Forkhead transcription factor regulates prostate-specific gene expression and prostatic morphogenesis: a functional interaction with androgen signaling

Abstract

Comparative genome analysis implied that a progressively more elaborate regulation of gene expression rather than invention of new genes is responsible for the organismal complexity, and that a constrained organization of metazoan enhancers is essential for the precise patterns of gene expression during development. This project was focused on studying of the transcriptional regulatory complex that directs tissue-specific gene expression and organ development in the prostate. Genetic manipulation of the rat probasin (Pbsn) gene promoter in transgenic mice uncovered a ~150 base pair core DNA fragment, which confers prostate-selectivity to the expression of this gene. Among a cluster of DNA sequence-specific transcription factors directly binding to this region, there are two cell-type limited proteins: the androgen receptor, a nuclear receptor, and Foxa1, a forkhead protein. Reporter assays indicated the forkhead response elements are crucial for AR-mediated transcriptional regulation. In vitro and in vivo binding assays established that the two proteins can bind adjacent DNA sequence concomitantly while the binding is not interdependent. A physical interaction occurs between Foxa1 and AR when the ligand for AR is present. This interaction is mediated through the forkhead domain and the AR DNA binding domain. Alignment of the forkhead and the androgen response elements resulted in the identification of a shared cis-regulatory code that is present in a variety of prostate-specific enhancers across species. The impact of Foxa1 loss-of-function and haploinsufficiency on mouse prostatic organogenesis was analyzed. Foxa1-/- prostates showed drastic morphologic alteration including a disorganized epithelial pattern resembling primitive epithelial cords, and an expansion in the mesenchymal smooth muscle layer. Cell type and ultrastructural studies demonstrated that Foxa1-/- epithelium is predominantly arrested as immature basal cells, consistent with a failure of luminal determination. Foxa1-/- basal cells actively express Sonic hedgehog (Shh), Ptc1 and Foxa2, proteins that are normally elevated during embryonic ductal budding in the UGS. Alteration of these signals correlates with the expansion of precursor cells and the modified epithelial-stromal pattern in Foxa1-/- prostates. In contrast, expression of the Nkx3.1 homeobox protein is absent in these Foxa1-/- cells. In addition a haploinsufficient phenotype was observed in heterozygous dorsal prostates showing a similar ductal morphologic defect. A novel Foxa1 target gene was identified. Upon examination of its promoter, critical forkhead response elements were identified to be immediately flanked by androgen response elements. We propose that Foxa1 regulates genes involved in prostatic ductal morphogenesis, and promotes epithelial cell maturation through balancing the effects of Shh, Foxa2, and Nkx3.1.