pmc.ncbi.nlm.nih.gov

Retinoic acid regulates sex-specific timing of meiotic initiation in mice

Abstract

In mammals, meiosis is initiated at different time points in males and females, but the mechanism underlying this difference is unknown. Female germ cells begin meiosis during embryogenesis. In males, embryonic germ cells undergo G0/G1 mitotic cell cycle arrest, and meiosis begins after birth. In mice, the Stimulated by Retinoic Acid Gene 8 (Stra8) has been found to be required for the transition into meiosis in both female and male germ cells. Stra8 is expressed in embryonic ovaries just before meiotic initiation, whereas its expression in testes is first detected after birth. Here we examine the mechanism underlying the sex-specific timing of Stra8 expression and meiotic initiation in mice. Our work shows that signaling by retinoic acid (RA), an active derivative of vitamin A, is required for Stra8 expression and thereby meiotic initiation in embryonic ovaries. We also discovered that RA is sufficient to induce Stra8 expression in embryonic testes and in vitamin A-deficient adult testes in vivo. Finally, our results show that cytochrome p450 (CYP)-mediated RA metabolism prevents premature Stra8 expression in embryonic testes. Treatment with an inhibitor specific to RA-metabolizing enzymes indicates that a cytochrome p450 from the 26 family (CYP26) is responsible for delaying Stra8 expression in embryonic testes. Sex-specific regulation of RA signaling thus plays an essential role in meiotic initiation in embryonic ovaries and precludes its occurrence in embryonic testes. Because RA signaling regulates Stra8 expression in both embryonic ovaries and adult testes, this portion of the meiotic initiation pathway may be identical in both sexes.

Keywords: sex determination

Embryonic development of the mammalian gonad is a dynamic process, during which both germ and somatic cells acquire sex-specific characteristics. In mice, differences between the somatic components of ovaries and testes are microscopically evident by embryonic day 12.5 (E12.5). Germ cells, however, remain indistinguishable between the sexes until E13.5, when ovarian germ cells initiate prophase of meiosis and testicular germ cells arrest in the G0/G1 stage of the mitotic cell cycle (1). In this study, we address the question of why ovarian germ cells initiate meiosis embryonically, whereas testicular germ cells postpone meiosis until after birth. In this context, we examine the role of retinoic acid (RA) signaling in meiotic initiation.

Our studies are based on the recent finding that Stimulated by Retinoic Acid Gene 8 (Stra8) is required for meiotic initiation in both sexes (A. E. Baltus, D.B.M., Y. C. Hu, M. L. Goodheart, A. E. Carpenter, D. G. de Rooij, and D.C.P., unpublished work). Stra8 expression in ovarian germ cells occurs over a period of 4 days starting at E12.5, just 1 day before meiotic germ cells with characteristically condensed chromatin can be observed (2). Stra8 expression in the ovary occurs in an anterior-to-posterior wave, which is followed by anterior-to-posterior waves of expression of meiotic markers such as Dmc1 (a meiotic recombinase) and Scp3 (a synaptonemal complex protein) (2–4). In contrast, embryonic testes do not express Stra8 (2). Instead, Stra8 is first expressed in the mitotic germ cells of postnatal testes and subsequently in the premeiotic germ cells (spermatogonia) of adult testes (2, 5).

Stra8 expression is stimulated by RA in embryonal carcinoma cells and embryonic stem cells in culture (5). We wondered whether RA also regulates endogenous Stra8 expression in the germ cells of embryonic ovaries and postnatal testes. RA is an important regulator of embryonic patterning and development (6). It is generated by a series of oxidative reactions from dietary vitamin A [retinol (ROL)] (6). Local levels of RA are regulated by retinaldehyde dehydrogenases, which catalyze the last step of RA synthesis and by the cytochrome p450, 26 family of enzymes (CYP26), which degrade RA (7–9). RA serves as a ligand to a family of nuclear hormone receptors known as RA receptors (RARs), which bind to RA response elements (RAREs) in the regulatory regions of target genes (10). Three major RAR isotypes have been identified (RAR-α, -β, and -γ), and these exhibit overlapping expression patterns and functional redundancy in many tissues (10, 11).

RARs are expressed in the gonads of both sexes (12–14), and several lines of evidence suggest that RA may play a role in the regulation of meiosis in both postnatal testes and embryonic ovaries. First, spermatogenesis is blocked in males fed a vitamin A-deficient (VAD) diet (15). Their testicular tubules are depleted of all meiotic cells, and the only remaining germ cells are undifferentiated spermatogonia and preleptotene spermatocytes (16, 17). Because administration of RA to VAD animals is sufficient to restore and synchronize spermatogenesis in all seminiferous tubules, it has been suggested that RA is necessary for proper meiotic prophase (17–19). Second, RA treatment of embryonic rat ovaries in culture hastens meiotic prophase (20). Here, we examine the role of RA in the regulation of Stra8, a gene necessary for meiotic initiation, and in this way explore the mechanism behind the sex-specific timing of meiotic initiation.

Results

Stra8 Expression in Embryonic Ovaries Requires RAR Signaling.

To determine whether RA is required for Stra8 expression in embryonic ovaries, we assayed the gene’s expression in gonads cultured in the presence of RAR panantagonist BMS-204493. This compound, which antagonizes all three RAR isotypes, prevents RA signaling by stabilizing interactions between RARs and corepressor proteins, thereby silencing expression of genes normally up-regulated by RA (21). Ovaries dissected at E11.5 and cultured for 2 days in control medium expressed Stra8, whereas testes did not (Fig. 1A), recapitulating the endogenous expression pattern. In contrast, ovaries cultured in the presence of RAR antagonist BMS-204493 did not express Stra8 (Fig. 1B), although expression of a control germ cell marker, Oct4, was maintained (Fig. 1C). We conclude that RAR signaling is required for Stra8 expression in embryonic ovaries.

Fig. 1.

Fig. 1.

RAR signaling is required for Stra8 expression in embryonic ovaries. (A and B) Whole-mount in situ hybridization with Stra8 probe in gonads dissected at E11.5 and cultured for 2 days (A) in control medium or (B) with RAR panantagonist BMS-204493. (C) Whole-mount in situ hybridization with Oct4 probe in gonads dissected at E11.5 and cultured for 2 days with BMS-204493. (D–F) Photomicrographs of hematoxylin/eosin-stained sections of (D) wild-type ovary dissected at E11.5 and cultured for 4 days in control medium or (E and F) Stra8−/− ovaries cultured (E) in control medium or (F) with RA. The long arrow indicates representative cell with meiotic prophase condensation (D). We observed a few pyknotic nuclei in all samples (D–F, short arrows).

These results, taken together with the observation that Stra8 is required for meiotic initiation in embryonic ovaries in vivo (A. E. Baltus, D.B.M., Y. C. Hu, M. L. Goodheart, A. E. Carpenter, D. G. de Rooij, and D.C.P., unpublished work), imply that meiotic initiation there requires RAR signaling. To confirm that RA acts through Stra8 to induce meiotic prophase condensation in embryonic ovaries, we cultured Stra8−/− ovaries with and without RA. As expected, germ cells in wild-type ovaries dissected at E11.5 and cultured for 4 days in control medium displayed chromatin condensation (Fig. 1D), indicative of early meiotic prophase (20, 22). No such condensation was observed in Stra8−/− ovaries cultured in control medium (Fig. 1E) or with RA (Fig. 1F). Thus, RA failed to rescue the loss of meiotic prophase condensation in Stra8−/− ovaries. We conclude that RA induction of Stra8 expression in ovarian germ cells is essential to achieve the chromatin condensation of meiotic prophase.

Exogenous RA Induces Stra8 Expression in Embryonic Testes.

Testicular germ cells do not express Stra8 until after birth (2), despite the fact that testicular cells express RARs during embryonic development (13). Perhaps the germ cells of embryonic testes are not exposed to RA in vivo. To determine whether exogenous RA can induce Stra8 expression in embryonic testes, we assayed the gene’s expression in gonads cultured with RA added to the medium. As expected, testes dissected at E12.5 and cultured for 2 days in control medium expressed no Stra8 (Fig. 2A). By contrast, testes cultured with RA displayed abundant expression of Stra8 (Fig. 2B).

Fig. 2.

Fig. 2.

RA and agonists to all three RARs induce Stra8 expression in testes. Whole-mount in situ hybridization with Stra8 probe in gonads dissected at E12.5 and cultured for 2 days (A) in control medium, (B) with RA, (C) with RA after busulfan treatment, (D) with RAR-α agonist BMS-194753, (E) with RAR-β agonist BMS-213309, or (F) with RAR-γ agonist BMS-270394.

Although this RA-induced expression of Stra8 appeared to be limited to embryonic testis cords, we wanted to determine whether it was restricted to germ cells, where Stra8 is normally expressed in adult testes and embryonic ovaries. Embryonic gonads were depleted of germ cells by busulfan treatment (23) and then cultured with RA added to the medium. Stra8 expression was lost in both embryonic testes and ovaries treated in this manner (Fig. 2C). The simplest interpretation of these findings is that RA signaling induces Stra8 expression only in the germ cell lineage. (Although less likely, we cannot exclude the possibility that, in this experimental context, Stra8 is expressed in somatic cells, but only in the presence of germ cells.) Similar experiments were performed with testes from Wv/Wv embryos, which are severely depleted of germ cells due to a point mutation in the c-kit tyrosine kinase receptor (24). Like busulfan-treated testes, testes from Wv/Wv embryos failed to express Stra8 when cultured with RA (data not shown). Again, the simplest interpretation of these findings is that exogenous RA induces Stra8 expression in the germ cells but not in the somatic cells of embryonic testes.

In rats, all three RAR isotypes are known to be expressed in embryonic testes (13, 25), and, in theory, any one might mediate the premature induction of Stra8 observed in mouse embryonic testes treated with exogenous RA. To address this question empirically, we cultured embryonic gonads in the presence of RAR agonists selective for each of the three RAR isotypes: BMS-194753 (RAR-α), BMS-213309 (RAR-β), and BMS-270394 (RAR-γ). Treatment with any of these selective RAR agonists induced Stra8 expression in embryonic testes (Fig. 2 D–F). These results suggest that, in embryonic testes, all three RAR isotypes are capable of activating Stra8 transcription.

RA Induces Stra8 Expression in VAD Adult Testes.

Does RA also regulate Stra8 expression in the premeiotic cells of adult testes? If so, we might expect Stra8 expression to be low in the testes of VAD male mice but to increase after injection of such mice with ROL (vitamin A), the precursor of RA. To test this prediction, we removed a single testis from each of several adult VAD males, after which each of the mice was injected with all-trans ROL in oil or with oil alone. The second testis from each animal was harvested 24 h later, and Stra8 transcript levels, pre- and postinjection, were compared by quantitative RT-PCR. Stra8 expression increased dramatically in the testes of VAD males after injection of ROL but not after injection of oil alone (Fig. 3). This result demonstrates that RA signaling can induce Stra8 expression in adult testes in vivo and thus represents a shared element of the regulatory pathway required for meiotic initiation in adult testes and embryonic ovaries.

Fig. 3.

Fig. 3.

Quantitative RT-PCR analysis of Stra8 expression in ROL-injected or control (oil-injected) adult VAD testes compared to preinjection, contralateral testes. (Error bars indicate standard deviation; P < 0.022, paired t test, two sided.)

CYP-Mediated Metabolism of RA Prevents Stra8 Expression in Embryonic Testes.

The absence of Stra8 expression in embryonic testes in vivo implies that RA signaling there differs from that in embryonic ovaries. In theory, embryonic testes might synthesize less RA, metabolize RA more efficiently, or both. Lower levels of RA synthesis in embryonic testes have not been reported. However, several CYP enzymes, including some that may degrade RA, are known to be expressed in embryonic testes but not embryonic ovaries (26–28). One of these CYP enzymes could metabolize RA and thereby shield testicular germ cells from RA signaling.

To determine whether a testis-specific CYP enzyme could be responsible for the lack of Stra8 expression in embryonic testes, we assayed Stra8 expression in gonads cultured in the presence of ketoconazole, a potent but nonspecific CYP inhibitor (29, 30). As expected, testes dissected at E12.5 and cultured for 2 days in control medium did not express Stra8 (Fig. 4A). However, embryonic testes cultured in the presence of ketoconazole displayed robust expression of Stra8 (Fig. 4B). Ketoconazole had no detectable effect on embryonic ovaries. Busulfan-treated testes failed to express Stra8 after treatment with ketoconazole (Fig. 4C), suggesting that ketoconazole induced Stra8 expression only in germ cells.

Fig. 4.

Fig. 4.

CYP26 activity prevents Stra8 expression in embryonic testes. (A–D and F–I) Whole-mount in situ hybridization with Stra8 probe in gonads dissected at E12.5 and cultured for 2 days in control medium (A and F), with CYP inhibitor ketoconazole (B), with ketoconazole after busulfan treatment (C), with RAR panantagonist BMS-204493 and ketoconazole (D), with CYP26 inhibitor R115866 (G), with R115866 after busulfan treatment (I), or with BMS-204493 and R115866. (E and J) In situ hybridization with Oct4 probe in gonads dissected at E12.5 and cultured for 2 days with BMS-204493 and ketoconazole (E) or with BMS-204493 and R115866 (J).

Given the diversity of CYP enzymes inhibited by ketoconazole, it was important to confirm that the compound’s effects on Stra8 expression required RAR activation. Accordingly, we dissected testes at E12.5 and cultured them for 2 days in the presence of both ketoconazole and RAR panantagonist BMS-204493. As expected, the RAR antagonist blocked the effect of ketoconazole in embryonic testes (Fig. 4D). This dual treatment was not toxic to germ cells, as judged by expression of the germ cell marker Oct4 (Fig. 4E). Taken together, the experiments involving ketoconazole strongly suggest that, in embryonic testes, germ-cell transcription of Stra8 is prevented by a CYP enzyme that metabolizes RA.

Implicating the CYP26 Family in RA Degradation in Embryonic Testes.

Several CYP enzymes are capable of metabolizing RA, but enzymes of the CYP26 family are particularly specific and efficient (31). Although ketoconazole is a nonspecific inhibitor of CYP enzymes, the compound R115866 is a highly selective inhibitor of CYP26-mediated metabolism of RA in vivo (32). To test whether CYP26-mediated RA metabolism in embryonic testes prevents Stra8 expression, we dissected embryonic gonads at E12.5 and cultured them for 2 days in the presence of R115866. This CYP26 inhibitor induced Stra8 expression in embryonic testes but had no detectable effect on embryonic ovaries (Fig. 4G) when compared with controls (Fig. 4F). As expected, R115866 failed to induce Stra8 expression in testes from busulfan-treated and Wv/Wv embryos (Fig. 4H; data not shown), suggesting that only germ cells express Stra8 in response to inhibition of CYP26. Embryonic testes cultured in the presence of both the CYP26 inhibitor and the RAR panantagonist failed to express Stra8 (Fig. 4I), confirming that RARs mediate the effects of the CYP26 inhibitor. This dual treatment did not affect the control germ cell marker Oct4 (Fig. 4J). Based on all these findings, we conclude that a CYP26 enzyme degrades RA in embryonic testes and thereby precludes germ-cell transcription of Stra8.

Cyp26b1 Is Expressed in Embryonic Testes but Not in Embryonic Ovaries.

In mice, the Cyp26 gene family has three members, only one of which, Cyp26b1, is expressed in embryonic gonads (26, 33, 34). The Cyp26b1 gene was shown previously to be expressed in somatic cells of embryonic testes (26). We confirmed and extended these findings by whole-mount in situ hybridization of Cyp26b1 transcripts in embryonic gonads. Cyp26b1 expression in testes begins by E11.5 and is maintained until at least E15.5 (Fig. 5). No expression of Cyp26b1 in ovaries was seen during the same period (Fig. 5). Based on Cyp26b1’s unique expression pattern and our finding that inhibiting CYP26 activity induces Stra8 expression in embryonic testes, we hypothesize that CYP26B1 is the RA-metabolizing enzyme responsible for delaying Stra8 expression in embryonic testicular germ cells.

Fig. 5.

Fig. 5.

Developmental time course of Cyp26b1 expression in embryonic gonads as revealed by whole-mount in situ hybridization.

When a Y chromosome derived from Zalende mice is crossed onto a C57BL/6 genetic background, many of the resulting XYZAL embryos are partially sex-reversed, with ovotestes (2, 35, 36). An XYZAL ovotestis is typically comprised of a central region exhibiting testicular histology and two polar regions displaying ovarian histology (35, 36). Whole-mount in situ hybridization to XYZAL ovotestes revealed mutually exclusive realms of expression for Stra8 and Cyp26b1 (Fig. 6). Stra8 expression is confined to the ovarian (polar) portions of the ovotestis, whereas Cyp26b1 expression is limited to the testicular (central) region. Thus, Stra8 and Cyp26b1 appear to be expressed in nonoverlapping domains, consistent with the hypothesis that Cyp26b1 activity prevents transcription of Stra8.

Fig. 6.

Fig. 6.

Stra8 and Cyp26b1 expressed in nonoverlapping domains in E14.5 XYZAL ovotestes. Stra8 is expressed in the polar ovarian regions. Cyp26b1 is expressed in the central testicular region.

Discussion

Based on the findings reported here, we propose a model in which RA signaling and metabolism regulate whether female and male germ cells initiate meiosis during embryogenesis (Fig. 7). The model posits that, in embryonic ovaries, RA induces germ cells to express Stra8, which in turn leads to initiation of meiosis. In embryonic testes, an enzyme of the CYP26 family, likely CYP26B1, degrades RA and thereby prevents expression of Stra8 and precludes initiation of meiosis. According to our model, and as discussed below, both production and degradation of RA occur outside germ cells, in the somatic cells and tissues of the embryo. Thus, we propose that germ cells respond to sex differences in the RA environments offered by embryonic ovaries and testes.

Fig. 7.

Fig. 7.

A proposed model for regulation of meiotic initiation in embryonic gonads.

Further, our findings in VAD mice suggest that meiotic initiation in adult testes may be regulated by a pathway similar to that operating in embryonic ovaries. In testes of VAD adult males, Stra8 is significantly up-regulated in response to ROL, a precursor to RA (Fig. 3). Thus, at least a portion of the meiotic initiation pathway appears to be identical in the adult testis and embryonic ovary. The spermatogenic arrest observed in VAD adult males may be due in part to the absence of Stra8 expression. Conversely, testis-wide induction of Stra8 by RA may contribute to the spermatogenic synchronization observed in VAD animals after administration of ROL (17).

In the 1970s, Byskov and Saxen (37) suggested that a somatically generated “meiosis inducing substance” is present in the embryonic ovary, and that it is required for germ cells there to initiate meiotic prophase. However, other investigators noted that ectopic germ cells localized not in ovaries but in the adrenal glands, in the mesonephros, or even in embryonic lung aggregates, also initiated meiosis (1, 22, 38). These observations led investigators to argue that, if a meiosis-inducing substance existed, it must be present in all of these somatic contexts (22, 38). Alternatively, they suggested, germ cells in embryonic ovaries may initiate meiosis autonomously, without reference to somatic cues. Our findings suggest that RA is the somatically generated meiosis-inducing substance. RA is known to be present in many embryonic tissues, including adrenal gland (39, 40), mesonephros (41), and lung (42). Thus, our proposal that RA is the meiosis-inducing substance readily accounts for the ability of embryonic germ cells to initiate meiosis in diverse extragonadal sites.

Several findings lead us to suggest that RA, although somatically produced, acts directly on embryonic germ cells to induce expression of Stra8. As shown by other investigators, ectopic germ cells in diverse extragonadal sites initiate meiosis in the absence, or near absence, of gonadal somatic cells (22, 38). The simplest interpretation of these observations, together with our present findings, is that RA acts via RARs located in germ cells rather than signaling indirectly via gonadal somatic cells. In embryonic ovaries, RARs are readily detected in germ cells but are expressed at very low levels, if at all, in somatic cells (12). Indeed, the promoter of the Stra8 gene contains two putative RAREs, raising the possibility that RA up-regulates Stra8 transcription by binding to RARs directly engaged at the Stra8 promoter.

We do not yet know the tissue source of the RA that (i) induces Stra8 expression in embryonic ovaries and (ii) is eliminated by CYP26 activity in embryonic testes. Although the gonads themselves could synthesize RA, at least two possible extragonadal sources merit consideration. First, the embryonic adrenal gland, located just anterior to the gonad, is a site of robust RA synthesis during embryogenesis (39, 40). Could diffusion of RA from the embryonic adrenal gland account for the anterior-to-posterior wave of Stra8 expression observed in the developing ovary (2)? Second, by studying mice carrying a lacZ transgene controlled by RAREs (43), we detected RA in the mesonephroi of both sexes at E11.5 and E14.5 (Fig. 8, which is published as supporting information on the PNAS web site), corroborating other investigators’ findings (41). (Our assay did not reveal RA in the gonads of either sex at these time points, suggesting that RA concentrations are low in embryonic gonads or, perhaps more likely given our findings, that RARE-lacZ detection of RA is ineffective there.)

We discovered that germ cells in cultured embryonic testes will express Stra8 if exposed to RA, to RAR agonists, or to inhibitors of CYP26 activity (Figs. 2 and 4). An important but unresolved question is whether such treatments of embryonic testes will induce premature meiotic prophase. For example, in embryonic testes cultured in the presence of CYP26 inhibitor R115866, we observed some degree of chromatin condensation in a few cells (data not shown), but this was typically accompanied by and might potentially be confused with apoptosis. Further experiments will be required to determine whether inhibition of CYP26 activity induced bona fide meiotic prophase.

McLaren and Southee’s results (22) suggested that the somatic cells of the embryonic testis inhibit germ cells from entering meiotic prophase. Our findings substantiate this hypothesis and identify CYP26-catalyzed degradation of RA as a central mechanism of meiotic inhibition. McLaren and Southee (22) have noted that the embryonic testes’ ability to inhibit meiosis is lost if the architectural integrity of the testis cords, rings of somatic cells that enclose the germ cells, is mechanically disrupted. This observation accords with published data as to the site of Cyp26b1 expression in testes. In early postnatal and adult testes, Cyp26b1 expression is confined to the myoid cells that encircle the seminiferous tubules (14). Similarly, in embryonic testes, Cyp26b1 is expressed in cells outside the developing testis cords (26). Taken together, these observations suggest the possibility that CYP26B1-expressing cells surrounding the testis cords form a catabolic barrier that prevents RA generated outside the cords from reaching the germ cells within. A similar barrier has been described in the adult testis (14).

How does our model (Fig. 7) relate to prevailing understandings of mammalian sex determination, specifically with respect to the role of Sry, the sex-determining gene on the Y chromosome? Viewed from the perspective of our present studies and model, Sry must function to ensure that cells surrounding the testis cords express Cyp26b1. Sry has been shown to function in the supporting cell lineage, which gives rise to the intratubular Sertoli cells (44, 45). Thus, we surmise that Sry’s role in regulating Cyp26b1 expression is indirect, executed via signaling between somatic cells of different types.

Materials and Methods

Mice.

CD1 random-bred mice (Charles River Laboratories) were used in all embryonic gonad culture experiments, except where mutant strains are specified. Stra8−/− embryos were generated by mating heterozygotes (A. E. Baltus, D.B.M., Y. C. Hu, M. L. Goodheart, A. E. Carpenter, D. G. de Rooij, and D.C.P., unpublished work). XYZAL embryos with ovotestes were obtained by crossing Zalende/Ei males (The Jackson Laboratory) with C57BL/6 females to generate F1 hybrids. F1 males were then backcrossed to C57BL/6 females to create N2 XYZAL embryos with feminized gonads, as reported (2, 36). Embryos carrying a RARE-lacZ transgene were obtained from matings of RARE-lacZ−/+ males (43) with CD-1 females. Wv/Wv (The Jackson Laboratory) embryos were obtained by mating heterozygotes.

Embryo Collection and Sexing.

To establish timed matings, female mice were housed with male mice overnight. Noon of the day when a vaginal plug was evident was considered E0.5. E11.5 embryos were sexed by isolating DNA and using a multiplex PCR reaction involving two primer pairs. One pair of primers amplifies a 245-bp fragment of myogenin as an internal positive control (primer sequences: 5′-TTACGTCCATCGTGGACAGCAT-3′ and 5′-TGGGCTGGGTGTTAGTCTTAT-3′). The second pair of primers amplifies a 342-bp segment of a Y-chromosome repeat sequence (5′-CAGTTACCAATCAACACATCAC-3′ and 5′-CTGGAGCTCTACAGTGATGA-3′).

Germ Cell Depletion.

At 9.5 days postcoitum, pregnant females were injected i.p. with 200 μl of 8 mg/ml busulfan (Sigma) in 50% dimethyl sulfoxide (23). Embryos from these busulfan-treated mothers were dissected at E11.5 or E12.5.

Embryonic Gonad Cultures and Treatments.

Pregnant mice were killed by cervical dislocation, and embryonic gonads were dissected out in PBS solution. Gonads were cultured atop 1.5- to 1.7-ml agar blocks, as described (46). The following compounds were dissolved in ethanol and stored in aliquots at −80°C before use: all-trans RA (Sigma); RAR panantagonist BMS-204493; and selective RAR agonists BMS-194753, BMS-213309, and BMS-270394 (all gifts from Bristol-Myers Squibb). Ketoconazole (Sigma) and CYP26 inhibitor R115866 (a gift from Johnson and Johnson, New Brunswick, NJ) were dissolved in dimethyl sulfoxide and stored at room temperature before use. All compounds were added to the culture media surrounding the agar blocks at least 1 hour before freshly dissected gonads were deposited on the blocks. The concentrations of compounds in culture media were as follows: BMS-204493, 5 μM; all-trans RA, 0.7 μM; BMS-194753, 0.5 μM; BMS-213309, 0.5 μM; BMS-270394, 0.5 μM; ketoconazole, 0.7 μM; and R115866, 0.7 μM. Control cultures were treated with ethanol and/or dimethyl sulfoxide, as appropriate.

Histology and in Situ Hybridization.

Embryonic gonads for histological examination were dissected in PBS solution and cultured as described above. Gonads were then fixed in Bouin’s solution at 4°C overnight, sectioned, and stained with hematoxylin/eosin.

For in situ hybridizations, embryonic gonads were dissected in PBS solution and fixed in 4% paraformaldehyde at 4°C overnight. Whole-mount in situ hybridizations were performed as reported (47). Digoxigenin-labeled RNA probes for Stra8 and Oct4 (2) were detected by using an alkaline phosphatase-conjugated antidigoxigenin antibody and staining with BM purple alkaline phosphatase substrate (Roche Applied Science, Indianapolis).

Analysis of Stra8 Expression in VAD Testes.

Adult female mice (129/SvJ, The Jackson Laboratory) were maintained on a VAD diet (Harlan Teklad, Indianapolis) for at least 2 weeks before mating and throughout pregnancy. Their male offspring were fed a VAD diet for 13–14 weeks. One testis was removed from each animal and cut into two pieces. One half was fixed in Bouin’s solution for histological assessment of spermatogenesis. The other half was placed in TRIzol (Invitrogen) for RNA extraction and served as a preinjection control in RT-PCR analysis. Incisions were sealed and the animals allowed to recover for 24 h. Five animals with similarly deficient spermatogenesis (as judged by preinjection testicular histology) were selected for subsequent analysis. Three (experimental) animals were injected s.c. with 100 μl of 15 mg/ml all-trans ROL acetate (Sigma) in 10% ethanol and sesame oil. Two (control) animals were injected with 100 μl of 10% ethanol and sesame oil. The experimental and control animals’ remaining testes were harvested 24 h after injection and placed in TRIzol for RNA extraction. Quantitative RT-PCR analysis was performed on a Prism 7000 System (Applied Biosystems) by using SYBR-green labeling. Each RNA sample was analyzed, in triplicate with Stra8 primers that amplify a 151-bp product (primers: 5′-GTTTCCTGCGTGTTCCACAAG-3′ and 5′-CACCCGAGGCTCAAGCTTC-3′) and control Rps2 primers that amplify a 112-bp product (primers: 5′-CTGACTCCCGACCTCTGGAAA-3′ and 5′-GAGCCTGGGTCCTCTGAACA-3′). Stra8 expression was normalized to Rps2 expression and compared in pre- and postinjection testes from each mouse.

Supplementary Material

Supporting Figure

Acknowledgments

We thank Janet Rossant (Mt. Sinai Hospital, Toronto) and Wellington Cardoso (Boston University, Boston) for RARE-lacZ reporter mice; Winston Bellott, Yueh-Chiang Hu, and Steve Rozen for help with statistical analysis; and Erika Anderson, Andrew Baltus, Leo DiNapoli, Mark Gill, Terry Hassold, Patricia Hunt, Julian Lange, Yanfeng Lin, Jacob Mueller, and Steve Rozen for critical reading of the manuscript. We thank Chris Zusi (Bristol-Myers Squibb) for helpful suggestions, Bristol-Myers Squibb for RAR antagonists and agonists, and Johnson & Johnson for R115866. This research was supported by the Howard Hughes Medical Institute and National Institute of Child Health and Human Development Grants HD39963 (to B.C.) and HD10808 (to M.D.G.).

Abbreviations

En

embryonic day n

RA

retinoic acid

RAR

RA receptor

RARE

RA response element

CYP

cytochrome p450

ROL

retinol

VAD

vitamin A-deficient

CYP26

cytochrome p450 from the 26 family of enzymes.

A preliminary report of these findings was presented at the Germ Cells meeting, October 2004, Cold Spring Harbor, NY.

Conflict of interest statement: No conflicts declared.

References

  • 1.McLaren A. Symp. Soc. Exp. Biol. 1984;38:7–23. [PubMed] [Google Scholar]
  • 2.Menke D. B., Koubova J., Page D. C. Dev. Biol. 2003;262:303–312. doi: 10.1016/s0012-1606(03)00391-9. [DOI] [PubMed] [Google Scholar]
  • 3.Yao H. H., DiNapoli L., Capel B. Development (Cambridge, U.K.) 2003;130:5895–5902. doi: 10.1242/dev.00836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bullejos M., Koopman P. Mol. Reprod. Dev. 2004;68:422–428. doi: 10.1002/mrd.20105. [DOI] [PubMed] [Google Scholar]
  • 5.Oulad-Abdelghani M., Bouillet P., Decimo D., Gansmuller A., Heyberger S., Dolle P., Bronner S., Lutz Y., Chambon P. J. Cell Biol. 1996;135:469–477. doi: 10.1083/jcb.135.2.469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Ross S. A., McCaffery P. J., Drager U. C., De Luca L. M. Physiol. Rev. 2000;80:1021–1054. doi: 10.1152/physrev.2000.80.3.1021. [DOI] [PubMed] [Google Scholar]
  • 7.Duester G., Mic F. A., Molotkov A. Chem. Biol. Interact. 2003;143–144:201–210. doi: 10.1016/s0009-2797(02)00204-1. [DOI] [PubMed] [Google Scholar]
  • 8.Abu-Abed S., Dolle P., Metzger D., Beckett B., Chambon P., Petkovich M. Genes Dev. 2001;15:226–240. doi: 10.1101/gad.855001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Niederreither K., Abu-Abed S., Schuhbaur B., Petkovich M., Chambon P., Dolle P. Nat. Genet. 2002;31:84–88. doi: 10.1038/ng876. [DOI] [PubMed] [Google Scholar]
  • 10.Chambon P. FASEB J. 1996;10:940–954. [PubMed] [Google Scholar]
  • 11.Kastner P., Mark M., Chambon P. Cell. 1995;83:859–869. doi: 10.1016/0092-8674(95)90202-3. [DOI] [PubMed] [Google Scholar]
  • 12.Morita Y., Tilly J. L. Endocrinology. 1999;140:2696–2703. doi: 10.1210/endo.140.6.6826. [DOI] [PubMed] [Google Scholar]
  • 13.Dolle P., Ruberte E., Leroy P., Morriss-Kay G., Chambon P. Development (Cambridge, U.K.) 1990;110:1133–1151. doi: 10.1242/dev.110.4.1133. [DOI] [PubMed] [Google Scholar]
  • 14.Vernet N., Dennefeld C., Rochette-Egly C., Oulad-Abdelghani M., Chambon P., Ghyselinck N. B., Mark M. Endocrinology. 2005;147:96–110. doi: 10.1210/en.2005-0953. [DOI] [PubMed] [Google Scholar]
  • 15.Wolbach S. B., Howe P. R. J. Exp. Med. 1925;42:753–777. doi: 10.1084/jem.42.6.753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Thompson J. N., Howell J. M., Pitt G. A. Proc. R. Soc. London Ser. B; 1964. pp. 510–535. [DOI] [PubMed] [Google Scholar]
  • 17.Morales C., Griswold M. D. Endocrinology. 1987;121:432–434. doi: 10.1210/endo-121-1-432. [DOI] [PubMed] [Google Scholar]
  • 18.van Pelt A. M., de Rooij D. G. Endocrinology. 1991;128:697–704. doi: 10.1210/endo-128-2-697. [DOI] [PubMed] [Google Scholar]
  • 19.de Rooij D.G., van Pelt A. M. M., van de Kant H. J. G., van der Saag P. T., Peters A. H. F. M., Heyting C., Boer P. In: Function of Somatic Cells in the Testis. Bartke A., editor. New York: Springer; 1994. pp. 345–361. [Google Scholar]
  • 20.Livera G., Rouiller-Fabre V., Valla J., Habert R. Mol. Cell. Endocrinol. 2000;165:225–231. doi: 10.1016/s0303-7207(00)00271-9. [DOI] [PubMed] [Google Scholar]
  • 21.Germain P., Iyer J., Zechel C., Gronemeyer H. Nature. 2002;415:187–192. doi: 10.1038/415187a. [DOI] [PubMed] [Google Scholar]
  • 22.McLaren A., Southee D. Dev. Biol. 1997;187:107–113. doi: 10.1006/dbio.1997.8584. [DOI] [PubMed] [Google Scholar]
  • 23.Merchant H. Dev. Biol. 1975;44:1–21. doi: 10.1016/0012-1606(75)90372-3. [DOI] [PubMed] [Google Scholar]
  • 24.Nocka K., Tan J. C., Chiu E., Chu T. Y., Ray P., Traktman P., Besmer P. EMBO J. 1990;9:1805–1813. doi: 10.1002/j.1460-2075.1990.tb08305.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cupp A. S., Dufour J. M., Kim G., Skinner M. K., Kim K. H. Endocrinology. 1999;140:2343–2352. doi: 10.1210/endo.140.5.6720. [DOI] [PubMed] [Google Scholar]
  • 26.Abu-Abed S., MacLean G., Fraulob V., Chambon P., Petkovich M., Dolle P. Mech. Dev. 2002;110:173–177. doi: 10.1016/s0925-4773(01)00572-x. [DOI] [PubMed] [Google Scholar]
  • 27.Menke D. B., Page D. C. Gene Expr. Patterns. 2002;2:359–367. doi: 10.1016/s1567-133x(02)00022-4. [DOI] [PubMed] [Google Scholar]
  • 28.Nef S., Schaad O., Stallings N. R., Cederroth C. R., Pitetti J. L., Schaer G., Malki S., Dubois-Dauphin M., Boizet-Bonhoure B., Descombes P., et al. Dev. Biol. 2005;287:361–377. doi: 10.1016/j.ydbio.2005.09.008. [DOI] [PubMed] [Google Scholar]
  • 29.Higashi Y., Omura M., Suzuki K., Inano H., Oshima H. Endocrinol. Jpn. 1987;34:105–115. doi: 10.1507/endocrj1954.34.105. [DOI] [PubMed] [Google Scholar]
  • 30.Van Wauwe J. P., Coene M. C., Goossens J., Van Nijen G., Cools W., Lauwers W. J. Pharmacol. Exp. Ther. 1988;245:718–722. [PubMed] [Google Scholar]
  • 31.White J. A., Beckett-Jones B., Guo Y. D., Dilworth F. J., Bonasoro J., Jones G., Petkovich M. J. Biol. Chem. 1997;272:18538–18541. doi: 10.1074/jbc.272.30.18538. [DOI] [PubMed] [Google Scholar]
  • 32.Stoppie P., Borgers M., Borghgraef P., Dillen L., Goossens J., Sanz G., Szel H., Van Hove C., Van Nyen G., Nobels G., et al. J. Pharmacol. Exp. Ther. 2000;293:304–312. [PubMed] [Google Scholar]
  • 33.Tahayato A., Dolle P., Petkovich M. Gene Expr. Patterns. 2003;3:449–454. doi: 10.1016/s1567-133x(03)00066-8. [DOI] [PubMed] [Google Scholar]
  • 34.MacLean G., Abu-Abed S., Dolle P., Tahayato A., Chambon P., Petkovich M. Mech. Dev. 2001;107:195–201. doi: 10.1016/s0925-4773(01)00463-4. [DOI] [PubMed] [Google Scholar]
  • 35.Eicher E. M., Washburn L. L., Whitney J. B., 3rd, Morrow K. E. Science. 1982;217:535–537. doi: 10.1126/science.7089579. [DOI] [PubMed] [Google Scholar]
  • 36.Eicher E. M., Washburn L. L. J. Exp. Zool. 1983;228:297–304. doi: 10.1002/jez.1402280213. [DOI] [PubMed] [Google Scholar]
  • 37.Byskov A. G., Saxen L. Dev. Biol. 1976;52:193–200. doi: 10.1016/0012-1606(76)90239-6. [DOI] [PubMed] [Google Scholar]
  • 38.Zamboni L., Upadhyay S. J. Exp. Zool. 1983;228:173–193. doi: 10.1002/jez.1402280204. [DOI] [PubMed] [Google Scholar]
  • 39.Haselbeck R. J., Ang H. L., Deltour L., Duester G. Endocrinology. 1997;138:3035–3041. doi: 10.1210/endo.138.7.5274. [DOI] [PubMed] [Google Scholar]
  • 40.Haselbeck R. J., Duester G. Dev. Dyn. 1998;213:114–120. doi: 10.1002/(SICI)1097-0177(199809)213:1<114::AID-AJA11>3.0.CO;2-2. [DOI] [PubMed] [Google Scholar]
  • 41.Shen S., van den Brink C. E., Kruijer W., van der Saag P. T. Int. J. Dev. Biol. 1992;36:465–476. [PubMed] [Google Scholar]
  • 42.Malpel S., Mendelsohn C., Cardoso W. V. Development (Cambridge, U.K.) 2000;127:3057–3067. doi: 10.1242/dev.127.14.3057. [DOI] [PubMed] [Google Scholar]
  • 43.Rossant J., Zirngibl R., Cado D., Shago M., Giguere V. Genes Dev. 1991;5:1333–1344. doi: 10.1101/gad.5.8.1333. [DOI] [PubMed] [Google Scholar]
  • 44.Albrecht K. H., Eicher E. M. Dev. Biol. 2001;240:92–107. doi: 10.1006/dbio.2001.0438. [DOI] [PubMed] [Google Scholar]
  • 45.Palmer S. J., Burgoyne P. S. Development (Cambridge, U.K.) 1991;112:265–268. doi: 10.1242/dev.112.1.265. [DOI] [PubMed] [Google Scholar]
  • 46.Martineau J., Nordqvist K., Tilmann C., Lovell-Badge R., Capel B. Curr. Biol. 1997;7:958–968. doi: 10.1016/s0960-9822(06)00415-5. [DOI] [PubMed] [Google Scholar]
  • 47.Wilkinson D. G., Nieto M. A. Methods Enzymol. 1993;225:361–373. doi: 10.1016/0076-6879(93)25025-w. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Figure