Retinoic acid prevents germ cell mitotic arrest in mouse fetal testes

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

[Cell Cycle 7:5, 656-664; 1 March 2008]; ©2008 Landes Bioscience

656 Cell Cycle 2008; Vol. 7 Issue 5

During mouse fetal development, meiosis is initiated in female germ cells only, with male germ cells undergoing mitotic arrest. Retinoic acid (RA) is degraded by Cyp26b1 in the embryonic testis but not in the ovary where it initiates the mitosis/meiosis transition. However the role of RA status in fetal germ cell prolif-eration has not been elucidated. As expected, using organ cultures, we observed that addition of RA in 11.5 days post-conception (dpc) testes induced Stra8 expression and meiosis. Surprisingly, in 13.5 dpc testes although RA induced Stra8 expression it did not promote meiosis. On 11.5 and 13.5 dpc, RA prevented male germ cell mitotic arrest through PI3K signaling. Therefore 13.5 dpc testes appeared as an interesting model to investigate RA effects on germ cell proliferation/differentiation independently of RA effect on the meiosis induction. At this stage, RA delayed SSEA-1 extinc-tion, p63γ expression and DNA hypermethylation which normally occur in male mitotic arrested germ cells. In vivo, in the fetal male gonad, germ cells cease their proliferation and loose SSEA-1 earlier than in female gonad and RA administration maintained male germ cell proliferation. Lastly, inhibition of endogenous Cyp26 activity in 13.5 dpc cultured testes also prevented male germ cell mitotic arrest. Our data demonstrate that the reduction of RA levels, which occurs specifically in the male fetal gonad and was known to block meiosis initiation, is also necessary to allow the establishment of the germ cell mitotic arrest and the correct further differentiation of the fetal germ cells along the male pathway.

Introduction

Mammalian gonads acquire sex-specific features during fetal development (reviewed in ref. 1). Testes and ovaries differentiate from morphologically identical, bipotential gonads. In mice, primordial

germ cells (PGC) colonize the bipotential gonad about 10.5 days post conception (dpc) in mice and proliferate actively. Once in the gonad, PGC are termed gonocytes. Specifically in the fetal XY gonad some somatic cells proliferate and differentiate to form the Sertoli cells that enclose gonocytes in testicular cords from 12.5 dpc onwards. The germ cells remain very similar in male and female gonads until about 13.5 dpc. At this age, female germ cells initiate prophase I of meiosis and rapidly go through the leptotene and zygotene stages characterized by the generation of DNA double-strand breaks and the establishment of the synaptonemal complex.2 At about the same stage, male germ cells undergo mitotic arrest in G0/G1 (‘quiescence’), with all gonocytes having entered the quiescence phase by 15.5 dpc.3,4 At this stage, male germ cells no longer express markers of undifferentiated germ cells, such as the stage-specific embryonic antigen-1 (SSEA-1),5 and instead express new specific markers, including p63γ, a member of the p53 family.6 The paternal genomic imprint seems to be imposed during quiescence, as the nuclei of male germ cells display strong methylation from 16.5 dpc onwards.7 The mechanisms regulating the quiescence phase experienced by male germ cells remain unknown. Male germ cells resume mitosis shortly after birth and differentiate into spermatogonia.

Under experimental conditions, the germ cells comprising the bipotential gonad may initiate meiosis regardless of their genetic sex.8 After 12.5 dpc, XY germ cells cannot be persuaded anymore to enter meiosis. Retinoic acid (RA) has been identified as a key factor in the commitment of fetal germ cells toward the female or the male pathway. We have previously shown that RA can accelerate the initiation of meiosis in rat fetal ovaries9 and RA has recently been shown to induce meiosis in mouse fetal germ cells from both XX and XY gonads.10-13 This effect involves the expression of Stra8, a gene stimulated by RA, in the premeiotic germ cells.11,12 Gonocytes in the fetal testis are protected against the effects of RA by CYP26b1, a male-specific enzyme that degrades RA, possibly explaining why meiosis is not initiated in male gonocytes.10

Retinoids have also been reported to be involved in germ cell proliferation and apoptosis. RA and retinol have long been known to be essential for adult male reproductive function.14 One of the many functions of RA in the adult testis is the stimulation of premeiotic

*Correspondence to: Gabriel Livera; Unit of Gametogenesis and Genotoxicity; Route du Panorama-BP6; Fontenay aux Roses Cedex 92265 France; Tel.: +33.1.46.54.99.36; Fax: +33.1.46.54.99.06; Email: [email protected]

Submitted: 09/25/07; Revised: 12/10/07; Accepted: 12/21/07

Previously published online as a Cell Cycle E-publication: http://www.landesbioscience.com/journals/cc/article/5482

Report

Retinoic acid prevents germ cell mitotic arrest in mouse fetal testesEmilie Trautmann,1-3,† Marie-Justine Guerquin,1-3,† Clotilde Duquenne,1-3 Jean-Baptiste Lahaye,1-3 René Habert1-3 and Gabriel Livera1-3,*

1Univ Paris 7—Denis Diderot; Laboratory of Differentiation and Radiobiology of the Gonads; Unit of Gametogenesis and Genotoxicity; 2CEA; DSV/DRR/SEGG/LDRG; 3INSERM; U566; Fontenay aux Roses, France

†These authors contributed equally to this work.

Abbreviations: BrdU, bromodeoxyuridine; CYP26b1, cytochrome P450, subfamily 26, polypeptide 1; Dpc, day post-conception; PGC, primordial germ cell; PI3K, phosphoinositide 3-kinase; PTEN, phosphatase and TENsin homolog; RA, retinoic acid; Scp3, synaptonemal complex protein 3; SSEA-1, stage-specific embryonic antigen-1; Stra 8, stimulated by retinoic acid 8

Key words: developing gonads, germ cells, retinoic acid, meiosis, quiescence, proliferation, PI3K

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

Retinoids actions in the embryonic testis

www.landesbioscience.com Cell Cycle 657

spermatogonial proliferation.15 The effects of RA have also been extensively studied in rat fetal gonads and RA has been reported to induce germ cell apoptosis.16-19 In mice deficient for Cyp26b1, fetal male germ cell apoptotic extinction has also been reported.13

While RA involvement in meiosis initiation in the fetal ovary is clearly established, no study has yet been performed to elucidate the role of the RA drop in male germ cell differentiation. Because meiosis initiation in the fetal ovary is nearly concomitant with the mitotic arrest in the fetal testis, we hypothesized here that the decrease of RA levels in the testis is involved in the male germ cell quiescence and differentiation. In the present study, we found that RA prevented male germ cell mitotic arrest and differentiation in the mouse fetal testis. This revealed an original role for RA in male germ cells as on the contrary in many somatic cell lines RA treatment induces differ-entiation.20-22

Figure 1. Effect of RA on the initiation of meiosis at 11.5 dpc. Testes from 11.5 dpc fetuses were cultured for 4 days with (RA) or without (CTR) 1 μM RA. After culture, testes were fixed and embedded in paraffin. Sections (5 μm) were cut and stained with hematoxylin (A, bar: 10 μm). In control cultures, germ cells presented specific features of quiescent gonocytes (left). In RA-treated 11.5 dpc XY gonads, several germ cells presenting chromatin condensation typical of the leptotene (middle) and zygotene (right) stages were observed. Meiotic cells were also detected using whole-mount immuno-fluorescence for Scp3, a protein of the synaptonemal complex, in 11.5 dpc XX gonads cultured for four days in control medium (XX) and in 11.5 dpc XY gonads cultured with RA (XY RA). White arrows indicate Scp3-stained cells. Scp3 was absent from XY gonads cultured in control conditions (XY CTR) (B, bar: 100 μm, t, testis; m, mesonephros).

Figure 2. Effect of RA on the initiation of meiosis at 11.5 and 13.5 dpc. Testes from 11.5 and 13.5 dpc fetuses were cultured for 4 or 2 days, respec-tively, either with (RA) or without (CTR) RA or with (Keto) or without (CTR) ketoconazole. Entry into meiosis was evaluated by immunohistochemical staining for γH2AX. White arrows indicate unstained germ cells and black arrows indicate γH2AX-stained germ cells (A, bar: 10 μm). At the end of the culture period, total RNA was extracted and Dmc1 expression was measured by QRT-PCR to confirm the presence of meiotic cells in 11.5 dpc RA-treated gonads (B). Stra8 expression was also assessed by QRT-PCR (C). mvh was used as a reference. Values are means ± SEM of four to six determinations. *p < 0.05 in paired Student’s t test.

Results

Retinoic acid initiates meiosis in early fetal germ cells. In cultures of 11.5 and 12.5 dpc XY gonads with control medium, germ cells present the typical morphology of resting gonocyte and no meiosis was observed. In 11.5 and 12.5 dpc XY gonads cultured with 10-6 M of RA, about 20% of the germ cells had histological charac-teristics typical of the leptotene and zygotene stages, such as marked chromatin condensation (Fig. 1A). This effect of RA on the initiation of meiosis was confirmed by the detection of Scp3. Indeed, Scp3-stained cells were abundant in RA-treated 11.5 dpc male gonads and were not detected in control conditions (Fig. 1B).

Retinoic acid does not initiate meiosis on 13.5 dpc. In 13.5 dpc testes, RA did not produce meiotic figures and no Scp3-stained cells were detected in RA-treated testes (data not shown). This limitation of the effects of RA to the period prior to 13.5 dpc was confirmed by the immunostaining for γH2AX (Fig. 2A). RA induced γH2AX expression in large numbers of germ cells in 11.5 dpc XY gonads

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.



whereas no stained cells were observed in 13.5 dpc control and RA-treated testes. Similarly, no meiosis was observed in 13.5 dpc testes treated with ketoconazole, an inhibitor of the CYP family, although there was substantial induction of meiosis in 11.5 dpc XY gonads subjected to the same treatment (Fig. 2A).

The expression of Dmc1, a meiotic recombinase, was measured by QRT-PCR. As previously described,26 a weak expression of Dmc1 was observed in male gonads, both in 11.5 dpc and in 13.5 dpc testes cultured in control medium. RA significantly increased Dmc1 expres-sion only in 11.5 dpc gonads (Fig. 2B). We also determined the level of expression of Stra8 by QRT-PCR in 11.5 and 13.5 dpc control and RA-treated gonads. RA strongly increased Stra8 expression at both ages (Fig. 2C). Similar results were obtained using either βactin or mvh to normalize the expression (data not shown).

To further ensure that RA did not induce meiosis in 13.5 dpc testes, we investigated in more detail the development of male germ cells in RA-treated gonads. Germ cells were traced by adding BrdU for 3 hours on the second day of culture. Cultures were then stopped or continued for 24 additional hours. The percentage of BrdU-posi-tive germ cells did not change between the second and the third day of culture (Fig. 3A). On the third day, these labeled cells did not have the features of meiotic germ cells observed in 13.5 dpc ovaries 24

hours after BrdU incorporation (Fig. 3B). We also cultured 13.5 dpc testes for 4 days with or without RA, and no meiosis was detected at the end of the culture period (data not shown).

Retinoic acid increases proliferation and decreases differentia-tion of male germ cells in vitro. To investigate the effect of RA on germ cell proliferation, we cultured testes from 11.5, 12.5, 13.5 and 14.5 dpc fetuses for 4, 3, 2 and 1 days, respectively, so that all reached a comparable development stage, corresponding to about 15.5 dpc. Consistently with our findings in vivo (see later Fig. 7 and ref. 24), germ cell proliferation levels decreased throughout the culture in control conditions (Fig. 4A). Addition of 10-6 M of RA reduced or prevented this decrease if added at 11.5, 12.5 or 13.5 dpc, but not if added at 14.5 dpc, when most germ cells were already in mitotic arrest. Similar results were obtained for a concentration of 10-5 M of RA (data not shown). As BrdU incorporation in germ cells also marks pre-meiotic DNA replication, we measured the percentage of germ cell mitotic figures. The number of germ cell mitotic figures in 13.5 dpc testes was increased by culture with RA (Fig. 4B). It may be noted that RA also increased germ cell proliferation in 11.5 dpc XX gonads after 36 hours of culture (31.5 ± 1.8% in control and 40.3 ± 2.8% in RA-treated gonads).

To evaluate the effect of RA on male germ cell differentiation we measured the expression of markers of undifferentiated (SSEA-1) and differentiated (DNA methylation and p63γ) male germ cells. The percentage of SSEA-1-positive germ cells was 80% higher in RA-treated than in control 13.5 dpc testes cultured for two days (Fig. 4C). Furthermore, testes explanted at 13.5 dpc, a stage without methylation, exhibited a strong staining for 5methylcytosine (5mc) after 4 days of culture in control medium, whereas in RA-treated testes, about half of the germ cells remained unstained (Fig. 4D). Lastly, RT-PCR analysis after culture of 11.5 and 13.5 dpc testes showed that RA also repressed expression of p63γ (Fig. 4E).

RA signaling involves PI3K. To investigate the involvement of the PI3K pathway, which plays a major role in cell cycle progres-sion, we cultured 12.5 and 13.5 dpc testes in the presence of RA for three and two days, respectively, with or without LY294002, a specific inhibitor of PI3K. This inhibitor completely abolished the effect of RA on germ cell proliferation (Fig. 5A and B). The addi-tion of LY294002 in 12.5 dpc RA-treated testes did not change the number of meiotic germ cells detected by γH2AX staining (data not shown). We were unable to study the involvement of this inhibitor at 11.5 dpc because it induced specific cytotoxic effects on the meso-nephros.

Retinoic acid induces germ cell apoptosis. Apoptosis was measured by determining the percentage of cleaved caspase3-positive germ cells. In the presence of RA, germ cell apoptosis was inceased by 10 fold both in 11.5 and 13.5 dpc cultured testes (Fig. 6).

Retinoic acid administration in vivo prevents germ cell mitotic arrest. We wondered whether RA administration in vivo would affect male germ cell proliferation as previously observed in vitro. Consistently with our in vitro findings, we observed that a single oral dose of 100 mg/kg of RA at 13.5 dpc stimulated germ cell prolifera-tion in vivo (Fig. 7A and B). At 15.5 dpc testes from vehicle-treated fetuses contained barely any BrdU-positive germ cells while about a quarter of germ cells were stained for BrdU in testes from RA-treated fetuses.

Correlation between endogenous RA and germ cell prolifera-tion and differentiation. If RA physiologically acts in vivo on germ

Figure 3. Absence of meiosis in 13.5 dpc RA-treated testis. 13.5 dpc testes were cultured for two days as described in legend to Figure 2. Proliferating germ cells were then traced by BrdU incorporation. Control and RA-treated testes were maintained for an additional 24 hours after BrdU incorporation. The percentage of BrdU-positive germ cells was determined (A). **p < 0.01 in paired Student’s t test. In (B) 24 hours after BrdU-incorporation, the black arrow indicates a stained RA-treated male germ cell which did not enter meiosis (XY + RA) and the white arrow indicates a stained female germ cell which did (XX, CTR; bar: 10 μm).

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.



cell proliferation and differentiation, these activities should differ between male and female germ cells as RA is degraded specifically in the fetal testis. Therefore we investigated the timing of these processes in both sexes. The germ cell proliferation index was determined in gonads from 11.5 to 16.5 dpc animals injected with BrdU (Fig. 8A). Germ cell proliferation decreased in both testes and ovaries from 11.5 to 15.5 dpc. At 11.5 and 12.5 dpc, there was no significant difference in germ cell proliferation between the sexes. From 13.5 dpc onwards, male germ cell proliferation decreased considerably, whereas no such massive decrease was observed in ovaries. The decrease in female

germ cell proliferation was 24 hours behind that in male germ cell proliferation and more germ cell mitotic figures were also observed in 13.5 dpc ovaries than in 13.5 dpc testes (Fig. 8B). In the same line, the percentage of germ cells stained for SSEA-1 was 25% higher in ovaries than in testes at 13.5 and 15.5 dpc; this evidenced the lag between female and male differentiation (Fig. 8C).

Lastly, if the physiological degradation of RA in the testis is involved in the blockage of the gonocyte cell cycle, the inhibition of this degradation should increase germ cell proliferation. Indeed, in vitro treatment of 13.5 dpc testes with 40 μM of ketoconazole increased germ cell proliferation similarly to the addition of RA (Fig. 9A and B). Addition of RA together with ketoconazole did not increase further germ cell proliferation.

Discussion

RA has been reported to play a major role in determining the fate of germ cells by inducing meiosis in the fetal ovary whereas the drop of RA in the fetal testis protects male germ cells from initiating meiosis.10-13 In this study, we investigated RA effects on meiosis initiation in fetal male germ cells as a function of the developmental stage. Furthermore, we demonstrated for the first time that RA drop in the fetal testis is required to allow male germ cell mitotic arrest and differentiation.

RA effects on meiosis initiation. Histological observations and staining for meiotic markers showed that RA triggered entry into meiosis in 11.5 and 12.5 dpc XY gonads. This is consistent with recently published data.10,11 In addition to the meiotic markers used by these teams (Stra8, Scp1, Scp3 and Dmc1), we also evidenced the appearance of meiotic DNA double-strand breaks by γH2AX staining. The effects of RA on meiosis initiation depended on devel-opmental stage. Indeed, in culture of 13.5 dpc testes, no initiation of meiosis was observed. This observation was expected as it has previously reported that germ cell fate can be reversed only before 13.5 dpc.27

We detected a large increase in Stra8 expression in RA-treated gonads as previously described.11,12 Interestingly, we observed this

Figure 4. Effect of RA on male germ cell proliferation and differentiation. In (A) testes from 11.5 to 14.5 dpc were cultured for four to one day (D4 to D1) and BrdU was added during the last three hours of culture or at the time of explantation (D0). One testis from each fetus was cultured in control medium (CTR), and the other was cultured in medium supplemented with 1 μM RA (RA). The germ cell proliferation index was determined by counting stained and unstained gonocytes. Values are means ± SEM of five determina-tions. Different letters represent significantly different values as assessed by ANOVA, with p < 0.05. In (B) the percentage of germ cell mitotic figures in 13.5 dpc control and RA-treated testes was determined. Values are means ± SEM of five determinations. *p < 0.05 in paired Student’s t test. Sections stained with an antibody anti-MVH are presented (right). The white arrow indicates a quiescent germ cell and the black arrow indicates a germ cell in M-phase (bar: 10 μm). In (C) germ cell differentiation was evaluated by measuring the percentage of SSEA-1-positive cells in 13.5 dpc testes at the time of explantation (D0) or after 2 days of culture with or without RA (D2). Values are means ± SEM of three determinations (C). *p < 0.05 in paired Student’s t test. In (D) white arrows indicate unstained germ cells and black arrows indicate stained germ cells (bar: 10 μm). DNA methylation was studied by immunofluorescence staining of 5methylcytosine (5mc) in sections of 13.5 dpc testes cultured for four days. In (E) p63γ expression was studied by RT-PCR in 11.5 and 13.5 dpc XY gonad cultures. T15.5 stands for 15.5 dpc testes and O15.5 stands for 15.5 dpc ovaries.

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.



increase even when no germ cell entered meiosis (i.e., on 13.5 dpc). This indicates that although Stra8 expression has been proven to be required, it is not sufficient to commit any germ cells to meiosis.

RA effects on germ cell proliferation. RA prevented mitotic arrest in the germ cells of cultured testes when applied at 11.5, 12.5 or 13.5 dpc. Importantly, in vivo, RA administration at 13.5 dpc also prevented male germ cell mitotic arrest. We also observed that RA speeds up germ cell proliferation in the fetal ovary. Therefore, RA stimulation of germ cell proliferation is not specific to the fetal testis and, indeed, this phenomenon has been reported in every known mammalian mitotic germ cell, from PGC to spermatogonia and oogonia.28,29 Therefore, our observation that in 14.5 dpc testes, RA neither maintained nor stimulated germ cell mitosis is of particular interest. This indicates that fetal quiescence is irreversible and different from the spermatogonial mitotic arrest observed in cases of vitamin A deficiency (VAD) in the adult testis. Indeed, in cases of VAD, RA rapidly restores germ cell proliferation in the testis.29 Thus, the decrease in the amount of RA occurring physiologically in the fetal testis seems to be required to allow the entry into quiescence, and an additional mechanism independent of RA signaling probably lock germ cells into mitotic arrest.

We demonstrated that RA signaling rely upon PI3K activation. Indeed, a PI3K inhibitor counteracted the effect of RA on male germ cell mitosis. Similarly, RA has been shown to increase P85alpha (a PI3K subunit) levels and PI3K activity in F9 cells.30 The PI3K pathway is of particular interest as PI3K/Akt signaling plays key roles in cell cycle progression and cell differentiation. Moreover, the specific deletion in fetal germ cells of PTEN, a phosphatase that antagonizes PI3K, impairs mitotic arrest and results in the produc-tion of germ cells with immature characters.31 These effects are similar to the effects of RA we describe and provide support for the hypothesis that PI3K is one of the main targets of RA in germ cells.

Interestingly, we report here that RA stimulates proliferation without inducing meiosis in 13.5 dpc testis. This suggests that the effects of RA on germ cell proliferation and on meiosis initiation are independent. It must be noted that this demonstration was render possible by the use of the 13.5 dpc mouse testes that provides a unique model in which RA acts on male fetal germ cell prolifera-tion without initiating meiosis. This model differs from Cyp26b1 knock out mice10 in which, RA excess is induced from the first steps of testis organogenesis and meiosis is induced in male germ cells as in cultures of 11.5 XY gonads treated with RA. Therefore we provide here a unique model that may allow future investigations of the mechanisms specifically controlling male differentiation of fetal germ cells. Our proposal that RA effects on germ cell proliferation and meiosis induction are independent ones is strengthened by our observation that a PI3K inhibitor prevented RA action on germ cell proliferation whereas RA induction of meiosis was independent of PI3K signaling.

In vivo we also found that germ cell proliferation slowed down and stopped later in females than in males. As female fetal gonads contain more RA than male ones, this suggests that RA physiologi-cally maintains germ cell proliferation in the ovary. This finding is of particular interest as progression through the cell cycle is required for the initiation of meiosis. Possibly, RA induces meiosis in the fetal ovary at least partially by maintaining a proliferative status. On the other hand, using culture of 13.5 dpc testis which allows

Figure 5. Mechanism of action of RA on germ cell proliferation. Testes at 12.5 and 13.5 dpc were cultured with RA and with or without LY294002, a specific inhibitor of the PI3K pathway, and the germ cell proliferation index was determined. In (A) values are means ± SEM of six determina-tions. Different letters represent significantly different values as assessed by ANOVA, with p < 0.05. In (B) sections of 13.5 dpc testes cultured with RA in the absence (RA) or the presence (RA + LY) of LY294002 are presented. White arrows indicate unstained germ cells and black arrows indicate BrdU-stained germ cells (bar: 10 μm).

Figure 6. Effect of RA on germ cell apoptosis. XY gonads at 11.5 and 13.5 dpc were cultured with or without RA as described in the legend to Figure 2 and cleaved-caspase 3 was detected by immunohistochemistry. In (A) the percentage of caspase-positive germ cells was determined in four dif-ferent pairs of testis for each age. **p < 0.01 in paired Student’s t test. In (B) sections of 13.5 dpc testes cultured with and without RA are presented and white arrows indicate unstained germ cells and black arrows indicate cleaved-caspase 3-stained germ cells (bar: 10 μm).

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.



varying germ cell proliferation without inducing meiosis, we demon-strated that inhibition of endogenous Cyp26 activity was sufficient to prevent male germ cell mitotic arrest. This indicates that RA degradation in the fetal testis not only prevents the female pathway (i.e., meiosis induction) as previously described but is also required for inducing the male pathway (quiescence) in germ cells.

RA prevents male germ cell differentiation. Our finding that RA prevent male germ cell mitotic arrest is consistent with our observa-tion that RA (i) increased the production of SSEA-1, a marker of multipotent cells strongly expressed in proliferating fetal germ cells,32 and (ii) decreased p63γ expression and DNA hypermethylation; both these phenomena have been described as markers of germ cells in mitotic arrest.6,7 These various results indicate that RA prevented male fetal germ cell differentiation.

RA induces germ cell apoptosis. RA excess in the mouse fetal testis induces germ cell apoptosis both in vitro (this work) and in vivo in Cyp26b1-null mice.13 We report here that apoptosis occurs in RA-treated gonads even when no meiosis is observed. This is consistent with our previous studies in rat and human fetal testes in which RA induced apoptosis without meiosis.17,33 Thus, some germ cells undergo apoptosis without entering meiosis.

One may consider as odd that RA stimulates both prolifera-tion and apoptosis. We hypothesize that this may actually reflect a mechanism by which the fetal testis eliminates germ cells that do not differentiate correctly along the male pathway. The mechanisms beyond this phenomenon remained to be investigated but they might enlighten the genesis of the testicular cancer that is believed to originate from fetal germ cells that would have failed to differentiate correctly.34

In conclusion, we propose to add a new role for RA in determining the fetal germ cell fate: RA prevents germ cell entry into quiescence and maintained male germ cells in an undifferentiated stage (Fig. 10). This study shows for the first time that RA degradation in the fetal testis induces germ cell mitotic arrest and differentiation along the male pathway independently of blocking meiosis initiation.

Materials and Methods

Animals. NMRI mice were housed in controlled photoperiod conditions (lights on from 08:00 to 20:00) and were supplied with commercial food and tap water ad libitum. For accurate staging of 11.5 dpc embryos, the number of tail somites (ts) between the hind limb and the tip of the tail was determined at the time of dissec-tion and only embryos with 18 ± 2 ts were retained. The sex of the fetuses before 12.5 dpc was determined by PCR amplification of Sry as previously described.6 In vivo RA administration was performed as described by Morita and Tilly.23 Briefly, a single dose of 400 μl of vehicle (dimethylsulfoxide:corn oil, 1:4) or of RA (100 mg/kg) was administrated to 13.5 dpc pregnant female mice by oral gavage, and the mice were sacrificed on 15.5 dpc. All animal studies were conducted in accordance with the guidelines for the care and use of laboratory animals of the French Ministry of Agriculture.

Organ cultures and treatment. Pregnant mice were killed by cervical dislocation and organ culture was performed as previously described.24 We dissected out 11.5 dpc XY gonads associated with their mesonephros and 12.5 dpc to 14.5 dpc testes. The gonads were placed on Millicell CM filters (Millipore, Billerica, MA) and cultured on the top of 300 μl of Ham F12/DMEM (1:1) (Life Technologies Inc., Grand Island, NY) for one to four days at 37°C under a 5%

CO2/95% air atmosphere. Gonads from 11.5, 12.5, 13.5 and 14.5 dpc fetuses were respectively cultured for 4, 3, 2 and 1 days in order to reach a similar developmental stage. No serum was added to the culture medium. For each fetus, one testis was cultured in a medium containing all-trans-RA (10-6 to 10-5 M, Sigma, St. Louis, MO) and the other was cultured in a control medium. In some experiments, 1 μM of LY294002 (Sigma), a PI3K-specific inhibitor, was added to the culture media. LY294002 was added immediately after setting up both gonads from the same fetus in culture and one hour later RA was added only into the culture medium of one testis without removing LY294002. 40 μM of ketoconazole (Sigma), an inhibitor of the CYP26 family, was added to some cultures with or without RA. Either one testis was cultured in control medium and the other one from the same fetus in presence of ketoconazole, either one testis was cultured in presence of RA and the other one in presence of RA plus ketoconazole. Each culture was repeated at least three times independently with fetuses from different litters and for each treat-ment four to ten testes were analyzed.

Histology. Embryonic male gonads were fixed in Bouin’s fluid just after dissection or at the end of culture. The fixed testes were dehydrated, embedded in paraffin and cut into 5-μm sections, which were mounted on glass slides, dewaxed, rehydrated and stained with hematoxylin and eosin. The gonocytes were identified on the basis of their large, spherical, lightly stained nuclei and clearly visible cyto-plasmic membrane.25 Nuclei of gonocytes in mitotic arrest contained fine chromatin granules and globular nucleoli, whereas proliferating gonocytes nuclei contained patches of heterochromatin (personal observation). Meiotic cells displayed marked condensation of the chromatin forming distinct fine threads with a beaded appearance at the leptotene stage and presenting a characteristic criss-cross of coiled chromosome threads at the zygotene stage.9

Figure 7. Effect of RA administration in vivo. Pregnant female mice were given a single oral dose of vehicle or of RA at 13.5 dpc and injected with BrdU at 15.5 dpc. The testes from vehicle and RA-treated fetuses were fixed three hours later. In (A) germ cell proliferation was evaluated by determin-ing the percentage of BrdU-positive germ cells. Values are means ± SEM of eight to ten determinations; ***p < 0.001 in unpaired Student’s t test. In (B) sections of 15.5 dpc testes from vehicle-treated and RA-treated fetuses are presented. White arrows indicate unstained germ cells and black arrows indicate BrdU-stained germ cells (bar: 10 μm).

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.



Immunohistochemistry. Gamma-H2AX immunohistochemistry was used to reveal DNA double-strand breaks, a hallmark of meiosis. Sections were mounted on a single slide and heated in a permeabili-zation solution [0.01 M sodium citrate (pH 6)] for 30 min at 100°C, cooled and incubated with 10% normal horse serum (NHS, Vector Laboratories) and mouse anti γH2AX antibody (1:500, Euromedex). The primary antibody was detected by incubation with a biotinylated horse antimouse antibody (1:200; Vector Laboratories). Slides were incubated with an avidin-biotine-peroxidase complex (Vectastain Elite ABC kit; Vector Laboratories) and peroxidase activity was detected with DAB. SSEA-1, a marker of undifferentiated germ cells, was detected with a similar protocol. Briefly, sections were incubated with 1.5% Normal Goat Serum (NGS, Vector Laboratories), and then overnight with the mouse anti-SSEA1 antibody (1:5, Peptide Specialty Laboratories GmbH, Germany). The primary antibody was detected by incubation with a peroxidase-linked anti-mouse IgG (1:200, GE Healthcare). The percentage of SSEA-1-positive germ cells was determined by a blind counting of at least 500 stained and unstained germ cells. Immunohistochemistry for cleaved-caspase 3, a marker of apoptosis, was carried out as previously described.25

Immunofluorescence. Scp3, a protein of the synaptonemal complex, was detected by whole-mount immunofluorescence. After culture, treated and untreated testes were fixed in 4% paraformal-dehyde at 4°C, washed in PBS, blocked by incubation in 3% BSA supplemented with 0.1% Triton, incubated with rabbit anti-Scp3 antibody (1:200, Peptide Specialty Laboratories GmbH) and donkey anti-rabbit-Cy3-conjugated antibody (1:500). Slides were mounted in Vectashield (Vector Laboratories) and observed with a confocal microscope (Laser Scanning Microscope LSM 5 Pascal, Carl Zeiss). For DNA methylation studies, the gonads were fixed in Carnoy’s fluid, sectioned and treated as previously described.7 Briefly, DNA methylation was detected with an anti-5mc antibody (1:100, Eurogentec, Seraing, Belgium).

Measurement of germ cell proliferation. The percentage of germ cells in S-phase was evaluated by measuring the BrdU (5-bromo-2'-deoxyuridine) incorporation, by immunohistochemical methods, using the Cell Proliferation Kit (GE Healthcare, Buckinghamshire, UK) according to the manufacturer’s recommendations. BrdU (1%) was either added at the end of the culture period or was injected intraperitoneally into pregnant mice 3 hours before tissue fixation. For some in vitro experiments, BrdU was washed and cultures were maintained for one additional day before tissue fixation. Briefly, five randomly selected sections were mounted, rehydrated, incubated for 1 h with anti-BrdU antibody and then with a peroxidase-linked anti-mouse IgG. Finally, peroxidase activity was visualized using 3,3'-diaminobenzidine (DAB, Vector Laboratories, Burlingame, CA). The BrdU incorporation index was determined by blind counting of at least 500 stained and unstained germ cells. Histolab analysis software was used for counting (Microvision Instruments, Evry, France).

Measurement of germ cell mitotic figures. The percentage of germ cells mitotic figures was evaluated using mouse vasa homolog (MVH) immunostaining to detect germ cells. The immunohistochemical procedure used was identical to that for the detection of BrdU, with an additional blocking step [5% Normal Goat Serum (NGS, Vector Laboratories) and 2% bovine serum albumin (BSA, Sigma)]. The primary antibody was a rabbit anti-MVH (1/750, Abcam) and was

Figure 8. Germ cell proliferation and differentiation in vivo. Three hours after BrdU injection, male (XY) and female (XX) gonads from 11.5 to 16.5 dpc were dissected out and fixed. In (A) germ cell proliferation was evaluated using BrdU immunohistochemistry and meiosis initiation was measured in XX gonads using γH2AX staining. Means ± SEM of three to seven determina-tions are shown. In (B) the percentage of mitotic figures in 13.5 dpc testes and ovaries was measured using MVH (Vasa) staining to identify germ cells. Values are means ± SEM of five determinations. **p < 0.01 and ***p < 0.001 in paired Student’s t test. In (C) germ cell differentiation was evaluated by immunohistochemical staining for SSEA-1, a marker of undifferentiated germ cells, in 13.5 and 15.5 dpc testes and ovaries. Values are means ± SEM of five determinations.

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.



detected with a biotinylated goat anti-rabbit antibody (1:200). Slides were incubated with an avidin-biotin-peroxidase complex (Vectastain Elite ABC kit) and peroxidase activity was detected with DAB. The percentage of mitotic figures was determined by blind counting of at least 500 stained and unstained germ cells.

Reverse transcription and real-time PCR. At the end of the culture period, total RNA was extracted using the RNeasy mini-kit (Qiagen) and reverse transcription was carried out with the Omniscript kit (Invitrogen), according to the manufacturer’s instructions. Expression of Stra8 (stimulated by retinoic acid), a premeiotic marker, was assessed by quantitative real-time RT-PCR using the SYBR Green Universal PCR Master Mix 2X (Applied Byosystems, Foster City, CA). Primers for Stra8 were: F: 5'-GCCTGGAGACCTTTGACGA-3' and R: 5'-GGCTTTTGGAAGCAGCCTTT-3'. Reactions were carried out and fluorescence was detected on an ABI Prism 7000 apparatus (Applied Biosystems, Foster City, CA). The amount of each cDNA detected was normalized using βactin (F: 5'-TGACCCAGATCATGTTTGAGA-3' and R: 5'-TACGACCAGAGGCA TACAGG-3') and mvh (F: 5'-GAAGAAATCCAGAGGTTGGC-3' and R: 5'-GAAGGATCGTCTGCTGAACA-3'). Expression of Dmc1, a meiotic marker, was assessed by quantitative real-time RT-PCR using Taqman gene expression assay (primers and probes: Mm00494485-m1) and were amplified with the TaqMan PCR Master Mix (Applied Biosystems). PCR efficiency was 2.09 for βactin, 2.01 for mvh, 2.05 for Stra8 and 1.95 for Dmc1.

Expression of each p63γ, a marker of germ cell mitotic arrest was studied by RT-PCR. Primers for p63γ were F: 5'-CAGCACCA GCACCTACTTC-3' and R: 5'-CGTCAGACTGTGTCGGAG-3'. PCR were performed as previously described.6

Statistical analysis. All values are expressed as means ± SEM. The significance of differences between mean values for the treated

and untreated testes was evaluated using Student’s paired t-test or ANOVA followed by Tukey’s comparison test.

Acknowledgements

We thank V. Neuvillle and S. Leblay for taking care of the animals and A. Gouret for her secretarial help. We are grateful to A. Chicheportiche for providing the antibody anti-Scp3. English was reviewed by A. Edelman.

This work was supported by INSERM, Commissariat à l’Energie Atomique (CEA) and Université Paris 7—Denis Diderot. E.T. holds a fellowship from the CEA. MJ.G. holds a fellowship from the Ministère de l’Education Nationale, de la Recherche et de la Technologie.

References 1. Kim Y, Capel B. Balancing the bipotential gonad between alternative organ fates: a new

perspective on an old problem. Dev Dyn 2006; 235:2292-300. 2. De Felici M, Klinger FG, Farini D, Scaldaferri ML, Iona S, Lobascio M. Establishment

of oocyte population in the fetal ovary: primordial germ cell proliferation and oocyte pro-grammed cell death. Reprod Biomed Online 2005; 10:182-91.

3. 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.

4. Moreno SG, Dutrillaux B, Coffigny H. Study of the gonocyte cell cycle in irradiated TP53 knockout mouse foetuses and newborns. Int J Radiat Biol 2002; 78:703-9.

5. Gomperts M, Garcia Castro M, Wylie C, Heasman J. Interactions between primordial germ cells play a role in their migration in mouse embryos. Development 1994; 120:135-41.

6. Petre-Lazar B, Livera G, Moreno SG, Trautmann E, Duquenne C, Hanoux V, Habert R, Coffigny H. The role of p63 in germ cell apoptosis in the developing testis. J Cell Physiol 2007; 210:87-98.

7. Coffigny H, Bourgeois C, Ricoul M, Bernardino J, Vilain A, Niveleau A, Malfoy B, Dutrillaux B. Alterations of DNA methylation patterns in germ cells and Sertoli cells from developing mouse testis. Cytogenet Cell Genet 1999; 87:175-81.

8. McLaren A. Entry of mouse embryonic germ cells into meiosis. Developmental Biology 1997; 187:107-13.

9. Livera G, Rouiller Fabre V, Valla J, Habert R. Effects of retinoids on the meiosis in the fetal rat ovary. Mol Cell Endocrinol 2000; 165:225-31.

10. 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.

Figure 9. Effect of Cyp26 inhibition on germ cell proliferation. Testes at 13.5 dpc were cultured with 1 μM RA and/or 40 μM Ketoconazole. In A, germ cell proliferation was evaluated by determining the percentage of BrdU-posi-tive germ cells. **p < 0.01 in paired Student’s t test. In (B) sections of 13.5 dpc testes cultured with and without ketoconazole and/or RA are presented. White arrows indicate unstained germ cells and black arrows indicate BrdU-stained germ cells (bar: 10 μm).

Figure 10. Proposed model of the effects of RA on fetal germ cells. We suggest that RA has two main and distinct roles in germ cell sexual determi-nation: (1) RA stimulates meiosis initiation; this effect depends of Stra8 but not of PI3K and (2) RA prevents germ cell entry into quiescence; this effect depends of PI3K. Multipotent germ cells express markers of undifferentiated cells, for example SSEA-1, and proliferate actively from 11.5 dpc onwards. From 13.5 dpc onwards, the RA degradation in the testis directs the gono-cytes toward quiescence and toward the male pathway. In the ovary, high RA concentrations maintain germ cell proliferation, increase Stra8 expres-sion and induce meiosis committing these proliferating cells to the female pathway. The new RA effect hereby described is presented in red.

©2008 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.



11. Koubova J, Menke DB, Zhou Q, Capel B, Griswold MD, Page DC. Retinoic acid regulates sex-specific timing of meiotic initiation in mice. Proc Natl Acad Sci USA 2006; 103:2474-9.

12. Baltus AE, Menke DB, Hu YC, Goodheart ML, Carpenter AE, de Rooij DG, Page DC. In germ cells of mouse embryonic ovaries, the decision to enter meiosis precedes premeiotic DNA replication. Nat Genet 2006; 38:1430-4.

13. Maclean G, Li H, Metzger D, Chambon P, Petkovich M. Apoptotic extinction of germ cells in testes of Cyp26b1 knockout mice. Endocrinology 2007; 148:4560-7.

14. Livera G, Rouiller Fabre V, Pairault C, Levacher C, Habert R. Regulation and perturbation of testicular functions by vitamin A. Reproduction 2002; 124:173-80.

15. Chung SS, Wolgemuth DJ. Role of retinoid signaling in the regulation of spermatogenesis. Cytogenet Genome Res 2004; 105:189-202.

16. Marinos E, Kulukussa M, Zotos A, Kittas C. Retinoic acid affects basement membrane for-mation of the seminiferous cords in 14-day male rat gonads in vitro. Differentiation 1995; 59:87-94.

17. Livera G, Rouiller Fabre V, Durand P, Habert R. Multiple effects of retinoids on the devel-opment of Sertoli, germ and Leydig cells of fetal and neonatal rat testis in culture. Biol Reprod 2000; 62:1303-14.

18. Livera G, Pairault C, Lambrot R, Lelievre Pegorier M, Saez JM, Habert R, Rouiller Fabre V. Retinoid-sensitive steps in steroidogenesis in fetal and neonatal rat testes: in vitro and in vivo studies. Biol Reprod 2004; 70:1814-21.

19. Cupp A, Dufour J, Kim G, Skinner M, Kim K. Action of retinoids on embryonic and early postnatal testis development. Endocrinology 1999; 140:2343-52.

20. Rochette Egly C, Chambon P. F9 embryocarcinoma cells: a cell autonomous model to study the functional selectivity of RARs and RXRs in retinoid signaling. Histol Histopathol 2001; 16:909-22.

21. Verlinden L, Verstuyf A, Mathieu C, Tan BK, Bouillon R. Differentiation induction of HL60 cells by 1,25(OH)2D3, all trans retinoic acid, rTGF-beta2 and their combinations. J Steroid Biochem Mol Biol 1997; 60:87-97.

22. Soprano DR, Teets BW, Soprano KJ. Role of retinoic Acid in the differentiation of embryo-nal carcinoma and embryonic stem cells. Vitam Horm 2007; 75:69-95.

23. Morita Y, Tilly JL. Segregation of retinoic acid effects on fetal ovarian germ cell mitosis versus apoptosis by requirement for new macromolecular synthesis. Endocrinology 1999; 140:2696-703.

24. Livera G, Delbes G, Pairault C, Rouiller Fabre V, Habert R. Organotypic culture, a pow-erful model for studying rat and mouse fetal testis development. Cell Tissue Res 2006; 324:507-21.

25. Delbes G, Levacher C, Pairault C, Racine C, Duquenne C, Krust A, Habert R. Estrogen receptor {beta}-mediated inhibition of male germ cell line development in mice by endog-enous estrogens during perinatal life. Endocrinology 2004; 145:3395-403.

26. Chuma S, Nakatsuji N. Autonomous transition into meiosis of mouse fetal germ cells in vitro and its inhibition by gp130-mediated signaling. Dev Biol 2001; 229:468-79.

27. Adams IR, McLaren A. Sexually dimorphic development of mouse primordial germ cells: switching from oogenesis to spermatogenesis. Development 2002; 129:1155-64.

28. Koshimizu U, Watanabe M, Nakatuji N. Retinoic acid is a potent growth activator of Mouse primordial germ cells in vitro. Dev Biol 1995; 168:683-5.

29. Gaemers IC, Sonneveld E, van Pelt AM, Schrans BH, Themmen AP, van der Saag PT, de Rooij DG. The effect of 9-cis-retinoic acid on proliferation and differentiation of a sper-matogonia and retinoid receptor gene expression in the vitamin A-deficient mouse testis. Endocrinology 1998; 139:4269-76.

30. Bastien J, Plassat JL, Payrastre B, Rochette-Egly C. The phosphoinositide 3-kinase/Akt pathway is essential for the retinoic acid-induced differentiation of F9 cells. Oncogene 2006; 25:2040-7.

31. Kimura T, Suzuki A, Fujita Y, Yomogida K, Lomeli H, Asada N, Ikeuchi M, Nagy A, Mak TW, Nakano T. Conditional loss of PTEN leads to testicular teratoma and enhances embryonic germ cell production. Development 2003; 130:1691-700.

32. Kudo T, Kaneko M, Iwasaki H, Togayachi A, Nishihara S, Abe K, Narimatsu H. Normal embryonic and germ cell development in mice lacking alpha 1,3-fucosyltransferase IX (Fut9) which show disappearance of stage-specific embryonic antigen 1. Mol Cell Biol 2004; 24:4221-8.

33. Lambrot R, Coffigny H, Pairault C, Donnadieu AC, Frydman R, Habert R, Rouiller-Fabre V. Use of organ culture to study the human fetal testis development: effect of retinoic acid. J Clin Endocrinol Metab 2006.

34. Rajpert-De Meyts E. Developmental model for the pathogenesis of testicular carcinoma in situ: genetic and environmental aspects. Hum Reprod Update 2006; 12:303-23.

Retinoic acid prevents germ cell mitotic arrest in mouse fetal testes

Documents