of androgen signaling, respectively. The Hedgehog pathway may represent another androgen-regulated pathway as deletion of Shh during sexual differentiation led to hypospadias (Fig. 1C) (11–13). In the current study, Zheng et al. (3) use a candidate approach and examine the expression of 88 genes in the Hedgehog, Bmp, Wnt, and Fgf pathways, and compare their expression to either antiandrogen (flutamide) -treated or estrogen (estradiol benzoate) -treated GT. Another ligand of the Hedgehog pathway, IHH, is one of only two molecules whose expression is down-regulated by both compounds. To test whether Ihh is required for masculinization of the male genitalia, Zheng et al. (3) used Hoxa13Cre to conditionally deleted Ihh in both the urethal epithelium and mesenchyme of the developing GT. The resulting mutant penis is underdeveloped and shows much reduced penile structures, including the fibrocartilaginous male urogenital mating protuberance (MUMP) and the MUMP ridge. During sexual differentiation, Ihh is strongly expressed in urethral epithelium and genital mesenchyme of male but not female GT, and its expression is greatly repressed by either flutamide or estradiol benzoate in male GT (3). Combined, these data convincingly demonstrate the function of Ihh in genital masculinization (Fig. 1C). Given the well-documented role for Ihh in cartilage and bone development (14), it is tempting to speculate that Ihh expression is elevated in male mouse genitalia to promote penian bone (os penis) and cartilaginous MUMP development because there is no cartilage tissue in the female clitoris and the bone segment (os clitoris) is very small (8). If this hypothesis is correct, it would then be interesting to examine genital Ihh expression in other mammalian species whose genitalia do not contain bone or cartilage, such as those in marsupials and hyenas, as well as in human beings. Such studies may provide additional insight into the function of Ihh in genital masculinization.

Epidemiological studies, as well as studies using experimental animals, have shown that a cadre of EDCs can lead to development of hypospadias. These include chemicals that interfere with androgen-signaling, such as phthalates and fungicides, compounds with estrogenic activities, such as diethylstilbestrol, and finally progestins. Despite some noticeable anatomical and morphogenetic differences between rodent and human penises (2), most pathways governing penile development and differentiation should be conserved. Unfortunately, our knowledge of genitalia development and differentiation is still in its infancy. Because genome-wide association studies have only yielded a handful of candidate genes for hypospadias (15, 16), perhaps more effort should be devoted to model organism studies exemplified by the current study to better understand the normal biological process of genitalia outgrowth and differentiation.