Aldosterone and RU28318 were purchased from Steraloids (Newport, RI, USA). Platelet-derived growth factor BB (PDGF-BB) was purchased from Sigma (St. Louis, MO, USA). DMEM/F-12 and Medium 199 culture media were purchased from Invitrogen (Carlsbad, CA, USA). LH was a gift of NIDDK (US). EDS was purchased from Pterosaur Biotech (Hangzhou, China).

Ninety-day-old (adult) male Sprague-Dawley rats were purchased from Shanghai Laboratory Animal Co. Ltd. (Shanghai, China) and were raised in a 12 h dark/light cycle temperatures at 23 ± 2°C, and relative humidity of 45–55%. Eighteen rats were adjusted for a week before they were subjected to the intraperitoneal injection (i.p.) of EDS (75 mg/kg). This dose was used to completely deplete Leydig cells according to the previous study (10). Animal procedures were approved by the Institutional Animal Care and Use Committee of Tongji University and were performed in accordance with the Guide for the Care and Use of Laboratory Animals by the National Research Council.

Seven days after single i.p. of EDS when all Leydig cells were eliminated (11), seminiferous tubules were mechanically separated from the interstitium as previously described (11). The seminiferous tubules were distributed randomly into 12-well plates, with each well containing tubule fragments of equal total length (about 3 cm). Tubules were cultured at 34°C and 5% CO₂ in a basal medium (BM), which has 1:1 mixture of DMEM/F-12 and Medium 199 (pH 7.2), sodium bicarbonate (2.2 mg/mL), bovine serum albumin (1 mg/mL), and penicillin (100 U/mL)/streptomycin (100 µg/mL). ALDO (0–100 nM), and/or RU28318 (1 µM) were added to BM to investigate the effects on the proliferation of SLCs. RU28318 is a specific antagonist of NR3C2 (5).

The Leydig cell differentiation-inducing medium (LIM) was constructed by including 5 mM ITS (insulin, transferrin, and selenium), 5 ng/ml LH, and 5 mM lithium chloride as described (11). The previous study showed that seminiferous tubules in LIM were cultured for 14–21 days and SLCs on the surface of seminiferous tubules were fully differentiated into the Leydig cell lineage (8, 11, 12). In this LIM, addition of ALDO (0–100 nM) and/or RU28318 (1 µM) (a NR3C2 antagonist) was performed to investigate the effects of ALDO on SLC differentiation and its mechanism. Concentrations of ALDO and RU28318 were selected according to the previous studies (5, 13). ALDO or RU21318 was dissolved in DMSO and an aliquot of ALDO or RU21318 was added to medium with the final concentration of DMSO as 0.1%. Medium containing 0.1% DMSO served as the control.

Duplicated wells were used at each time point, and each experiment was replicated at least three times.

Medium concentrations of testosterone were measured by IMMULITE^® 2000 Total Testosterone Immunoassay Kit from Siemens (Gwynedd, UK). The duplicate assay for measurement of testosterone concentration in each medium was performed. The lower detection limit of testosterone was 0.2 ng/mL. Blank medium and blank medium containing 100 nM ALDO or 1 µM RU21318, or 100 nM ALDO + 1 µM RU21318 were measured to check whether they influence the measurement of testosterone levels. No influence was found. The intra-assay and inter-assay CVs were 5.75 and 7.53%, respectively.

Seminiferous tubules were cultured in BM or LIM for 14 days, and tubules then were air dried. Leydig cells on the surface of the tubules were evaluated by histochemical staining for HSD3B1 activity as described previously (14). Briefly, an aliquot of staining solution (25 µL), containing 0.4 mM etiocholanolone as the steroid substrate and 2 mM NAD⁺ as the cofactor and nitro blue tetrazolium chloride as the staining substrate, was added on seminiferous tubules. The samples were incubated at room temperature for 30 min. After staining, seminiferous tubules were washed using PBS and fixed with 4% paraformaldehyde at room temperature for 30 min. The tubules were visualized in a light microscope.

Stem Leydig cell adheres to the surface and its DNA synthesis was measured by the Click-iT^® EdU (EDU) Alexa Fluor^® 488 Imaging Kit (Life Technologies, OR, USA). A previous study demonstrated that PDGF-BB (10 ng/mL) robustly stimulated DNA synthesis in SLCs (15). Therefore, PDGF-BB (10 ng/mL) was used to increase DNA synthesis in SLCs. The isolated seminiferous tubules were cultured in BM with ALDO and/or RU28318 for the additional 5 days without (control) or with 10 ng/ml PDGF-BB. Then, EdU was supplied to the culture well and the samples were incubated for 24 h. Tubules were washed, fixed in 4% paraformaldehyde, and incubated with reaction solution. EdU incorporation in SLCs was visualized under a fluorescence microscope (Olympus, Tokyo, Japan) and images were captured. EdU-positive cells were counted and calculated by the flat surface area of the seminiferous tubules using the Image Pro Plus 6.0 software (Media Cybernetics, Inc., MD, USA). A total of 20 seminiferous tubule images were counted. Some seminiferous tubules were embedded in 2% argose gel, fixed in 4% paraformaldehyde, and cross sections (10 µm) were cut. Sections were incubated with mouse monoclonal antibody against α-smooth muscle actin (SMA, Sigma, A2547, 1:200) and the conjugated secondary antibody (Donkey anti-Mouse IgG Secondary Antibody, Alexa Fluor^® 594, Invitrogen, A-21203, 1:500) was used to visualize SMA. The EdU staining procedure was performed as above. The sections were visualized under immunofluorescence microscope.

Seminiferous tubules were treated with or without ALDO and/or RU28318 in LIM for 14 days and were embedded in 2% agarose gel, and cross sections (10 µm) were cut as previously described (11). In brief, the sections were later fixed in 4% paraformaldehyde. Sections were washed and then incubated with mouse monoclonal anti-SMA antibody (1:200) and rabbit monoclonal antibody against HSD11B1 (Pterosaur Biotech, PB 10021, 1:500) for 60 min followed by incubation with conjugated secondary antibody (Alexa Fluor^® 594 for SMA) and (donkey anti-rabbit Dylight 488, Bioworld, BS10018, 1:100 for HSD11B1) for 30 min. HSD11B1 was used to identify Leydig cells at the advanced stage because it begins expressed in rat Leydig cells starting on postnatal day 28 (16), when immature Leydig cells are emerged (17). SMA was used for labeling peritubular myoid cells (11). The sections were counterstained with DAPI (Sigma, D8417, 2 µg/mL). HSD11B1-positive Leydig cells per tubule cross section were counted.

A previous study showed that CD90⁺ cells on the surface of the seminiferous tubules were SLCs (11). Seminiferous tubules were treated with or without 1–100 nM ALDO in BM for 5 days. The seminiferous tubules were washed using PBS containing 1% BSA. Then, SLCs were stained using PE-conjugated CD90 antibody (Thermo Fisher Scientific, 12-0900-83, 1:100) in the dark place at 4°C for 45 min. The tubules were mounted on a slide for visualization under a fluorescence microscope (Olympus, Japan) and images were captured. CD90-positive cells were counted and calculated by counting SLC number per mm² flat surface area using the ImageProPlus 7.0 software (Media Cybernetics, USA).

Seminiferous tubules were treated with or without ALDO and/or RU28318 in LIM for 14 days. Then tubules were washed and submerged to 0.5 mL Trizol^® Reagent (Life Technologies, CA, USA), and total RNAs were extracted according to the manufacturer’s instruction as described (11). The Leydig cell genes and their primers were used as described previously (18–20). Primers used for qPCR had at least one span intron. These genes include Lhcgr, Scarb1, Star, Cyp11a1, Hsd3b1, Cyp17a1, Hsd17b3, Srd5a1, and Hsd11b1. The relative mRNA levels of targeted genes were normalized to Rps16 (internal control gene). The RNA was reversely transcribed into cDNA using Reverse Transcription System (Promega, WI, USA) according to the manufacturer’s instruction. The qPCR was carried out in a 25-µl reaction volume with SYBR Green detection system (Bio-Rad Laboratories, Inc., CA, USA). Light Cycler^®480 SYBR Green I Master was purchased from Roche Diagnostics (IN, USA). Reactions were run on a Bio-Rad qPCR system (Bio-Rad Laboratories, Inc., CA, USA) for up to 40 cycles and the melting curves were routinely checked afterward. The Ct value was recorded and the standard curve method was used to calculate the gene expression levels as previously described (21).

Seminiferous tubules were treated with or without ALDO and/or RU28318 in LIM for 14 days. Tubules were washed with PBS and submerged to the Radioimmunoprecipitation Assay Buffer (Beyotime Biotechnology, Shanghai, China) and homogenized to obtain total proteins. The protein concentrations of samples were measured with an Enhanced BCA Protein Assay Kit (Beyotime Biotechnology, Shanghai, China). An aliquot (50 µg of proteins) of sample was added to gel well and electrophoresed on 10% polyacrylamide gels containing sodium dodecyl sulfate and proteins were transferred. Then, the membranes were incubated overnight at 4°C with primary antibodies against the following antigens: SCARB1 (MultiSciences, 70-ab1967-050, 1:1,000), LHCGR (MultiSciences, 70-ab7496-050, 1:1,000), and β-actin (ACTB, Beyotime, AA128, 1:1,000). ACTB (house-keeping protein) served as the internal control. The membranes were then washed and incubated with HRP-conjugated anti-rabbit IgG secondary antibody (MultiSciences, 70-GAR0072, 1:2,000) or HRP-conjugated anti-mouse IgG secondary antibody (MultiSciences, 70-GAM0072, 1:2,000) for 2 h at room temperature and washed. The immunoreactive bands were visualized by chemiluminescence using Western Bright^® ECL (Advansta, CA, USA). The intensity of band was analyzed with Image J software 1.51j8 (NIH, USA). The intensity was adjusted and calculated relative to the control (set as 100%).

Values are expressed as mean ± SD, and data were analyzed by the GraphPad Prism 6 (GraphPad Software Inc., CA, USA). After the normal distribution is confirmed, multiple groups were performed by one-way ANOVA followed by ad hoc Tukey’s comparison of all columns compared with the control column. Mean value comparisons between two groups were performed by t-test. Differences were considered significant at P < 0.05.

To investigate the potential effects of ALDO on the proliferation of SLCs, we adopted an in vitro seminiferous tubule culture system. A previous study demonstrated that PDGF-BB (10 ng/ml) significantly increased EdU incorporation into the nuclei of SLCs (11). Isolated seminiferous tubules were cultured for 5 days with PDGF-BB (10 ng/mL) in the presence of 0–100 nM ALDO and/or a NR3C2 antagonist RU28318, then, EdU incorporation into SLCs on the surface of seminiferous tubules was investigated. As shown in Figure 1A, in BM, there are a few EdU-incorporated SLCs present on the surface of the seminiferous tubules. PDGF-BB (10 ng/ml) remarkably increased EdU-incorporated number of SLCs (Figure 1B), which were located on the surface of the seminiferous tubule (outside SMA-positive myoid cells, Figure 1B inset). When PDGF-BB plus 1–100 nM ALDO was used, ALDO concentration-dependently decreased PDGF-BB-stimulated EdU-incorporated SLCs (Figures 1C–E), with 100 nM ALDO being significant difference compared to the control (Figure 1E). RU28318 (the NR3C2 antagonist) alone did not affect PDGF-BB-mediated EdU incorporation (Figure 1G), while it reversed the effect of 100 nM ALDO (Figure 1F). This suggests that ALDO suppresses SLC proliferation via binding to NR3C2. Quantitative data of EdU incorporated cell number was summarized (Figure 1H).

We cultured the seminiferous tubules in BM (control) or LIM for 14 days. No HSD3B1-positive Leydig cells were formed in the control (Figure 2A), while many HSD3B1-positive Leydig cells were formed on the surface of the seminiferous tubule after culture with LIM (Figure 2B). Apparently, significantly higher level of testosterone was produced by the Leydig cells after culture with LIM (Figure 2D). This confirms the previous observation that LIM is an inducer for SLC differentiation.

To further examine whether ALDO can affect the pool of SLCs, the seminiferous tubules were cultured in BM with various concentrations of ALDO (0–100 nM) and/or RU28318 for 7 days and then the seminiferous tubules were switched into LIM for additional 14 days to induce these SLCs into the Leydig cells. Testosterone levels in the medium were significantly decreased at 100 nM ALDO (Figure 3A), and its effect was reversed by RU28318, indicating the effect was acted via binding to NR3C2 (Figure 3A). A previous study demonstrated that CD90-positive cells were SLCs (11). After 5 days of treatment with 0–100 nM ALDO in BM, we found that ALDO (100 nM) significantly lowered CD90-positive SLC number and the effect was reversed by RU28318 (Figures 3B–E).

The seminiferous tubules were continuously cultured in LIM for 14 days to induce SLCs into adult Leydig cells, thus producing testosterone as in Figure 2D. When various concentrations of ALDO were added in LIM and the seminiferous tubules were cultured for 14 days. ALDO decreased the HSD11B1-positive Leydig cell numbers at ≥10 nM and RU28318 reversed the ALDO-mediated effects (Figures 4A–F). ALDO decreased medium testosterone levels at ≥10 nM and RU28318 (1 µM) alone did not affect the SLC differentiation (Figure 4G). After adjustment by the Leydig cell numbered, it was found that 10 nM ALDO increased medium testosterone levels (Figure 4H). This indicates that ALDO stimulates Leydig cell differentiation of the existing stem cells.

We examined the effects of ALDO on the expression levels of Leydig cell-specific genes (Figure 5A). Statistically, we found ALDO significantly reduced the Scarb1 (encoding SCARB1) level at ≥10 nM and Lhcgr (encoding LHCGR) level ≥ 100 nM. However, ALDO did not affect other seven gene mRNA levels. However, we further calculated Lhcgr, Scarb1, Star, Cyp11a1, Hsd3b1, Cyp17a1, Hsd17b3, Srd5a1, and Hsd11b1 levels after adjustment of Leydig cell number (Figure 6) and we found that ALDO actually increased the expression levels of all these genes at 10 nM and higher. In parallel, we also selected two important Leydig-cell specific proteins, SCARB1 and LHCGR, and we found the levels of these two proteins were reduced by ALDO (Figure 5B or Figure S1–S3 adapted to ImageJ Software). However, ALDO increased SCARB1 and LHCGR protein levels per cell after adjustment of Leydig cells (Figure 6). RU28318 reversed ALDO-mediated effects. These results indicate that ALDO inhibits the proliferation of SLCs by lowering Leydig cell number but induces the differentiation of SLCs via increasing the Leydig cell specific gene expression.