POF was induced in mice according to a previously described method (3–5). Briefly, 8-week-old female C57BL/6 mice (n=40) were purchased from the Experimental Animal Center of Shanghai University of Traditional Chinese Medicine (Shanghai, China). Mice were housed in a temperature-controlled environment under standard light-dark cycles with ad libitum access to food and water as previously described (3). Mice were randomly divided into two groups (20 mice/group). Mice in the POF group were injected intraperitoneally with 70 mg/kg cyclophosphamide (Sigma-Aldrich, St. Louis, MO, USA), followed by intraperitoneal injections of 30 mg/kg cyclophosphamide every three days for three weeks, to establish the POF mouse model. Mice in the control group were injected intraperitoneally with an equal volume of normal saline every three days for three weeks. All mice were sacrificed by cervical dislocation at three weeks and the ovaries were harvested for subsequent analysis. The study was approved by the Ethics Committee of Shanghai Geriatric Institute of Chinese Medicine (Shanghai, China; ref. SHAGESYDW2015026). All experiments conformed to standards set by the Laboratory Animal Regulation of the State Scientific and Technological Commission (www.slarc.org.cn).

H&E staining was performed according to a previously described method (3). Briefly, fresh tissues were immersed in 4% paraformaldehyde (Sigma-Aldrich) for 30 min at room temperature to allow fixation. Tissues were dehydrated using an ethanol gradient, embedded in paraffin, sectioned (thickness, 6 µm) and deparaffinized in xylene (Sigma-Aldrich). Tissue sections were stained with H&E (Sigma-Aldrich), cleared in xylene and mounted on slides using neutral balsam (Sigma-Aldrich). The number of atretic follicles was counted per microscope field, in three non-overlapping fields of ovary sections from each mouse.

Immunofluorescence staining was performed as previously described (4). Briefly, fresh tissues were fixed and sectioned as for H&E staining. Tissue sections were incubated with immunofluorescence blocking solution (Beyotime Institute of Biotechnology, Haimen, China) for 30 min at 37°C. Subsequently, sections were washed three times in immunofluorescence wash solution (Beyotime Institute of Biotechnology) for 5 min at room temperature. Sections were incubated with primary antibodies (Table I) for 45 min at 37°C. The washing step was repeated, and sections were incubated with secondary antibodies (Table II) for 45 min at 37°C. Following a final repeat of the washing step, sections were mounted using Kisser's Mounting Medium (Beyotime Institute of Biotechnology).

Immunohistochemical staining was performed as previously described (4). Briefly, ovarian tissue samples were washed three times with phosphate-buffered saline (PBS; Sigma-Aldrich), fixed with 4% paraformaldehyde (Sigma-Aldrich) for 30 min, dehydrated through a graded series of ethanol, cleared in xylene and embedded in paraffin. Subsequently, serial 6-µm thick sections were cut, rinsed with 3% phosphate buffer (Sigma-Aldrich), and subjected to heat retrieval in a microwave. The sections were then incubated with primary antibodies (Table I) for 45 min at 37°C. Subsequently, sections were washed three times in immunohistochemistry wash solution (Beyotime Institute of Biotechnology) for 5 min at room temperature. Sections were incubated with secondary antibodies (Table II) for 45 min at 37°C. Following a repeat of the washing step, ABC chromogenic reagent (Sigma-Aldrich) was added to visualize bound antibody. PBS (pH 7.4) was used as a negative control for primary antibody. Sections were mounted using neutral balsam (Beyotime Institute of Biotechnology). Five randomly selected fields of view (magnification, ×200; Olympus BX43; Olympus Corporation, Tokyo, Japan) were assessed for each tissue section and analyzed using Integrated Performance Primitives software version 4.0 (Intel Corporation, Santa Clara, CA, USA).

Mouse blood was obtained by retro-orbital blood collection at sacrifice (~100 µl blood from each mouse) and centrifuged at 453 × g at 4°C for 10 min. The supernatants were collected and these serum samples were used for ELISA. ELISAs were performed according to the manufacturer's instructions for the mouse E2 and FSH ELISA kits (Westang Bio-Tech Co., Ltd., Shanghai, China). Briefly, 100 µl serum, along with mouse E2 standards (at concentrations of 8,000, 4,000, 2,000, 1,000, 500, 250 and 125 pg/ml) and mouse FSH standards (at concentrations of 10, 5, 2.5, 1.25, 0.625, 0.312 and 0.156 ng/ml), was added to pre-coated 96-well detection plates and incubated for 60 min at 37°C. The solution was discarded and the plates were washed three times with wash solution. Horseradish peroxidase-conjugated detection antibodies were added and incubated for 60 min at 37°C. Subsequently, stop solution was added and the optical density of each well in the 96-well plate was measured at a wavelength of 450 nm.

Cells were isolated and flow cytometry was performed according to a previously described method (5,10). Briefly, the murine ovaries were minced into small pieces (~1 mm³) and washed three times with PBS at 4°C. The samples were digested using 10 mg/ml collagenase type I (Sigma-Aldrich) and 0.1% hyaluronidase (Sigma-Aldrich) in PBS for 5 min on an orbital shaker at 37°C, and filtered through 40-µm diameter cell strainers. These granulosa cells from ovaries were fixed in 4% paraformaldehyde for 30 min at room temperature. Following centrifugation at 453 × g for 5 min at 37°C, the supernatant was discarded and the cell precipitate was resuspended in blocking solution (Beyotime Institute of Biotechnology) for 30 min at 37°C. Cells were centrifuged, the blocking solution was discarded, and primary antibodies (Table I) were added and incubated for 45 min at 37°C. Following centrifugation, the antibodies were discarded and wash solution (Beyotime Institute of Biotechnology) was added to the cells for 5 min at room temperature. Cells were then centrifuged, and secondary antibodies (Table II) were added and incubated for 30 min at 37°C. Following centrifugation, the antibodies were discarded and wash solution (Beyotime Institute of Biotechnology) was added to the cells for 5 min at room temperature. Positive cells were detected by flow cytometry (Quanta SC; Beckman Coulter, Inc., Brea, CA, USA).

Samples were fixed and embedded according to a previous study (23) and the protocol provided by KingMed Center for Clinical Laboratory Co., Ltd. (Guangzhou, China). Tissue samples were fixed in 1% glutaraldehyde (Sigma-Aldrich) for 4 h followed by 1% osmium tetroxide (Sigma-Aldrich) for 1 h. Samples were then dehydrated in acetone and embedded in Eponate 12 resin (Ted Pella, Inc., Redding, CA, USA). Samples were cut into ultrathin sections (thickness, 70 nm), placed on copper grids and stained using 1% uranyl acetate and 1% lead citrate (Sigma-Aldrich). The sections were observed and images captured using a JEM-1230 transmission electron microscope (JEOL, Ltd., Tokyo, Japan).

Statistical analyses were performed in GraphPad Prism software version 5.00 (GraphPad Software, Inc., La Jolla, CA, USA). The data are presented as the mean ± standard error where applicable; differences were evaluated using Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Histological analysis with H&E staining revealed that the number of atretic follicles in the ovaries of cyclophosphamide-treated mice was significantly increased compared with the control group (P=0.018; Fig. 1A). In addition, the texture of ovarian tissues was dense in control mice. Various stages of follicles were observed in ovaries, including primordial follicles (Fig. 1B; pink arrows), antral follicles (Fig. 1B; red arrows), cumulus oophorus (Fig. 1B; blue arrows) and mature eggs (Fig. 1B; black arrows). Following treatment with cyclophosphamide for three weeks, ovarian tissues demonstrated significant atrophy, the ovarian stroma had empty space, and apoptosis and necrosis in ovarian granulosa cells was significant (Fig. 1B; yellow arrows). In addition, it was difficult to identify the various stages of follicles and mature eggs, and atretic follicles and luteal tissue atrophy were abundant (Fig. 1B). Immunohistochemistry results (Fig. 1C) indicated that the expression of the ovarian granulosa cell markers anti-Müllerian hormone (AMH) and inhibinα/β in the ovaries of POF mice was reduced compared with control mice; however, no difference was observed in FSH receptor (FSHR) staining between the two groups. The weight of ovarian tissues from cyclophosphamide-treated mice was significantly reduced compared with that of control mice (P=0.022; Fig. 2A). ELISA results demonstrated that the peripheral blood E2 level in mice treated with cyclophosphamide was significantly reduced compared with the control group (P=0.015; Fig. 2B), and the FSH level was significantly increased compared with the control group (P=0.027; Fig. 2C). Histological and hormone level detection results therefore suggested that cyclophosphamide induced ovarian granulosa cell injury and apoptosis as well as the development of POF.

Transmission electron microscopy was performed to confirm that telocytes were present in ovarian tissues. A unique type of cell body was observed in the ovarian stroma in control mice. Although the cell volume was small, it had very long telopodes, which had a typical dichotomous branching structure and alternating thick and thin segments (Fig. 3). However, cells with these morphological features were not observed in ovarian tissues from POF mice. Following a review of the literature, it was hypothesized that these cells were telocytes.

Flow cytometry detection of telocyte markers revealed that the number of CD34/PDGFRα, CD34/PDGFRβ and CD34/vimentin double-positive cells in the ovaries of POF mice was significantly reduced compared with control mice (P=0.031; Fig. 4), whereas the number of CD34/c-kit double-positive cells was not significantly different between the two groups (P=0.061; Fig. 4). In addition, immunofluorescence staining results indicated that the number of CD34/PDGFRα, CD34/PDGFRβ and CD34/vimentin double-positive cells in the ovarian tissues of POF mice was decreased compared with control mice (Fig. 5). These results suggested that the number of telocytes significantly decreased in ovarian tissues of mice following cyclophosphamide treatment.