Cell type-autonomous and non-autonomous requirements for Dmrt1 in postnatal testis differentiation

doi:10.1016/j.ydbio.2007.04.046

Dev Biol. Author manuscript; available in PMC 2008 July 15.

Published in final edited form as:

Dev Biol. 2007 July 15; 307(2): 314–327.

Published online 2007 May 3. doi: 10.1016/j.ydbio.2007.04.046

PMCID: PMC1995593

NIHMSID: NIHMS27420

Cell type-autonomous and non-autonomous requirements for Dmrt1 in postnatal testis differentiation

Shinseog Kim,^1,² Vivian J. Bardwell,^1,^2,³ and David Zarkower^1,^2,³^*

¹Department of Genetics, Cell Biology, and Development, Minneapolis, MN 55455, USA

²Graduate Program in Biochemistry, Molecular Biology, and Biophysics, Minneapolis, MN 55455, USA

³Cancer Center University of Minnesota, Minneapolis, MN 55455, USA

^*Author for correspondence: Department of Genetics, Cell Biology, and Development 6-160 Jackson Hall 321 Church St. SE Minneapolis, MN 55455 email: zarko001/at/umn.edu phone: 612-625-9450 fax: 612-626-7031

The publisher's final edited version of this article is available at Dev Biol

See other articles in PMC that cite the published article.

Publisher's Disclaimer

Abstract

Genes containing the DM domain, a conserved DNA binding motif first found in Doublesex of Drosophila and mab-3 of C. elegans, regulate sexual differentiation in multiple phyla. The DM domain gene Dmrt1 is essential for testicular differentiation in vertebrates. In the mouse, Dmrt1 is expressed in pre-meiotic germ cells and in Sertoli cells, which provide essential support for spermatogenesis. Dmrt1 null mutant mice have severely dysgenic testes in which Sertoli cells and germ cells both fail to differentiate properly after birth. Here we use conditional gene targeting to identify the functions of Dmrt1 in each cell type. We find that Dmrt1 is required in Sertoli cells for their postnatal differentiation, and for germ line maintenance and for meiotic progression. Dmrt1 is required in germ cells for their radial migration to the periphery of the seminiferous tubule where the spermatogenic niche will form, for mitotic reactivation and for survival beyond the first postnatal week. Thus Dmrt1 activity is required autonomously in the Sertoli and germ cell lineages, and Dmrt1 activity in Sertoli cells also is required non-autonomously to maintain the germ line. These results demonstrate that Dmrt1 plays multiple roles in controlling the remodeling and differentiation of the juvenile testis.

Keywords: DMRT1, conditional targeting, testis, Sertoli, gonocyte, spermatogenesis, autonomous

INTRODUCTION

The seminiferous epithelium of the mammalian testis is a complex and highly dynamic tissue comprised of germ cells and Sertoli cells surrounded by a basal lamina and a layer of peritubular myoid cells. During early postnatal development and puberty the Sertoli and germ cells undergo tremendous changes and interact closely to establish a germ line stem cell population, supported by the Sertoli cells, that can produce mature spermatozoa for the duration of reproductive life.

Primordial germ cells (PGCs) appear around 7 days post-coitum (dpc) in the extraembryonic mesoderm (Ginsburg et al., 1990) and migrate through the allantois to the hindgut, reaching the genital ridge by 9.5-11 dpc (Anderson et al., 2000). After PGCs arrive in the gonadal primordium they are enclosed by pre-Sertoli cells, which coalesce during this period to form testis cords that later elongate into seminiferous tubules (McLaren, 1998). Sertoli cells play an important role in controlling germ cells in the embryo, arresting male germ cell mitosis in G1 and blocking meiotic initiation (Adams and McLaren, 2002; Bowles et al., 2006; McLaren and Southee, 1997). The mitotically-arrested PGCs differentiate into gonocytes, which remain mitotically quiescent until around birth in the mouse (de Rooij and Russell, 2000; Nagano et al., 2000; Vergouwen et al., 1991).

During the first postnatal week, the seminiferous tubules are remodeled in preparation for spermatogenesis. At birth, the tubules have the embryonic arrangement, with Sertoli cells at the periphery and gonocytes in the center. Starting perinatally, the gonocytes migrate radially from their initial central location to the basal lamina at the periphery (Nagano et al., 2000). Explant culture experiments suggest that the c-KIT receptor and its ligand may play a role in this process (Orth et al., 1997). Expression patterns and transgenic studies also have implicated ALCAM and short-type PB-cadherin as potential mediators of gonocye migration (Ohbo et al., 2003; Wu et al., 2005), but so far no gene has been shown to be essential in gonocytes for their postnatal migration. By postnatal day 6 (P6), ~ 90% of gonocytes are located at the basal lamina in what will become the spermatogonial stem cell niche (McLean et al., 2003; Nagano et al., 2000). Germ cells that complete the radial migration rapidly differentiate into germ line stem cells (A spermatogonia) that are first apparent by P6 in the mouse, whereas germ cells that fail to migrate die (de Rooij, 2001). The spermatogonia undergo proliferative cell divisions and begin to differentiate into spermatocytes that commence meiosis.

The early postnatal period also is a critical time for Sertoli cell development. Sertoli cells proliferate only in the fetal and pre-pubertal periods, and by the end of the second postnatal week they begin to undergo maturation, losing their proliferative capacity and acquiring functions that will enable them to support spermatogenesis (Sharpe et al., 2003). Sertoli cell support is crucial for germ cell differentiation, meiosis and spermiogenesis, and signals from the Sertoli cells are essential for the maintenance of the spermatogonial stem cell niche in adults (Griswold, 1995; Hess et al., 2006). Although the cellular behaviors of germ cells and Sertoli cells in the early postnatal testis that lead to the establishment of the spermatogenic stem cell niche are well described, very little is known of how these processes are regulated. However, as described below, the Dmrt1 gene is required for these events and provides an entry point for the elucidation of their control.

Dmrt1 (doublesex and mab-3 related transcription factor 1) is a critical and conserved regulator of postnatal testis differentiation (Raymond et al., 2000). Dmrt1 is expressed in the testis of all vertebrates so far examined, starting at the genital ridge stage and continuing throughout adult life (Kettlewell et al., 2000; Marchand et al., 2000; Raymond et al., 1999a; Raymond et al., 2000; Smith et al., 1999). In humans, deletion of the region of chromosome 9p containing DMRT1 results in testicular dysgenesis (Crocker et al., 1988; Ion et al., 1998; Ogata et al., 1997; Raymond et al., 1999b; Raymond et al., 1998), while amplification of DMRT1 is associated with a form of testicular germ cell cancer, spermatocytic seminoma (Looijenga et al., 2006). Dmrt1 mutant mice have severely abnormal testis development starting at about P2 and resulting in severely dysgenic testes resembling those of 9p deletion patients (Fahrioglu et al., 2007; Raymond et al., 2000). Dmrt1 encodes a protein with a DM domain, a DNA binding motif first identified in the sexual regulators Doublesex of Drosophila and the MAB-3 of C. elegans (Erdman and Burtis, 1993; Raymond et al., 1998). DM domain-encoding genes have been shown to regulate various aspects of sexual differentiation in insects, nematodes, and mammals, suggesting an ancient involvement of these genes in sexual regulation (Zarkower, 2001).

Targeted deletion of Dmrt1 in the mouse demonstrated that the gene is essential for postnatal testis differentiation (Raymond et al., 2000). Dmrt1 is unusual in that it is expressed specifically in both Sertoli cells and in germ cells, starting as soon as the genital ridge forms, and its loss affects the development of both cell types. Germ line defects in Dmrt1 null mutant mice include the failure of germ cells to undergo radial migration, to reactivate mitosis, to enter meiosis, and to survive beyond P10. Sertoli cell defects include developmental arrest and over-proliferation (Fahrioglu et al., 2007; Raymond et al., 2000). In addition, Dmrt1 mutant adults are incompletely virilized (J. Balciuniene and D.Z., unpublished). Based on mRNA expression analysis, loss of Dmrt1 affects both Sertoli cells and germ cells at least as early as P1 (Fahrioglu et al., 2007).

Although Dmrt1 is expressed in Sertoli cells and germ cells and is required for their postnatal development, the essential role of Sertoli cells in supporting germ cell development has made it unclear in which cells Dmrt1 performs which of its functions. To find out, we have conditionally targeted Dmrt1 in the Sertoli cell and germ cell lineages using cell type-specific Cre transgenes. This approach reveals specific requirements for Dmrt1 in both cell lineages as well as a cell non-autonomous role for Dmrt1 in germ cell development. We find that Dmrt1 is required in germ cells but not Sertoli cells for radial migration of gonocytes to the basal lamina, for their mitotic reactivation, and for germ cell survival beyond the first postnatal week. Dmrt1 activity in Sertoli cells is required for their proper organization and differentiation, and for survival and differentiation of wild type germ cells beyond the second postnatal week.

MATERIALS AND METHODS

Generation of SCDmrt1KO and GCDmrt1KO mutant mice

Cell type-specific Dmrt1 knockout mice were generated using Cre-mediated recombination of loxP sites in a functional Dmrt1 allele, Dmrt1^floxed, whose construction was previously described (Raymond 2000). Dmrt1^+/− animals (129/Sv and C57BL/6 mixed background) were crossed either with Desert hedgehog–Cre (Dhh-Cre) mice expressing Cre selectively in Sertoli cells (Lindeboom et al., 2003), or with Tissue Non-specific Alkaline Phosphatase Cre (TNAP-Cre) animals expressing Cre in primordial germ cells (PGCs) (Lomeli et al., 2000). Dmrt1^{floxed/floxed} animals were crossed with Dhh-Cre; Dmrt1^+/− to generate SCDmrt1KO mutants (Dhh-Cre; Dmrt1^floxed/−) or with TNAP-Cre; Dmrt1^+/− animals to produce GCDmrt1KO mutants (TNAP-Cre; Dmrt1^floxed/−). As controls, we used Dmrt1^floxed/− or Dmrt1^floxed/+ littermates.

Genotyping

Genotyping was performed as described previously (Fahrioglu et al., 2007; Raymond et al., 2000). Tail-clip DNA was amplified for 35 cycles. The wild-type Dmrt1 allele (Dmrt1⁺) was detected by PCR with CR92/CR99, with an annealing temperature of 55°C. The Dmrt1^floxed allele was detected by PCR with KOS2/KOS3N, with an annealing temperature of 62°C. The deleted allele Dmrt1⁻ was detected with KOS1N/KOS3N with an annealing temperature 65°C. Cre transgenes were detected by PCR with CreF/CreR, with annealing temperature of 62°C. PCR with CR92/CR99 contained 10% DMSO.

Primers

CreF: 5'-CCTGATGGACATGTTCAGGGATCG -3'
CreR: 5'-TCCATGAGTGAACGAACCTGGTCG -3'
CR92: 5'-CAGCTCCATGGCGAACGACGACACATTCGG-3'
CR99: 5'- CTGCAGCGAGCGCATTTGGGCAGC-3'
KOS2: 5'- TGCACACGTGCACCCTCGCCATCG -3'
KOS1N: 5' GATCTATCTGGAGCCAGGTGGTAG -3'
KOS3N: 5'- TCATGGCAGCTCTCCCAGTGGAGC -3'

Histological analysis

Dissected testes were fixed in Bouin's fixative or phosphate-buffered formalin overnight at 4°C, progressively dehydrated in a graded ethanol series and embedded in paraffin wax. Sections (6 μm) were deparaffinized, rehydrated, and stained with hematoxylin and eosin.

Tissue immunofluorescent staining

Slides with paraffin sections were washed in PBT (0.1% Tween 20 in phosphate buffered saline [PBS]) and autoclaved in 10 mM citric acid (pH 6.0) to retrieve antigenicity. Slides were blocked in 5% serum (matched to the species of the secondary antibody) in PBS for 1 h at room temperature and incubated with primary antibodies overnight at 4°C prior to detection with secondary antibodies.

Antibodies

Primary antibodies used for immunofluorescence were rat anti-GATA1 (1:200, Santa Cruz Biotechnology, sc-265), goat anti-GATA4 (1:200, Santa Cruz Biotechnology, sc-1237), rabbit anti-AR (1:50, Santa Cruz Biotechnology, sc-816), rabbit anti phosphohistone H3 (1:200, Upstate 06-570), rat anti-TRA98 (1:200, gift of H. Tanaka and Y. Nishimune), rat anti-BC7 (1:100, gift of H. Tanaka and Y. Nishimune), rat anti-TRA369 (1:1000, gift of H. Tanaka and Y. Nishimune) and rabbit anti-DMRT1 (1:400; (Raymond et al., 2000)). Secondary antibodies used were goat anti-rabbit Alexa 488, goat anti-rabbit Alexa 594, and goat anti-rat Alexa 594 (Molecular Probes), all used at 1:250. Donkey anti-goat FITC (Jackson) and donkey anti-rabbit Texas Red (Jackson) were used at 1:50 according to the manufacturer's instructions.

RESULTS

Generation of Sertoli cell-specific Dmrt1 knockout mice

To test whether Dmrt1 is required autonomously in each of the two cell lineages that express the gene, we generated cell-type specific Dmrt1 knockout mice. We started by targeting Dmrt1 in the Sertoli cell lineage. These experiments employed mice with the null deletion allele Dmrt1⁻ (Raymond et al., 2000), or a conditional allele, Dmrt1^floxed, in which the same segment that is deleted in Dmrt1⁻ is instead flanked by loxP sites. Cre-mediated recombination of these loxP sites removes the transcriptional start site and proximal promotor and the first exon, which contains the DM domain, and results in a protein- and mRNA-null mutation (Raymond et al., 2000). For conditional targeting we used the Dhh-Cre transgene (gift of Dr. Dies Meijer), which has been shown to recombine reporter transgenes in pre-Sertoli cells very efficiently by 18 dpc (Lindeboom F, Gillemans N et al., 2003). The breeding scheme is described in Materials and Methods. In each experiment we compared Dhh-Cre;Dmrt1^floxed/− (SCDmrt1KO) animals, with Dmrt1 deleted in Sertoli cells, to Dmrt1^floxed/− or Dmrt1^floxed/+ control animals containing one or two functional copies of the gene in Sertoli cells. Comparison of these controls with each other and with Dmrt1^+/+ revealed no phenotypic differences (data not shown).

To confirm that Dmrt1 is efficiently and specifically deleted in Sertoli cells of SCDmrt1KO mice, we stained P5 testis sections with antibodies to DMRT1 and either the Sertoli cell marker GATA4 or the germ cell marker TRA98 (Figure 1). GATA4 normally is expressed in juvenile Sertoli cells but not in germ cells, and is expressed in Dmrt1 mutant Sertoli cells (Sharpe, R et al., 2003; Raymond et al, 2000). DMRT1 expression was absent from virtually all Sertoli cells in SCDmrt1KO mice, while remaining in the GATA4-negative germ cells (Figure 1A). Staining with the germ cell-specific TRA98 antibody and anti-DMRT1 confirmed that DMRT1 expression was unaffected in virtually all germ cells of SCDmrt1KO mice (Figure 1B). Together these results confirm that Dmrt1 is deleted efficiently and specifically in the Sertoli cells of SCDmrt1KO testes.

Figure 1

Specific and efficient deletion of Dmrt1 in Sertoli cells of SCDmrt1KO mutants

Abnormal organization of seminiferous tubules in SCDmrt1KO testes

To evaluate testicular development in SCDmrt1KO mice, we first histologically examined the testes of SCDmrt1KO mice and control littermates during the initial period of spermatogenesis, from P5 to P28. By P5, prior to meiosis, most germ cells normally have migrated radially from their initial position in the center of the seminiferous tubule into close apposition with the basal lamina (Figure 2A; closed arrowheads). In SCDmrt1KO testes, a few germ cells could be found in the center of tubules (Figure 2A, open arrowheads), but most had successfully migrated to the periphery (closed arrowheads). By P7, the approximate time of meiotic initiation, germ cells were exclusively peripheral both in control and SCDmrt1KO testes (Figure 2B), suggesting that Sertoli cell Dmrt1 expression is not required for germ cell radial migration, although the process may be slightly delayed in SCDmrt1KO testes. At P9, in control testes the Sertoli cells were organized in a ring either just inside the ring of germ cells at the periphery or mixed among the germ cells at the periphery, but in SCDmrt1KO mutants the Sertoli cells were more randomly distributed (Figure 2C). By P14, control tubules had begun to form lumens and had accumulated primary spermatocytes in the adluminal compartment (Figure 2D). In SCDmrt1KO testes at P14, however, the pattern was very different, with no luminal space apparent and tubules filled with a poorly organized mix of Sertoli cells and differentiating germ cells (Figure 2D). By P28, the control seminiferous tubules contained polarized Sertoli cells properly adjoining the basal lamina, and normally developing germ cells, including spermatogonia, primary spermatocytes, round spermatids, and elongated spermatids. In contrast, the SCDmrt1KO tubules at P28 contained no apparent germ cells and were filled with randomly localized unpolarized Sertoli cells of immature morphology (Figure 2E).

Figure 2

Abnormal seminiferous tubule morphology in SCDmrt1KO testes

Dmrt1 activity in Sertoli cells is not required for germ cell mitotic reactivation

Germ cells in Dmrt1 null testes fail to migrate to the tubule periphery, do not reactivate mitosis or initiate meiosis, and die before P14 (Fahrioglu et al., 2007; Raymond et al., 2000). As just described, although germ cells in SCDmrt1KO testes migrate to the basal lamina in the first postnatal week and initiate meiosis, they are lost by P28. We therefore examined other aspects of early germ cell development in SCDmrt1KO mutants. Specifically, we asked whether wild type germ cells surrounded by mutant Sertoli cells can reactivate mitosis, and we more closely examined their meiotic progression (see below).

To examine mitosis, we used an antibody to the mitotic marker phospho-histone H3 (P-H3; Figure 3). P-H3 also is expressed in meiotic cells, so we also stained for the meiotic marker BC7 (not shown). We found that control and SCDmrt1KO testes had similar numbers of mitotic germ cells both at P5 (not shown) and P7 (Figure 3). This contrasts with Dmrt1 null mutants, in which germ cell mitosis was absent at these stages (Fahrioglu et al., 2007). Because gonocyte mitotic reactivation requires Dmrt1 but is unaffected by elimination of Dmrt1 in Sertoli cells, we conclude that the requirement is likely to be germ line-autonomous.

Figure 3

Germ cell mitotic reactivation does not require Dmrt1 in Sertoli cells

Meiotic progression, but not initiation, requires Dmrt1 activity in Sertoli cells

To examine the onset of the defects in SCDmrt1KO testes, we used antibodies to TRA98 and GATA4 to follow the fates of germ cells and Sertoli cells, respectively, between P5 and P14 (Figure 4). As shown in Figure 2, germ cell migration appears slightly delayed at P5 in SCDmrt1KO mutants (Figure 4A), but it does occur by P7 (Figure 4B), and otherwise little difference was apparent between control and SCDmrt1KO testes up to P7. However, by P9 the distribution of TRA98-positive germ cells was highly abnormal in SCDmrt1KO mutants, with many more cells in the center and many fewer at the periphery of the tubules relative to controls (Figure 4C). This confirms the histology in Figure 2 showing that the mutant germ cells, although able to migrate to the periphery, fail to maintain proper organization with Sertoli cells. At P14 in control tubules Sertoli cell nuclei were intercalated among spermatogonia at the periphery, and there was an adluminal compartment contained differentiating meiotic germ cells. This indicates that Sertoli cell polarization was occurring and separate populations of spermatogonia (peripheral) and spermatocytes (adluminal) were being established. In contrast, in the SCDmrtr1KO seminiferous tubules there was no organized peripheral ring of Sertoli cells and spermatogonia; instead the tubules were filled with TRA98-positive germ cells and lacked a lumen, and many Sertoli cells were located away from the basal lamina (Figure 4D). In addition, some tubules had very few spermatogonia at the periphery at P14, suggesting that Dmrt1 in Sertoli cells may play a role in helping maintain the spermatogonial population.

Figure 4

Dmrt1 is not required in Sertoli cells for germ cell radial migration and meiotic initiation

Although meiosis can initiate in SCDmrt1KO testes, germ cells are absent by P28, as shown earlier. To examine meiotic progression, we used two stage-specific germ cell antibody markers: BC7, which recognizes leptotene to zygotene spermatocytes (Koshimizu et al., 1993); and TRA369, which recognizes pachytene spermatocytes through elongated spermatids (Watanabe et al., 1992). At P14, some SCDmrt1KO tubules contained many BC7-positive cells, indicating that initiation of meiotic prophase does not require Dmrt1 in Sertoli cells (Figure 5A). However, very few cells in SCDmrt1KO testes were TRA369-positive, suggesting a possible meiotic prophase arrest (Figure 5B). We also examined P21 testes, in case meiotic progression was delayed rather than arrested, but also found very few TRA369-positive cells (data not shown). From these results we conclude that germ cells in SCDmrt1KO testes efficiently enter meiosis, but most of them arrest at or before pachynema.

Figure 5

Progression through meiotic prophase requires Dmrt1 in Sertoli cells

The results described so far indicate that Sertoli cells are present in SCDmrt1KO mutants and that germ cells move and develop appropriately until approximately P7 and can reactivate mitosis and enter meiotic prophase in approximately normal numbers. After this stage, however, SCDmrt1KO testes begin to exhibit a number of abnormalities. These defects include aberrant organization of germ cells and Sertoli cells, failure to establish a spermatogonial population at the periphery, failure of Sertoli cell polarization and lumen formation, and meiotic arrest leading to germ cell death.

Dmrt1 is required in Sertoli cells for their differentiation

In Dmrt1 null mutants, Sertoli cells over-proliferate and fail to adopt a differentiated morphology (Raymond et al., 2000). Genome-wide mRNA expression analysis in Dmrt1 mutant testes has identified a number of Sertoli cell markers that are underexpressed at P1 and P2, further indicating that Sertoli cell differentiation is abnormal in Dmrt1 mutants (Fahrioglu et al., 2007). As described above, Sertoli cells in SCDmrt1KO mutants also appear incompletely differentiated and fail to support germ cell development.

We further investigated Sertoli cell differentiation in SCDmrt1KO mutants using two maturation markers, the androgen receptor (AR) (Figure 6) and GATA1 (Figure 7). In rodents AR protein is expressed in peritubular myoid cells at all stages, and is up-regulated in Sertoli cells prior to their maturation (Bremner et al., 1994). This up-regulation occurred between P5 and P9 in Sertoli cells of control testes but not in those of SCDmrt1KO testes (Figure 6). However, at P21 some Sertoli cell expression of AR was present, indicating that Sertoli cell maturation was delayed. Consistent with this conclusion, GATA4 expression was elevated in Sertoli cells at P21 and P28 in SCDmrt1KO testes (not shown). GATA1 was activated in Sertoli cells of SCDmrt1KO mutants, but later than in controls, and GATA1 expression was lost in adulthood (14 weeks) in the mutants (Figure 7). GATA1 staining at P21 also clearly revealed the abnormal arrangement of Sertoli cell nuclei in SCDmrt1KO tubules described earlier.

Figure 6

Dmrt1 is required for normal activation of androgen receptor in Sertoli cells

Figure 7

Dmrt1 is required for normal activation and for maintenance of GATA1 expression in Sertoli cells

Both the histological analysis and antibody staining indicated that Sertoli cell defects observed in SCDmrt1KO mutants closely resemble those we previously observed in Dmrt1 null mutants. We conclude that Dmrt1 activity is required autonomously in the Sertoli cell lineage, and that Dmrt1 activity in germ cells has little overt effect on Sertoli cell differentiation.

Dmrt1 is required cell-autonomously for germ cell radial migration

The results described so far demonstrate that Dmrt1 activity in Sertoli cells is required for their postnatal differentiation and also for meiotic progression and survival of germ cells. In principle, Dmrt1 activity in Sertoli cells could explain much of the defective germ cell development we previously observed in Dmrt1 null mutants. However, germ cells in Dmrt1 null mutants are more severely affected than those in SCDmrt1KO mutants, strongly suggesting that germ cell development requires Dmrt1 activity in the germ cells as well as in the Sertoli cells.

To test the requirement for Dmrt1 in germ cells directly, we conditionally targeted Dmrt1 to generate germ cell-specific Dmrt1 knockout (GCDmrt1KO) mice. We used a TNAP-Cre (Tissue Non-specific Alkaline Phosphatase-Cre) transgene, which is active in primordial germ cells prior to their colonization of the gonad (Lomeli et al., 2000). Although TNAP-Cre activity is less efficient and cell type-specific than that of Dhh-Cre, comparison of germ cell phenotypes between SCDmrt1KO and GCDmrt1KO mutants should nevertheless reveal germ line-autonomous requirements for Dmrt1.

GCDmrt1KO animals and littermate controls were generated by a breeding scheme analogous to that for SCDmrt1KO mice. To confirm the selective deletion of Dmrt1 in GCDmrt1KO mutant germ cells, we first double-stained with antibodies to DMRT1 and GATA4 (Figure 8A). At P7, this revealed a number of tubule sections in GCDmrt1KO testes in which no Dmrt1-positive germ cells were present and nearly all Sertoli cells retained DMRT1 expression. Thus TNAP-Cre was able to selectively eliminate Dmrt1 from germ cells. Next we stained P7 GCDmrt1KO testes for DMRT1 and TRA98 (Figure 8B). As expected, we observed tubule sections lacking DMRT1-positive germ cells. In contrast to the germ cells in control testes, many germ cells lacking DMRT1 had failed to migrate to the periphery (mutant: 63% failed to migrate, 265 cells counted in 60 tubule sections; control: 16% failed to migrate, 1003 cells counted in 68 tubule sections). We conclude from these results that Dmrt1 is required in germ cells but not in Sertoli cells for germ cell radial migration, as suggested by the SCDmrt1KO results described earlier. In addition, the reduced number of germ cells in GCDmrt1KO testes is consistent with the expected failure of mutant germ cells to reactivate mitosis.

Figure 8

Dmrt1 is required in germ cells for their radial migration

Dmrt1 is required germ line-autonomously for germ cell survival

In Dmrt1 null mutants, germ cells are virtually absent by P14 and do not enter meiosis. This contrasts with SCDmrt1KO mutants, in which many germ cells were still present at P14 and had efficiently entered meiotic prophase. The likely explanation is that Dmrt1 is required cell-autonomously in germ cells for survival beyond P10 and for meiotic initiation. However, we could not exclude an alternative possibility: the more severe germ cell defects in Dmrt1 null mutations might result exclusively from an embryonic requirement for Dmrt1 in pre-Sertoli cells, prior to the time of Dhh-Cre action. We therefore examined GCDmrt1KO mutants at P14, asking whether the germ cell phenotype more closely resembles that of Dmrt1 null mutants or SCDmrt1KO mutants (Figure 9). Histological analysis at P14 revealed that GCDmrt1KO mutants had some tubules completely lacking germ cells, and these contained normally polarized Sertoli cells (Figure 9A, arrowheads). This was confirmed by double-staining with antibodies to DMRT1 and TRA98 (Figure 9B): in control testes all tubule sections contained many TRA98-positive germ cells; but some tubule sections of GCDmrt1KO had no germ cells (arrowheads). The identity of the surviving cells was confirmed by staining for DMRT1 and GATA1 (Figure 9C), which revealed tubule sections in which all cells were GATA1-positive Sertoli cells. These data demonstrate that Dmrt1 loss in the germ line causes germ cells to die much earlier than does Dmrt1 loss in Sertoli cells.

Figure 9

Germ cell survival and meiosis require Dmrt1 in germ cells

Based on analysis of GCDmrt1KO mutants, we conclude that Dmrt1 is required cell-autonomously in germ cells for radial migration and survival to P14, but Dmrt1 expression in germ cells is not required for Sertoli cell polarization.

DISCUSSION

In this study we have used conditional gene targeting to dissect the functions of Dmrt1 in Sertoli cells and germ cells, comparing the phenotypes of animals lacking Dmrt1 function in each of these cell lineages to each other and to null mutant animals lacking Dmrt1 in all cells. This approach has uncovered cell type-autonomous requirements for Dmrt1 during early postnatal testis development in both Sertoli cells and germ cells, as well as a cell non-autonomous requirement for Dmrt1 in Sertoli cells, as summarized in Table 1 and discussed below.

Table 1

Deletion of Dmrt1 in Sertoli cells was particularly informative. We found that Dmrt1 mutant Sertoli cells in SCDmrt1KO animals failed to become closely associated with the tubule periphery. Instead they were distributed throughout the tubules, eventually filling the entire tubule. This closely resembles the Sertoli cell phenotype of Dmrt1 null mutants. Mutant Sertoli cells in SCDmrt1KO animals also failed to up-regulate the androgen receptor and polarize, indicating a developmental block at an immature stage, again very similar to the phenotype of Dmrt1 null mutants. We conclude that these defects are indicative of lineage-autonomous functions for Dmrt1 in Sertoli cells.

Although Sertoli cells in SCDmrt1KO mice initially activated GATA1 expression by P21, suggesting that they had undergone partial maturation, this expression was lost by 14 weeks. We speculate that this might indicate “de-differentiation” of mutant Sertoli cells, which may help explain the disintegration of seminiferous tubules and changes in Sertoli cell morphology that occur in Dmrt1 null mutants between 6 weeks and about four months (Raymond et al., 2000) (DZ, unpublished data). The basement membrane surrounding the semiferous tubules is produced jointly by peritubular myoid (PTM) cells and Sertoli cells (Skinner et al., 1985; Tung and Fritz, 1987). De-differentiation of mutant Sertoli cells in adults may result in failure to maintain the basement membrane and eventual loss of tubule integrity.

Sertoli cell-specific targeting also revealed that Dmrt1 is required in Sertoli cells for germ cell development. In SCDmrt1KO mice, germ cells underwent radial migration and mitotic reactivation, and initiated meiosis, but they failed to complete meiosis and eventually died. This demonstrates that Dmrt1 activity in Sertoli cells is required for proper support of the germ line during meiosis, but apparently not for the events leading up to meiotic initiation. The eventual complete loss of germ cells, including spermatogonia, in SCDmrt1KO mutants suggests that Dmrt1 activity in Sertoli cells is required to establish a functional niche capable of supporting spermatogenesis. Germ cell failure in SCDmrt1KO mutants might result directly from the mutant Sertoli cells failing to produce a specific signal(s), such as a secreted signaling molecule, or from a more general failure of Sertoli cells to functionally associate with and nutritionally support germ cells, or both. Several candidate signaling pathways have been identified by mRNA expression profiling at P1 and P2, which suggested possible disruption in Dmrt1 null mutants of several signaling pathways, including GDNF, FSH, and retinoic acid (Fahrioglu et al., 2007).

Deletion of the androgen receptor (AR) in Sertoli cells (in “SCARKO” mutant mice), like deletion of Dmrt1 in Sertoli cells, has been shown to affect Sertoli cell maturation and germ cell development (Chang et al., 2004; De Gendt et al., 2004; Holdcraft and Braun, 2004; Tan et al., 2005). Thus the delayed activation of AR in Dmrt1 mutant Sertoli cells may partially account for the Sertoli cell and germ cell phenotypes of SCDmrt1KO mutants. However, we observed additional defects in SCDmrt1KO testes, including the failure of Sertoli cells to localize properly and to polarize, and a more severe disruption of germ cell development in SCDmrt1KO mutants versus SCARKO mutants. These phenotypic differences indicate that Dmrt1 cannot function exclusively via the AR pathway.

The germ cell defects in SCDmrt1KO mutants were less severe than in Dmrt1 null mutants, suggesting that Dmrt1 also is required autonomously in germ cells, and we confirmed this interpretation using GCDmrt1KO animals. Dmrt1 mutant germ cells in GCDmrt1KO mice were defective in radial migration and died between P7 and P14, just as in Dmrt1 null mutants. This contrasts with the wild type germ cells in SCDmrt1KO animals, which migrated and were still present at P14. The relatively low targeting efficiency of the TNAP-Cre transgene limited the analysis possible in GCDmrt1KO mutants. However, comparison of germ cells in Dmrt1 null mutants and GCDmrt1KO animals versus SCDmrt1KO animals clearly showed that Dmrt1 is autonomously required in germ cells for radial migration and survival. To our knowledge this is the first demonstration of a genetic requirement in germ cells for postnatal migration.

It is clear from SCDmrt1KO mutants that Dmrt1 activity is required in Sertoli cells for germ cell survival and differentiation, as discussed above. We did not observe any evidence of the converse: that Dmrt1 activity in germ cells is required for Sertoli cell differentiation. This conclusion is somewhat limited by the caveat that tubule sections in GCDmrt1KO mutants with all germ cells negative for Dmrt1 were uncommon, but in such tubules the wild type Sertoli cells were properly organized at the periphery and were polarized.

As mentioned earlier, Dmrt1 is unusual in being expressed in germ cells and Sertoli cells. These cell types are derived from unrelated progenitor populations in different parts of the early embryo and only come together later as a result of long-range migration by the germ cells. Thus it is perhaps unsurprising that most other testis regulators are expressed only in one cell type or the other. It is unknown in which cells the ancestral function of Dmrt1 likely arose, or even whether the expression pattern of Dmrt1 in the mouse is typical, because a variety of Dmrt1 expression patterns have been reported in vertebrates. In humans, fetal DMRT1 mRNA has been described only in Sertoli cells (Moniot et al., 2000), while DMRT1 protein is expressed only in germ cells in the adult testis (Looijenga et al., 2006). Similarly, in a lizard, Dmrt1 mRNA is expressed initially in pre-Sertoli cells and then in germ cells (Sreenivasulu et al., 2002). In fish, Dmrt1 mRNA has been reported to be expressed only in Sertoli cells, or only in germ cells, or in both (Kobayashi et al., 2004; Marchand et al., 2000; Xia et al., 2007). However, most of these studies examined limited developmental stages, so expression in both Sertoli cells and germ cells cannot be excluded in any group of vertebrates. Although it is unclear how common combined Sertoli and germ cell expression of Dmrt1 may be among vertebrates, our results demonstrate that Dmrt1 function is essential in both cell types for testicular differentiation and fertility in the mouse.

In conclusion, the work described here demonstrates that Dmrt1 function is required in Sertoli cells and germ cells for their respective postnatal development, and also has a function in Sertoli cells that is required for germ cell differentiation and survival. These data better define what this essential and highly conserved testicular regulator does in each cell type, which is important for eventually understanding how the remodeling and differentiation of the juvenile testis is controlled.

ACKNOWLEDGEMENTS

We thank members of the Zarkower and Bardwell laboratories for many helpful discussions, Drs. D. Meijer and A. Nagy for Cre transgenic mice, and Drs. H. Tanaka and Y. Nishimune for generous gifts of antibodies. We particularly thank Dr. C. Raymond for assistance in starting this project and many helpful discussions. This work was supported by the NIH (GM59152), the Minnesota Medical Foundation, and the University of Minnesota Developmental Biology Center.

Footnotes

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

Adams IR, McLaren A. Sexually dimorphic development of mouse primordial germ cells: switching from oogenesis to spermatogenesis. Development. 2002;129:1155–64. [PubMed]
Anderson R, Copeland TK, Scholer H, Heasman J, Wylie C. The onset of germ cell migration in the mouse embryo. Mech Dev. 2000;91:61–8. [PubMed]
Bowles J, Knight D, Smith C, Wilhelm D, Richman J, Mamiya S, Yashiro K, Chawengsaksophak K, Wilson MJ, Rossant J, Hamada H, Koopman P. Retinoid signaling determines germ cell fate in mice. Science. 2006;312:596–600. [PubMed]
Bremner WJ, Millar MR, Sharpe RM, Saunders PT. Immunohistochemical localization of androgen receptors in the rat testis: evidence for stage-dependent expression and regulation by androgens. Endocrinology. 1994;135:1227–34. [PubMed]
Chang C, Chen YT, Yeh SD, Xu Q, Wang RS, Guillou F, Lardy H, Yeh S. Infertility with defective spermatogenesis and hypotestosteronemia in male mice lacking the androgen receptor in Sertoli cells. Proc Natl Acad Sci U S A. 2004;101:6876–81. [PubMed]
Crocker M, Coghill SB, Cortinho R. An unbalanced autosomal translocation (7;9) associated with feminization. Clin Genet. 1988;34:70–3. [PubMed]
De Gendt K, Swinnen JV, Saunders PT, Schoonjans L, Dewerchin M, Devos A, Tan K, Atanassova N, Claessens F, Lecureuil C, Heyns W, Carmeliet P, Guillou F, Sharpe RM, Verhoeven G. A Sertoli cell-selective knockout of the androgen receptor causes spermatogenic arrest in meiosis. Proc Natl Acad Sci U S A. 2004;101:1327–32. [PubMed]
de Rooij DG. Proliferation and differentiation of spermatogonial stem cells. Reproduction. 2001;121:347–54. [PubMed]
de Rooij DG, Russell LD. All you wanted to know about spermatogonia but were afraid to ask. J Androl. 2000;21:776–98. [PubMed]
Erdman SE, Burtis KC. The Drosophila doublesex proteins share a novel zinc finger related DNA binding domain. Embo J. 1993;12:527–35. [PubMed]
Fahrioglu U, Murphy MM, Zarkower D, Bardwell VJ. mRNA expression analysis and the molecular basis of neonatal testis defects in Dmrt1 mutant mice. Sexual Development. 2007;1:42–58. [PubMed]
Ginsburg M, Snow MH, McLaren A. Primordial germ cells in the mouse embryo during gastrulation. Development. 1990;110:521–8. [PubMed]
Griswold MD. Interactions between germ cells and Sertoli cells in the testis. Biol Reprod. 1995;52:211–6. [PubMed]
Hess RA, Cooke PS, Hofmann MC, Murphy KM. Mechanistic insights into the regulation of the spermatogonial stem cell niche. Cell Cycle. 2006;5:1164–70. [PMC free article] [PubMed]
Holdcraft RW, Braun RE. Androgen receptor function is required in Sertoli cells for the terminal differentiation of haploid spermatids. Development. 2004;131:459–67. [PubMed]
Ion R, Telvi L, Chaussain JL, Barbet JP, Nunes M, Safar A, Rethore MO, Fellous M, McElreavey K. Failure of testicular development associated with a rearrangement of 9p24.1 proximal to the SNF2 gene. Hum Genet. 1998;102:151–6. [PubMed]
Kettlewell JR, Raymond CS, Zarkower D. Temperature-dependent expression of turtle Dmrt1 prior to sexual differentiation. Genesis. 2000;26:174–8. [PubMed]
Kobayashi T, Matsuda M, Kajiura-Kobayashi H, Suzuki A, Saito N, Nakamoto M, Shibata N, Nagahama Y. Two DM domain genes, DMY and DMRT1, involved in testicular differentiation and development in the medaka, Oryzias latipes. Dev Dyn. 2004;231:518–26. [PubMed]
Koshimizu U, Watanabe D, Sawada K, Nishimune Y. A novel stage-specific differentiation antigen is expressed on mouse testicular germ cells during early meiotic prophase. Biol Reprod. 1993;49:875–84. [PubMed]
Lindeboom F, Gillemans N, Karis A, Jaegle M, Meijer D, Grosveld F, Philipsen S. A tissue-specific knockout reveals that Gata1 is not essential for Sertoli cell function in the mouse. Nucleic Acids Res. 2003;31:5405–12. [PMC free article] [PubMed]
Lomeli H, Ramos-Mejia V, Gertsenstein M, Lobe CG, Nagy A. Targeted insertion of Cre recombinase into the TNAP gene: excision in primordial germ cells. Genesis. 2000;26:116–7. [PubMed]
Looijenga LH, Hersmus R, Gillis AJ, Pfundt R, Stoop HJ, van Gurp RJ, Veltman J, Beverloo HB, van Drunen E, van Kessel AG, Pera RR, Schneider DT, Summersgill B, Shipley J, McIntyre A, van der Spek P, Schoenmakers E, Oosterhuis JW. Genomic and expression profiling of human spermatocytic seminomas: primary spermatocyte as tumorigenic precursor and DMRT1 as candidate chromosome 9 gene. Cancer Res. 2006;66:290–302. [PubMed]
Marchand O, Govoroun M, D'Cotta H, McMeel O, Lareyre J, Bernot A, Laudet V, Guiguen Y. DMRT1 expression during gonadal differentiation and spermatogenesis in the rainbow trout, Oncorhynchus mykiss. Biochim Biophys Acta. 2000;1493:180–7. [PubMed]
McLaren A. Gonad development: assembling the mammalian testis. Curr Biol. 1998;8:R175–7. [PubMed]
McLaren A, Southee D. Entry of mouse embryonic germ cells into meiosis. Dev Biol. 1997;187:107–13. [PubMed]
McLean DJ, Friel PJ, Johnston DS, Griswold MD. Characterization of spermatogonial stem cell maturation and differentiation in neonatal mice. Biol Reprod. 2003;69:2085–91. [PubMed]
Moniot B, Berta P, Scherer G, Sudbeck P, Poulat F. Male specific expression suggests role of DMRT1 in human sex determination. Mech Dev. 2000;91:323–5. [PubMed]
Nagano R, Tabata S, Nakanishi Y, Ohsako S, Kurohmaru M, Hayashi Y. Reproliferation and relocation of mouse male germ cells (gonocytes) during prespermatogenesis. Anat Rec. 2000;258:210–20. [PubMed]
Ogata T, Muroya K, Matsuo N, Hata J, Fukushima Y, Suzuki Y. Impaired male sex development in an infant with molecularly defined partial 9p monosomy: implication for a testis forming gene(s) on 9p. J Med Genet. 1997;34:331–4. [PMC free article] [PubMed]
Ohbo K, Yoshida S, Ohmura M, Ohneda O, Ogawa T, Tsuchiya H, Kuwana T, Kehler J, Abe K, Scholer HR, Suda T. Identification and characterization of stem cells in prepubertal spermatogenesis in mice small star, filled. Dev Biol. 2003;258:209–25. [PubMed]
Orth JM, Qiu J, Jester WF, Jr., Pilder S. Expression of the c-kit gene is critical for migration of neonatal rat gonocytes in vitro. Biol Reprod. 1997;57:676–83. [PubMed]
Raymond CS, Kettlewell JR, Hirsch B, Bardwell VJ, Zarkower D. Expression of Dmrt1 in the genital ridge of mouse and chicken embryos suggests a role in vertebrate sexual development. Dev Biol. 1999a;215:208–20. [PubMed]
Raymond CS, Murphy MW, O'Sullivan MG, Bardwell VJ, Zarkower D. Dmrt1, a gene related to worm and fly sexual regulators, is required for mammalian testis differentiation. Genes Dev. 2000;14:2587–95. [PubMed]
Raymond CS, Parker ED, Kettlewell JR, Brown LG, Page DC, Kusz K, Jaruzelska J, Reinberg Y, Flejter WL, Bardwell VJ, Hirsch B, Zarkower D. A region of human chromosome 9p required for testis development contains two genes related to known sexual regulators. Hum Mol Genet. 1999b;8:989–96. [PubMed]
Raymond CS, Shamu CE, Shen MM, Seifert KJ, Hirsch B, Hodgkin J, Zarkower D. Evidence for evolutionary conservation of sex-determining genes. Nature. 1998;391:691–5. [PubMed]
Sharpe RM, McKinnell C, Kivlin C, Fisher JS. Proliferation and functional maturation of Sertoli cells, and their relevance to disorders of testis function in adulthood. Reproduction. 2003;125:769–84. [PubMed]
Skinner MK, Tung PS, Fritz IB. Cooperativity between Sertoli cells and testicular peritubular cells in the production and deposition of extracellular matrix components. J Cell Biol. 1985;100:1941–7. [PMC free article] [PubMed]
Smith CA, McClive PJ, Western PS, Reed KJ, Sinclair AH. Conservation of a sex-determining gene. Nature. 1999;402:601–2. [PubMed]
Sreenivasulu K, Ganesh S, Raman R. Evolutionarily conserved, DMRT1, encodes alternatively spliced transcripts and shows dimorphic expression during gonadal differentiation in the lizard, Calotes versicolor. Mech Dev. 2002;119(Suppl 1):S55–64. [PubMed]
Tan KA, De Gendt K, Atanassova N, Walker M, Sharpe RM, Saunders PT, Denolet E, Verhoeven G. The role of androgens in sertoli cell proliferation and functional maturation: studies in mice with total or Sertoli cell-selective ablation of the androgen receptor. Endocrinology. 2005;146:2674–83. [PubMed]
Tung PS, Fritz IB. Morphogenetic restructuring and formation of basement membranes by Sertoli cells and testis peritubular cells in co-culture: inhibition of the morphogenetic cascade by cyclic AMP derivatives and by blocking direct cell contact. Dev Biol. 1987;120:139–53. [PubMed]
Vergouwen RP, Jacobs SG, Huiskamp R, Davids JA, de Rooij DG. Proliferative activity of gonocytes, Sertoli cells and interstitial cells during testicular development in mice. J Reprod Fertil. 1991;93:233–43. [PubMed]
Watanabe D, Sawada K, Koshimizu U, Kagawa T, Nishimune Y. Characterization of male meiotic germ cell-specific antigen (Meg 1) by monoclonal antibody TRA 369 in mice. Mol Reprod Dev. 1992;33:307–12. [PubMed]
Wu J, Jester WF, Jr., Orth JM. Short-type PB-cadherin promotes survival of gonocytes and activates JAK-STAT signalling. Dev Biol. 2005;284:437–50. [PubMed]
Xia W, Zhou L, Yao B, Li CJ, Gui JF. Differential and spermatogenic cell-specific expression of DMRT1 during sex reversal in protogynous hermaphroditic groupers. Mol Cell Endocrinol. 2007;263:156–72. [PubMed]
Zarkower D. Establishing sexual dimorphism: conservation amidst diversity? Nat Rev Genet. 2001;2:175–85. [PubMed]