Journal of Steroids & Hormonal Science

Anti-Müllerian hormone (AMH) is also known as Müllerian inhibiting substance (MIS), and is a member of the TGF-β superfamily [1]. AMH plays an essential role in the regression of Müllerian ducts and descent of the testes during sexual differentiation of male fetuses [2,3]. Fetal Sertoli cells secrete AMH until birth, even after Müllerian ducts regression [4,5]. Moreover, male transgenic mice overexpressing Amh manifest feminized genitalia [6], in accordance with decreased Leydig cell numbers and testosterone secretion [7]. Conversely, Amh knockout mice manifested Leydig cell hyperplasia and infertility [8,9]. It is therefore likely that AMH plays an important role in fetal Leydig cell function [10,11].

Postnatally, expression of Amh and Amhr2 has been observed in rat testes and ovaries [12,13]. During postnatal maturation of the testis, Sertoli cells change morphologically and functionally from immature type to mature type cells, although slightly immature Sertoli cells persist even in mature testes [14].

However, there were few reports that examine AMH-producing cells of postnatal mature testes in detail. Recently, we were the first to show expression of AMH and AMH type II receptor (AMHR2) in spermatocytes from rats aged 21 days (21d) using immunohistochemical (IHC) staining and in situ hybridization (ISH). In agreement, germ cells from 21d rat testes expressed Amh and Amhr2 [15].

During the development of spermatogenesis in maturing rat testis, germ cells of neonatal testes are only spermatogonia, and spermatocytes appear during weaning prior to formation of round and elongated spermatids [16]. Spermatogonia and spermatocytes are present in 21d rat testis, whereas round and elongated spermatids are not. Moreover, in postnatal maturing rat testis, Sertoli cells and spermatocytes express and synthesize AMH and AMHR2 [15].

AMH signals are initiated by binding to an AMHR2, and can activate other signaling pathways such as those of β-catenin or NFkB [17]. In previous reports of AMH-AMHR2 signal transduction, AMH was shown to bind AMHR2 and activate the SMAD 1, 5, and 8 signaling pathway through one or several type 1 receptors [18].

SMAD proteins are intracellular mediators of TGF signaling and in postnatal mouse testes; SMAD1 was detected in spermatocytes and round spermatids [19], whereas SMAD5 was detected mainly in spermatogonia [20]. AMH activates the SMAD1, 5, and 8 signaling pathway through AMHR2 and inhibits SMAD6 [20]. We also reported that 21d testes and isolated germ cells expressed Smad1, 5, and 8 [15].

These data indicated that AMH is involved in spermatogenesis. However, although AMH is present in seminiferous tubules, its physiological roles in mature testis are not entirely clear. Thus, in this study we characterized expression patterns of AMH and AMHR2 in germ cells from mature rat testis during the spermatogenic cycle.

Animals

All experiments were approved by the Animal Care and Use Committee of the University of Yamanashi, Japan.

Normal Sprague-Dawley (SD) rats aged 5, 10, 16, 30, and 44 days (d) were purchased from Japan SLC and were housed for 5 days in an air-conditioned animal room at an ambient temperature of 22°C and a relative humidity of 55% ± 10% under a 12 h light/12 h dark cycle. Male rats were bred with maternal rats until postnatal day 20 (20d). The day of birth was designated as 0d and testes were collected from 10, 15, 21, 35 and 49d rats for use in assays.

In situ hybridization (ISH)

To determine the localization of Amh and Amhr2 in 49d rat testis, ISH was performed using antisense probes comprising 55-mer oligonucleotides, which were labeled at the 5′-end using digoxigenin (DIG). Two non-overlapping probes were designed based on the Amh and Amhr2 sequences and were commercially synthesized (BEX, Tokyo, Japan). Amh and Amhr2 antisense probe sequences were previously reported [15]. Paraffin-embedded rat testis sections of 8 μm in thickness, were prepared and mounted on glass slides, which were pre-coated with 3-aminopropyltriethoxysilane (Shin-Etsu Chemical Co., Tokyo, Japan). All hybridization steps, including washing and visualization, were performed using a commercial kit (IsHyb In situ Hybridization Kit; Biochain Institute, Newark, CA) according to the manufacturer’s instructions. Probe signals were detected with an alkaline phosphatase-conjugated anti-DIG antibody and nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl-phosphate (NBT/BCIP) solution. ISH was performed by mixing two individual mRNA probes (AS1 and AS2), and specificity of hybridization was confirmed using sense oligonucleotide probes located at the same position as the antisense probes, which were used as a negative controls.

AMH and AMHR2 antibodies

AMH and AMHR2 polyclonal antibodies [15] were raised by Operon Biotechnologies K.K. custom synthesis service (Tokyo, Japan). Briefly, the AMH peptides, C+KFQEPPPGGASRWE (143-156) and CAWPQSDRNPRYGNH (485-499) were synthesized, and mixed, and rabbits were immunized four times with the mixture and with the carrier protein KLH every week. In identical procedures, the antigen peptides, CDPVPRAHPSPSSTL (92-106) and CSATDPRGLRELLED (470-484) were used to raise the AMHR2 antibody in rabbits. Polyclonal antisera were purified using affinity column chromatography with synthesized antigens.

Immunohistochemical (IHC) staining

Testes from 10d-49d rats were fixed in buffered neutral formalin, were embedded in paraffin, and were sectioned (5-μm, thickness) for IHC staining, following the procedure prescribed by Morpho Technology (Sapporo, Japan). IHC procedures were performed as previously reported [15]. Briefly, fixed tissues were deparaffinized, rehydrated, and washed with tris-buffered solution (TBS). Antigen retrieval was then performed using a high temperature antigen unmasking technique, in which sections were incubated with EDTA buffer solution (pH 9.0) at 95°C for 20 min. Sections were then rinsed with TBS, and endogenous peroxidase activity was inactivated with 0.1% H₂O₂ at room temperature for 10 min. Tissue sections were then immunostained using an automated staining system (Autostainer Plus, Dako). Sections were incubated with anti-AMH antibody (1:100 dilution) or anti-AMHR2 antibody (1:100 dilution) at room temperature for 60 min and were then washed with TBS. Sections were then incubated with Histofine® Simple Stain MAX-PO(R) (Nichirei, Tokyo) at room temperature for 30 min, and were visualized with DAB solution at room temperature for 5 min. Sections were counterstained with hematoxylin for 1.5 min and negative controls were processed simultaneously by omitting the primary antibodies.

Germ-cell isolation

Germ cells were isolated from 49d testes to examine the effects of the AMH antibody on AMH expression. Testicular cells were isolated as described previously by Galardo et al. [21] with minor modifications [15]. Germ cells were isolated from 4 testes of 49d rats in each experiment. Testes were decapsulated and treated with 0.1% collagenase type 2 (Worthington Biochemical Corp., Lakewood, NJ) and 10 U/ml DNase I (D7291, Sigma-Aldrich, St Louis, MO) in 20 ml of Hanks' balanced salt solution (HBSS) for 10 min at 35°C in a Bio Shaker (BR-30L, Taitec, Tokyo) at 100 oscillations per min. Collagenase solution was diluted with 20 ml of HBSS and seminiferous tubules were allowed to sediment for 2 min. Supernatants were then discarded and the tubular pellets were digested with 20 ml of 0.1% trypsin (Gibco) in HBSS for 20 min at 35°C in a Bio Shaker at 100 oscillations per min. Trypsin-digested solutions were supplemented with 10 ml of 10% fetal calf serum (FCS) in HBSS and were filtered with a Steriflip 100 μm nylon net filter (Millipore) and then centrifuged at 400 × g for 5 min. Cell pellets were re-suspended in a 1:1 mixture of DMEM-Ham's F-12 medium supplemented with 10% FCS, 15-mM NaHCO₃, 100 IU/ml penicillin, 2.5 mg/ml amphotericin B, and 20-mM HEPES (pH 7.4; DMEM-F12). Isolated cells from seminiferous tubules were seeded in 10 culture plates (10 cm²/plate) and were cultured at 35°C in a 5% CO₂ incubator for 6 h. During this period, Sertoli cells were attached to the surface of the plate. Attached cells (Sertoli cells) were washed twice with HBSS, and cultured again with DMEM-F12 for 12 h. These cells were then washed twice with HBSS, scraped, centrifuged, and frozen for total RNA isolation. Floating germ cells were collected by carefully removing the medium and centrifuging at 400 × g for 5 min. Germ cell pellets were resuspended in HBSS and were loaded on a discontinuous four-layer (20%, 25%, 32%, and 37%) Percoll density gradient (Percoll PLUS, GE Healthcare Bio-Sciences) for gradient centrifugation at 800 × g for 30 min at 4°C. The fraction between the 25%-32% Percoll interface was collected. To remove Percoll, 4 volumes of HBSS were added to the recovery solution, and were centrifuged at 400 × g for 5 min at 4°C. Germ cell pellets were re-suspended in DMEM-F12, and were then cultured in 10 cm² plates at 35°C in a CO₂ incubator for 12 h. Few Sertoli cells remained in the germ cell population, and these were attached to the dish surface. Floating germ cells were obtained by carefully removing the medium and were precipitated from the medium by centrifugation at 400 × g for 5 min. Germ cell pellets were re-suspended in 60 ml of DMEM-F12 (without FCS), and cell suspensions were then divided into 6 plates (10 ml/plate). Subsequently, 50 μl of anti-AMH antibody was added to 3 of 6 plates (final dilution 200X). Re-suspended germ cells in these 6 plates were cultured at 35°C in a CO₂ incubator for 2 h, and were then centrifuged. Germ cell pellets collected from 6 plates were frozen for total RNA isolation and subsequent quantitative real-time polymerase chain reaction (qRT-PCR). Experiments were performed in triplicate.

Quantitative real-time polymerase chain reaction

Total RNA was extracted from germ cells and Sertoli cells that were isolated from 49d rat testes, using RNeasy Mini Kit (QIAGEN, Tokyo). Subsequently, cDNA was synthesized from 1 μg of total RNA template using a High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). For qRT-PCR, triplicated samples containing 9 μl of cDNA with 10 μl of Taqman Universal PCR Master Mix (Applied Biosystems) and 1 μl o f 20X A s s ays-o n-dem an d G en e Expression Product (Applied Biosystems) were pre-incubated at 50°C for 2 min and then at 95°C for 10 min. Amplification was performed by 40 cycles of reaction at 95°C for 15 s and 60°C for 1 min. Fluorescence intensity was quantitatively analyzed using an ABI Prism 7500 Sequence Detection System (Applied Biosystems). Genes that were analyzed using PCR included Amh, Amhr2, Scp1, Rsbn1, Ngfr, Rhox5, and Smads 1, 5, and 8 (Table 1). Testicular expression of individual genes in whole testis from 49 d rats was set at 1.0. Relative expression levels of genes in isolated germ cells were normalized to that of β-actin .

Table 1: List of candidate genes.

Statistical analysis

Relative expression levels of Amh, Amhr2, Scp1, Rsbn1, and Smads 1, 5, and 8 in isolated germ cells were log-transformed, and the effects of anti-AMH antibody were analyzed separately using linear regression models that were adjusted for the order of experiments. P-values of less than 0.05 were considered indicative of statistical significance.

ISH of Amh and Amhr2

Figure 1 demonstrates signals of Amh and Amhr2 in stages I-IV, VIII and IX-XI in ISH analyses of 49d seminiferous tubules. Although signal intensities of Amh and Amhr2 varied from weak to strong, Amh and Amhr2 signals in spermatocytes were detected at all stages of the spermatogenic cycle and especially during stage VIII (Figures 1b and 1e). Signals of Amh and Amhr2 in round and elongated spermatids were negative at all stages, but were unclear in spermatogonia, and were faint in Sertoli cells.

Figure 1: In situ hybridization of Amh and Amhr2 in seminiferous tubules from 49 day old (49d) rat testis. I-IV, VIII, and IX-XI indicate stages of spermatogenic cycle. Solid arrows indicate positive spermatocytes. Solid arrowheads indicate Sertoli cells. Open arrowheads indicate negative round (a, b, d, e) and elongated (c, f) spermatids. Amh and Amhr2 expression levels varied from weak staining (a, d), through mild (c, f), to strong (b,e).

IHC staining of AMH and AMHR2

Figure 2 demonstrates age-related changes in AMH and AMHR2 IHC staining. Only spermatogonia and Sertoli cells were present in the seminiferous tubules of 10d testis, whereas few spermatocytes appeared in those of 15d testes. In 10d testes, AMH and AMHR2 stainings were strong in Sertoli cells, but were not present in spermatogonia (Figures 2a and 2e). When spermatocytes appeared in the seminiferous tubules of 15d testes, AMH and AMHR2 staining’s observed in Sertoli cells and were weakly present in spermatocytes, but were absent in spermatogonia (Figures 2b and 2f). In 21d testis, many spermatogonia had differentiated into spermatocytes, but no spermatids were observed. Strong AMH and AMHR2 staining’s observed in spermatocytes, but were absent in Sertoli cells and spermatogonia (Figures 2c and 2g). Spermatocytes had differentiated into spermatids in 35d seminiferous tubules, and AMH staining was strong in spermatocytes, weak in Sertoli cells, and not present in elongated spermatids and spermatogonia (Figure 2d). In contrast, AMHR2 staining was strong in spermatocytes and elongated spermatids, but was unclear in Sertoli cells (Figure 2h).

Figure 2: Immunohistochemical (IHC) staining of AMH and AMHR2 in seminiferous tubules of 10d, 15d, 21d, and 35d testes (IX-XI indicates stage of spermatogenic cycle). Solid arrowheads indicate positive Sertoli cells (a, b, c, d, e, and f). Open arrowheads indicate negative spermatogonia (a-h). Solid arrows indicate positive spermatocytes (b, c, d, g, and h). Open arrows indicate negative elongated spermatids (d). Red solid arrows indicate positive elongated spermatids (h).

Figures 3 and 4 shows the spermatogenic stage-specific IHC staining of AMH and AMHR2 in 49d seminiferous tubules. In spermatocytes, AMH staining was negative or weak at stage I-IV (Figure 3a), but was strong at stages VII-XIV. The strongest AMH staining was observed at stage VIII (Figure 3c and Figure 4a). In contrast, AMH staining was clearly observed in round spermatids at stages I-IV (Figure 3a), but was decreased at stage VII, and was diminished in elongated spermatids at stages VIII-XIV (Figures 3c, 3d and 3e. Figure 4a). AMHR2 staining was negative or weak in spermatocytes at stages I-IV (Figure 3f), but was strong at stages VIIXIV. Unlike the staining of AMH in spermatids, AMHR2 staining was present in round and elongated spermatids at all stages, and was especially strong in cell membranes of round spermatids at stage VIIVIII (Figures 4b and 4c). Moreover, AMHR2 staining was observed in residual bodies (Figure 4c). AMH and AMHR2 staining’s were absent in spermatogonia at all stages, and were clear in Sertoli cells at stages IIV.

Figure 3: IHC staining of AMH and AMHR2 at spermatogenic stages in semuniferous tubules of a 49D testis. Sg: spermatogonia; sc: spermatocyte; r: round spermatid; e: elongated spermatid; se: Sertoli cell. IHC stainings of AMH and AMHR2 in spermatocytes were negative or weak at stages I-IV (a, f), but were strong at stages VII-XIV (b-e, g-j). AMH staining of round spermatids was strong at stages I-IV (a), and was faint at stage VII (b). AMH staining of elongated spermatids was negative at all stages (a, d, e). AMHR2 stainings of round and elongated spermatids were strong at all stages from I-XIV. AMH and AMHR2 staining of Sertoli cells was weakly positive at stages I-IV and VII.

Figure 4: IHC staining of AMH and AMHR2 at spermatigenic stage VIII. AMH staining (a) was strong in spermatocytes (solid arrow), was faint in Sertoli cells (solid arrow head), and negative in round spermatids (red arrow). AMHR2 staining (b,c) was strong in spermatocytes (solid arrow), was faint in Sertoli cells (solid arrow head) and was strong in cell membranes of round spermatids (green arrow). AMHR2 staining (c) was positive in residual bodies (green arrow head).

Expression levels of Amh , Amhr2 , Scp1 , Rsbn1 , Ngfr , Rhox5 , and Smads1 , 5 , and 8 in germ cells and Sertoli cells isolated from 49d testes

Figure 5 shows expression levels of Amh , Amhr2 , Ngfr , and Rhox5 in isolated germ cells and Sertoli cells. Expression of Ngfr (immature Sertoli cell specific gene) and Rhox5 (mature Sertoli cell specific gene) was not observed in isolated germ cells. Isolated Sertoli cells expressed Amh , Ngfr , and Rhox5 , but not Amhr2 .

Figure 5: Relative expression of Amh , Amhr2 , Ngfr and Rhox1 in isolated germ cells and Sertoli cells of 49d testes. Testicular expression at 49d was set at 1.0; Data are presented as means and standard deviations.

As shown in Figure 6, germ cells from 49d testes expressed Amh, Amhr2, Scp1 and Rsbn1, and Smads 1, 5, and 8. After incubation of isolated germ cells with anti-AMH antibody, expression levels of Amh, Amhr2, Scp1, Smad1 and Smad5 were significantly increased, whereas that of Rsbn1 was significantly decreased.

Figure 6: Effect of AMH antibody on expression levels of Amh, Amhr2, Smads (1 , 5 , 8 ), Scp1 and Rsbn1 in isolated germ cells of 49d testes. Testicular expression at 49d was set at 1.0; Data are presented as means and standard deviations.

Josso et al. were the first to report evidence that Sertoli cells in fetal testes produce AMH [22]. Based on studies of fetal testes, it was thought that AMH was produced only by Sertoli cells in males. During postnatal spermatogenesis, Sertoli cells and spermatogonia are present in 10d rat testes, and whereas spermatocytes appear after 15d testis, round spermatids do not appear in testes until 21d. Recently, we showed that spermatocytes of rat testes synthesize AMH and AMHR2 at 21d [15]. Similarly, several recently reports show AMH expression in oocytes and primordial follicles [23,24], although other studies show that AMH is absent in oocytes of all follicular categories [25,26]. Hence, although the present experiments indicate that germ cells of matured gonads express AMH, the physiological role of AMH in gonads remains unclear.

In this study, we first clarified that spermatocytes expressed and synthesized AMH and AMHR2 during spermatogenic cycle in 49 d rat testes. The stage-specific expression profiles of AMH and AMHR2 in spermatocytes, spermatids and Sertoli cells are showed in Table 2.

Table 2: Stage-specific expression of AMH and AMHR2 during spermatogenic cycle.

Spermatocytes expressed Amh and Amhr2, and synthesized AMH and AMHR2. IHC staining of AMH in spermatocytes was negative or weak during stage I-IV of the spermatogenic cycle, but became strong during stages VII-XIV. Moreover, the intensity of AMH staining in spermatocytes during the spermatogenic cycle corresponded with that of Amh expression in spermatocytes. In contrast, Amh expression was not detected in spermatids, despite positive staining of AMH protein, suggesting that during differentiation of spermatocytes to round spermatids, AMH is transferred from spermtocytes to round spermatids. Accordingly, staining of the transferred AMH in round spermatids decreased with the progression of the spermatogenic cycle, was diminished at stage VIII, and was not observed in elongated spermatids. AMH staining was also weak in Sertoli cells at stage I to IV. Taken together, these data indicate that AMH is predominantly produced by spermatocytes rather than Sertoli cells in matured testes.

In this study, we first clarified that spermatocytes expressed and synthesized AMH and AMHR2 during spermatogenic cycle in 49 d rat testes. The stage-specific expression profiles of AMH and AMHR2 in spermatocytes, spermatids and Sertoli cells are showed in Table 2.

Spermatocytes expressed Amh and Amhr2, and synthesized AMH and AMHR2. IHC staining of AMH in spermatocytes was negative or weak during stage I-IV of the spermatogenic cycle, but became strong during stages VII-XIV. Moreover, the intensity of AMH staining in spermatocytes during the spermatogenic cycle corresponded with that of Amh expression in spermatocytes. In contrast, Amh expression was not detected in spermatids, despite positive staining of AMH protein, suggesting that during differentiation of spermatocytes to round spermatids, AMH is transferred from spermtocytes to round spermatids. Accordingly, staining of the transferred AMH in round spermatids decreased with the progression of the spermatogenic cycle, was diminished at stage VIII, and was not observed in elongated spermatids. AMH staining was also weak in Sertoli cells at stage I to IV. Taken together, these data indicate that AMH is predominantly produced by spermatocytes rather than Sertoli cells in matured testes.

Moreover, expression patterns of Amhr2 and AMHR2 in spermatocytes closely resembled those of Amh and AMH, with high expression during stages VII-XIV. Amhr2 expression was not detected in spermtids, despite positive staining of AMHR2, suggesting that AMHR2 was transferred from spermatocytes to round spermatids. However, unlike the expression pattern of AMH, AMHR2 was observed in round spermatids and elongated spermatids. AMHR2 staining was clearly observed at cell membranes of round spermatids, especially during stages VII-VIII. Moreover, AMHR2 was present in residual bodies at stage VIII. Because spermatocytes synthesize both AMH and AMHR2, these data indicate that 1) AMH acts on spermatocytes in an autocrine fashion during stages VII-XIV, 2) AMH is mainly produced by spermatocytes and acts on round and elongated spermatids through AMHR2 in a paracrine fashion from stages VIII to XIV, especially at stage VIII, and 3) AMH from Sertoli cells acts on elongated spermatids during the early stages I-IV. These observations suggest that AMH contributes to morphological changes from round to elongated spermatid forms at stages VIII-IX, and to morphological changes in elongated spermatids.

In our experiments with isolated testicular cells, expression of Sertoli cell specific genes (Ngfr and Rhox5 ) was not observed in isolated germ cells. The result indicates that cultured germ cells did not have the contamination of Sertoli cells. In cultured germ cells isolated from 49d testes, expression levels of Amh, Amhr2, Smad1, and Smad5 were significantly increased by the addition of AMH antibody. In a previous study, AMH through AMHR2 activated the SMAD1/5/8 signaling pathway and inhibited SMAD6 [20]. SMAD1 was also detected in spermatocytes and round spermatids [19], and SMAD5 was detected mainly in spermatogonia [20]. In the present study, although spermatogonia did not express Amhr2 and AMHR2, expression levels of Smads 1 and 5 were increased in isolated germ cells, indicating that SMADs1 and 5 are present in spermatocytes and spermatids. Moreover, treatment with anti-AMH antibody led to increased Scp1 expression and decreased Rsbn expression in isolated germ cells. Scp1 is pachytene spermatocyte specific gene and Rsbn1 is round spermatids specific gene. These results may suggest that AMH mainly secreted by spermatocytes controls differentiation from spermatocytes to round spermatids and morphological changes from round spermatids to elongated form.

In AMHR2 expressing ovarian granulose cell tumor cells, exogenous AMH activated SMADs 1 and 5, and augmented caspase-3 activity, a marker of apoptosis induction [27]. Caspase-3 is present in spermatocytes and residual bodies and may be associated with germ cell apoptosis [28]. During terminal morphological spermatogenic changes, spermatid cytoplasm’s are eliminated and finally collected in residual bodies that display several apoptotic features. Male mice with a targeted deletion of the septin4 gene (Sept4 ) are sterile due to abnormal spermatozoa with attached residual cytoplasm. In accordance, caspase inhibitor XIAP was induced following mutation of Sept4 [29]. These reports suggest that AMH activates caspase-3 through AMHR2-SMAD signaling in germ cells, and that caspase-3 contributes to cytoplasmic apoptosis in elongated spermatids.

In conclusion, the present data show that the spermatocytes stagespecifically express and synthesize AMH and AMHR2 during the spermatogenic cycle, and that AMH and AMHR2 are transferred to spermatids. AMH was expressed in spermatocytes at stages VII-XIV, and AMHR2 was most strongly expressed in round spermatids at stage VIII, and then in elongated spermatids and residual bodies. Finally, Smad 1 and 5 expressions were increased in response to AMH in cultured isolated germ cells, indicating that the AMH-SMAD pathway may be associated with morphological changes from round to elongated spermatids at stage VIII. However, the inferred cytoplasmic apoptosis of spermatids remains to be fully established. Future studies are required to define further details of the physiological role of AMH during spermatogenic cycles.

The authors wish to thank Takashi Okamoto (Laboratory of Veterinary Biochemistry, Faculty of Agriculture, Tottori University) and Yoshiko Tanabe (Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi) for their excellent technical assistance.

This work was supported by Grant-in-aid for Scientific Research (C 24591504) of Japan Society for the Promotion of Science.

None of the authors have any potential conflicts of interest associated with this research.