|Home | About | Journals | Submit | Contact Us | Français|
The study by Agbor et al. in this issue of Biology of Reproduction provides a novel approach to determine the functions of Sertoli cell and germ cell-expressed Dmrt1.
DMRT1 (doublesex and mab-3-related transcription factor 1) is an evolutionarily conserved regulator of postnatal testis differentiation (reviewed in Zarkower ). DMRT1 encodes a protein with a DM domain, a DNA-binding motif first described in the sexual regulators doublesex of Drosophila and MAB-3 of Caenorhabditis elegans [2, 3]. DM domain-encoding genes regulate various aspects of sexual differentiation in nematodes and insects. Most reproductive regulators are restricted to a particular phylum; in contrast, DMRT1 is expressed in the testis of all vertebrates [4–8]. Despite prominent specific expression of Dmrt1 in the early gonadal primordia, Dmrt1 loss-of-function mutation has no discernible effect on sex determination or on subsequent differentiation of mouse embryonic testes [8, 9]. However, shortly after birth, Dmrt1 mutant mice develop highly dysplastic testes resembling those of men with a 9p24 chromosomal deletion. Human DMRT1 is localized to this chromosomal region, whose hemizygosity is associated with testicular dysgenesis [3, 8–12].
In mammals, Dmrt1 expression emerges shortly after formation of the genital ridge and can easily be detected in the two principal cell types of the seminiferous epithelium: Sertoli cells and germ cells. This expression in adjacent and highly symbiotic cell types complicates the interpretation of the phenotype that arises as a result of global gonadal Dmrt1 deficiency. Given the principal role of Sertoli cells in supporting the germ cell population, the abnormal differentiation of Sertoli cells could be sufficient to derail germ cell development. Therefore, the function of Dmrt1 activity in Sertoli (SCDmrt1) and germ cells (GCDmrt1) warrants assessment. Kim et al.  previously provided important insights by conditionally deleting Dmrt1 in Sertoli cell and germ cell lineages using cell type-specific Cre transgenes. This approach revealed requirements for Dmrt1 in both cell lineages as well as a cell nonautonomous role for Dmrt1 in germ cell development. Dmrt1-null mutant germ cells do not survive beyond Postnatal Day 10 (P10) and fail to undergo radial migration, reactivate mitosis, or enter meiosis. Subsequent research confirmed a dual role for DMRT1 in promoting spermatogonial development and restricting meiosis . Loss of Dmrt1 in Sertoli cells leads to their overproliferation and developmental arrest in this cell population [8, 9].
The study by Agbor et al.  in this issue of Biology of Reproduction provides a complementary approach to determine the functions of SCDmrt1 and GCDmrt1. The article describes generation of Wt1-Dmrt1 transgenic (Dmrt1+/−;tg) mice, which express rat Dmrt1 cDNA in gonadal supporting cells by directing Dmrt1 expression from the Wilms tumor locus in a yeast artificial chromosome. A mouse model of DMRT1-deficient germ cells (Dmrt1−/−;tg) was obtained by breeding Dmrt1-null (Dmrt1−/−) mice with the Wt1-Dmrt1 transgenic animals.
The authors provide solid evidence that transgenic DMRT1 is amply expressed in Sertoli cells of the testis shortly after birth. Based on a previous study , one could safely predict that SCDmrt1 would not complement the germ cells, as they are habituated to their own DMRT1 source. Indeed, similar to Dmrt1−/− mice, male Dmrt1−/− transgenic mice lost the Dmrt1-null germ cells that failed to complete radial migration before P15. However, it was considerably more surprising that SCDmrt1 turned out to be so inefficient in restoring functionality to the Sertoli cell population in which it was re-expressed. Here, earlier data supported the hypothesis that DMRT1 activity in germ cells had little overt effect on Sertoli cell differentiation . The resulting phenotype and analysis of testis gene expression demonstrated that, by most measures, Dmrt1−/−;tg testis resembled the Dmrt1-null rather than normal Dmrt1+/+ gonads. The evaluation of Dmrt1−/−;tg Sertoli cells by electron microscopy further confirmed that their maturation was mostly abnormal. In summary, these observations strongly suggest that an entirely Dmrt1-deficient germ cell milieu impedes the normal development of both the Sertoli cells and the seminiferous epithelium.
Nonetheless, introducing Dmrt1 back into Sertoli cells improved the overall organization of the Dmrt1−/− testis. While Dmrt1-null gonads remain largely dysplastic, Dmrt1+/−;tg transgenic testes harbor a clearly identifiable lattice of seminiferous tubules. The authors demonstrated that SCDmrt1 partially corrects the integrity of Sertoli intercell tight junctions and identified Sertoli cell polarity markers, ESPIN and NECTIN-2, as critically dependent on SCDmrt1. Importantly, the restoration of tight junctions was incomplete, in contrast with the previous hypothesis that Dmrt1 expression in germ cells is not required for Sertoli cell polarization.
The article by Agbor et al.  provides incontrovertible evidence that troubleshooting Dmrt1 expression restricted solely to Sertoli cells is vastly inadequate (Fig. 1). This work illustrates that to thoroughly elucidate gene function in complex scenarios, both gene deletion and gene recovery are necessary, as they provide complementary cues. Additionally, genetic tools to create gene loss could make a difference: the Cre-mediated deletion of Dmrt1 in a subset of germ cells  results in a different phenotype versus testes in which Dmrt1-positive Sertoli cells arise side by side with Dmrt1-null germ cells. Using multiple available mouse models, further research could explore mammalian Dmrt1 deficiency in several settings. For example, we recently learned that Dmrt1 loss de-represses Foxl2 expression and reprograms Sertoli cells into ovarian granulosa cells. Dmrt1 gene deletion solely in Sertoli cells is sufficient to activate this sex switch . The timing of this sex reversal, approximately 2 wk postnatally, many days after Dmrt1 loss, remains unexplained. It will be very informative to examine whether in the new model described by Agbor et al.  Sertoli cells preserve their male identity and are no longer positive for female markers. It may be technically easier to explore some of the aspects of transdifferentiation in the Dmrt1 transgenics, as they do not rely on Cre excision. Undoubtedly, an updated version of Dmrt1 operating instructions is due in the near future.
1Supported by NIH grant 2R01HD042751 to S.G.T.