We observed that graded estrogen exposure increases ZEB1 mRNA and protein expression in vitro in human foreskin fibroblast cells. Furthermore, we found that ZEB1 is over expressed in preputial tissue of patients with severe hypospadias.
ZEB1 is expressed throughout the developing embryo and is necessary for survival.13
Most research on the function of ZEB1 has focused on the effect of ZEB1 knockouts in development or ZEB1 overexpression in carcinogenesis. To our knowledge, there have not been any reports of human congenital genitourinary anomalies associated with ZEB1 overexpression. Therefore, it is necessary to extrapolate from the known functions of ZEB1 to explain and provide the context for the association we found between ZEB1 expression levels and hypospadias.
The urethral abnormalities seen in hypospadias can be viewed as a failure of epithelial cell adhesion. In hypospadias the urethra terminates proximal to its orthotopic location in the glans penis as a result of failure of the urethral plate and abortive corpus spongiosum to fuse in the midline. We hypothesize that ZEB1 overexpression decreases cellular adhesion in the developing male urethra and ventral penile skin, which results in the abortive penile development seen in hypospadias. It is known that ZEB1 overexpression decreases intercellular adhesion by repressing multiple cellular adhesion molecules such as E-cadherin, which is a transmembrane glycoprotein that mediates cellular adhesion between epithelial cells through interaction with other proteins such as β
To our knowledge, no studies have investigated the expression patterns of E-cadherin in normal and hypospadic urethras. However, poor cellular adhesion is a likely causative factor in hypospadias as mice with mutations of ephrin-B2, a cell surface molecule important in cellular adhesion, manifest severe hypospadias and cloacal anomalies.18
It is plausible that the increased ZEB1 expression in the basal layer of preputial skin in boys with hypospadias reflects decreased expression of cell adhesion molecules in the urethra such that fusion of the urethral folds is aborted prematurely. However, it is possible that the molecular and cellular events occurring in the prepuce, which is of ectodermal origin, do not reflect events occurring in the urethra, which is of endodermal origin.
An alternative, although less likely, explanation is that ZEB1 overexpression decreases urethral collagen. In osteoblasts ZEB1 has been shown to down-regulate type I collagen.19
If ZEB1 also represses collagen I genes in the urethra, it is possible that deficient urethral collagen formation contributes to hypospadias. However, to our knowledge, ZEB1 mediated decreased collagen formation has not been demonstrated in tissues other than cartilage and bone. Additionally there are no differences in the amount or distribution of collagen I or III in hypospadic and normal penises.20
We found a direct relationship between estrogen concentration and ZEB1 foreskin fibroblast expression in vitro. We also found that boys with severe hypospadias have higher levels of ZEB1 mRNA and protein levels than controls. Although the urethral tissue was not directly examined, it is possible that there is also a dose response effect of ZEB1 on urethral development. Chamberlain and Sanders found that estrogen increases ZEB1 transcription through direct interaction of the estrogen receptor complex with the ZEB1 gene.21
Our findings support these results and suggest that the enhancement of ZEB1 transcription by estrogen is concentration dependent. Additionally Graham et al reported that ZEB1 binds to the AR promoter, ZEB1 regulates the response of the AR to androgens, and there is reciprocal suppression between AR and ZEB1.22
These results were observed in an aggressive breast cancer cell line, and it is unknown how ZEB1 and AR interact during normal and abnormal embryonic development. However, when viewed in this context, our findings raise the possibility that ZEB1 mediates some of the effects of sex hormones on penile development.23–25
Alterations in sex hormones during the critical period of external genitalia formation have been hypothesized to increase the risk of hypospadias.26,27
Boys born to women exposed to diethylstilbestrol in utero have higher rates of hypospadias than boys born to women who are not exposed to diethylstilbestrol.4
To our knowledge, no published study has revealed an association between the dose of estrogen exposure and hypospadias severity. However, in a population based cohort boys who were small for gestational age were significantly more likely to have moderate to severe hypospadias but not mild hypospadias compared to normal children.28
It is possible that ZEB1 overexpression links the associated conditions of small for gestational age and severe hypospadias, since ZEB1 represses production of types I and II collagen in osteoblasts and chondrocytes, respectively.19,29
In this study Hs68 cells were used for the in vitro experiments. This cell line has important strengths, including being derived from the foreskin of a newborn male, and previous studies in which Hs68 cells were exposed to estrogen have supported findings in human tissue samples and mouse models.30
However, this cell line has limitations. Hs68 cells might not reflect the gene expression profile or cellular properties of the developing urethral plate, as they are endodermal in origin and derived from neonatal rather than fetal tissue. Given that Hs68 cells are fibroblasts, we cannot definitively state that estrogen increases ZEB1 expression in human foreskin epithelium in vitro. However, since ZEB1 levels vary by tissue type, we believed it was of primary importance to perform the in vitro studies in cells that originated from human neonatal male genital skin. We are not aware of an epithelial cell line that meets these criteria. Despite these limitations, we believe the Hs68 cell line is an acceptable, albeit imperfect, model for investigating hypospadias at a cellular level, given that the foreskin and urethra are affected in hypospadias, and human subject studies support the in vitro experiments.
Given the responsiveness of ZEB1 to estrogen and the association of estrogen exposure and androgen disruption with hypospadias, the mechanism by which ZEB1 mediates the effect of estrogen in the pathogenesis of hypospadias merits further investigation. Future experiments using murine models to investigate the effect of estrogen on ZEB1 expression and the development of hypospadias should help elucidate whether ZEB1 overexpression can cause hypospadias.