Follicular Cells Acquire Sertoli Cell Characteristics after Oocyte Loss
http://www.100md.com
内分泌学杂志 2005年第7期
Laboratoire de Physiologie et Physiopathologie (C.J.G., N.C., S.M.-G., S.M.), Centre National de la Recherche Scientifique-Unité Mixte de Recherche 7079, Université Paris VI, Paris, France; and Laboratoire de Pathologie Hormonale Moléculaire (M.G.F.), Institut National de la Santé et de la Recherche Médicale EA3739, Hopital Debrousse, 69322 Lyon cedex 05, France
Address all correspondence and requests for reprints to: Dr. S. Magre, Laboratoire de Physiologie et Physiopathologie, Centre National de la Recherche Scientifique-Unité Mixte de Recherche 7079, Université Paris VI, 7 Quai Saint Bernard, 75005 Paris, France. E-mail: solange.magre@snv.jussieu.fr.
Abstract
Although it has been suggested that in mammals the loss of female germ cells may induce the masculinization of the ovarian compartment, there has been as yet no conclusive demonstration. To directly address that question, the present study has been designed to determine the fate of follicular cells after oocyte loss. Using -irradiation to selectively deplete oocytes in nongrowing follicles in female rats, we show that follicular cells in oocyte-depleted follicles survive, proliferate, and subsequently acquire morphological characteristics of Sertoli cells: elongated cytoplasm, basal location of the nucleus, and specific Sertoli cell junctions, the ectoplasmic specializations. These Sertoli-like cells express, however, the female-specific marker FOXL2 (Forkhead L2) but not the male sex-specific marker SOX-9 (Sry-type high-mobility-group box transcription factor-9) underlying the maintenance of molecular characteristics of granulosa cells. Before transdifferentiating into Sertoli-like cells, follicular cells of oocyte-depleted follicles initiate the expression of anti-Mullerian hormone and inhibin -subunit that are typically synthesized by granulosa cells from the onset of follicular growth. Experimental modifications of the endocrine balance of the irradiated females show that there is a close relationship between plasma FSH levels and the occurrence of Sertoli-like cells. In addition to providing experimental evidence for the crucial role of the oocyte in granulosa cell phenotype maintenance, these results emphasize that the transdifferentiation of granulosa cells into Sertoli cells occurs in a multistep fashion, requiring the maturation of granulosa cells and depending on the endocrine milieu.
Introduction
THE GONAD ARISES as a bipotential primordium with the ability to develop into either an ovary or a testis. Both somatic and primordial germ cells (PGCs) follow a dimorphic fate once sex determination occurs. In mammals, testicular differentiation is governed by the Sry (sex-determining region, Y chromosome) gene, which initiates the differentiation of Sertoli cells and their organization into seminiferous cords (1). In the absence of Sry, the gonad differentiates into an ovary (2, 3). However, the hypothesis arguing that the ovarian development is the default pathway of gonadal differentiation is now controversial (4, 5, 6).
PGCs develop according to their sex somatic cell environment. It has been proposed that whereas in females PGCs enter meiosis in a cell autonomous manner, in males an as-yet-unidentified signal produced by Sertoli cells prevents PGCs from entering meiosis (7, 8). In addition, although germ cells in males are dispensable for the formation of testis cords, there are several lines of evidence suggesting that they may well reinforce Sertoli cell differentiation (9, 10) by acting via a feedback loop mediated by paracrine factors (11). Moreover, the active inhibition of meiosis would be essential because PGCs committed to meiosis could antagonize the differentiation of testis cords (12). In females, germ cells are essential for directing the ovarian pathway because no follicle differentiates in their absence (13, 14). If germ cells are destroyed or lacking, as in busulfan-treated rats (15), or in We/We mice (16), respectively, the female gonad is constituted by undifferentiated pregranulosa cells arranged in cords that subsequently degenerate. In some cases, the loss of female germ cells either at the mitotic or meiotic stage has been associated with the differentiation of testis-like cords within fetal ovaries (14, 17).
Ovarian sex reversal was first reported in the germ cell-depleted gonad of the freemartin bovine (18), which is a female exposed in utero to the testicular hormones, i.e. anti-Müllerian hormone (AMH) and testosterone, of her male cotwin. Differentiation of testis-like cords in association with germ cell depletion was also observed in rat fetal ovaries cultured in the presence of fetal testes (19) or the presumptive masculinisating factor AMH (20) as well as in aging rat ovaries (21) and grafts of fetal mouse ovaries under the renal capsule of adult hosts (22, 23). To our knowledge, however, there is still no direct evidence that female germ cell loss induces ovarian sex reversal. One may hypothesize that the loss of germ cells could be the consequence of the transdifferentiation of the ovarian supporting cells rather than its cause. The purpose of the present study was to determine the fate of follicular cells in the absence of meiotic germ cells and in particular whether under such conditions follicular cells transdifferentiate into Sertoli-like cells. Taking advantage of a rat experimental model in which oocytes in nongrowing follicles are selectively depleted by -irradiation at 5 d postnatal (dpn) (24), we studied the subsequent development of the surrounding follicular cells at both morphological and molecular levels. We report that after oocyte loss, follicular cells mature comparably to follicular cells in early growing follicles and thereafter acquire some morphological characteristics of immature Sertoli cells, presumably under the control of the endocrine environment and more specifically of FSH.
Materials and Methods
Animals and irradiation
Pregnant female rats from the Sprague Dawley strain were purchased from Charles River Laboratories (L’Arbresle, France). Males at 0 dpn and females at 5 dpn were whole body exposed to a -irradiation using either Cobalt or Cesium source at a dose rate of 0.8 Gy/min, with a final dose of 1.5 Gy. The exposure of males at 0 dpn induces the death of most germ cells and leads to the development of Sertoli cell-only tubules (SCO) (25). From 4 wk, animals were weaned and housed at four to five per cage. Animals were maintained in standard lighting (12 h light, 12 h dark) and temperature (22 C) with food and water ad libitum. All the animal procedures were approved and performed according to the 1986 European Communities Council Directives and those of the Ministère de l’Agriculture et de la Forêt.
Tissue collection
Females and males were killed by decapitation in the afternoon. For adult control and irradiated females, daily vaginal smears were performed. Animals were killed and blood was collected in the afternoon of the estrus. Blood was collected in heparinized tubes, centrifuged at 3000 x g for 15 min, and collected plasma was stored at –20 C until assayed for hormones. The ovaries and testes were either snap frozen in liquid nitrogen and stored at –80 C or processed for further morphological studies.
Tissue processing
For histological examination of ovarian and testicular morphology, organs were placed in Bouin’s fixative for at least 24 h. Sections of 5 μm thickness were stained according to the protocol of Tuchmann du Plessis (26). For in situ hybridization, immunohistochemistry, and terminal deoxynucleotidyltransferase-mediated deoxyuridine 5-triphosphate-fluorescein nick end labeling (TUNEL) assay, ovaries and testes were fixed in 2% paraformaldehyde-PBS and further processed as described (24). For immunohistochemistry using 5-bromo-2'-deoxyuridine (BrdU) and AMH antibodies, ovaries were fixed overnight in 2% paraformaldehyde-PBS, rinsed several times in PBS buffer, and dehydrated in alcohol. They were then embedded in paraffin wax and sectioned at 5 μm thickness.
For ultrastructural studies, ovaries of females aged 5 dpn and 1, 2, 8, 12, and 24 wk were cut into small pieces and fixed in 2% paraformaldehyde and 0.5% glutaraldhehyde in 0.1 M cacodylate buffer (pH 7.2) overnight at 4 C. They were rinsed in the same buffer, postfixed for 2 h in 2% osmium tetroxide in 0.1 M cacodylate buffer at room temperature, dehydrated through a graded series of alcohol, and embedded in epoxy embedding resin (Fluka, Sigma-Aldrich Corp., St. Louis, MO). Semithin sections (1 μm) were stained with methylene blue. Thin sections were mounted on grids and stained with uranyl acetate and lead citrate. Electron micrographs were taken on a EM906 electron microscope (Zeiss, G?ttinger, Germany).
In situ hybridization
Digoxigenin-labeled sense and antisense riboprobe synthesis as well as in situ hybridizations were performed as previously described (24). Primers used for the synthesis of the riboprobes are reported in Table 1.
TABLE 1. Primers used for in situ hybridization and real time PCR
BrdU injections
Control and irradiated females received three ip injections of 50 μl BrdU (Sigma-Aldrich) dissolved in saline (5 mg/ml) (one injection every 2 d) in the prepubertal period (from 8 to 14 dpn) or in adulthood (from 90 to 96 dpn) and were killed the day after the last injection.
Immunohistochemistry and TUNEL assay
Immunohistochemistry studies were performed on frozen sections except those for BrdU and AMH that were performed on paraffin sections. Paraffin sections were deparaffinized in toluene and rehydrated in decreasing concentrations of alcohol. Tissue sections were then boiled for 5 min in 0.1 M citrate buffer (pH 6.0). For both frozen and paraffin sections, endogenous peroxidase activity was blocked by treatment in H2O2 for 10 min (blocking reagents, Dako Corp., Carpinteria, CA). Nonspecific binding was then blocked in PBS/10% BSA for 20 min. Sections were incubated overnight at 4 C with the primary antibody (Table 2) diluted in blocking solution (PBS/0.05% BSA), followed by incubation for 1 h with the appropriate secondary biotinylated antibody (Vector Laboratories, Burlingame, CA) and 1 h with a peroxidase-conjugated streptavidin horseradish complex (ABC Elite kit, Vector). The reaction product was developed using 3, 3'-diaminobenzidine tetrahydrochloride (Dako). For negative controls (not shown), primary antibody was omitted. For double immunofluorescence, sections were blocked for 20 min. in PBS/10% BSA and then incubated with the antibody against cytokeratin 8 (CK-8) diluted in blocking solution overnight at 4 C. Sections were washed three times in PBS for 5 min each and then incubated for 2 h with fluorescein isothiocyanate-conjugated mouse secondary antibodies (dilution 1:100; Amersham, Les Ulis, France) at room temperature. Sections were rinsed three times and incubated at room temperature for 2 h with the primary antibody against the tyrosine kinase receptor KIT, rinsed, and incubated for 1 h with rabbit biotinylated secondary antibodies (dilution 1:300, Amersham), rinsed and incubated for 1 h in Extravidin (dilution 1:100; Sigma-Aldrich). Sections were rinsed and mounted in Vectashield (Vector). Fluorescence labeling was observed with a Zeiss epifluorescence microscope.
TABLE 2. Primary antibodies used for immunohistochemistry
TUNEL assay was performed on frozen sections as previously described (24).
RNA isolation, reverse transcription (RT), and real-time PCR
Total RNA was isolated from ovaries and testes using the RNeasy Protect miniextraction kit (QIAGEN SA, Courtaboeuf, France) following the manufacturer’s instructions. The quality of extracted RNA was verified by spectrophotometry and electrophoresis on agarose gel. For each sample, RT was performed with 0.5 μg total RNA incubated with random primers (Promega Corp., Madison, MI) using a Superscript II kit (Invitrogen Corp., Carlsbad, CA) according to the manufacturer’s instructions.
Real-time PCR analysis of oligodendrocyte-specific protein gene (Osp) and the ribosomal gene L19 (Rpl19), used as an internal control, was performed using the iCycler system (Bio-Rad Laboratories, Hercules, CA). Briefly, PCR was performed in 96-well microtiter plates with the QuantiTect SYBR Green PCR Kit (QIAGEN), using 5 μl of the cDNA from the RT reactions diluted at 1:100. PCR was performed as follows: 15 min initial HotStart enzyme activation at 95 C, followed by 45 cycles of 95 C denaturation for 30 sec, annealing at 58 C for 30 sec, and extension at 72 C for 30 sec. At the end of the PCR, melting curve analysis was performed to verify product specificity by increasing the temperature from 55 to 95 C at 0.1 C/sec. Melting point analysis demonstrated a single melting breakpoint in all experiments performed. In each reaction a standard curve was obtained from serial dilutions of a mix of RT products from ovaries and testes. For each experiment, all samples were run in triplicate. The cycle threshold values obtained from each sample were collected and data were analyzed by normalizing the levels of Osp levels to those of the internal control Rpl19. The specificity of the primers for both Osp and Rpl19 was verified after migration of cDNA on agarose gel. A single product of 339 and 196 bp were amplified for Osp and Rpl19, respectively, and the cDNA were sequenced. No bands were seen in mock reactions in which reverse transcriptase has been omitted.
Histological counts
Counts of the epithelial structures observed in the irradiated ovaries (see Results) were carried out on every fifth ovarian section from the prepubertal period until adulthood with careful examination of neighboring sections to avoid double counting of the same epithelial structure. Epithelial structures were scored as oocyte-depleted follicles (ODFs) when they displayed cuboidal somatic cells enveloped by a basement membrane (BM) with no germ cell and they were scored as transdifferentiated oocyte depleted follicles (tODFs) when they displayed polarized cells with an elongated cytoplasm and a nucleus located toward the BM.
Hormone assays
Plasma 17?-estradiol (E2) and testosterone (T) as well as LH and FSH were assayed by specific RIA as previously described (27, 28). Plasma E2 and T were assayed by specific RIA after ethyl ether extraction, followed by chromatographic purification on Celite columns. Plasma LH and FSH were assayed using RIA kits purchased from the National Institute of Diabetes and Digestive and Kidney Diseases (Baltimore, MD) with highly purified rat LH (rLH-I-10) and FSH (rFSH-I-9) for iodination, reference preparations (rLH-RP3 and rFSH-RP2), and appropriate antisera (anti-rLH-S11 and anti-rFSH-S11). Plasma inhibin A and B were determined using a two-site ELISA specific for each peptide (Oxford Bio-Innovation, Oxon, UK) as previously described (29). Intraassay coefficients of variation were 4.3, 4.9, 5.1, 7, and 5% and interassay coefficients of variation were 10, 12, 11.2, 12.5, and 9–10% for inhibin A, inhibin B FSH, LH, and steroids, respectively. The sensitivities of the assays were 7 and 10 pg/ml, 0.1 and 3.0 ng/ml, and 3 and 5 pg/tube for inhibin A, inhibin B, FSH, LH, E2, and T, respectively.
Hormonal treatments
Adult-irradiated females were divided into five groups (n = 5 females/group). At 8 and 10 wk of age, females in groups 1–4 received an im injection of the long-acting GnRH agonist triptorelin (Decapeptyl, generously provided by Beaufour-Ipsen Industrie, Dreux, France; 300 μg/kg body weight), known to reduce durably the gonadotropin levels (30). In addition, females in groups 1–4 were injected sc every 2 d with: group 1, saline; group 2, 2 IU recombinant human FSH (r-hFSH, Puregon; generously provided by Organon, Serifontaine, France); group 3, 5 IU human chorionic gonadotropin (hCG Endo Pregnyl, generously provided by Organon). Females of group 4 received twice weekly a sc injection of 100 μg testosterone (Sigma-Aldrich) (31) dissolved in olive oil (Sigma-Aldrich). Females in group 5 received only saline injections. At the end of the experiment (12 wk of age), ovaries and blood were collected and processed as described above.
Statistical analysis
Data collected from at least three different animals were expressed as the mean ± SEM and analyzed using GBStat 6.0 software (Dynamic Microsystems, Inc., Silver Spring, MD). Data were compared by one-way or two-way ANOVA in conjunction with a post hoc multiple comparison test (Scheffé’s test) when necessary. Statistical significance was considered when P < 0.05. Transformation of data in log values was sometimes required when high differences in means between groups resulted in nonproportional variances.
Results
Survival of follicular cells after oocyte loss
At the time of irradiation, i.e. at 5 dpn, the ovary contained primordial follicles located in the cortex as well as growing follicles at the primary and preantral stages mainly distributed in the core of the ovary (data not shown). We showed previously that the irradiation at 5 dpn induced within 24 h the depletion of nearly all oocytes in primordial follicles and to a lesser extent in small primary follicles, whereas the growing follicles were unaffected (24).
Double-immunocytochemical studies for KIT used as an oocyte marker (32) and for CK-8, which is characteristic of granulosa cells of follicles at the primordial/primary stages (33) in control and irradiated ovaries, carried out from 6 to 15 dpn, confirmed the disappearance of oocytes in the cortical area of irradiated ovaries (shown at 6 dpn, Fig. 1, A–D). In control ovaries, CK-8/KIT immunolabeling was localized in primordial/primary follicles in the cortical area at all studied ages (Fig. 1, A and C). In the irradiated ovaries from 6 dpn onward, in which no KIT labeling was detected in the cortical area as a result of oocyte depletion, a strong CK-8 labeling was observed (Fig. 1, B and D). This result indicates that despite oocyte loss, granulosa cells in primordial/early primary follicles persisted in the irradiated ovaries.
FIG. 1. Maintenance of primordial/primary follicles devoid of oocyte. A–D, Double immunofluorescence studies of KIT (A and B) and CK-8 (C and D) in control and irradiated ovaries at 6 dpn. In the control ovary, the antigen KIT is localized in the oocyte, whereas CK-8 is present in granulosa cells of primordial/primary follicles (arrowheads). In the irradiated ovary, the depletion of the oocyte and the maintenance of granulosa cells of primordial/primary follicles is reflected by the negative staining for KIT and the positive staining for CK-8 (arrowheads). In both groups, growing follicles (GF) with KIT immunolabeled oocytes and no CK-8-positive granulosa cells are present toward the center of the ovaries. E and F, Histological sections of control and irradiated ovaries at 8 dpn. In the cortical area containing the primordial follicles in the control ovary (arrowheads in E, inset), numerous ODFs, which are composed of cuboidal cells surrounded by a BM are observed in the irradiated ovary (arrowheads in F, inset). In both cases, growing follicles (GF) are located in the center of the ovary. G and H, Electron microscopic comparison between a normal primordial follicle (G) and an ODF (H) at 15 dpn. G, Granulosa cell; Oo, oocyte; Cy, Cytoplasm; N, nucleus; Nu, nucleolus. Scale bars in A–D, 30 μm; E and F, 100 μm; insets, 20 μm. Magnification in G and H, x2560.
Observations at the histological (Fig. 1, E and F) and ultrastructural (Fig. 1, G and H) levels showed that similar to primordial/early primary follicles of control ovaries (Fig. 1, E and G), the ODFs observed in irradiated ovaries were composed of two to eight granulosa cells per section, limited by a continuous BM and surrounded by conjunctive cells (Fig. 1, F and H). They also exhibited an irregular nucleus containing one nucleolus as well as clusters of heterochromatin located mainly along the inner membrane of the nuclear envelope (Fig. 1, G and H). As a result of oocyte loss, however, granulosa cells in ODFs displayed an expanded cytoplasm and a cuboidal shape when compared with squamous granulosa cells in primordial/early primary follicles of control ovaries (Fig. 1, E–H).
Morphological modifications of follicular cells after oocyte depletion
The evolution of ODFs was followed in subsequent ages until 24 wk. As soon as 12 dpn, some ODFs exhibited cells with an elongated cytoplasm and a round nucleus located near the BM (shown at 2 wk in Fig. 2A). With advancing age, the elongation of the cytoplasm became more and more pronounced and the number of cells increased, ranging from around eight cells per section in prepubertal females to around 30 cells in 24-wk females (Fig. 2B). Within these epithelial structures, cells were most commonly arranged in a single layer and no lumen was observed at any ages studied. No epithelial structures containing both cuboidal and elongated cells were observed. This would indicate that, in a given ODF, all granulosa cells were submitted simultaneously to the morphological modifications. These atypical epithelial structures, which were never observed in control ovaries (Fig. 2, C and D), displayed gross morphological characteristics of SCO tubules of sterile prepubertal male rats (Fig. 2E).
FIG. 2. Morphological alterations of follicular cells after oocyte loss. A–D, Histological sections of irradiated and control ovaries at 2 (A and C) and 18 (B and D) wk. At both ages in irradiated ovaries, tODFs with cells exhibiting an elongated cytoplasm and a basally located nucleus (arrows in A and B) are observed in the vicinity of ODFs (arrowheads in A and B). tODFs are delineated by a continuous BM and surrounded by conjunctive cells (A and B). Note that the size and the polarization of cells within tODFs increase with age (compare A and B). In control females, the ovaries that contain primordial (arrowheads) and growing follicles (GF) do not possess ODFs or tODFs. E, Histological sections of SCO tubules at 2 wk. Note the close morphological resemblance between immature Sertoli cells and cells contained within tODFs. F, Semithin section of a tODF in a 24-wk ovary, also shown at the ultrastructural level in G and H. G and H, The BM surrounding the tODFs is indicated (G). The ectoplasmic specialization (ES) in G is shown at a higher magnification in H. It is formed by tight junction (TJ) and parallel-arranged laminae of rough endoplasmic reticulum (RER). N, Nucleus; Nu, nucleolus; TJ, tight junction, RER, rough endoplasmic reticulum. Scale bars, A, B, and E, 25 μm and insets, 100 μm. C, 30 μm; D, 60 μm. Magnification in F, x350; G, x2,156; in H, x21,560.
At the ultrastructural level, the ovoid to elongated nuclei of the atypical ovarian cells essentially located in the basal part of the epithelial structures (Fig. 2, F and G) exhibited heterochromatin clusters along the nuclear membrane and a large well-defined nucleolus frequently located in the center (Fig. 2 and G). Specialized cell junctions were observed in the oldest studied females, i.e. at 24 wk (Fig. 2, G and H). They resembled ectoplasmic specializations (ES), which are typical Sertoli/Sertoli cell junctions, participating to the blood-testis barrier (BTB) and formed by occluding junctions surrounded by circumferential actin filaments in the cytoplasmic side and bordered by a rough endoplasmic reticulum cistern (34, 35). Because they share similar morphological features with Sertoli cells, the atypical elongated ovarian cells will be referred to as Sertoli-like-cells and the ODFs containing such cells as tODFs.
Molecular markers of granulosa/Sertoli cell phenotypes
To further characterize the phenotype of cells within tODF, we undertook a study of molecular markers expressed by the somatic lineage of the gonad, specific of the ovary [Forkhead L2 (FOXL2)] (36, 37), the testis [Sry-type high-mobility-group box transcription factor (SOX-9), Osp] (38, 39, 40) or gonads of both sexes (AMH) (41).
FOXL2 is expressed in somatic cells of the ovary as soon as in fetal life (35, 36). After birth, it is found in granulosa cells of follicles from the primordial follicle stage onward and, to a lesser extent, in mesenchymal cells (36, 37) (Fig. 3A and Table 3). In contrast, its expression is undetectable in both fetal and postnatal testis of different animal species (36, 37, 42) and in the rat (our own data, not shown). A substantial nuclear staining was observed in follicular cells contained within ODFs (Fig. 3B and Table 3) and also in Sertoli-like cells of tODF (Fig. 3C and Table 3). The expression of Sox-9, a factor that is characteristic of Sertoli cells from the onset of sexual differentiation (38, 39) (Fig. 3D), was studied at the mRNA and protein levels, using in situ hybridization (ISH) and immunohistochemistry. Its expression was detected in neither tODFs nor ODFs (Fig. 3E and Table 3 and data not shown). The expression of Osp (or claudin-11), encoding a protein involved in the edification of the BTB (40), was examined using RT-PCR and ISH. At all studied ages, quantitative RT-PCR analyses showed that the relative levels of Osp transcripts were low in both control and irradiated ovaries when compared with SCO tubules of immature males (Fig. 3F). In the irradiated ovaries at 24 wk, however, the relative levels of Osp transcripts were increased up to 4- to 5-fold, compared with the ovarian levels at previous ages (Fig. 3F; P < 0.05). When studied by ISH, Osp transcripts were widely expressed in SCO tubules of immature rats (Fig. 3G) and were absent from control ovaries (Table 3 and data not shown). Osp transcripts were detected in the Sertoli-like cells of the largest tODFs observed in the ovary from 20 wk onward (Fig. 3H and Table 3) and were absent from ODFs (Table 3 and data not shown).
FIG. 3. Expression of follicular and Sertoli cell markers. A–C, Immunocytochemistry for FOXL2 in control (A) and irradiated ovaries (B and C). FOXL2 is detected in the granulosa cell nuclei of primordial (arrowhead, A) and growing follicles (GF) and follicular cells of ODFs (arrowheads, B) and Sertoli-like cells of tODFs (open arrow, C). D and E, ISH for Sox-9 in SCO tubules of a prepubertal male (D) and an adult irradiated ovary (E). Sox-9 transcripts are observed in Sertoli cells of sterile male (open arrows, D) but not Sertoli-like cells of tODFs (open arrows, E). F–H, Studies of the expression of Osp. Quantitative PCR analysis of Osp mRNA levels (F) in control (C) and irradiated (I) ovaries at the indicated ages and in SCO tubules of immature males (M). Each point represents the result obtained for a sample, and the bar represents the mean of data in each group. Localization of Osp mRNA by ISH (G and H). Osp transcripts are observed in Sertoli cells of immature SCO tubules (G, arrow) and Sertoli-like cells of the largest tODFs (H, arrow) but not ODFs and small tODFs (H, arrowhead) in irradiated ovaries of females aged 20 wk. I, TUNEL assay at 6 dpn and J–M, immunohistochemistry for AMH at 6 dpn (J), 2 wk (K and L), and adult females (M). In micrographs I and J, TUNEL assay and AMH immunocytochemistry were performed on the same tissue section. At 6 dpn, i.e. 24 h after irradiation, follicles with a degenerating TUNEL-positive oocyte (arrowhead and dotted lines, I and J) or ODF (small open arrow and dotted lines in I and J) do not express AMH, in contrast to GF with a healthy oocyte (arrow, I and J). At 2 wk, AMH expression is absent in small ODFs (arrowheads, K) and appears in larger ones (arrows, L). Sertoli-like cells of tODFs show a strong staining for AMH (large open arrows, M). Scale bars, 20 μm.
TABLE 3. Summary of the expression pattern of markers in ODF and tODF studied by ISH or immunohistochemistry
In the course of gonadal development, AMH is expressed by fetal and immature Sertoli cells as well as by granulosa cells from the primary follicle stage onward (43, 44) (Table 3 and Fig. 3J). In the first 24 h after irradiation, follicles devoid of oocytes or displaying a degenerating oocyte as visualized by TUNEL assay did not express AMH (Fig. 3, I and J). In subsequent days, AMH expression was not detected in the smallest ODFs (Fig. 3K and Table 3) and appeared as ODFs grew in size (seven to eight cells per section) (Fig. 3L). In Sertoli-like cells of tODFs, AMH was widely expressed (Fig. 3M and Table 3). Thus, AMH expression was initiated in follicular cells of ODFs after they had reached a certain developmental stage, and it was maintained thereafter through the transdifferentiation process.
Dynamics of ODF and tODF
Histological counts of ODF and tODF were performed through the prepubertal period and after puberty until 24 wk (Fig. 4). At 2 wk, irradiated ovaries contained numerous ODFs, with a marked variability in their number between females, ranging from around 300 to 1000 per ovary. Very few tODFs were encountered at this age.
FIG. 4. Evolution of the population of ODFs and tODFs. Histological counts of ODFs () and tODFs () were performed from 2 to 24 wk as described in Materials and Methods. Bars, Means ± SEM of three to five ovaries. Different letters indicate significant differences between ages (two-way ANOVA). The gray line represents the evolution of tODF proportion, expressed as the number of tODFs divided by the total number of both ODFs and tODFs.
The number of ODFs decreased progressively with age (r = 0.8; P < 0.001) and reached very low values at 24 wk, compared with those observed in younger animals (P < 0.05). Conversely, the number of tODFs increased progressively as the animal aged (r = 0.85, P < 0.001). There was a significant inverse correlation between ODFs and tODFs numbers (r = –0.65, P = 0.0021, n = 24) that resulted in the increase in the proportion of tODFs, expressed as the number of tODFs divided by the total number of tODFs plus ODFs, as animals advanced in age (Fig. 4, gray line).
To determine whether oocyte loss affected proliferation of both follicular and Sertoli-like cells, cellular proliferation was assessed by immunocytochemistry after a 1-wk BrdU administration in either the prepubertal period or adulthood. BrdU-positive granulosa cells were observed in several primordial/primary follicles in control ovaries (Fig. 5A) and all growing follicles in both control and irradiated ovaries (Fig. 5, A–C). In contrast, a very limited number of ODFs displayed BrdU-positive cells in prepubertal (Fig. 5B) and adult animals (data not shown). In addition, tODFs displayed only occasional BrdU-positive Sertoli-like-cells in prepubertal (Fig. 5C) and adult animals (data not shown). The process of cellular apoptosis was examined by TUNEL assay. As expected, TUNEL-positive granulosa cells were observed in follicles beyond the preantral follicle stage and particularly in antral follicles in both control and irradiated females (data not shown). No TUNEL-positive cells were observed in the ODFs and tODFs examined (data not shown).
FIG. 5. Markers of follicular functionality in ODFs and tODFs. A–C, Immunocytochemistry for BrdU in control (A) and irradiated ovaries (B and C). BrdU labeling is widely observed in granulosa cells of growing follicles (GF) and scarce in primordial follicles (arrowheads, A). Most of the cells in ODFs are BrdU negative (B, arrowheads), and some cells in tODFs (C, open arrow) are BrdU positive. D–F, ISH for inh in control (D) and irradiated ovaries (E and F). Inh is absent or faintly expressed in primordial follicles (arrowheads, D) but is unequivocally present from the primary follicle stage (arrows, D). Inh staining is near the background level in small ODFs (arrowheads, E) and moderate to high in larger ones (arrows, E). In contrast, all tODFs exhibit inh staining (open arrow, F). G–I, Immunocytochemistry for ER? in control (G) and irradiated ovaries (H and I). ER? is expressed in growing follicles (GF in G and H) from the preantral stage onward. Diffuse cytoplasmic staining is observed in oocytes and granulosa cells of both primordial/primary follicles (arrowhead, G) and ODFs (arrowhead, H). Diffuse cytoplasmic and nuclear staining is detected in cells within tODFs (open arrow, I). J–L, Immunocytochemistry for AR in control (J) and irradiated ovaries (K and L). AR is absent from primordial/early primary follicles (arrowheads, J) and appears in growing follicles (GF) from the large primary stage. AR is absent from ODF (arrowheads, K), but it is detected in some cells of tODFs (open arrow, L). Scale bars, 20 μm.
Functionality of ODFs and tODFs
The phenotypic alteration of follicular cells after oocyte loss led us to investigate their functionality. In situ analyses of the tissue localization of the transcripts of inhibin -subunit (inh), activin/inhibin ?A-subunit (inh?A), and Cyp19a1 (encoding the enzyme P450-aromatase) as well as the proteins 3?-hydroxysteroid dehydrogenase (HSD), - and ?-isoforms of estrogen receptors (ER and ER?), and androgen receptor (AR), which are all known to be expressed in relation to folliculogenesis progression, were carried out by ISH or immunohistochemistry studies (Fig. 5, D–L and Table 3). Inh was found in granulosa cells from the primary stage onward (Fig. 5D), as expected (45). In ODFs, the expression pattern of inh ranged from negative to marked staining (Fig. 5E). In contrast, all the tODFs examined were stained for inh (Fig. 5F). The inh?A subunit, Cyp19a1, and 3?-HSD, which were all expressed in granulosa cells from the antral stage onward, were absent from ODFs and tODFs (Table 3 and data not shown).
ER? was strongly expressed in the nucleus of granulosa cells in growing follicles from the preantral follicle stage onward, whereas a weak immunostaining was observed in the cytoplasm of granulosa cells of primordial/primary follicles and in oocytes (Fig. 5G and Table 3), as previously described (46). Some cells within ODFs (Fig. 5H and Table 3) displayed cytoplasmic immunostaining for ER?. A diffused immunostaining for ER? was found in the nucleus and the cytoplasm of cells in some tODFs (Fig. 5I and Table 3). ER was expressed in the nuclei of both thecal and interstitial cells (Table 3 and data not shown), as previously described (46). No specific immunostaining was observed within follicles, ODFs or tODFs (Table 3 and data not shown). AR was widely expressed in the ovary and granulosa cells from the large primary follicle stage onward as well as interstitial cells (Fig. 5J and Table 3) as previously described (47). No staining was observed in ODFs (Fig. 5K and Table 3); in contrast, some cells within tODFs displayed marked AR nuclear staining (Fig. 5L and Table 3).
Involvement of the endocrine status
As a result of primordial follicle depletion in the neonatal period, irradiated females displayed premature ovarian failure (POF) (24). This impaired reproductive capacity was accompanied by a progressive increase in plasma FSH levels, which were significantly elevated from 8 wk onward when compared with those of control females and younger irradiated females (Fig. 6A). Plasma LH levels displayed great variability within each group, and no significant differences were observed between irradiated and control females (Fig. 6B). Plasma inhibin A and B levels were markedly decreased in irradiated females when compared with control females at the two ages studied, i.e. 12 and 24 wk (Fig. 6, C and D). Plasma E2 levels were unaltered at all ages studied in irradiated females when compared with control females (Fig. 6E). The levels of circulating T, assayed at 24 wk, were slightly elevated in irradiated females when compared with those of controls (Fig. 6F).
FIG. 6. Plasma concentrations of FSH (A), LH (B), inhibin A (C), inhibin B (D), E2 (E), and testosterone (F) in control () and irradiated () females. FSH and LH were assayed from 4 to 24 wk, inhibin A (InhA) and B (InhB) at 12 and 24 wk, E2 from 8 to 24 wk, and T at 24 wk. Bars, Means ± SEM of plasma samples assayed from four to 11 females. Asterisks, Significant differences between control and irradiated females; different letters, significant differences within a group (P < 0.05, two-way ANOVA in A–E; one-way ANOVA in F).
We investigated the possible involvement of the endocrine environment and more specifically of elevated FSH levels in the transdifferentiation process. From 8 wk onward, irradiated females were injected with a long-acting GnRH agonist (aGnRH), known to durably reduce gonadotropin levels (30). When studied at 12 wk, aGnRH-injected females displayed a 2-fold decrease in plasma FSH levels when compared with saline-injected females (20.4 ± 1.4 ng/ml in aGnRH-injected females vs. 55.6 ± 5.7, respectively, n = 5 female per group; P < 0.05), indicating that the treatment was effective. Reduction in gonadotropin levels accelerated the depletion of growing follicles, and very few healthy growing follicles or corpora lutea were observed when compared with those in saline-injected irradiated females (Fig. 7A, insets). Strikingly, the treatment induced a significant reduction in the proportion of tODF when compared with that in saline-injected females (Fig. 7B). The administration of T to aGnRH-treated females did not significantly change the proportion of tODF (Fig. 7B). In aGnRH-injected females repeatedly treated with hCG, the proportion of tODFs was not significantly different from either that of aGnRH-injected or saline-injected females (Fig. 7B). Interestingly, when aGnRH-injected females were repeatedly treated with r-hFSH, the proportion of tODFs was significantly increased, compared with that of females injected with aGnRH alone, and reached the same levels as those in saline-treated females (Fig. 7B). Moreover in saline- and aGnRH-injected females, there was a significant positive correlation between the levels of circulating FSH and the number of tODFs (r = 0.480; P < 0.05; n = 11). The correlation was even stronger when FSH plasma levels were plotted against the proportion of tODFs (r = 0.79; P < 0.05; n = 11).
FIG. 7. Influence of the endocrine balance on tODF proportion. A, Histology showing ovarian morphology at 12 wk of saline-injected and aGnRH-injected females. In saline-injected females, the ovaries contain growing follicles and corpora lutea (inset). In the cortical area, numerous tODFs (arrows) and a few ODFs (arrowhead) are observed. In aGnRH-injected females, the ovaries are small and occasionally contain follicles (inset). In the cortical area, ODFs predominate (arrowheads) and tODFs are of small size (arrow). Proportion of tODFs in saline-injected females () and after administration of aGnRH alone (–) or in combination with T, hCG, or r-hFSH (). Bars, Means ± SEM. Different letters indicate significant differences between treatment groups (P < 0.05, two-way ANOVA). Scale bars, A: 20 μm; insets, 100 μm.
Discussion
Up to the present study, only indirect evidence supported the hypothesis that in mammals, oocyte loss leads to the development of seminiferous-like cords (14, 17). We have directly addressed this question by examining the consequences of a selective oocyte-depletion in primordial/small primary follicles.
Acquisition of characteristics of immature Sertoli cells by follicular cells after oocyte loss
No morphological alterations of follicular cells from ODFs were observed until 1 wk after oocyte loss. From around 2 wk, histological and ultrastructural analyses have shown that follicular cells in a number of ODFs had acquired several morphological features of immature Sertoli cells. These cells displayed an elongated cytoplasm and a basally located nucleus; with advancing age, they became highly polarized as a result of expanding cytoplasmic volume. In their nuclei, the nucleolus was prominent but never showed a tripartite structure, which is normally observed in Sertoli cells of rat testes from around 2–5 wk and characterizes the final phase of nuclear differentiation (48). Ectoplasmic specializations (ESs), which are specific tight junctions of Sertoli cells developing from 2–3 wk in prepubertal males (49), were observed in the oldest studied females (24 wk). In line with this observation are the results obtained in our study of Osp, encoding a protein known to be constitutive of ES (40). Osp transcripts were detected in Sertoli-like cells of the largest tODFs from 20 wk onward, a result that is consistent with those obtained in quantitative RT-PCR showing that in irradiated ovaries their levels significantly increased between 12 and 24 wk. Altogether, these results show that follicular cells survive after oocyte loss and that they subsequently transdifferentiate into immature Sertoli-like-cells maturing slowly as animals advance in age.
Recruitment and maturation of follicular cells after oocyte loss
Throughout their development, follicular cells of ODFs and Sertoli-like-cells of tODFs express the protein encoded by FoxL2, an early sexually dimorphic gene expressed in the ovary that is considered to be involved in the regulation of granulosa cell differentiation at the primordial follicle stage (36, 37, 42, 50). The expression of FOXL2 in Sertoli-like cells of tODFs indicates, thus, that these cells maintained a follicular phenotype.
The expression of AMH, which is normally initiated from the onset of follicular growth (44), was observed in follicular cells of the largest ODFs and thereafter maintained in Sertoli-like cells of tODFs. Given that irradiation induced the death of oocytes contained in nongrowing follicles, the expression of AMH in the largest ODFs may well indicate that their follicular cells mature similarly to those of follicles entering the growing pool. As females advanced in age, there was a constant decline in ODF number (in parallel to the increase in tODF number), which is reminiscent of the progressive exhaustion of the nongrowing follicle pool taking place in normal ovaries consecutively to both follicular recruitment and atresia. The evolution in the number of both ODFs and tODFs demonstrates that the transdifferentiation of follicular cells did not occur immediately after oocyte loss and that only a fraction of ODFs was converting into tODFs at any given time. Thus, our findings strongly suggest that oocyte loss by itself is not sufficient to promote cellular switching from a granulosa to a Sertoli cell phenotype. The exit of ODFs from a nongrowing state to a growing state would be an additional prerequisite. This is in agreement with a study carried out on mouse ovarian grafts showing that pregranulosa cells must attain a certain state of maturation before converting into Sertoli-like cells (23).
As shown by our experiments of BrdU uptake, few ODFs in contrast to nongrowing follicles showed mitotic follicular cells, indicating that the proliferation capacity of follicular cells was impaired after oocyte loss. These results are in line with reports demonstrating that close interactions between oocyte and granulosa cells are essential to direct follicular development (51). During their development, ODFs acquired both inh and AMH expression but did not express AR and ER? present in growing follicles from the large primary (AR) and preantral stage (ER?). In tODFs, Sertoli-like cells expressed all these factors but not those expressed in follicles from the antral stage onward (inh?A subunit, 3?-HSD, and Cyp19a1).
Altogether, our results provide direct evidence that in the absence of the oocyte, follicular cells of nongrowing follicles (primordial/early primary stage) can initiate their differentiation and acquire some functional characteristics of follicular cells of early growing follicles (large primary/preantral stage). Thus, follicular cell maturation proceeds, at least to a certain extent, in a cell autonomous manner. However, it is also possible that paracrine factors produced by the remaining growing follicles could be involved in the maturation of follicular cells of ODFs. It has been proposed, for example, that the regulation of primordial follicle recruitment could operate in an interfollicular fashion via the production of AMH by growing follicles (52).
The transdifferentiation process in oocyte depleted follicles is not dependent on SOX-9
In our model, in agreement with the hypothesis essentially based on freemartin cattle and ovarian graft studies (14, 17), oocyte loss is associated with the acquisition of morphological characteristics of Sertoli cells by granulosa cells. In mammals, one of the first genes activated in male gonads after Sry expression is Sox9, which is both sufficient and required for male testis differentiation and can induce the formation of testis cords in XX fetal gonad (53, 54, 55). Sox9 is required for the expression of Amh by Sertoli cells (56, 57), but additional functions during sex determination remain elusive. The presence of Sox9 transcripts has been reported in several cases of ovarian sex reversal, as in transplants of fetal mouse ovaries (39) and postnatal sex reversed ovaries of mice deficient in both types of functional ERs, ER and ER? (ER? knockout mice) (58, 59), as well as females incapable of synthesizing estrogens due to disruption of the Cyp19a1 gene encoding the P450-aromatase enzyme (ArKO mice) (60). In addition to confirming the presence of Sertoli cells, these results suggested that SOX-9 could mediate the process of ovarian sex reversal (61, 62). The lack of SOX-9 expression in both follicular cells of ODFs and Sertoli-like cells of tODFs indicates that in our model the transdifferentiation process is not mediated by SOX-9 itself. It raises the possibility that other factors, perhaps acting downstream SOX-9, are implicated in the transdifferentiation process induced by oocyte loss. It has long been thought that AMH had a masculinizing influence on the ovarian somatic lineage as in its presence testis-like cords develop within fetal ovaries (20, 63). In our study, we have shown that AMH was expressed in follicular cells of ODFs and Sertoli-like cells of tODFs. AMH, however, is not a male sex-specific protein because it is expressed in the postnatal ovary (41). In addition, it is now commonly accepted that AMH is involved indirectly in ovarian sex reversal by exerting a lethal effect on female germ cells (14, 17, 64, 65). It is thus unlikely that in ODFs AMH promotes granulosa cell transdifferentiation, but we cannot rule out that AMH has an effect on the differentiation of follicular cells in the absence of oocyte.
It is noticeable that oocyte loss does not always lead to the transdifferentiation of granulosa cells into Sertoli cells. For instance, in antral or preovulatory follicles the degeneration or the experimental removal of the oocyte induces the luteinization of granulosa cells (66, 67, 68). Similarly, a mutation of the growth differentiation factor 9 gene (Gdf9) by homologous recombination, which leads to the loss of the oocyte in primary follicles, results in the morphological and functional luteinization of the surrounding granulosa cells (69, 70). Granulosa cells of Gdf9–/– primary follicles fail to proliferate but acquire before oocyte loss follicular markers normally expressed from the preantral stage onwards (70). In Fecundity X Inverdale sheep carrying a point mutation in the bone morphogenetic protein-15 coding sequence, after the degeneration of the oocyte in primary follicles no alteration in the morphology of granulosa cells has been reported (71, 72). Granulosa cells in primary follicles of Inverdale ewes fail to proliferate as in Gdf9–/– mice, yet they acquire follicular markers of preantral follicles only after oocyte loss (72). These differences in the effect of oocyte loss on granulosa cells could well depend on the animal species and/or also the state of differentiation of the granulosa cells at the time of oocyte loss.
FSH as a potential activator of granulosa cell transdifferentiation
As a result of the early depletion of oocytes contained in primordial follicles, irradiated females displayed POF from the very beginning of reproductive life (24). We report here that the levels of circulating FSH were positively correlated with the proportion of tODFs. Interestingly, GnRH-deficient hypogonadal mice displaying very low plasma levels of gonadotropins exhibit 1 yr after irradiation ovaries composed of very numerous single-layered follicles devoid of oocytes (73), which morphologically resemble the ODFs described in the present study. Our preliminary data together with the results obtained in irradiated GnRH-deficient hypogonadal mice have led us to further investigate the potential influence of the endocrine imbalance and particularly high levels of FSH on the transdifferentiation process. We found that in irradiated females with lowered circulating gonadotropin levels after administration of a long-acting aGnRH, the proportion of tODFs was significantly decreased, compared with that of saline-treated females. In contrast, the repeated administration of r-hFSH, but neither that of hCG nor testosterone, to aGnRH-treated females efficiently prevented the decrease in tODF proportion. Thus, our findings give strong evidence that high levels of circulating FSH occurring during the POF could play a role in the transdifferentiation process. Nevertheless, our results cannot demonstrate whether FSH acts directly or indirectly by either promoting the cellular switching of follicular cells into Sertoli-like cells or preventing apoptosis of Sertoli-like cells. In a complementary analysis of aGnRH-treated ovaries by TUNEL assay, no apoptotic Sertoli-like cells were detected (data not shown). It is possible, however, that cell death occurred at the beginning of the treatment and was not detected at the time of observation.
The mechanism of action of FSH on follicular cells of ODFs and Sertoli-like cells of tODFs remains to be elucidated. In males, FSH is not involved in the fetal differentiation of Sertoli cells but stimulates their proliferation in perinatal life (74). Moreover, FSH regulates the maturation of Sertoli cells in prepubertal life (75). Its role, however, on the differentiation of the BTB has not been clearly established (76, 77). In females, early follicular growth can occur in the absence of FSH (78, 79) and FSH receptors are absent until follicles reach the primary to small preantral stages (80, 81). Given that ODF and tODF cells displayed maturational characteristics of granulosa cells of growing follicles, they might also develop FSH responsiveness. Because FSH stimulates growth, differentiation, and survival of follicles from the preantral follicular stage (82), it is tempting to speculate that FSH exerts such effects on maturing cells of ODFs and tODFs.
Is there a role for estrogen signaling pathway in our model?
Whereas the involvement of estrogens in the ovarian determining pathway has not been clearly established in eutherian mammals, the presence of Sertoli-like cells in the postnatal ovaries of ER? knockout and ArKO mice (58, 59, 60, 61, 62) underlines that at least in mice, estrogens may play a critical role in the maintenance of the phenotype of the female somatic compartment. Consistent with these findings, phytoestrogen diet or E2 administration prevents or reverses the transdifferentiation of Sertoli-like cells in ArKO mice (60, 62). In both ArKO and ER? knockout mice, the endocrine environment is affected, as shown by the increase in FSH, LH, and T plasma levels in ArKO and the increase in LH and T in ER? knockout mice (83, 84). This raises the possibility that the lack of estrogen signaling could be indirectly involved in the development of Sertoli-like cells for instance by altering endocrine parameters that are normally regulated by estrogens.
In irradiated females, the levels of E2 remained unaltered and, thus, they are unlikely to account for the increase in tODF number with age. As our immunocytochemistry studies revealed no expression of ER and ER? isoforms in follicular cells of ODFs, it may be hypothesized that oocyte loss and the lack of estrogen could together be involved in triggering transdifferentiation of follicular cells into Sertoli-like cells. Accordingly, in both ArKO and ER? mice, there are no germ cells in the tubular structures containing Sertoli-like cells, and in ER? knockout mice, it has been reported that Sertoli-like cells develop within atretic follicles, in which oocyte/granulosa cell interaction are disrupted (59, 60, 61). On the other hand, it must be underlined that several different gonadal features, besides Sox9 expression, exist between our model and both ArKO and ER? knockout mice. For instance, Leydig-like cells differentiate in ovaries of ArKO females (60) and Sertoli-like cells develop from granulosa cells of antral follicle in ER? knockout mice (61).
Conclusion
Altogether, our results provide evidence that in the absence of the oocyte, follicular cells of nongrowing follicles transdifferentiate into immature Sertoli-like cells. The recruitment of follicular cells of nongrowing follicles and their transdifferentiation into Sertoli-like cells was initiated in prepubertal life and continued throughout adulthood. The highly variable interval between oocyte loss and the conversion of follicular cells into Sertoli-like cells between ODFs underlines that cellular switching is not triggered by oocyte loss alone but may require a multistep process, which includes the recruitment and maturation of follicular cells within ODF and involves FSH regulation (Fig. 8).
FIG. 8. Presumptive mechanism leading to the transdifferentiation of granulosa cells into Sertoli-like cells after oocyte loss. In normal primordial follicles, recruitment is regulated by close interactions between oocytes and their surrounding granulosa cells as well as factors produced by growing follicles. The onset of follicular growth is characterized by the onset of AMH and inh expression in granulosa cells. In oocyte-depleted follicles, granulosa cells survive and, as indicated by AMH and inh expression, enter growth, a process that could be regulated by factors produced by the remaining growing follicles. Then granulosa cells transdifferentiate into Sertoli-like cells, possibly under regulation by high plasma FSH levels. GF, Growing follicle; Ooc., oocyte.
This ability of follicular cells to switch into their testis counterparts may reflect the similar fetal origin of these two cell types as has been suggested in other cases of cellular plasticity (85). The mechanism(s) involved in the transdifferentiation of granulosa cells into Sertoli-like cells after oocyte loss remain(s) to be resolved. One may hypothesize that the oocyte produces yet-unidentified factors that act on follicular cells to prevent their conversion into Sertoli-like cells. In line with this hypothesis are the findings suggesting that during fetal life, meiotic germ cells could reinforce the ovarian pathway by antagonizing the testis differentiation pathway (12). Further insights into the mechanisms involved in bidirectional communication between the oocyte and the companion somatic cells as well as the discovery of new oocyte factors will possibly bring answers concerning the mechanisms involved in the maintenance of the granulosa cell phenotype.
Acknowledgments
The authors thank O. Locquet, M. P. Monneret, M. Delacroix, M. Robin, C. Salvard, and E. Etienne for helpful technical assistance; C. Roulin and Dr. C. Bleux for -irradiation (Institut Curie); Drs. J. L. Lefaix and H. Coffigny for -irradiation (La Ferme Atomique); Dr. S. Brailly-Tabard for inhibin dosages; and Professor F. Jaubert for critical advices. The authors also thank Dr. S. Carreau for providing AR antibodies, Dr. M. Fellous for FOXL2 antibodies, Dr. B. Vigier for AMH antibodies, Dr. R. Guennoun for 3?-HSD antibodies, and Dr. F. Poulat for SOX-9 antibodies.
References
Koopman P, Gubbay J, Vivian N, Goodfellow P, Lovell-Badge R 1991 Male development of chromosomally female mice transgenic for SRY. Nature 351:117–121
Gubbay J, Collignon J, Koopman P, Capel B, Economou A, Munsterberg A, Vivian N, Goodfellow P, Lovell-Badge R 1990 A gene mapping to the sex-determining region of the mouse Y chromosome is a member of a novel family of embryonically expressed genes. Nature 346:245–250
Hawkins JR 1993 Mutational analysis of SRY in XY females. Hum Mut 2:347–350
McElreavey K, Vilain E, Abbas N, Herskowitz I, Fellous M 1993 A regulatory cascade hypothesis for mammalian sex determination: SRY represses a negative regulator of male development. Proc Natl Acad Sci USA 90:3368–3372
Vainio S, Heikkila M, Kispert A, Chin N, McMahon AP 1999 Female development in mammals is regulated by Wnt-4 signalling. Nature 397:405–409
Pailhoux E, Vigier B, Chaffaux S, Servel N, Taourit S, Furet JP, Fellous M, Grosclaude F, Cribiu EP, Cotinot C, Vaiman D 2001 An 11.7-kb deletion triggers intersexuality and polledness in goats. Nat Genet 29:453–458
Dolci S, De Felici M 1990 A study of meiosis in chimeric mouse fetal gonads. Development 109:37–40
McLaren A, Southee D 1997 Entry of mouse embryonic germ cells into meiosis. Dev Biol 187:107–113
Nagamine C, Carlisle C 1996 The dominant white spotting oncogene allele Kit(W-42J) exacerbates XY(DOM) sex reversal. Development 122:3597–3605
Nagamine CM, Capehart J, Carlisle C, Chang D 1998 Ovotestes in B6-XXSxr sex-reversed mice. Dev Biol 196:24–32
Adams I, McLaren A 2002 Sexually dimorphic development of mouse primordial germ cells: switching from oogenesis to spermatogenesis. Development 129:1155–1164
Yao HH, DiNapoli L, Capel B 2003 Meiotic germ cells antagonize mesonephric cell migration and testis cord formation in mouse gonads. Development 130:5895–5902
Burgoyne PS, Baker TG 1985 Perinatal oocyte loss in XO mice and its implications for the aetiology of gonadal dysgenesis in XO women. J Reprod Fertil 75:633–645
McLaren A 1991 Development of the mammalian gonad: the fate of the supporting cell lineage. Bioessays 13:151–156
Merchant H 1975 Rat gonadal and ovarian organogenesis with and without germ cells. An ultrastructural study. Dev Biol 44:1–21
Merchant-Larios H, Centeno B 1981 Morphogenesis of the ovary from the sterile W/Wv mouse. Prog Clin Biol Res 59B:383–392
Whitworth DJ 1998 XX germ cells: the difference between an ovary and a testis. Trends Endocrinol Metab 9:2–6
Jost A, Vigier B, Prépin J, Perchellet JP 1973 Studies on sex differentiation in mammals. Rec Progr Horm Res 29:1–41
Charpentier G, Magre S 1990 Masculinizing effect of testes on developing rat ovaries in organ culture. Development 110:839–849
Vigier B, Watrin F, Magre S, Tran D, Josso N 1987 Purified bovine AMH induces a characteristic freemartin effect in fetal rat prospective ovaries exposed to it in vitro. Development 100:43–55
Crumeyrolle-Arias M, Scheib D, Aschheim P 1976 Light and electron microscopy of the ovarian interstitial tissue in the senile rat: normal aspect and response to HCG of "deficiency cell" and "epithelial cords." Gerontology 22:185–204
Taketo T, Merchant-Larios H, Koide SS 1984 Induction of testicular differentiation in a fetal mouse ovary by transplantation into adult male mice. Proc Soc Exp Biol Med 176:148–153
Taketo-Hosotani T, Merchant-Larios H, Thau RB, Koide SS 1985 Testicular cell differentiation in fetal mouse ovaries following transplantation into adult male mice. J Exp Zool 236:229–237
Guigon CJ, Mazaud S, Forest MG, Brailly-Tabard S, Coudouel N, Magre S 2003 Unaltered development of the initial follicular waves and normal pubertal onset in female rats after neonatal deletion of the follicular reserve. Endocrinology 144:3651–3662
Beaumont H 1960 Changes in the radiosensitivity of the testis during foetal development. Int J Radiat Biol 2:247–256
Tuchmann-Duplessis H 1947 Une technique de coloration commode pour la numeration des cellules glandulaires de l’hypophyse. Bulletin Hist Appl 9:180–181
Lozach A, Garrel G, Lerrant Y, Berault A, Counis R 1998 GnRH-dependent up-regulation of nitric oxide synthase I level in pituitary gonadotrophs mediates cGMP elevation during rat proestrus. Mol Cell Endocr 143:43–51
Forest MG, David M, Lecoq A, Jeune M, Bertrans J 1980 Kinetics of the HCG-induced steroidogenic response of the human testis. III. Studies in children of the plasma levels of testosterone and HCG: rationale for testicular stimulation test. Pediatr Res 14:819–824
Sharpe RM, Turner KJ, McKinnell C, Groome NP, Atanassova N, Millar MR, Buchanan DL, Cooke PS 1999 Inhibin B levels in plasma of the male rat from birth to adulthood: effect of experimental manipulation of Sertoli cell number. J Androl 20:94–101
Lerrant Y, Kottler M, Bergametti F, Moumni M, Blumberg-Tick J, Counis R 1995 Expression of gonadotropin-releasing hormone (GnRH) receptor gene is altered by GnRH agonist desensitization in a manner similar to that of gonadotropin ?-subunit genes in normal and castrated rat pituitary. Endocrinology 136:2803–2808
Bukovsky A, Ayala ME, Dominguez R, Keenan JA, Wimalasena, McKenzie PP, Caudle MR 2000 Postnatal androgenization induces premature aging of rat ovaries. Steroids 65:190–205
Manova K, Nocka K, Besmer P, Bachvarova R 1990 Gonadal expression of c-kit encoded at the W locus of the mouse. Development 110:1057–1069
Fridmacher V, Locquet O, Magre S 1992 Differential expression of acidic cytokeratins 18 and 19 during sexual differentiation of the rat gonad. Development 115:503–517
Gilula N, Fawcett D, Aoki A 1976 The Sertoli cell occluding junctions and gap junctions in mature and developing mammalian testis. Dev Biol 50:142–168
Russell L 1977 Observations on rat Sertoli ectoplasmic (‘junctional’) specializations in their association with germ cells of the rat testis. Tissue Cell 9:475–498
Cocquet J, Pailhoux E, Jaubert F, Servel N, Xia X, Pannetier M, De Baere E, Messiaen L, Cotinot C, Fellous M, Veitia RA 2002 Evolution and expression of FOXL2. J Med Genet 39:916–922[Free Full Text]
Schmidt D, Ovitt CE, Anlag K, Fehsenfeld S, Gredsted L, Treier AC, Treier M 2004 The murine winged-helix transcription factor Foxl2 is required for granulosa cell differentiation and ovary maintenance. Development 131:933–942
Kent J, Wheatley SC, Andrews JE, Sinclair AH, Koopman P 1996 A male-specific role for SOX9 in vertebrate sex determination. Development 122:2813–2822
Morais da Silva S, Hacker A, Harley V, Goodfellow P, Swain A, Lovell-Badge R 1996 Sox9 expression during gonadal development implies a conserved role for the gene in testis differentiation in mammals and birds. Nat Genet 14:62–68
Gow A, Southwood C, Li J, Pariali M, Riordan G., Brodie S, Danias J, Bronstein J, Kachar B, Lazzarini R 1999 CNS myelin and Sertoli cell tight junction strands are absent in Osp/claudin-11 null mice. Cell 99:649–659
Vigier B, Picard JY, Tran D, Legeai L, Josso N 1984 Production of anti-Müllerian hormone: another homology between Sertoli and granulosa cells. Endocrinology 114:1315–1320
Loffler KA, Zarkower D, Koopman P 2003 Etiology of ovarian failure in blepharophimosis ptosis epicanthus inversus syndrome: FOXL2 is a conserved, early-acting gene in vertebrate ovarian development. Endocrinology 144:3237–3243
Josso N, Cate RL, Picard JY, Vigier B, di Clemente N, Wilson C, Imbeaud S, Pepinsky RB, Guerrier D, Boussin L, Legeai L, Carre-Eusebe D 1993 Anti-mullerian hormone: the Jost factor. Rec Progr Horm Res 48:1–59
Baarends WM, Uilenbroek JT, Kramer P, Hoogerbrugge JW, van Leeuwen EC, Themmen AP, Grootegoed JA 1995 Anti-mullerian hormone and anti-mullerian hormone type II receptor messenger ribonucleic acid expression in rat ovaries during postnatal development, the estrous cycle, and gonadotropin-induced follicle growth. Endocrinology 136:4951–4962
Drummond A, Dyson M, Mercer J, Findlay J 1996 Differential responses of post-natal rat ovarian cells to FSH and activin. Mol Cell Endocrinol 30:21–32
Sar M, Welsch F 1999 Differential expression of estrogen receptor-? and estrogen receptor- in the rat ovary. Endocrinology 140:963–971
Tetsuka M, Whitelaw P, Bremner W, Millar M, Smyth CD, Hillier SG 1995 Developmental regulation of androgen receptor in rat ovary. J Endocrinol 145:535–543
Ramos AJ, Dym M 1979 Ultrastructural differentiation of rat Sertoli cells. Biol Reprod 21:909–922
Meyer R, Posalky Z, McGinley D 1977 Intercellular junction development in maturing rat seminiferous tubules. J Ultrastruct Res 61:271–283
Uda M, Ottolenghi C, Crisponi L, Garcia JE, Deiana M, Kimber W, Forabosco A, Cao A, Schlessinger D, Pilia G 2004 Foxl2 disruption causes mouse ovarian failure by pervasive blockage of follicle development. Hum Mol Gen 13:1171–1181
Eppig J 2001 Oocyte control of ovarian follicular development and function in mammals. Reproduction 122:829–838
Durlinger AL, Gruijters MJ, Kramer P, Karels B, Ingraham HA, Nachtigal MW, Uilenbroek JT, Grootegoed JA, Themmen AP 2002 Anti-Mullerian hormone inhibits initiation of primordial follicle growth in the mouse ovary. Endocrinology 143:1076–1084
Bishop CE, Whitworth DJ, Qin Y, Agoulnik AI, Agoulnik IU, Harrison WR, Behringer RR, Overbeek PA 2000 A transgenic insertion upstream of sox9 is associated with dominant XX sex reversal in the mouse. Nat Genet 26:490–494
Vidal VP, Chaboissier MC, de Rooij DG, Schedl A 2001 Sox9 induces testis development in XX transgenic mice. Nat Genet 28:216–217
Chaboissier MC, Kobayashi A, Vidal VI, Lutzkendorf S, Van D, Kant HJ, Wegner M, De Rooij DG, Behringer RR, Schedl A 2004 Functional analysis of Sox8 and Sox9 during sex determination in the mouse. Development 131:1891–1901
De Santa Barbara P, Bonneaud N, Boizet B, Desclozeaux M, Moniot B, Sudbeck P, Scherer G, Poulat F, Berta P 1998 Direct interaction of SRY-related protein SOX9 and steroidogenic factor 1 regulates transcription of the human anti-Mullerian hormone gene. Mol Cell Biol 18:6653–6665
Arango NA, Lovell-Badge R, Behringer RR 1999 Targeted mutagenesis of the endogenous mouse Mis gene promoter: in vivo definition of genetic pathways of vertebrate sexual development. Cell 99:409–419
Couse JF, Hewitt SC, Bunch DO, Sar M, Walker VR, Davis BJ, Korach KS 1999 Postnatal sex reversal of the ovaries in mice lacking estrogen receptors and ?. Science 286:2328–2331
Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M 2000 Effect of single and compound knockouts of estrogen receptors (ER) and ? (ER?) on mouse reproductive phenotypes. Development 127:4277–4291
Britt KL, Kerr J, O’Donnell L, Jones ME, Drummond AE, Davis SR, Simpson ER, Findlay JK 2002 Estrogen regulates development of the somatic cell phenotype in the eutherian ovary. FASEB J 16:1389–1397
Dupont S, Dennefeld C, Krust A, Chambon P, Mark M 2003 Expression of Sox9 in granulosa cells lacking the estrogen receptors, ER and ER?. Dev Dyn 226:103–106
Britt K, Stanton P, Misso M, Simpson E, Findlay J 2004 The effects of estrogen on the expression of genes underlying the differentiation of somatic cells in the murine gonad. Endocrinology 145:3950–3960
Behringer RR, Cate RL, Froelick GJ, Palmiter RD, Brinster RL 1990 Abnormal sexual development in transgenic mice chronically expressing Müllerian inhibiting substance. Nature 345:167–170
Taketo T, Saeed J, Manganaro T, Takahashi M, Donahoe PK 1993 Müllerian inhibiting substance production associated with loss of oocytes and testicular differentiation in the transplanted mouse XX gonadal primordium. Biol Reprod 49:13–23
Whitworth DJ, Shaw G, Renfree MB 1996 Gonadal sex reversal of the developing marsupial ovary in vivo and in vitro. Development 122:4057–4063
Pincus G, Enzmann EV 1935 The comparative behavior of mammalian eggs in vivo and in vitro I. The activation of ovarian eggs. J Exp Med 62:655–675
Falck B 1959 Site of production of oestrogen in rat ovary as studied in microtransplants. Acta Physiol Scand 47:1–101
El-Fouly MA, Cook B, Nekola M, Nalbandov AV 1970 Role of the ovum in follicular luteinization. Endocrinology 87:288–293
Dong J, Albertini DF, Nishimori K, Kumar TR, Lu N, Matzuk MM 1996 Growth differentiation factor-9 is required during early ovarian folliculogenesis. Nature 383:531–535
Elvin JA, Yan C, Wang P, Nishimori K, Matzuk MM 1999 Molecular characterization of the follicle defects in the growth differentiation factor 9-deficient ovary. Mol Endocrinol 13:1018–1034
Braw-Tal R, McNatty KP, Smith P, Heath DA, Hudson NL, Phillips DJ, McLeod BJ, Davis GH 1993 Ovaries of ewes homozygous for the X-linked Inverdale gene (FecXI) are devoid of secondary and tertiary follicles but contain many abnormal structures. Biol Reprod 49:895–907
Juengel JL, Quirke LD, Tisdall DJ, Smith P, Hudson NL, McNatty KP 2000 Gene expression in abnormal ovarian structures of ewes homozygous for the inverdale prolificacy gene. Biol Reprod 62:1467–1478
Tennent B, Beamer W 1986 ovarian tumors not induced by irradiation and gonadotropins in hypogonadal (hpg) mice. Biol Reprod 34:751–760
Orth J 1984 The role of follicle-stimulating hormone in controlling Sertoli cell proliferation in testes of fetal rats. Endocrinology 115:1248–1255
Johnston H, Baker P, Abel M, Charlton H, Jackson G, Fleming L, Kumar TR, O Shaughnessy PJ 2004 Regulation of Sertoli cell number and activity by follicle-stimulating hormone and androgen during postnatal development in the mouse. Endocrinology 145:318–329
De Kretser D, Burger H, Fortune D, Hudson B, Long A, Paulsen C, Taft H 1972 Hormonal, histological and chromosomal studies in adult males with testicular disorders. J Clin Endocrinol Metab 35:392–401
Johnson M 1970 The pituitary and the blood-testis barrier. J Reprod Fertil 22:181–182
Kumar TR, Wang Y, Lu N, Matzuk MM 1997 Follicle stimulating hormone is required for ovarian follicle maturation but not male fertility. Nat Genet 15:201–204
Abel M, Wootton A, Wilkins V, Huhtaniemi I, Knight P, Charlton H 2000 The effect of a null mutation in the follicle-stimulating hormone receptor gene on mouse reproduction. Endocrinology 141:1795–1803
Mayerhofer ADG, Costa ME, Ojeda SR 1997 A role for neurotransmitters in early follicular development: induction of functional follicle-stimulating hormone receptors in newly formed follicles of the rat ovary. Endocrinology 138:3320–3329
Rannikki A, Zhang F, Huhtaniemi I 1995 Ontogeny of follicle-stimulating hormone receptor gene expression in the rat testis and ovary. Mol Cell Endocrinol 107:199–208
McGee EA, Perlas E, LaPolt PS, Tsafriri A, Hsueh AJ 1997 Follicle-stimulating hormone enhances the development of preantral follicles in juvenile rats. Biol Reprod 57:990–998
Fisher CR, Graves KH, Parlow AF, Simpson ER 1998 Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp 19 gene. Proc Natl Acad Sci USA 95:6965–6970
Couse JF, Yates MM, Walker VR, Korach KS 1999 Characterization of the hypothalamo-pituitary-gonadal axis in estrogen receptor (ER) null mice reveals hypergonadism and endocrine sex reversal in females lacking ER but not ER?. Mol Endocrinol 17:1039–1053
Tosh D, Slack J 2002 How cells change their phenotype. Nat Rev Mol Cell Biol 3:187–194
Lane E, Goodman S, Trejdosiewicz L 1982 Disruption of the keratin filament network during epithelial cell division. EMBO J 1:1365–1372(Céline J. Guigon, No?llin)
Address all correspondence and requests for reprints to: Dr. S. Magre, Laboratoire de Physiologie et Physiopathologie, Centre National de la Recherche Scientifique-Unité Mixte de Recherche 7079, Université Paris VI, 7 Quai Saint Bernard, 75005 Paris, France. E-mail: solange.magre@snv.jussieu.fr.
Abstract
Although it has been suggested that in mammals the loss of female germ cells may induce the masculinization of the ovarian compartment, there has been as yet no conclusive demonstration. To directly address that question, the present study has been designed to determine the fate of follicular cells after oocyte loss. Using -irradiation to selectively deplete oocytes in nongrowing follicles in female rats, we show that follicular cells in oocyte-depleted follicles survive, proliferate, and subsequently acquire morphological characteristics of Sertoli cells: elongated cytoplasm, basal location of the nucleus, and specific Sertoli cell junctions, the ectoplasmic specializations. These Sertoli-like cells express, however, the female-specific marker FOXL2 (Forkhead L2) but not the male sex-specific marker SOX-9 (Sry-type high-mobility-group box transcription factor-9) underlying the maintenance of molecular characteristics of granulosa cells. Before transdifferentiating into Sertoli-like cells, follicular cells of oocyte-depleted follicles initiate the expression of anti-Mullerian hormone and inhibin -subunit that are typically synthesized by granulosa cells from the onset of follicular growth. Experimental modifications of the endocrine balance of the irradiated females show that there is a close relationship between plasma FSH levels and the occurrence of Sertoli-like cells. In addition to providing experimental evidence for the crucial role of the oocyte in granulosa cell phenotype maintenance, these results emphasize that the transdifferentiation of granulosa cells into Sertoli cells occurs in a multistep fashion, requiring the maturation of granulosa cells and depending on the endocrine milieu.
Introduction
THE GONAD ARISES as a bipotential primordium with the ability to develop into either an ovary or a testis. Both somatic and primordial germ cells (PGCs) follow a dimorphic fate once sex determination occurs. In mammals, testicular differentiation is governed by the Sry (sex-determining region, Y chromosome) gene, which initiates the differentiation of Sertoli cells and their organization into seminiferous cords (1). In the absence of Sry, the gonad differentiates into an ovary (2, 3). However, the hypothesis arguing that the ovarian development is the default pathway of gonadal differentiation is now controversial (4, 5, 6).
PGCs develop according to their sex somatic cell environment. It has been proposed that whereas in females PGCs enter meiosis in a cell autonomous manner, in males an as-yet-unidentified signal produced by Sertoli cells prevents PGCs from entering meiosis (7, 8). In addition, although germ cells in males are dispensable for the formation of testis cords, there are several lines of evidence suggesting that they may well reinforce Sertoli cell differentiation (9, 10) by acting via a feedback loop mediated by paracrine factors (11). Moreover, the active inhibition of meiosis would be essential because PGCs committed to meiosis could antagonize the differentiation of testis cords (12). In females, germ cells are essential for directing the ovarian pathway because no follicle differentiates in their absence (13, 14). If germ cells are destroyed or lacking, as in busulfan-treated rats (15), or in We/We mice (16), respectively, the female gonad is constituted by undifferentiated pregranulosa cells arranged in cords that subsequently degenerate. In some cases, the loss of female germ cells either at the mitotic or meiotic stage has been associated with the differentiation of testis-like cords within fetal ovaries (14, 17).
Ovarian sex reversal was first reported in the germ cell-depleted gonad of the freemartin bovine (18), which is a female exposed in utero to the testicular hormones, i.e. anti-Müllerian hormone (AMH) and testosterone, of her male cotwin. Differentiation of testis-like cords in association with germ cell depletion was also observed in rat fetal ovaries cultured in the presence of fetal testes (19) or the presumptive masculinisating factor AMH (20) as well as in aging rat ovaries (21) and grafts of fetal mouse ovaries under the renal capsule of adult hosts (22, 23). To our knowledge, however, there is still no direct evidence that female germ cell loss induces ovarian sex reversal. One may hypothesize that the loss of germ cells could be the consequence of the transdifferentiation of the ovarian supporting cells rather than its cause. The purpose of the present study was to determine the fate of follicular cells in the absence of meiotic germ cells and in particular whether under such conditions follicular cells transdifferentiate into Sertoli-like cells. Taking advantage of a rat experimental model in which oocytes in nongrowing follicles are selectively depleted by -irradiation at 5 d postnatal (dpn) (24), we studied the subsequent development of the surrounding follicular cells at both morphological and molecular levels. We report that after oocyte loss, follicular cells mature comparably to follicular cells in early growing follicles and thereafter acquire some morphological characteristics of immature Sertoli cells, presumably under the control of the endocrine environment and more specifically of FSH.
Materials and Methods
Animals and irradiation
Pregnant female rats from the Sprague Dawley strain were purchased from Charles River Laboratories (L’Arbresle, France). Males at 0 dpn and females at 5 dpn were whole body exposed to a -irradiation using either Cobalt or Cesium source at a dose rate of 0.8 Gy/min, with a final dose of 1.5 Gy. The exposure of males at 0 dpn induces the death of most germ cells and leads to the development of Sertoli cell-only tubules (SCO) (25). From 4 wk, animals were weaned and housed at four to five per cage. Animals were maintained in standard lighting (12 h light, 12 h dark) and temperature (22 C) with food and water ad libitum. All the animal procedures were approved and performed according to the 1986 European Communities Council Directives and those of the Ministère de l’Agriculture et de la Forêt.
Tissue collection
Females and males were killed by decapitation in the afternoon. For adult control and irradiated females, daily vaginal smears were performed. Animals were killed and blood was collected in the afternoon of the estrus. Blood was collected in heparinized tubes, centrifuged at 3000 x g for 15 min, and collected plasma was stored at –20 C until assayed for hormones. The ovaries and testes were either snap frozen in liquid nitrogen and stored at –80 C or processed for further morphological studies.
Tissue processing
For histological examination of ovarian and testicular morphology, organs were placed in Bouin’s fixative for at least 24 h. Sections of 5 μm thickness were stained according to the protocol of Tuchmann du Plessis (26). For in situ hybridization, immunohistochemistry, and terminal deoxynucleotidyltransferase-mediated deoxyuridine 5-triphosphate-fluorescein nick end labeling (TUNEL) assay, ovaries and testes were fixed in 2% paraformaldehyde-PBS and further processed as described (24). For immunohistochemistry using 5-bromo-2'-deoxyuridine (BrdU) and AMH antibodies, ovaries were fixed overnight in 2% paraformaldehyde-PBS, rinsed several times in PBS buffer, and dehydrated in alcohol. They were then embedded in paraffin wax and sectioned at 5 μm thickness.
For ultrastructural studies, ovaries of females aged 5 dpn and 1, 2, 8, 12, and 24 wk were cut into small pieces and fixed in 2% paraformaldehyde and 0.5% glutaraldhehyde in 0.1 M cacodylate buffer (pH 7.2) overnight at 4 C. They were rinsed in the same buffer, postfixed for 2 h in 2% osmium tetroxide in 0.1 M cacodylate buffer at room temperature, dehydrated through a graded series of alcohol, and embedded in epoxy embedding resin (Fluka, Sigma-Aldrich Corp., St. Louis, MO). Semithin sections (1 μm) were stained with methylene blue. Thin sections were mounted on grids and stained with uranyl acetate and lead citrate. Electron micrographs were taken on a EM906 electron microscope (Zeiss, G?ttinger, Germany).
In situ hybridization
Digoxigenin-labeled sense and antisense riboprobe synthesis as well as in situ hybridizations were performed as previously described (24). Primers used for the synthesis of the riboprobes are reported in Table 1.
TABLE 1. Primers used for in situ hybridization and real time PCR
BrdU injections
Control and irradiated females received three ip injections of 50 μl BrdU (Sigma-Aldrich) dissolved in saline (5 mg/ml) (one injection every 2 d) in the prepubertal period (from 8 to 14 dpn) or in adulthood (from 90 to 96 dpn) and were killed the day after the last injection.
Immunohistochemistry and TUNEL assay
Immunohistochemistry studies were performed on frozen sections except those for BrdU and AMH that were performed on paraffin sections. Paraffin sections were deparaffinized in toluene and rehydrated in decreasing concentrations of alcohol. Tissue sections were then boiled for 5 min in 0.1 M citrate buffer (pH 6.0). For both frozen and paraffin sections, endogenous peroxidase activity was blocked by treatment in H2O2 for 10 min (blocking reagents, Dako Corp., Carpinteria, CA). Nonspecific binding was then blocked in PBS/10% BSA for 20 min. Sections were incubated overnight at 4 C with the primary antibody (Table 2) diluted in blocking solution (PBS/0.05% BSA), followed by incubation for 1 h with the appropriate secondary biotinylated antibody (Vector Laboratories, Burlingame, CA) and 1 h with a peroxidase-conjugated streptavidin horseradish complex (ABC Elite kit, Vector). The reaction product was developed using 3, 3'-diaminobenzidine tetrahydrochloride (Dako). For negative controls (not shown), primary antibody was omitted. For double immunofluorescence, sections were blocked for 20 min. in PBS/10% BSA and then incubated with the antibody against cytokeratin 8 (CK-8) diluted in blocking solution overnight at 4 C. Sections were washed three times in PBS for 5 min each and then incubated for 2 h with fluorescein isothiocyanate-conjugated mouse secondary antibodies (dilution 1:100; Amersham, Les Ulis, France) at room temperature. Sections were rinsed three times and incubated at room temperature for 2 h with the primary antibody against the tyrosine kinase receptor KIT, rinsed, and incubated for 1 h with rabbit biotinylated secondary antibodies (dilution 1:300, Amersham), rinsed and incubated for 1 h in Extravidin (dilution 1:100; Sigma-Aldrich). Sections were rinsed and mounted in Vectashield (Vector). Fluorescence labeling was observed with a Zeiss epifluorescence microscope.
TABLE 2. Primary antibodies used for immunohistochemistry
TUNEL assay was performed on frozen sections as previously described (24).
RNA isolation, reverse transcription (RT), and real-time PCR
Total RNA was isolated from ovaries and testes using the RNeasy Protect miniextraction kit (QIAGEN SA, Courtaboeuf, France) following the manufacturer’s instructions. The quality of extracted RNA was verified by spectrophotometry and electrophoresis on agarose gel. For each sample, RT was performed with 0.5 μg total RNA incubated with random primers (Promega Corp., Madison, MI) using a Superscript II kit (Invitrogen Corp., Carlsbad, CA) according to the manufacturer’s instructions.
Real-time PCR analysis of oligodendrocyte-specific protein gene (Osp) and the ribosomal gene L19 (Rpl19), used as an internal control, was performed using the iCycler system (Bio-Rad Laboratories, Hercules, CA). Briefly, PCR was performed in 96-well microtiter plates with the QuantiTect SYBR Green PCR Kit (QIAGEN), using 5 μl of the cDNA from the RT reactions diluted at 1:100. PCR was performed as follows: 15 min initial HotStart enzyme activation at 95 C, followed by 45 cycles of 95 C denaturation for 30 sec, annealing at 58 C for 30 sec, and extension at 72 C for 30 sec. At the end of the PCR, melting curve analysis was performed to verify product specificity by increasing the temperature from 55 to 95 C at 0.1 C/sec. Melting point analysis demonstrated a single melting breakpoint in all experiments performed. In each reaction a standard curve was obtained from serial dilutions of a mix of RT products from ovaries and testes. For each experiment, all samples were run in triplicate. The cycle threshold values obtained from each sample were collected and data were analyzed by normalizing the levels of Osp levels to those of the internal control Rpl19. The specificity of the primers for both Osp and Rpl19 was verified after migration of cDNA on agarose gel. A single product of 339 and 196 bp were amplified for Osp and Rpl19, respectively, and the cDNA were sequenced. No bands were seen in mock reactions in which reverse transcriptase has been omitted.
Histological counts
Counts of the epithelial structures observed in the irradiated ovaries (see Results) were carried out on every fifth ovarian section from the prepubertal period until adulthood with careful examination of neighboring sections to avoid double counting of the same epithelial structure. Epithelial structures were scored as oocyte-depleted follicles (ODFs) when they displayed cuboidal somatic cells enveloped by a basement membrane (BM) with no germ cell and they were scored as transdifferentiated oocyte depleted follicles (tODFs) when they displayed polarized cells with an elongated cytoplasm and a nucleus located toward the BM.
Hormone assays
Plasma 17?-estradiol (E2) and testosterone (T) as well as LH and FSH were assayed by specific RIA as previously described (27, 28). Plasma E2 and T were assayed by specific RIA after ethyl ether extraction, followed by chromatographic purification on Celite columns. Plasma LH and FSH were assayed using RIA kits purchased from the National Institute of Diabetes and Digestive and Kidney Diseases (Baltimore, MD) with highly purified rat LH (rLH-I-10) and FSH (rFSH-I-9) for iodination, reference preparations (rLH-RP3 and rFSH-RP2), and appropriate antisera (anti-rLH-S11 and anti-rFSH-S11). Plasma inhibin A and B were determined using a two-site ELISA specific for each peptide (Oxford Bio-Innovation, Oxon, UK) as previously described (29). Intraassay coefficients of variation were 4.3, 4.9, 5.1, 7, and 5% and interassay coefficients of variation were 10, 12, 11.2, 12.5, and 9–10% for inhibin A, inhibin B FSH, LH, and steroids, respectively. The sensitivities of the assays were 7 and 10 pg/ml, 0.1 and 3.0 ng/ml, and 3 and 5 pg/tube for inhibin A, inhibin B, FSH, LH, E2, and T, respectively.
Hormonal treatments
Adult-irradiated females were divided into five groups (n = 5 females/group). At 8 and 10 wk of age, females in groups 1–4 received an im injection of the long-acting GnRH agonist triptorelin (Decapeptyl, generously provided by Beaufour-Ipsen Industrie, Dreux, France; 300 μg/kg body weight), known to reduce durably the gonadotropin levels (30). In addition, females in groups 1–4 were injected sc every 2 d with: group 1, saline; group 2, 2 IU recombinant human FSH (r-hFSH, Puregon; generously provided by Organon, Serifontaine, France); group 3, 5 IU human chorionic gonadotropin (hCG Endo Pregnyl, generously provided by Organon). Females of group 4 received twice weekly a sc injection of 100 μg testosterone (Sigma-Aldrich) (31) dissolved in olive oil (Sigma-Aldrich). Females in group 5 received only saline injections. At the end of the experiment (12 wk of age), ovaries and blood were collected and processed as described above.
Statistical analysis
Data collected from at least three different animals were expressed as the mean ± SEM and analyzed using GBStat 6.0 software (Dynamic Microsystems, Inc., Silver Spring, MD). Data were compared by one-way or two-way ANOVA in conjunction with a post hoc multiple comparison test (Scheffé’s test) when necessary. Statistical significance was considered when P < 0.05. Transformation of data in log values was sometimes required when high differences in means between groups resulted in nonproportional variances.
Results
Survival of follicular cells after oocyte loss
At the time of irradiation, i.e. at 5 dpn, the ovary contained primordial follicles located in the cortex as well as growing follicles at the primary and preantral stages mainly distributed in the core of the ovary (data not shown). We showed previously that the irradiation at 5 dpn induced within 24 h the depletion of nearly all oocytes in primordial follicles and to a lesser extent in small primary follicles, whereas the growing follicles were unaffected (24).
Double-immunocytochemical studies for KIT used as an oocyte marker (32) and for CK-8, which is characteristic of granulosa cells of follicles at the primordial/primary stages (33) in control and irradiated ovaries, carried out from 6 to 15 dpn, confirmed the disappearance of oocytes in the cortical area of irradiated ovaries (shown at 6 dpn, Fig. 1, A–D). In control ovaries, CK-8/KIT immunolabeling was localized in primordial/primary follicles in the cortical area at all studied ages (Fig. 1, A and C). In the irradiated ovaries from 6 dpn onward, in which no KIT labeling was detected in the cortical area as a result of oocyte depletion, a strong CK-8 labeling was observed (Fig. 1, B and D). This result indicates that despite oocyte loss, granulosa cells in primordial/early primary follicles persisted in the irradiated ovaries.
FIG. 1. Maintenance of primordial/primary follicles devoid of oocyte. A–D, Double immunofluorescence studies of KIT (A and B) and CK-8 (C and D) in control and irradiated ovaries at 6 dpn. In the control ovary, the antigen KIT is localized in the oocyte, whereas CK-8 is present in granulosa cells of primordial/primary follicles (arrowheads). In the irradiated ovary, the depletion of the oocyte and the maintenance of granulosa cells of primordial/primary follicles is reflected by the negative staining for KIT and the positive staining for CK-8 (arrowheads). In both groups, growing follicles (GF) with KIT immunolabeled oocytes and no CK-8-positive granulosa cells are present toward the center of the ovaries. E and F, Histological sections of control and irradiated ovaries at 8 dpn. In the cortical area containing the primordial follicles in the control ovary (arrowheads in E, inset), numerous ODFs, which are composed of cuboidal cells surrounded by a BM are observed in the irradiated ovary (arrowheads in F, inset). In both cases, growing follicles (GF) are located in the center of the ovary. G and H, Electron microscopic comparison between a normal primordial follicle (G) and an ODF (H) at 15 dpn. G, Granulosa cell; Oo, oocyte; Cy, Cytoplasm; N, nucleus; Nu, nucleolus. Scale bars in A–D, 30 μm; E and F, 100 μm; insets, 20 μm. Magnification in G and H, x2560.
Observations at the histological (Fig. 1, E and F) and ultrastructural (Fig. 1, G and H) levels showed that similar to primordial/early primary follicles of control ovaries (Fig. 1, E and G), the ODFs observed in irradiated ovaries were composed of two to eight granulosa cells per section, limited by a continuous BM and surrounded by conjunctive cells (Fig. 1, F and H). They also exhibited an irregular nucleus containing one nucleolus as well as clusters of heterochromatin located mainly along the inner membrane of the nuclear envelope (Fig. 1, G and H). As a result of oocyte loss, however, granulosa cells in ODFs displayed an expanded cytoplasm and a cuboidal shape when compared with squamous granulosa cells in primordial/early primary follicles of control ovaries (Fig. 1, E–H).
Morphological modifications of follicular cells after oocyte depletion
The evolution of ODFs was followed in subsequent ages until 24 wk. As soon as 12 dpn, some ODFs exhibited cells with an elongated cytoplasm and a round nucleus located near the BM (shown at 2 wk in Fig. 2A). With advancing age, the elongation of the cytoplasm became more and more pronounced and the number of cells increased, ranging from around eight cells per section in prepubertal females to around 30 cells in 24-wk females (Fig. 2B). Within these epithelial structures, cells were most commonly arranged in a single layer and no lumen was observed at any ages studied. No epithelial structures containing both cuboidal and elongated cells were observed. This would indicate that, in a given ODF, all granulosa cells were submitted simultaneously to the morphological modifications. These atypical epithelial structures, which were never observed in control ovaries (Fig. 2, C and D), displayed gross morphological characteristics of SCO tubules of sterile prepubertal male rats (Fig. 2E).
FIG. 2. Morphological alterations of follicular cells after oocyte loss. A–D, Histological sections of irradiated and control ovaries at 2 (A and C) and 18 (B and D) wk. At both ages in irradiated ovaries, tODFs with cells exhibiting an elongated cytoplasm and a basally located nucleus (arrows in A and B) are observed in the vicinity of ODFs (arrowheads in A and B). tODFs are delineated by a continuous BM and surrounded by conjunctive cells (A and B). Note that the size and the polarization of cells within tODFs increase with age (compare A and B). In control females, the ovaries that contain primordial (arrowheads) and growing follicles (GF) do not possess ODFs or tODFs. E, Histological sections of SCO tubules at 2 wk. Note the close morphological resemblance between immature Sertoli cells and cells contained within tODFs. F, Semithin section of a tODF in a 24-wk ovary, also shown at the ultrastructural level in G and H. G and H, The BM surrounding the tODFs is indicated (G). The ectoplasmic specialization (ES) in G is shown at a higher magnification in H. It is formed by tight junction (TJ) and parallel-arranged laminae of rough endoplasmic reticulum (RER). N, Nucleus; Nu, nucleolus; TJ, tight junction, RER, rough endoplasmic reticulum. Scale bars, A, B, and E, 25 μm and insets, 100 μm. C, 30 μm; D, 60 μm. Magnification in F, x350; G, x2,156; in H, x21,560.
At the ultrastructural level, the ovoid to elongated nuclei of the atypical ovarian cells essentially located in the basal part of the epithelial structures (Fig. 2, F and G) exhibited heterochromatin clusters along the nuclear membrane and a large well-defined nucleolus frequently located in the center (Fig. 2 and G). Specialized cell junctions were observed in the oldest studied females, i.e. at 24 wk (Fig. 2, G and H). They resembled ectoplasmic specializations (ES), which are typical Sertoli/Sertoli cell junctions, participating to the blood-testis barrier (BTB) and formed by occluding junctions surrounded by circumferential actin filaments in the cytoplasmic side and bordered by a rough endoplasmic reticulum cistern (34, 35). Because they share similar morphological features with Sertoli cells, the atypical elongated ovarian cells will be referred to as Sertoli-like-cells and the ODFs containing such cells as tODFs.
Molecular markers of granulosa/Sertoli cell phenotypes
To further characterize the phenotype of cells within tODF, we undertook a study of molecular markers expressed by the somatic lineage of the gonad, specific of the ovary [Forkhead L2 (FOXL2)] (36, 37), the testis [Sry-type high-mobility-group box transcription factor (SOX-9), Osp] (38, 39, 40) or gonads of both sexes (AMH) (41).
FOXL2 is expressed in somatic cells of the ovary as soon as in fetal life (35, 36). After birth, it is found in granulosa cells of follicles from the primordial follicle stage onward and, to a lesser extent, in mesenchymal cells (36, 37) (Fig. 3A and Table 3). In contrast, its expression is undetectable in both fetal and postnatal testis of different animal species (36, 37, 42) and in the rat (our own data, not shown). A substantial nuclear staining was observed in follicular cells contained within ODFs (Fig. 3B and Table 3) and also in Sertoli-like cells of tODF (Fig. 3C and Table 3). The expression of Sox-9, a factor that is characteristic of Sertoli cells from the onset of sexual differentiation (38, 39) (Fig. 3D), was studied at the mRNA and protein levels, using in situ hybridization (ISH) and immunohistochemistry. Its expression was detected in neither tODFs nor ODFs (Fig. 3E and Table 3 and data not shown). The expression of Osp (or claudin-11), encoding a protein involved in the edification of the BTB (40), was examined using RT-PCR and ISH. At all studied ages, quantitative RT-PCR analyses showed that the relative levels of Osp transcripts were low in both control and irradiated ovaries when compared with SCO tubules of immature males (Fig. 3F). In the irradiated ovaries at 24 wk, however, the relative levels of Osp transcripts were increased up to 4- to 5-fold, compared with the ovarian levels at previous ages (Fig. 3F; P < 0.05). When studied by ISH, Osp transcripts were widely expressed in SCO tubules of immature rats (Fig. 3G) and were absent from control ovaries (Table 3 and data not shown). Osp transcripts were detected in the Sertoli-like cells of the largest tODFs observed in the ovary from 20 wk onward (Fig. 3H and Table 3) and were absent from ODFs (Table 3 and data not shown).
FIG. 3. Expression of follicular and Sertoli cell markers. A–C, Immunocytochemistry for FOXL2 in control (A) and irradiated ovaries (B and C). FOXL2 is detected in the granulosa cell nuclei of primordial (arrowhead, A) and growing follicles (GF) and follicular cells of ODFs (arrowheads, B) and Sertoli-like cells of tODFs (open arrow, C). D and E, ISH for Sox-9 in SCO tubules of a prepubertal male (D) and an adult irradiated ovary (E). Sox-9 transcripts are observed in Sertoli cells of sterile male (open arrows, D) but not Sertoli-like cells of tODFs (open arrows, E). F–H, Studies of the expression of Osp. Quantitative PCR analysis of Osp mRNA levels (F) in control (C) and irradiated (I) ovaries at the indicated ages and in SCO tubules of immature males (M). Each point represents the result obtained for a sample, and the bar represents the mean of data in each group. Localization of Osp mRNA by ISH (G and H). Osp transcripts are observed in Sertoli cells of immature SCO tubules (G, arrow) and Sertoli-like cells of the largest tODFs (H, arrow) but not ODFs and small tODFs (H, arrowhead) in irradiated ovaries of females aged 20 wk. I, TUNEL assay at 6 dpn and J–M, immunohistochemistry for AMH at 6 dpn (J), 2 wk (K and L), and adult females (M). In micrographs I and J, TUNEL assay and AMH immunocytochemistry were performed on the same tissue section. At 6 dpn, i.e. 24 h after irradiation, follicles with a degenerating TUNEL-positive oocyte (arrowhead and dotted lines, I and J) or ODF (small open arrow and dotted lines in I and J) do not express AMH, in contrast to GF with a healthy oocyte (arrow, I and J). At 2 wk, AMH expression is absent in small ODFs (arrowheads, K) and appears in larger ones (arrows, L). Sertoli-like cells of tODFs show a strong staining for AMH (large open arrows, M). Scale bars, 20 μm.
TABLE 3. Summary of the expression pattern of markers in ODF and tODF studied by ISH or immunohistochemistry
In the course of gonadal development, AMH is expressed by fetal and immature Sertoli cells as well as by granulosa cells from the primary follicle stage onward (43, 44) (Table 3 and Fig. 3J). In the first 24 h after irradiation, follicles devoid of oocytes or displaying a degenerating oocyte as visualized by TUNEL assay did not express AMH (Fig. 3, I and J). In subsequent days, AMH expression was not detected in the smallest ODFs (Fig. 3K and Table 3) and appeared as ODFs grew in size (seven to eight cells per section) (Fig. 3L). In Sertoli-like cells of tODFs, AMH was widely expressed (Fig. 3M and Table 3). Thus, AMH expression was initiated in follicular cells of ODFs after they had reached a certain developmental stage, and it was maintained thereafter through the transdifferentiation process.
Dynamics of ODF and tODF
Histological counts of ODF and tODF were performed through the prepubertal period and after puberty until 24 wk (Fig. 4). At 2 wk, irradiated ovaries contained numerous ODFs, with a marked variability in their number between females, ranging from around 300 to 1000 per ovary. Very few tODFs were encountered at this age.
FIG. 4. Evolution of the population of ODFs and tODFs. Histological counts of ODFs () and tODFs () were performed from 2 to 24 wk as described in Materials and Methods. Bars, Means ± SEM of three to five ovaries. Different letters indicate significant differences between ages (two-way ANOVA). The gray line represents the evolution of tODF proportion, expressed as the number of tODFs divided by the total number of both ODFs and tODFs.
The number of ODFs decreased progressively with age (r = 0.8; P < 0.001) and reached very low values at 24 wk, compared with those observed in younger animals (P < 0.05). Conversely, the number of tODFs increased progressively as the animal aged (r = 0.85, P < 0.001). There was a significant inverse correlation between ODFs and tODFs numbers (r = –0.65, P = 0.0021, n = 24) that resulted in the increase in the proportion of tODFs, expressed as the number of tODFs divided by the total number of tODFs plus ODFs, as animals advanced in age (Fig. 4, gray line).
To determine whether oocyte loss affected proliferation of both follicular and Sertoli-like cells, cellular proliferation was assessed by immunocytochemistry after a 1-wk BrdU administration in either the prepubertal period or adulthood. BrdU-positive granulosa cells were observed in several primordial/primary follicles in control ovaries (Fig. 5A) and all growing follicles in both control and irradiated ovaries (Fig. 5, A–C). In contrast, a very limited number of ODFs displayed BrdU-positive cells in prepubertal (Fig. 5B) and adult animals (data not shown). In addition, tODFs displayed only occasional BrdU-positive Sertoli-like-cells in prepubertal (Fig. 5C) and adult animals (data not shown). The process of cellular apoptosis was examined by TUNEL assay. As expected, TUNEL-positive granulosa cells were observed in follicles beyond the preantral follicle stage and particularly in antral follicles in both control and irradiated females (data not shown). No TUNEL-positive cells were observed in the ODFs and tODFs examined (data not shown).
FIG. 5. Markers of follicular functionality in ODFs and tODFs. A–C, Immunocytochemistry for BrdU in control (A) and irradiated ovaries (B and C). BrdU labeling is widely observed in granulosa cells of growing follicles (GF) and scarce in primordial follicles (arrowheads, A). Most of the cells in ODFs are BrdU negative (B, arrowheads), and some cells in tODFs (C, open arrow) are BrdU positive. D–F, ISH for inh in control (D) and irradiated ovaries (E and F). Inh is absent or faintly expressed in primordial follicles (arrowheads, D) but is unequivocally present from the primary follicle stage (arrows, D). Inh staining is near the background level in small ODFs (arrowheads, E) and moderate to high in larger ones (arrows, E). In contrast, all tODFs exhibit inh staining (open arrow, F). G–I, Immunocytochemistry for ER? in control (G) and irradiated ovaries (H and I). ER? is expressed in growing follicles (GF in G and H) from the preantral stage onward. Diffuse cytoplasmic staining is observed in oocytes and granulosa cells of both primordial/primary follicles (arrowhead, G) and ODFs (arrowhead, H). Diffuse cytoplasmic and nuclear staining is detected in cells within tODFs (open arrow, I). J–L, Immunocytochemistry for AR in control (J) and irradiated ovaries (K and L). AR is absent from primordial/early primary follicles (arrowheads, J) and appears in growing follicles (GF) from the large primary stage. AR is absent from ODF (arrowheads, K), but it is detected in some cells of tODFs (open arrow, L). Scale bars, 20 μm.
Functionality of ODFs and tODFs
The phenotypic alteration of follicular cells after oocyte loss led us to investigate their functionality. In situ analyses of the tissue localization of the transcripts of inhibin -subunit (inh), activin/inhibin ?A-subunit (inh?A), and Cyp19a1 (encoding the enzyme P450-aromatase) as well as the proteins 3?-hydroxysteroid dehydrogenase (HSD), - and ?-isoforms of estrogen receptors (ER and ER?), and androgen receptor (AR), which are all known to be expressed in relation to folliculogenesis progression, were carried out by ISH or immunohistochemistry studies (Fig. 5, D–L and Table 3). Inh was found in granulosa cells from the primary stage onward (Fig. 5D), as expected (45). In ODFs, the expression pattern of inh ranged from negative to marked staining (Fig. 5E). In contrast, all the tODFs examined were stained for inh (Fig. 5F). The inh?A subunit, Cyp19a1, and 3?-HSD, which were all expressed in granulosa cells from the antral stage onward, were absent from ODFs and tODFs (Table 3 and data not shown).
ER? was strongly expressed in the nucleus of granulosa cells in growing follicles from the preantral follicle stage onward, whereas a weak immunostaining was observed in the cytoplasm of granulosa cells of primordial/primary follicles and in oocytes (Fig. 5G and Table 3), as previously described (46). Some cells within ODFs (Fig. 5H and Table 3) displayed cytoplasmic immunostaining for ER?. A diffused immunostaining for ER? was found in the nucleus and the cytoplasm of cells in some tODFs (Fig. 5I and Table 3). ER was expressed in the nuclei of both thecal and interstitial cells (Table 3 and data not shown), as previously described (46). No specific immunostaining was observed within follicles, ODFs or tODFs (Table 3 and data not shown). AR was widely expressed in the ovary and granulosa cells from the large primary follicle stage onward as well as interstitial cells (Fig. 5J and Table 3) as previously described (47). No staining was observed in ODFs (Fig. 5K and Table 3); in contrast, some cells within tODFs displayed marked AR nuclear staining (Fig. 5L and Table 3).
Involvement of the endocrine status
As a result of primordial follicle depletion in the neonatal period, irradiated females displayed premature ovarian failure (POF) (24). This impaired reproductive capacity was accompanied by a progressive increase in plasma FSH levels, which were significantly elevated from 8 wk onward when compared with those of control females and younger irradiated females (Fig. 6A). Plasma LH levels displayed great variability within each group, and no significant differences were observed between irradiated and control females (Fig. 6B). Plasma inhibin A and B levels were markedly decreased in irradiated females when compared with control females at the two ages studied, i.e. 12 and 24 wk (Fig. 6, C and D). Plasma E2 levels were unaltered at all ages studied in irradiated females when compared with control females (Fig. 6E). The levels of circulating T, assayed at 24 wk, were slightly elevated in irradiated females when compared with those of controls (Fig. 6F).
FIG. 6. Plasma concentrations of FSH (A), LH (B), inhibin A (C), inhibin B (D), E2 (E), and testosterone (F) in control () and irradiated () females. FSH and LH were assayed from 4 to 24 wk, inhibin A (InhA) and B (InhB) at 12 and 24 wk, E2 from 8 to 24 wk, and T at 24 wk. Bars, Means ± SEM of plasma samples assayed from four to 11 females. Asterisks, Significant differences between control and irradiated females; different letters, significant differences within a group (P < 0.05, two-way ANOVA in A–E; one-way ANOVA in F).
We investigated the possible involvement of the endocrine environment and more specifically of elevated FSH levels in the transdifferentiation process. From 8 wk onward, irradiated females were injected with a long-acting GnRH agonist (aGnRH), known to durably reduce gonadotropin levels (30). When studied at 12 wk, aGnRH-injected females displayed a 2-fold decrease in plasma FSH levels when compared with saline-injected females (20.4 ± 1.4 ng/ml in aGnRH-injected females vs. 55.6 ± 5.7, respectively, n = 5 female per group; P < 0.05), indicating that the treatment was effective. Reduction in gonadotropin levels accelerated the depletion of growing follicles, and very few healthy growing follicles or corpora lutea were observed when compared with those in saline-injected irradiated females (Fig. 7A, insets). Strikingly, the treatment induced a significant reduction in the proportion of tODF when compared with that in saline-injected females (Fig. 7B). The administration of T to aGnRH-treated females did not significantly change the proportion of tODF (Fig. 7B). In aGnRH-injected females repeatedly treated with hCG, the proportion of tODFs was not significantly different from either that of aGnRH-injected or saline-injected females (Fig. 7B). Interestingly, when aGnRH-injected females were repeatedly treated with r-hFSH, the proportion of tODFs was significantly increased, compared with that of females injected with aGnRH alone, and reached the same levels as those in saline-treated females (Fig. 7B). Moreover in saline- and aGnRH-injected females, there was a significant positive correlation between the levels of circulating FSH and the number of tODFs (r = 0.480; P < 0.05; n = 11). The correlation was even stronger when FSH plasma levels were plotted against the proportion of tODFs (r = 0.79; P < 0.05; n = 11).
FIG. 7. Influence of the endocrine balance on tODF proportion. A, Histology showing ovarian morphology at 12 wk of saline-injected and aGnRH-injected females. In saline-injected females, the ovaries contain growing follicles and corpora lutea (inset). In the cortical area, numerous tODFs (arrows) and a few ODFs (arrowhead) are observed. In aGnRH-injected females, the ovaries are small and occasionally contain follicles (inset). In the cortical area, ODFs predominate (arrowheads) and tODFs are of small size (arrow). Proportion of tODFs in saline-injected females () and after administration of aGnRH alone (–) or in combination with T, hCG, or r-hFSH (). Bars, Means ± SEM. Different letters indicate significant differences between treatment groups (P < 0.05, two-way ANOVA). Scale bars, A: 20 μm; insets, 100 μm.
Discussion
Up to the present study, only indirect evidence supported the hypothesis that in mammals, oocyte loss leads to the development of seminiferous-like cords (14, 17). We have directly addressed this question by examining the consequences of a selective oocyte-depletion in primordial/small primary follicles.
Acquisition of characteristics of immature Sertoli cells by follicular cells after oocyte loss
No morphological alterations of follicular cells from ODFs were observed until 1 wk after oocyte loss. From around 2 wk, histological and ultrastructural analyses have shown that follicular cells in a number of ODFs had acquired several morphological features of immature Sertoli cells. These cells displayed an elongated cytoplasm and a basally located nucleus; with advancing age, they became highly polarized as a result of expanding cytoplasmic volume. In their nuclei, the nucleolus was prominent but never showed a tripartite structure, which is normally observed in Sertoli cells of rat testes from around 2–5 wk and characterizes the final phase of nuclear differentiation (48). Ectoplasmic specializations (ESs), which are specific tight junctions of Sertoli cells developing from 2–3 wk in prepubertal males (49), were observed in the oldest studied females (24 wk). In line with this observation are the results obtained in our study of Osp, encoding a protein known to be constitutive of ES (40). Osp transcripts were detected in Sertoli-like cells of the largest tODFs from 20 wk onward, a result that is consistent with those obtained in quantitative RT-PCR showing that in irradiated ovaries their levels significantly increased between 12 and 24 wk. Altogether, these results show that follicular cells survive after oocyte loss and that they subsequently transdifferentiate into immature Sertoli-like-cells maturing slowly as animals advance in age.
Recruitment and maturation of follicular cells after oocyte loss
Throughout their development, follicular cells of ODFs and Sertoli-like-cells of tODFs express the protein encoded by FoxL2, an early sexually dimorphic gene expressed in the ovary that is considered to be involved in the regulation of granulosa cell differentiation at the primordial follicle stage (36, 37, 42, 50). The expression of FOXL2 in Sertoli-like cells of tODFs indicates, thus, that these cells maintained a follicular phenotype.
The expression of AMH, which is normally initiated from the onset of follicular growth (44), was observed in follicular cells of the largest ODFs and thereafter maintained in Sertoli-like cells of tODFs. Given that irradiation induced the death of oocytes contained in nongrowing follicles, the expression of AMH in the largest ODFs may well indicate that their follicular cells mature similarly to those of follicles entering the growing pool. As females advanced in age, there was a constant decline in ODF number (in parallel to the increase in tODF number), which is reminiscent of the progressive exhaustion of the nongrowing follicle pool taking place in normal ovaries consecutively to both follicular recruitment and atresia. The evolution in the number of both ODFs and tODFs demonstrates that the transdifferentiation of follicular cells did not occur immediately after oocyte loss and that only a fraction of ODFs was converting into tODFs at any given time. Thus, our findings strongly suggest that oocyte loss by itself is not sufficient to promote cellular switching from a granulosa to a Sertoli cell phenotype. The exit of ODFs from a nongrowing state to a growing state would be an additional prerequisite. This is in agreement with a study carried out on mouse ovarian grafts showing that pregranulosa cells must attain a certain state of maturation before converting into Sertoli-like cells (23).
As shown by our experiments of BrdU uptake, few ODFs in contrast to nongrowing follicles showed mitotic follicular cells, indicating that the proliferation capacity of follicular cells was impaired after oocyte loss. These results are in line with reports demonstrating that close interactions between oocyte and granulosa cells are essential to direct follicular development (51). During their development, ODFs acquired both inh and AMH expression but did not express AR and ER? present in growing follicles from the large primary (AR) and preantral stage (ER?). In tODFs, Sertoli-like cells expressed all these factors but not those expressed in follicles from the antral stage onward (inh?A subunit, 3?-HSD, and Cyp19a1).
Altogether, our results provide direct evidence that in the absence of the oocyte, follicular cells of nongrowing follicles (primordial/early primary stage) can initiate their differentiation and acquire some functional characteristics of follicular cells of early growing follicles (large primary/preantral stage). Thus, follicular cell maturation proceeds, at least to a certain extent, in a cell autonomous manner. However, it is also possible that paracrine factors produced by the remaining growing follicles could be involved in the maturation of follicular cells of ODFs. It has been proposed, for example, that the regulation of primordial follicle recruitment could operate in an interfollicular fashion via the production of AMH by growing follicles (52).
The transdifferentiation process in oocyte depleted follicles is not dependent on SOX-9
In our model, in agreement with the hypothesis essentially based on freemartin cattle and ovarian graft studies (14, 17), oocyte loss is associated with the acquisition of morphological characteristics of Sertoli cells by granulosa cells. In mammals, one of the first genes activated in male gonads after Sry expression is Sox9, which is both sufficient and required for male testis differentiation and can induce the formation of testis cords in XX fetal gonad (53, 54, 55). Sox9 is required for the expression of Amh by Sertoli cells (56, 57), but additional functions during sex determination remain elusive. The presence of Sox9 transcripts has been reported in several cases of ovarian sex reversal, as in transplants of fetal mouse ovaries (39) and postnatal sex reversed ovaries of mice deficient in both types of functional ERs, ER and ER? (ER? knockout mice) (58, 59), as well as females incapable of synthesizing estrogens due to disruption of the Cyp19a1 gene encoding the P450-aromatase enzyme (ArKO mice) (60). In addition to confirming the presence of Sertoli cells, these results suggested that SOX-9 could mediate the process of ovarian sex reversal (61, 62). The lack of SOX-9 expression in both follicular cells of ODFs and Sertoli-like cells of tODFs indicates that in our model the transdifferentiation process is not mediated by SOX-9 itself. It raises the possibility that other factors, perhaps acting downstream SOX-9, are implicated in the transdifferentiation process induced by oocyte loss. It has long been thought that AMH had a masculinizing influence on the ovarian somatic lineage as in its presence testis-like cords develop within fetal ovaries (20, 63). In our study, we have shown that AMH was expressed in follicular cells of ODFs and Sertoli-like cells of tODFs. AMH, however, is not a male sex-specific protein because it is expressed in the postnatal ovary (41). In addition, it is now commonly accepted that AMH is involved indirectly in ovarian sex reversal by exerting a lethal effect on female germ cells (14, 17, 64, 65). It is thus unlikely that in ODFs AMH promotes granulosa cell transdifferentiation, but we cannot rule out that AMH has an effect on the differentiation of follicular cells in the absence of oocyte.
It is noticeable that oocyte loss does not always lead to the transdifferentiation of granulosa cells into Sertoli cells. For instance, in antral or preovulatory follicles the degeneration or the experimental removal of the oocyte induces the luteinization of granulosa cells (66, 67, 68). Similarly, a mutation of the growth differentiation factor 9 gene (Gdf9) by homologous recombination, which leads to the loss of the oocyte in primary follicles, results in the morphological and functional luteinization of the surrounding granulosa cells (69, 70). Granulosa cells of Gdf9–/– primary follicles fail to proliferate but acquire before oocyte loss follicular markers normally expressed from the preantral stage onwards (70). In Fecundity X Inverdale sheep carrying a point mutation in the bone morphogenetic protein-15 coding sequence, after the degeneration of the oocyte in primary follicles no alteration in the morphology of granulosa cells has been reported (71, 72). Granulosa cells in primary follicles of Inverdale ewes fail to proliferate as in Gdf9–/– mice, yet they acquire follicular markers of preantral follicles only after oocyte loss (72). These differences in the effect of oocyte loss on granulosa cells could well depend on the animal species and/or also the state of differentiation of the granulosa cells at the time of oocyte loss.
FSH as a potential activator of granulosa cell transdifferentiation
As a result of the early depletion of oocytes contained in primordial follicles, irradiated females displayed POF from the very beginning of reproductive life (24). We report here that the levels of circulating FSH were positively correlated with the proportion of tODFs. Interestingly, GnRH-deficient hypogonadal mice displaying very low plasma levels of gonadotropins exhibit 1 yr after irradiation ovaries composed of very numerous single-layered follicles devoid of oocytes (73), which morphologically resemble the ODFs described in the present study. Our preliminary data together with the results obtained in irradiated GnRH-deficient hypogonadal mice have led us to further investigate the potential influence of the endocrine imbalance and particularly high levels of FSH on the transdifferentiation process. We found that in irradiated females with lowered circulating gonadotropin levels after administration of a long-acting aGnRH, the proportion of tODFs was significantly decreased, compared with that of saline-treated females. In contrast, the repeated administration of r-hFSH, but neither that of hCG nor testosterone, to aGnRH-treated females efficiently prevented the decrease in tODF proportion. Thus, our findings give strong evidence that high levels of circulating FSH occurring during the POF could play a role in the transdifferentiation process. Nevertheless, our results cannot demonstrate whether FSH acts directly or indirectly by either promoting the cellular switching of follicular cells into Sertoli-like cells or preventing apoptosis of Sertoli-like cells. In a complementary analysis of aGnRH-treated ovaries by TUNEL assay, no apoptotic Sertoli-like cells were detected (data not shown). It is possible, however, that cell death occurred at the beginning of the treatment and was not detected at the time of observation.
The mechanism of action of FSH on follicular cells of ODFs and Sertoli-like cells of tODFs remains to be elucidated. In males, FSH is not involved in the fetal differentiation of Sertoli cells but stimulates their proliferation in perinatal life (74). Moreover, FSH regulates the maturation of Sertoli cells in prepubertal life (75). Its role, however, on the differentiation of the BTB has not been clearly established (76, 77). In females, early follicular growth can occur in the absence of FSH (78, 79) and FSH receptors are absent until follicles reach the primary to small preantral stages (80, 81). Given that ODF and tODF cells displayed maturational characteristics of granulosa cells of growing follicles, they might also develop FSH responsiveness. Because FSH stimulates growth, differentiation, and survival of follicles from the preantral follicular stage (82), it is tempting to speculate that FSH exerts such effects on maturing cells of ODFs and tODFs.
Is there a role for estrogen signaling pathway in our model?
Whereas the involvement of estrogens in the ovarian determining pathway has not been clearly established in eutherian mammals, the presence of Sertoli-like cells in the postnatal ovaries of ER? knockout and ArKO mice (58, 59, 60, 61, 62) underlines that at least in mice, estrogens may play a critical role in the maintenance of the phenotype of the female somatic compartment. Consistent with these findings, phytoestrogen diet or E2 administration prevents or reverses the transdifferentiation of Sertoli-like cells in ArKO mice (60, 62). In both ArKO and ER? knockout mice, the endocrine environment is affected, as shown by the increase in FSH, LH, and T plasma levels in ArKO and the increase in LH and T in ER? knockout mice (83, 84). This raises the possibility that the lack of estrogen signaling could be indirectly involved in the development of Sertoli-like cells for instance by altering endocrine parameters that are normally regulated by estrogens.
In irradiated females, the levels of E2 remained unaltered and, thus, they are unlikely to account for the increase in tODF number with age. As our immunocytochemistry studies revealed no expression of ER and ER? isoforms in follicular cells of ODFs, it may be hypothesized that oocyte loss and the lack of estrogen could together be involved in triggering transdifferentiation of follicular cells into Sertoli-like cells. Accordingly, in both ArKO and ER? mice, there are no germ cells in the tubular structures containing Sertoli-like cells, and in ER? knockout mice, it has been reported that Sertoli-like cells develop within atretic follicles, in which oocyte/granulosa cell interaction are disrupted (59, 60, 61). On the other hand, it must be underlined that several different gonadal features, besides Sox9 expression, exist between our model and both ArKO and ER? knockout mice. For instance, Leydig-like cells differentiate in ovaries of ArKO females (60) and Sertoli-like cells develop from granulosa cells of antral follicle in ER? knockout mice (61).
Conclusion
Altogether, our results provide evidence that in the absence of the oocyte, follicular cells of nongrowing follicles transdifferentiate into immature Sertoli-like cells. The recruitment of follicular cells of nongrowing follicles and their transdifferentiation into Sertoli-like cells was initiated in prepubertal life and continued throughout adulthood. The highly variable interval between oocyte loss and the conversion of follicular cells into Sertoli-like cells between ODFs underlines that cellular switching is not triggered by oocyte loss alone but may require a multistep process, which includes the recruitment and maturation of follicular cells within ODF and involves FSH regulation (Fig. 8).
FIG. 8. Presumptive mechanism leading to the transdifferentiation of granulosa cells into Sertoli-like cells after oocyte loss. In normal primordial follicles, recruitment is regulated by close interactions between oocytes and their surrounding granulosa cells as well as factors produced by growing follicles. The onset of follicular growth is characterized by the onset of AMH and inh expression in granulosa cells. In oocyte-depleted follicles, granulosa cells survive and, as indicated by AMH and inh expression, enter growth, a process that could be regulated by factors produced by the remaining growing follicles. Then granulosa cells transdifferentiate into Sertoli-like cells, possibly under regulation by high plasma FSH levels. GF, Growing follicle; Ooc., oocyte.
This ability of follicular cells to switch into their testis counterparts may reflect the similar fetal origin of these two cell types as has been suggested in other cases of cellular plasticity (85). The mechanism(s) involved in the transdifferentiation of granulosa cells into Sertoli-like cells after oocyte loss remain(s) to be resolved. One may hypothesize that the oocyte produces yet-unidentified factors that act on follicular cells to prevent their conversion into Sertoli-like cells. In line with this hypothesis are the findings suggesting that during fetal life, meiotic germ cells could reinforce the ovarian pathway by antagonizing the testis differentiation pathway (12). Further insights into the mechanisms involved in bidirectional communication between the oocyte and the companion somatic cells as well as the discovery of new oocyte factors will possibly bring answers concerning the mechanisms involved in the maintenance of the granulosa cell phenotype.
Acknowledgments
The authors thank O. Locquet, M. P. Monneret, M. Delacroix, M. Robin, C. Salvard, and E. Etienne for helpful technical assistance; C. Roulin and Dr. C. Bleux for -irradiation (Institut Curie); Drs. J. L. Lefaix and H. Coffigny for -irradiation (La Ferme Atomique); Dr. S. Brailly-Tabard for inhibin dosages; and Professor F. Jaubert for critical advices. The authors also thank Dr. S. Carreau for providing AR antibodies, Dr. M. Fellous for FOXL2 antibodies, Dr. B. Vigier for AMH antibodies, Dr. R. Guennoun for 3?-HSD antibodies, and Dr. F. Poulat for SOX-9 antibodies.
References
Koopman P, Gubbay J, Vivian N, Goodfellow P, Lovell-Badge R 1991 Male development of chromosomally female mice transgenic for SRY. Nature 351:117–121
Gubbay J, Collignon J, Koopman P, Capel B, Economou A, Munsterberg A, Vivian N, Goodfellow P, Lovell-Badge R 1990 A gene mapping to the sex-determining region of the mouse Y chromosome is a member of a novel family of embryonically expressed genes. Nature 346:245–250
Hawkins JR 1993 Mutational analysis of SRY in XY females. Hum Mut 2:347–350
McElreavey K, Vilain E, Abbas N, Herskowitz I, Fellous M 1993 A regulatory cascade hypothesis for mammalian sex determination: SRY represses a negative regulator of male development. Proc Natl Acad Sci USA 90:3368–3372
Vainio S, Heikkila M, Kispert A, Chin N, McMahon AP 1999 Female development in mammals is regulated by Wnt-4 signalling. Nature 397:405–409
Pailhoux E, Vigier B, Chaffaux S, Servel N, Taourit S, Furet JP, Fellous M, Grosclaude F, Cribiu EP, Cotinot C, Vaiman D 2001 An 11.7-kb deletion triggers intersexuality and polledness in goats. Nat Genet 29:453–458
Dolci S, De Felici M 1990 A study of meiosis in chimeric mouse fetal gonads. Development 109:37–40
McLaren A, Southee D 1997 Entry of mouse embryonic germ cells into meiosis. Dev Biol 187:107–113
Nagamine C, Carlisle C 1996 The dominant white spotting oncogene allele Kit(W-42J) exacerbates XY(DOM) sex reversal. Development 122:3597–3605
Nagamine CM, Capehart J, Carlisle C, Chang D 1998 Ovotestes in B6-XXSxr sex-reversed mice. Dev Biol 196:24–32
Adams I, McLaren A 2002 Sexually dimorphic development of mouse primordial germ cells: switching from oogenesis to spermatogenesis. Development 129:1155–1164
Yao HH, DiNapoli L, Capel B 2003 Meiotic germ cells antagonize mesonephric cell migration and testis cord formation in mouse gonads. Development 130:5895–5902
Burgoyne PS, Baker TG 1985 Perinatal oocyte loss in XO mice and its implications for the aetiology of gonadal dysgenesis in XO women. J Reprod Fertil 75:633–645
McLaren A 1991 Development of the mammalian gonad: the fate of the supporting cell lineage. Bioessays 13:151–156
Merchant H 1975 Rat gonadal and ovarian organogenesis with and without germ cells. An ultrastructural study. Dev Biol 44:1–21
Merchant-Larios H, Centeno B 1981 Morphogenesis of the ovary from the sterile W/Wv mouse. Prog Clin Biol Res 59B:383–392
Whitworth DJ 1998 XX germ cells: the difference between an ovary and a testis. Trends Endocrinol Metab 9:2–6
Jost A, Vigier B, Prépin J, Perchellet JP 1973 Studies on sex differentiation in mammals. Rec Progr Horm Res 29:1–41
Charpentier G, Magre S 1990 Masculinizing effect of testes on developing rat ovaries in organ culture. Development 110:839–849
Vigier B, Watrin F, Magre S, Tran D, Josso N 1987 Purified bovine AMH induces a characteristic freemartin effect in fetal rat prospective ovaries exposed to it in vitro. Development 100:43–55
Crumeyrolle-Arias M, Scheib D, Aschheim P 1976 Light and electron microscopy of the ovarian interstitial tissue in the senile rat: normal aspect and response to HCG of "deficiency cell" and "epithelial cords." Gerontology 22:185–204
Taketo T, Merchant-Larios H, Koide SS 1984 Induction of testicular differentiation in a fetal mouse ovary by transplantation into adult male mice. Proc Soc Exp Biol Med 176:148–153
Taketo-Hosotani T, Merchant-Larios H, Thau RB, Koide SS 1985 Testicular cell differentiation in fetal mouse ovaries following transplantation into adult male mice. J Exp Zool 236:229–237
Guigon CJ, Mazaud S, Forest MG, Brailly-Tabard S, Coudouel N, Magre S 2003 Unaltered development of the initial follicular waves and normal pubertal onset in female rats after neonatal deletion of the follicular reserve. Endocrinology 144:3651–3662
Beaumont H 1960 Changes in the radiosensitivity of the testis during foetal development. Int J Radiat Biol 2:247–256
Tuchmann-Duplessis H 1947 Une technique de coloration commode pour la numeration des cellules glandulaires de l’hypophyse. Bulletin Hist Appl 9:180–181
Lozach A, Garrel G, Lerrant Y, Berault A, Counis R 1998 GnRH-dependent up-regulation of nitric oxide synthase I level in pituitary gonadotrophs mediates cGMP elevation during rat proestrus. Mol Cell Endocr 143:43–51
Forest MG, David M, Lecoq A, Jeune M, Bertrans J 1980 Kinetics of the HCG-induced steroidogenic response of the human testis. III. Studies in children of the plasma levels of testosterone and HCG: rationale for testicular stimulation test. Pediatr Res 14:819–824
Sharpe RM, Turner KJ, McKinnell C, Groome NP, Atanassova N, Millar MR, Buchanan DL, Cooke PS 1999 Inhibin B levels in plasma of the male rat from birth to adulthood: effect of experimental manipulation of Sertoli cell number. J Androl 20:94–101
Lerrant Y, Kottler M, Bergametti F, Moumni M, Blumberg-Tick J, Counis R 1995 Expression of gonadotropin-releasing hormone (GnRH) receptor gene is altered by GnRH agonist desensitization in a manner similar to that of gonadotropin ?-subunit genes in normal and castrated rat pituitary. Endocrinology 136:2803–2808
Bukovsky A, Ayala ME, Dominguez R, Keenan JA, Wimalasena, McKenzie PP, Caudle MR 2000 Postnatal androgenization induces premature aging of rat ovaries. Steroids 65:190–205
Manova K, Nocka K, Besmer P, Bachvarova R 1990 Gonadal expression of c-kit encoded at the W locus of the mouse. Development 110:1057–1069
Fridmacher V, Locquet O, Magre S 1992 Differential expression of acidic cytokeratins 18 and 19 during sexual differentiation of the rat gonad. Development 115:503–517
Gilula N, Fawcett D, Aoki A 1976 The Sertoli cell occluding junctions and gap junctions in mature and developing mammalian testis. Dev Biol 50:142–168
Russell L 1977 Observations on rat Sertoli ectoplasmic (‘junctional’) specializations in their association with germ cells of the rat testis. Tissue Cell 9:475–498
Cocquet J, Pailhoux E, Jaubert F, Servel N, Xia X, Pannetier M, De Baere E, Messiaen L, Cotinot C, Fellous M, Veitia RA 2002 Evolution and expression of FOXL2. J Med Genet 39:916–922[Free Full Text]
Schmidt D, Ovitt CE, Anlag K, Fehsenfeld S, Gredsted L, Treier AC, Treier M 2004 The murine winged-helix transcription factor Foxl2 is required for granulosa cell differentiation and ovary maintenance. Development 131:933–942
Kent J, Wheatley SC, Andrews JE, Sinclair AH, Koopman P 1996 A male-specific role for SOX9 in vertebrate sex determination. Development 122:2813–2822
Morais da Silva S, Hacker A, Harley V, Goodfellow P, Swain A, Lovell-Badge R 1996 Sox9 expression during gonadal development implies a conserved role for the gene in testis differentiation in mammals and birds. Nat Genet 14:62–68
Gow A, Southwood C, Li J, Pariali M, Riordan G., Brodie S, Danias J, Bronstein J, Kachar B, Lazzarini R 1999 CNS myelin and Sertoli cell tight junction strands are absent in Osp/claudin-11 null mice. Cell 99:649–659
Vigier B, Picard JY, Tran D, Legeai L, Josso N 1984 Production of anti-Müllerian hormone: another homology between Sertoli and granulosa cells. Endocrinology 114:1315–1320
Loffler KA, Zarkower D, Koopman P 2003 Etiology of ovarian failure in blepharophimosis ptosis epicanthus inversus syndrome: FOXL2 is a conserved, early-acting gene in vertebrate ovarian development. Endocrinology 144:3237–3243
Josso N, Cate RL, Picard JY, Vigier B, di Clemente N, Wilson C, Imbeaud S, Pepinsky RB, Guerrier D, Boussin L, Legeai L, Carre-Eusebe D 1993 Anti-mullerian hormone: the Jost factor. Rec Progr Horm Res 48:1–59
Baarends WM, Uilenbroek JT, Kramer P, Hoogerbrugge JW, van Leeuwen EC, Themmen AP, Grootegoed JA 1995 Anti-mullerian hormone and anti-mullerian hormone type II receptor messenger ribonucleic acid expression in rat ovaries during postnatal development, the estrous cycle, and gonadotropin-induced follicle growth. Endocrinology 136:4951–4962
Drummond A, Dyson M, Mercer J, Findlay J 1996 Differential responses of post-natal rat ovarian cells to FSH and activin. Mol Cell Endocrinol 30:21–32
Sar M, Welsch F 1999 Differential expression of estrogen receptor-? and estrogen receptor- in the rat ovary. Endocrinology 140:963–971
Tetsuka M, Whitelaw P, Bremner W, Millar M, Smyth CD, Hillier SG 1995 Developmental regulation of androgen receptor in rat ovary. J Endocrinol 145:535–543
Ramos AJ, Dym M 1979 Ultrastructural differentiation of rat Sertoli cells. Biol Reprod 21:909–922
Meyer R, Posalky Z, McGinley D 1977 Intercellular junction development in maturing rat seminiferous tubules. J Ultrastruct Res 61:271–283
Uda M, Ottolenghi C, Crisponi L, Garcia JE, Deiana M, Kimber W, Forabosco A, Cao A, Schlessinger D, Pilia G 2004 Foxl2 disruption causes mouse ovarian failure by pervasive blockage of follicle development. Hum Mol Gen 13:1171–1181
Eppig J 2001 Oocyte control of ovarian follicular development and function in mammals. Reproduction 122:829–838
Durlinger AL, Gruijters MJ, Kramer P, Karels B, Ingraham HA, Nachtigal MW, Uilenbroek JT, Grootegoed JA, Themmen AP 2002 Anti-Mullerian hormone inhibits initiation of primordial follicle growth in the mouse ovary. Endocrinology 143:1076–1084
Bishop CE, Whitworth DJ, Qin Y, Agoulnik AI, Agoulnik IU, Harrison WR, Behringer RR, Overbeek PA 2000 A transgenic insertion upstream of sox9 is associated with dominant XX sex reversal in the mouse. Nat Genet 26:490–494
Vidal VP, Chaboissier MC, de Rooij DG, Schedl A 2001 Sox9 induces testis development in XX transgenic mice. Nat Genet 28:216–217
Chaboissier MC, Kobayashi A, Vidal VI, Lutzkendorf S, Van D, Kant HJ, Wegner M, De Rooij DG, Behringer RR, Schedl A 2004 Functional analysis of Sox8 and Sox9 during sex determination in the mouse. Development 131:1891–1901
De Santa Barbara P, Bonneaud N, Boizet B, Desclozeaux M, Moniot B, Sudbeck P, Scherer G, Poulat F, Berta P 1998 Direct interaction of SRY-related protein SOX9 and steroidogenic factor 1 regulates transcription of the human anti-Mullerian hormone gene. Mol Cell Biol 18:6653–6665
Arango NA, Lovell-Badge R, Behringer RR 1999 Targeted mutagenesis of the endogenous mouse Mis gene promoter: in vivo definition of genetic pathways of vertebrate sexual development. Cell 99:409–419
Couse JF, Hewitt SC, Bunch DO, Sar M, Walker VR, Davis BJ, Korach KS 1999 Postnatal sex reversal of the ovaries in mice lacking estrogen receptors and ?. Science 286:2328–2331
Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M 2000 Effect of single and compound knockouts of estrogen receptors (ER) and ? (ER?) on mouse reproductive phenotypes. Development 127:4277–4291
Britt KL, Kerr J, O’Donnell L, Jones ME, Drummond AE, Davis SR, Simpson ER, Findlay JK 2002 Estrogen regulates development of the somatic cell phenotype in the eutherian ovary. FASEB J 16:1389–1397
Dupont S, Dennefeld C, Krust A, Chambon P, Mark M 2003 Expression of Sox9 in granulosa cells lacking the estrogen receptors, ER and ER?. Dev Dyn 226:103–106
Britt K, Stanton P, Misso M, Simpson E, Findlay J 2004 The effects of estrogen on the expression of genes underlying the differentiation of somatic cells in the murine gonad. Endocrinology 145:3950–3960
Behringer RR, Cate RL, Froelick GJ, Palmiter RD, Brinster RL 1990 Abnormal sexual development in transgenic mice chronically expressing Müllerian inhibiting substance. Nature 345:167–170
Taketo T, Saeed J, Manganaro T, Takahashi M, Donahoe PK 1993 Müllerian inhibiting substance production associated with loss of oocytes and testicular differentiation in the transplanted mouse XX gonadal primordium. Biol Reprod 49:13–23
Whitworth DJ, Shaw G, Renfree MB 1996 Gonadal sex reversal of the developing marsupial ovary in vivo and in vitro. Development 122:4057–4063
Pincus G, Enzmann EV 1935 The comparative behavior of mammalian eggs in vivo and in vitro I. The activation of ovarian eggs. J Exp Med 62:655–675
Falck B 1959 Site of production of oestrogen in rat ovary as studied in microtransplants. Acta Physiol Scand 47:1–101
El-Fouly MA, Cook B, Nekola M, Nalbandov AV 1970 Role of the ovum in follicular luteinization. Endocrinology 87:288–293
Dong J, Albertini DF, Nishimori K, Kumar TR, Lu N, Matzuk MM 1996 Growth differentiation factor-9 is required during early ovarian folliculogenesis. Nature 383:531–535
Elvin JA, Yan C, Wang P, Nishimori K, Matzuk MM 1999 Molecular characterization of the follicle defects in the growth differentiation factor 9-deficient ovary. Mol Endocrinol 13:1018–1034
Braw-Tal R, McNatty KP, Smith P, Heath DA, Hudson NL, Phillips DJ, McLeod BJ, Davis GH 1993 Ovaries of ewes homozygous for the X-linked Inverdale gene (FecXI) are devoid of secondary and tertiary follicles but contain many abnormal structures. Biol Reprod 49:895–907
Juengel JL, Quirke LD, Tisdall DJ, Smith P, Hudson NL, McNatty KP 2000 Gene expression in abnormal ovarian structures of ewes homozygous for the inverdale prolificacy gene. Biol Reprod 62:1467–1478
Tennent B, Beamer W 1986 ovarian tumors not induced by irradiation and gonadotropins in hypogonadal (hpg) mice. Biol Reprod 34:751–760
Orth J 1984 The role of follicle-stimulating hormone in controlling Sertoli cell proliferation in testes of fetal rats. Endocrinology 115:1248–1255
Johnston H, Baker P, Abel M, Charlton H, Jackson G, Fleming L, Kumar TR, O Shaughnessy PJ 2004 Regulation of Sertoli cell number and activity by follicle-stimulating hormone and androgen during postnatal development in the mouse. Endocrinology 145:318–329
De Kretser D, Burger H, Fortune D, Hudson B, Long A, Paulsen C, Taft H 1972 Hormonal, histological and chromosomal studies in adult males with testicular disorders. J Clin Endocrinol Metab 35:392–401
Johnson M 1970 The pituitary and the blood-testis barrier. J Reprod Fertil 22:181–182
Kumar TR, Wang Y, Lu N, Matzuk MM 1997 Follicle stimulating hormone is required for ovarian follicle maturation but not male fertility. Nat Genet 15:201–204
Abel M, Wootton A, Wilkins V, Huhtaniemi I, Knight P, Charlton H 2000 The effect of a null mutation in the follicle-stimulating hormone receptor gene on mouse reproduction. Endocrinology 141:1795–1803
Mayerhofer ADG, Costa ME, Ojeda SR 1997 A role for neurotransmitters in early follicular development: induction of functional follicle-stimulating hormone receptors in newly formed follicles of the rat ovary. Endocrinology 138:3320–3329
Rannikki A, Zhang F, Huhtaniemi I 1995 Ontogeny of follicle-stimulating hormone receptor gene expression in the rat testis and ovary. Mol Cell Endocrinol 107:199–208
McGee EA, Perlas E, LaPolt PS, Tsafriri A, Hsueh AJ 1997 Follicle-stimulating hormone enhances the development of preantral follicles in juvenile rats. Biol Reprod 57:990–998
Fisher CR, Graves KH, Parlow AF, Simpson ER 1998 Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp 19 gene. Proc Natl Acad Sci USA 95:6965–6970
Couse JF, Yates MM, Walker VR, Korach KS 1999 Characterization of the hypothalamo-pituitary-gonadal axis in estrogen receptor (ER) null mice reveals hypergonadism and endocrine sex reversal in females lacking ER but not ER?. Mol Endocrinol 17:1039–1053
Tosh D, Slack J 2002 How cells change their phenotype. Nat Rev Mol Cell Biol 3:187–194
Lane E, Goodman S, Trejdosiewicz L 1982 Disruption of the keratin filament network during epithelial cell division. EMBO J 1:1365–1372(Céline J. Guigon, No?llin)