Differentiation of human adipose derived stem cells into Leydig‐like cells with molecular compounds

Abstract Leydig cells (LCs) are the primary source of testosterone in the testis, and testosterone deficiency caused by LC functional degeneration can lead to male reproductive dysfunction. LC replacement transplantation is a very promising approach for this disease therapy. Here, we report that human adipose derived stem cells (ADSCs) can be differentiated into Leydig‐like cells using a novel differentiation method based on molecular compounds. The isolated human ADSCs expressed positive CD29, CD44, CD59 and CD105, negative CD34, CD45 and HLA‐DR using flow cytometry, and had the capacity of adipogenic and osteogenic differentiation. ADSCs derived Leydig‐like cells (ADSC‐LCs) acquired testosterone synthesis capabilities, and positively expressed LC lineage‐specific markers LHCGR, STAR, SCARB1, SF‐1, CYP11A1, CYP17A1, HSD3B1 and HSD17B3 as well as negatively expressed ADSC specific markers CD29, CD44, CD59 and CD105. When ADSC‐LCs labelled with lipophilic red dye (PKH26) were injected into rat testes which were selectively eliminated endogenous LCs using ethylene dimethanesulfonate (EDS, 75 mg/kg), the transplanted ADSC‐LCs could survive and function in the interstitium of testes, and accelerate the recovery of blood testosterone levels and testis weights. These results demonstrated that ADSCs could be differentiated into Leydig‐like cells by few defined molecular compounds, which might lay the foundation for further clinical application of ADSC‐LC transplantation therapy.

epidemiological study showed that near 4.5 million men in USA may suffer from testosterone deficiency. 5,6 To some extent, the testosterone deficiency can be relieved by exogenous testosterone replacement therapy. 7,8 However, this treatment disrupts the normal function of hypothalamic-pituitarytesticular axis, has no circadian rhythm and may cause a number of adverse reactions such as cardiovascular disorders, prostate tumorigenesis, et al. 9,10 Moreover, as physiological requirements for testosterone vary in individuals, the supplementation of exogenous testosterone is difficult to meet the demands of individual-based treatment. 11 LC transplantation had been demonstrated as a longacting and ideal physiological system of testosterone delivery. 12 However, LC has the limited ability to proliferate, and account for only 2%-4% of the total adult human testicular cells, 13 limiting the clinical application of LC transplantation therapy.
Stem cells have significant pluripotency and self-renewal characteristics, and had been universally used as seed cells in regenerative medicine or tissue engineering in recent years. 14 Thus, the transplantation of stem cell-derived LCs may be the alternative promising therapy for testosterone deficiency. Although several researches had attempted to induce stem cells, such as induced pluripotent stem cells (iPSCs), 15 embryonic stem cells (ESCs), [16][17][18] and mesenchymal stem cells (MSCs) 19,20 into steroidogenic cells using exogenous gene transcription factor transfection, it is not so safe for further clinical application. Human adipose derived stem cells (ADSCs) have been investigated as ideal seed cells because of the ease of obtaining, low immunogenicity and higher differentiation capacity. 21 Currently, the differentiation of ADSCs into LCs using molecular compounds but not bringing in the exogenous genes had not been reported.
In our study, human ADSCs were induced into Leydig-like cells (ADSC-LCs) using a molecule-compound-based strategy.
Transplantation of these ADSC-LCs into the rat models treated with ethylene dimethanesulfonate (EDS) 22 could promote the recovery of blood testosterone levels as well as reproductive organ weights.
These findings will provide a new insight on the development of cell transplantation therapy for testosterone deficiency.

| Human adipose derived stem cell isolation and culture
The adipose tissues were obtained from the abdomen of five male donors with the mean age at 38 years by liposuction. The informed consent was acquired from every donor, and our research was approved by Human Research and Ethical Committee of Wenzhou Medical University. The fresh adipose tissues were microdissected into 1 mm 3 pieces using a stereoscope, and then washed five times with PBS containing streptomycin (300 mg/mL) and penicillin (600 U/mL) to remove local anaesthetics, blood clots and red blood cells. The rest of tissues were digested using 0.1% type I collagenase at 37°C shaking table for 1 hour. Then these tissues were filtered through the cell strainer (100 μm) and centrifuged at 300 × g for 10 minutes. The deposit was suspended using low-glucose Dulbecco's Modified Eagle's Medium (LG-DMEM, Gibco) containing 10% fatal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin (P/S, Gibco). The suspension was filtered through the nylon mesh (100 μm), and seeded into the 25 cm 2 Petri dishes at a 37°C, 5% CO 2 incubator. The medium was replaced with fresh medium every 3 days. When reaching 80%-90% confluence, the cells would be passaged with 0.25% EDTA-trypsin.

| Osteogenic induction
ADSCs were conducted the osteogenic induction at passage 2 (P2) as the reported approach. 23 ADSCs (1 × 10 6 cells/well) were seeded into six-well cell culture dishes. Then cells were cultured under the osteogenic differentiation medium for 2 weeks, namely LG-DMEM supplemented with 1 μM dexamethasone (Sigma), 50 μM ascorbate-2-phosphate (Sigma), 100 μM glycerophosphate (Sigma), 10% FBS and 1% P/S. Then cells were rinsed three times with PBS, and were fixed by 4% paraformaldehyde at room temperature for 10 minutes. Then cells were rinsed three times using PBS, and stained using Alkaline Phosphatase Kit (Sigma). Lastly, cells were gently washed three times with deionized water, and were imaged under the inverted fluorescence microscope (OLYMPUS).

| Human Leydig cell isolation and culture
LCs were acquired from five male donors with the mean age of 45 years by testes excision within 20 hours. Informed consent was obtained from every donor, and our research was approved by the Human Research and Ethical Committee of Wenzhou Medical University. These testes were used to isolate immature Leydig cells (ILCs), which can express all androgen synthases, 24 and are capable of differentiation and proliferation. 25 Briefly, the testes through testicular artery were perfused with M-199 buffer (Gibco, NY, USA) containing DNase (0.25 mg/mL, Sigma) and collagenase (0.25 mg/ mL, Sigma) for digestion about 15 minutes. Cell suspension was then filtered with nylon mesh (100 µm), and cells were separated using the Percoll Gradient (Sigma). The cells at the density of 1.070-1.088 g/ mL were harvested. The purity of ILCs was assessed by HSD3B1 immunohistochemical staining solution, which contained NAD + as a cofactor and 0.4 mM etiocholanolone (Sigma). 26 The purity of ILCs was more than 95%.

| Differentiation of human adipose derived stem cells into Leydig-like cells
The point at which ADSCs (1×10 5 cells/well) were seeded onto the six-well culture plates in the ADSC medium was defined as day −5.

| Testosterone measurement by radioimmunoassay
The blood and cell culture supernatants were harvested for the quantitative detection of testosterone levels. For cell supernatant, 10 ng/mL LH was added into the medium (only DMEM/F12) in advance at least 3 hours to stimulate the testosterone production of LCs or ADSC-LCs. The levels of testosterone were detected using a tritium based radioimmunoassay with the antibody of anti-testosterone as previously reported. 28 Standard samples containing testosterone ranging from 10 to 2000 pg/mL were prepared in triplicate.
Standard or experimental samples were treated with antibody and tracer at 4°C overnight, and charcoal dextran suspension was used to separate the free and bound steroids. The bound steroid was mixed with the scintillation buffer and counted in the β scintillation counter (PE, CA, USA). The minimum detectable concentration of testosterone was 5 pg/mL. The inter-assay and intra-assay coefficient of variation was within 10%.

| Immunofluorescence assay
Immunofluorescence was performed to identify ADSC-LCs as previously reported. 23 Cells were fixed by 4% paraformaldehyde (Sigma) for 15 minutes, and washed three times using PBS. Cells were then permeabilized using 0.1% TritonX-100 at room temperature for 15 minutes, and incubated with 3% (w/v) BSA at room temperature for 1 hour. Cells were then incubated by primary antibodies (Table 1) overnight at 4°C, and then with FITC-or Cy3-conjugated secondary antibodies (1:1000, Bioword, USA) at room temperature for 60 minutes. Lastly, cells were rinsed three times using PBS, and incubated with DAPI (Sigma) for nuclear staining for 15 minutes, and rinsed three times with PBS before examination under the inverted fluorescence microscope (OLYMPUS).

| Reverse transcription polymerase chain reaction (RT-PCR) and real time polymerase chain reaction (qPCR)
Total RNAs from the samples were extracted using TRIZOL reagent (Gibco). The RNAs were reversely transcribed into cDNAs by the Reverse Transcription Kit (TOYOBO, Japan). The cDNAs were diluted 1:10, which were then used to conduct RT-PCR or qPCR. RT-PCR was conducted using the Authorized Thermal Cycler (Eppendorf, Hamburg, Germany). After amplification, 6 μL of each PCR product and 2 μL of loading buffer were mixed, and were electrophoresed into the 2% agarose containing nucleic acid dye (Sigma). Gel was scanned for further analysis. qPCR was conducted by the Thunderbird SYBR qPCR Mix (Takara, Tokyo, Japan) as the product instruction. Signals were harvested using the Light Cycler 480 Detection System (Roche, Basel, Switzerland). The relative gene expressions were normalized to GAPDH. The quantification was performed with the comparative 2 -ΔΔCt approach. The sequences of primers were shown in Table 2.

| Western blotting
Cells were rinsed using cold PBS, and were then lysed using lysis buffer containing protease inhibitor/1% phosphatase inhibitor mixture (Roche). The each sample with 50 μg of protein was applied into the 10% SDS-PAGE, and was then transferred into the polyvinylidene difluoride membranes (PVDF, Sigma) using an electroblot apparatus. After blocked with the blocking solution (5% free-fat milk) for 2 hours at 4°C, the PVDF membranes were then incubated using primary antibodies (Table 1) at 4°C overnight. The PVDF membranes were then rinsed five times (10 minutes each) using TBST, and incubated with HRP-conjugated secondary antibody (1:3000, Bioword) at room temperature for 2 hours. The PVDF membranes were rinsed five times (10 minutes each) using TBS-T. Bands were imaged by enhanced chemiluminescence (ECL, Pierce, USA).

| Flow cytometry
Flow cytometry was conducted as the reported method. 21 Briefly, cell samples were fixed using 4% paraformaldehyde in PBS, and permeabilized using 0.1% TritonX-100 (Sigma). Cells were then labelled with primary or isotype control antibodies at 4°C for 30 minutes.
Primary or isotype control antibodies were labelled with fluorophore conjugated secondary antibody at 4°C for 30 minutes. The labelled samples were detected by flow cytometry analyser (BD, USA).

| Tagging ADSC derived Leydig-like cells (ADSC-LCs) with PKH26
The standard protocol was performed according to PKH26 Product Information Sheet (Sigma, MINI2). Briefly, the cell suspension with 2×10 7 cells was centrifuged (400 × g, 5 minutes), and then were washed once using fresh LG-DMEM without FBS. After centrifugation, the supernatant was removed, and 1 mL of Diluent C was added. Cells were resuspended with gentle pipetting to ensure complete dispersion. 2 × Dye Solution (4 × 10 -6 M) was prepared through adding 4 μL of PKH26 ethanolic dye solution into 1 mL of Diluent C, and mixed them well. Then, 1 mL of 2 × Dye Solution was rapidly added into the cell suspension. Final concentration was 2 × 10 -6 M PKH26 for 1×10 7 cells/well. The mixing suspension was incubated using periodic mixing at room temperature for 5 minutes. The staining was stopped through adding 2 mL of FBS. Then the suspension was centrifuged (400 × g, 10 minutes) and rinsed three times. Finally, the cells tagged with PKH26 were transfer to fresh wells and used for transplantation.

| Transplantation of ADSC derived Leydig-like cells (ADSC-LCs) in vivo
For assessment whether ADSC-LCs could facilitate the recovery of testosterone deficiency of rats, ADSC-LC transplantation was conducted according to the previous report with some modifications. 29 Sixty 49-day-old male SD rats (n = 5 for each group at each time point) were used in this study. Water containing 210 mg/L Cyclosporin A (Sigma) was given to these rats throughout the experiment to prevent allograft rejection. Before cell transplantation, male rats were performed a single intraperitoneal injection of EDS (75 mg/kg, Pterosaur Biotech Co., Ltd., Hangzhou, China). EDS treatment would eliminate LCs in the adult testes of rats. 30 Then, ADSC-LCs labelled with PKH26 (red) were resuspended manually in a 15 mL tube. Cells were then washed twice using PBS and centrifuged (200 × g, 5 minutes). Lastly, each cell pellets were resuspended in PBS for transplantation. Cells were loaded into a 1 mL syringe for injection into the testis of adult male SD rats with EDS treatment.

| Immunohistochemistry
One testis from each rat was used for immunohistochemistry. The rats were killed using the overdose of sodium pentobarbital (Sigma). Testes were removed, and were fixed with 4% paraformaldehyde at 4°C overnight. Then testes were dehydrated using a graded series of xylene and ethanol, and were then embedded into paraffin. Five-micrometer-thick sections were cut, de-waxed in water and then were mounted on the glass slides.
Antigen retrieval was conducted with the microwave irradiation in 10 mM citrate buffer (pH 6.0) for 10

| Enumeration of Leydig cell number by stereology
To enumerate CYP11A1 positive LC numbers, sampling of the testis was performed according to a fractionator method as our previous report. 32 Identification of all LC lineages was done by the staining of CYP11A1. About 10 testis sections per rat were sampled from each testis. The total number of LCs was calculated by multiplying the number of LCs counted in a known fraction of the testis by the inverse of the sampling probability.

| Statistical analysis
All experiments were performed at least thrice, and the data are presented as the mean ± SEM. Statistical analyses were evaluated using an unpaired Student's t test or one-way ANOVA for more than two groups. P < 0.05 was considered statistically significant.

| The isolation and identification of human adipose derived stem cells
Human ADSCs could be isolated from liposuction adipose tissue using type I collagenase digestion. Generally, after isolation for  Furthermore, they were passaged and cultured in LC Medium for the following experiments. The schematic illustration is showed in Figure 2A. ADSCs grew well and reached about 100% confluence on day 0. After differentiation for 7 days, the shape of cells became oval from long-spindle with strong stereoscopic impression, but on day 14, the stereoscopic impression of these oval cells was disappeared, and on day 18, the partial Leydig-like cells (ADSC-LCs) exhibited ellipse shapes. These cells were tended to grow together to form clusters ( Figure 2B). Under the stimulus with 10 ng/mL LH for 3 hours, the enrichment ADSC-LCs could secrete testosterone (T) into the medium, and the level of T was more than that of ADSCs, but less than that of LCs ( Figure 2C).

| Differentiation of human adipose derived stem cells into Leydig-like cells (ADSC-LCs)
These results suggested that our approach based on the molecular compound induction is able to differentiate the partial ADSCs into Leydig-like cells.

| Identification of Leydig-like cells derived from human adipose derived stem cells (ADSC -LCs)
After differentiation and enrichment, the immunofluorescence assay These results demonstrated that the partial ADSCs were successfully differentiated into Leydig-like cells based on the molecular compound induction.

| Identification of Leydig-like cells derived from human adipose derived stem cells (ADSC -LCs)
RT-PCR assay was also conducted to characterize the expressions  which were significantly less than those of ADSCs ( Figure 4B). The heat map was quantified to more visually exhibit the consequences of qPCR. Red means the gene level is high, and green means the gene expression level is low ( Figure 4C). These results also demonstrated that ADSCs could be differentiated into Leydig-like cells based on the induction method of molecular compounds.

| Identification of Leydig-like cells derived from human adipose derived stem cells (ADSC-LCs)
The flow cytometry histograms were employed to assess the population levels of LC biomarkers CYP11A1, HSD3B1, CYP17A1, and 5B).
Western blotting assay was also used to identify the expressions of LC or ADSC protein biomarkers in ADSC-LCs. The results showed that ADSC-LCs could positively express LC biomarkers such as STAR, SCARB1, SF-1, CYP11A1 and HSD17B3, which were androgen biosynthetic enzymes for testosterone synthesis, but negatively express ADSC biomarkers CD29 and CD59. These protein expressions in undifferentiated ADSCs were contrary to ADSC-LCs, and LCs were in consistent with ADSC-LCs ( Figure 5C).
Taken together, these results further suggested that ADSCs could be partially differentiated into Leydig-like cells based on the induction method of molecular compounds.

| Transplantation of Leydig-like cells derived from human adipose derived stem cells (ADSC -LCs) into the testes of rats with EDS treatment
To assess whether ADSC-LCs have the ability to survive and func-  Figure 6A). After 14 days of cell transplantation, the PKH26-labelled ADSC-LCs (red) were exclusively distributed in the interstitium of the testis, and expressed the LC-specific marker CYP11A1 (green). In EDS-treated rats with PBS administration, the CYP11A1-positive cells were very less in the interstitium, but PBS injected rats without EDS treatment (control) strongly expressed CYP11A1 ( Figure 6B). Furthermore, after EDS administration, the levels of blood testosterone were dramatically declined to the undetectable bottom on day 7 and recovered gradually, suggesting that EDS could specifically eliminate the testosterone-producing LCs in the adult rats. In the EDStreated rats, the levels of blood testosterone began to increase on day 14, which was recovered to about 20% of control rats.
Notably, the levels of blood testosterone in EDS treated rats with ADSC-LC transplantation were higher than that of EDS treated rats with PBS injection, but were even lower than that of control rats with PBS injection on day 14 and 21 ( Figure 6C). Testosterone plays an important role in maintaining the normal weights of reproductive organs. 35 After exposure to EDS, the weights of testes were dramatically reduced in the EDS-treated rats on day 7, but this decrease could be rescued by ADSC-LC transplantation.
Quantitative results showed that the absolute weights of the testes in EDS treated rats with ADSC-LC transplantation were higher than that of EDS treated rats with PBS injection, but both were lower than that of control rats with PBS injection on day 14 and 21 ( Figure 6D). After EDS treatment, the bodyweights of rats were also decreased on the first 7 days. Subsequently, the bodyweights in ADSC-LC transplanted rats or PBS injected rats started to recover, but both were still lower than that of control rats with PBS injection on day 14 and 21 ( Figure 6E). These results suggested that ADSC-LCs transplanted in rats had acquired some properties of LCs as they had the potential to restore the blood testosterone levels, and further recover the testis and bodyweights of EDS treated-rats.

| Enumeration of CYP11A1 positive Leydig cell number
All LCs could be identified by CYP11A1 staining because this en- was almost similar to that on day 21 (Fig. S1B). These data suggested that the increasing CYP11A1 positive cells might be mainly derived from the endogenous regenerated LCs, and ADSC-LCs almost had no effects on the endogenous LCs.

| D ISCUSS I ON
Androgen deficiency is a very common physical disorder that not only affects adults but also jeopardizes children's health. 36  very similar to MSCs in terms of surface antigens and they also possess multipotentiality. 43,44 As adipose tissue can be more easily obtained than other tissues, ADSC is a more reliable stem cell source for therapy. 45,46 Therefore, in this study, we will try to induce human to the NR5A subfamily, which is essential for sexual differentiation and formation of the primary steroidogenic tissues. 54  Compared with other organs, testis is immunologically privileged. 62 To investigate whether ADSC-LCs have the ability to survive and function in the interstitium of rat testes in vivo, we transplanted ADSC-LCs into an EDS-treated rat, an androgen deficiency model, as previously described. 63 EDS is an alkylating agent which has selective pro-apoptotic effects on LCs. 31 Approximately 2-3 weeks after a single dose of EDS, newly regenerated LCs could be observed within the testicular interstitium. 64 Approximately 8 to 10 weeks later, the LC population returned to its original size and had restored its ability to produce testosterone. 65 Based on these results, we collected the testes of the cell-transplanted rats on day 21 after EDS administration when the regenerated LCs appear a little to assess the state of the transplanted cells. As a result, we observed that the transplanted PKH26-labelled ADSC-LCs localized exclusively in the interstitium of the testis, and expressed LC-specific marker CYP11A1. Importantly, these PKH26-labelled ADSC-LCs could survive at least 2 weeks in vivo, which demonstrated that they successfully integrated into the host niche. In addition, the blood testosterone levels of the ADSC-LC transplanted rats remained higher than that of the EDS-treated rats up to 21 days.
As testosterone plays an important role in maintaining normal reproductive organs, 66 testosterone deficiency can cause atrophy of reproductive organs. 4 After exposure to EDS, the testis and bodyweights of EDS-treated rats were dramatically decreased on day 7. Subsequently, however, these indicators in rats receiving ADSC-LC transplantation could be restored faster than those of PBS injected rats with EDS treatment, but both were still lower than those of PBS injected rats without EDS treatment (control) on day 14 and 21.
In this study, we developed a novel differentiation protocol, to our knowledge, which is the first to demonstrate that ADSCs were able to be differentiated into testosterone-producing Leydig-like cells (ADSC-LCs) by few defined molecular compounds but not bringing in the exogenous transcription factors. In addition, when ADSC-LCs labelled with lipophilic red dye (PKH26) were transplanted into EDS treated rats, they could survive and function in the interstitium of testes, and accelerate the recovery of blood testosterone levels, testis weights and bodyweights. Our findings provide a new insight into the stem cell-derived Leydig cell replacement therapies for the treatment of the patients with testosterone deficiency or decline.

CO N FLI C T O F I NTE R E S T
The authors have declared that no conflict of interest exists.