Transplantation of Human Urine-Derived Stem Cells Transfected with Pigment Epithelium-Derived Factor to Protect Erectile Function in a Rat Model of Cavernous Nerve Injury

Abstract

The aim of this study was to investigate whether intracavernous injection of urine-derived stem cells (USCs) or USCs genetically modified with pigment epithelium-derived factor (PEDF) could protect the erectile function and cavernous structure in a bilateral cavernous nerve injury-induced erectile dysfunction (CNIED) rat model. USCs were cultured from the urine of six healthy male donors. Seventy-five rats were randomly divided into five groups (n = 15 per group): sham, bilateral cavernous nerve (CN) crush injury (BCNI), USC, USC^GFP+, and USC^GFP/PEDF+ groups. The sham group received only laparotomy without CN crush injury and intracavernous injection with phosphate-buffered saline (PBS). All of the other groups were subjected to BCNI and intracavernous injection with PBS, USCs, USCs^GFP+, or USCs^GFP/PEDF+, respectively. The total intracavernous pressure (ICP) and the ratio of ICP to mean arterial pressure (ICP/MAP) were recorded. The penile dorsal nerves, the endothelium, and the smooth muscle were assessed within the penile tissue. The USC and USC^GFP/PEDF+ groups displayed more significantly enhanced ICP and ICP/MAP ratio (p < 0.05) 28 days after cell transplantation. Immunohistochemistry (IHC) and Western blot analysis demonstrated that the protection of erectile function and the cavernous structure by USCs^GFP/PEDF+ was associated with an increased number of nNOS-positive fibers within the penile dorsal nerves, improved expression of endothelial markers (CD31 and eNOS) and a smooth muscle marker (smoothelin), an enhanced smooth muscle to collagen ratio, decreased expression of transforming growth factor-β1 (TGF-β1), and decreased cell apoptosis in the cavernous tissue. The paracrine effect of USCs and USCs^GFP/PEDF+ prevented the destruction of erectile function and the cavernous structure in the CNIED rat model by nerve protection, thereby improving endothelial cell function, increasing the smooth muscle content, and decreasing fibrosis and cell apoptosis in the cavernous tissue.

Keywords

Cavernous never injury Erectile dysfunction (ED)Urine-derived stem cells (USCs)Pigment epithelium-derived factor (PEDF)

Introduction

Erectile dysfunction (ED) following radical prostatectomy (RP) is a frequent complication for patients with clinically localized prostate cancer that has a significant negative impact on the quality of life of these patients and their partners¹. The main cause of ED after RP is neuropraxia, followed by hypoxia and fibrosis of penile tissues^2,3. New surgical techniques such as laparoscopic nerve-sparing or robotic-assisted surgery may improve the potency rates, but the results have not been proven with a methodologically sound study^4,5. Oral phosphodiesterase type-5 (PDE5) inhibitors are currently regarded as the first-line treatment for ED and have a success rate of greater than 80% in the total ED population⁶. However, the response rate of these drugs is much lower and varies distinctly in these patients due to the apoptosis of corporal smooth muscle cells (SMCs) and the accumulation of collagen⁷. The treatment for postoperative ED remains challenging. Therefore, new treatment strategies for ED after RP are highly desirable.

Recently, stem cell (SC) therapy has become a hot topic in research for ED treatment⁸. A variety of SCs, such as embryonic stem cells (ESCs), muscle-derived stem cells (MDSCs), bone marrow-derived mesenchymal stem cells (BM-MSCs), and adipose tissue-derived stem cells (ADSCs), have proven to be effective for the treatment of cavernous nerve (CN) injury-induced erectile dysfunction (CNIED)^9–12. However, it is difficult to identify these intracavernosal-injected SCs in penile tissues. Kendirci and colleagues demonstrated that BM-MSCs secreted multiple neuroprotective and vasoprotective growth factors and cytokines, including basic fibroblast growth factor (bFGF), nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), vascular endothelial growth factor (VEGF), and insulin-like growth factor 1 (IGF-1)¹⁰. Our previous study also demonstrated that SCs could improve the erectile function of diabetic ED rats. Similar to other studies, the engraftment and differentiation of SCs could not be observed in the penile tissues 28 days after cell transplantation^13,14. The paracrine effect is considered to be the major mechanism contributing to the therapeutic effects of SCs for ED. However, the exact paracrine mechanisms have not been fully elucidated¹⁵.

Urine-derived stem cells (USCs) are a population of SCs that can be obtained from human voided urine using a noninvasive and low-cost method^16–18. They can be easily expanded in vitro and have the capacity for multipotent differentiation, and thus may represent one of the most promising sources for cell therapy for the treatment of urological diseases¹⁹. We have previously determined that USCs and USCs genetically modified with bFGF (also called FGF2) improved erectile function in a diabetic ED rat model.

Our previous study demonstrated that USCs could secrete pigment epithelium-derived factor (PEDF) in vitro and enhance nerve regeneration²⁰. PEDF is a 50-kDa glycoprotein that belongs to the noninhibitory and multi-functional serpin group²¹. It was first discovered in the culture medium of human fetal retinal pigment epithelial cells as a neurotrophic factor²². Recent studies have shown that PEDF is widely expressed throughout fetal and adult tissues and has multiple important functions, including differentiation, neuroprotective effects, and antiangiogenic effects on different cells or tissues²³. Our previous study demonstrated that the insufficient secretion of PEDF within the cavernous tissue was involved in CNIED²⁴. In this study, we investigated the efficacy of USCs and USCs overexpressing green fluorescent protein (GFP) and PEDF (USCs^GFP/PEDF+) in the treatment of a CNIED rat model and the underlying mechanisms of each treatment.

Materials and Methods

Ethics Statement

Human urine samples were obtained from six healthy male, voluntary donors. All donors provided informed consent and did not receive any financial compensation. The procedure was approved by the Sun Yat-sen University Health Sciences Institutional Review Board (IRB). Seventy-five male Sprague–Dawley (SD) rats (8–10 weeks old) were purchased from the Animal Center of Sun Yat-sen University (Guangzhou, Guangdong Province, P.R. China). The animals were kept in a standard, pathogen-free environment on a 12-h light/12-h dark cycle with free access to laboratory chow and water. The animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Sun Yat-sen University.

Study Design

USCs were cultured from the urine of six healthy male donors. Seventy-five SD rats were randomly assigned to five groups (n = 15 per group): sham, bilateral CN crush injury (BCNI), USC, USC^GFP+, and USC^GFP/PEDF+ groups. The sham group received only laparotomy without CN crush injury and intracavernous injection with phosphate-buffered saline (PBS; Boster, Wuhan, P.R. China). BCNI, USC, USC^GFP+, and USC^GFP/PEDF+ groups all received BCNI and intracavernous injection with PBS, USCs, USCs^GFP+, or USCs^GFP/PEDF+, respectively. Fourteen days after cell transplantation, five rats from each group were euthanized, and the penile tissue was harvested to analyze the PEDF protein level and the inflammation in the cavernous tissue. At 28 days posttreatment, the erectile function of all groups was evaluated using the total intracavernous pressure (ICP) and the ratio of ICP to mean arterial pressure (ICP/MAP) before the penile tissues were harvested for further experiments. One rat in the BCNI group died on day 13, and one rat in the USC^GFP+ group died on day 21 after the cell transplantation; they were excluded in this study.

Isolation, Culture, and Identification of Human USCs

A total of 18 sterile urine samples were collected from six healthy male donors with an average age of 31.8 years. The fresh mid- and last-stream urine samples were centrifuged at 500 × g for 5 min. The cell pellets were resuspended with mixed medium composed of keratinocyte serum-free medium (KSFM; Invitrogen, Carlsbad, CA, USA) and progenitor cell medium (Lonza, Basel, Switzerland) in a 1:1 ratio as previously described¹⁸. Approximately 500 cells per well were then plated in a 24-well cell culture plate and incubated under standard conditions (5% CO₂ and 37°C) for 48 h. The culture medium was changed every other day. After 1 week, the cells were observed with an inverted phase-contrast microscope (Nikon Instruments Inc., Melville, NY, USA). Only independent clones derived from a single cell were trypsinized (Invitrogen) for subsequent passage and culture. The cells were passaged at approximately 80% confluence. The human USCs were identified according to our previously described methods¹⁴. Fluorescence-conjugated antibodies (see Table 1 for a complete list of antibodies and concentrations) were incubated with the USCs at 4°C for 30 min. The expression of USC cell surface antigens (passage 3) was measured by flow cytometry analysis (FACSCalibur; BD Biosciences, Franklin Lakes, NJ, USA). The osteogenic and adipogenic differentiation of the USCs were assessed by Alizarin red S (Santa Cruz Biotechnology, Santa Cruz, CA, USA) or Oil red O staining (Abcam, Cambridge, UK), respectively.

Table 1.

Primary Antibodies Used in the Flow Cytometry Analysis

Primary Antibody	Company/Catalog No.	Antibody Dilution
CD24-FITC	BD, 560992	20 μl/test
CD31-FITC	BD, 560984	20 μl/test
CD34-FITC	BD, 348053	20 μl/test
CD44-PE	BD, 550989	20 μl/test
CD45-PE	BD, 555483	20 μl/test
CD73-PE	BD, 550257	20 μl/test
CD90-APC	BD, 559869	5 μl/test
CD105-PE	BD, 560839	5 μl/test
CD133-PE	MACS, 130080801	1:11
CD146-PE	BD, 550315	20 μl/test

Transfection of Human USCs with the PEDF-GFP Lentivirus or GFP Lentivirus Alone

Lentiviruses expressing GFP and PEDF or GFP alone were generated and modified according to our previously described method²⁵. Briefly, to construct the PEDF expression vector, the precursor sequence for PEDF was amplified from human genomic DNA by polymerase chain reaction (PCR) including the overhang sequences from a 5′ BamHI and a 3′ EcoRI restriction sites and cloned into the pHBLV-CMVIE-ZsGreen-T2A-puro lentiviral vector (Hanbio, Shanghai, P.R. China). The construct was verified by sequencing and designated pHBLV-GFP-PEDF.

For the lentivirus envelope, recombinant lentiviruses were produced by cotransfection of purified pHBLV-GFP-PEDF or pHBLV-GFP with the packaging plasmids PSPAX2 and PMD2G using the LipFiter system (Hanbio). Lentivirus-containing supernatants were harvested from the cultures after 48 or 72 h and filtered using 0.22-μm cellulose acetate filters (Millipore, Billerica, MA, USA). The lentiviruses were then concentrated with Amicon Ultra-15 Centrifugal Filter Units (Millipore) and stored at −80°C¹⁴.

For human USC infection, the pHBLV-GFP-PEDF storage solution was diluted and added to 1 × 10⁶ USCs cultured at a multiplicity of infection (MOI) of 20. Polybrene (Sigma-Aldrich, St. Louis, MO, USA) was added to the USC culture at a concentration of 6 μg/ml to improve the transduction efficiency. USCs transduced with pHBLV-GFP were used as a negative control. After 48 h, puromycin (Sigma-Aldrich) at a concentration of 2 μg/ml was added to the USC culture for resistance screening. Three passages after screening, the USCs were examined under a fluorescence microscope (Leica Microsystems, Wetzlar, Germany).

For PEDF expression analysis, the USC supernatants were analyzed by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's instructions (R&D Systems, Minneapolis, MN, USA). Serum-free medium was used as a control. Briefly, 100 μl of the standard or samples was incubated with precoated 96-well plates for 2 h on a rocking platform at room temperature (RT), followed by incubation with the biotin-labeled antibody and horseradish peroxidase (HRP) for 1 h at 37°C. The optical density (OD) values were read at the 450-nm wavelength using a microplate reader (Model 680; Bio-Rad, Hercules, CA, USA). USCs (1 × 10⁶ cells) were also harvested to extract total proteins for the Western blot analysis.

Establishment of a CNIED Rat Model

A rat model of BCNI-induced ED was established according to our previously described methods²⁶. Briefly, adult male SD rats were anesthetized with diluted pentobarbital sodium (40 mg/kg) via an intraperitoneal (IP) injection. They were subsequently fastened in a supine position on an isothermic surgical pad. After the surgical area was prepared with an iodinated disinfectant, a lower midline abdominal incision was performed and the prostate gland was exposed and identified. The major pelvic ganglion (MPG) and CN were isolated posterolaterally on both sides of the prostate under a dissecting microscope (10× magnification). For the sham group, the abdomen was then closed without any further surgical manipulation. For the crush group, both sides of the CN, 5 mm distal to the MPG, were directly crushed with mosquito hemostatic forceps (J31020; Jinzhong, Shanghai, P.R. China) for 1 min. After the penis was exposed, a 1 × 10⁶ human USC (passage 3) suspension in 0.2 ml of PBS or 0.2 ml of PBS alone was injected into the corpus cavernosa at the middle level at the same time as the nerve crush, according to our previous studies^14,24. An elastic band was fixed at the base of the penis before the injection of cells and was maintained for 2 min after the cell injection. All rats were not given any immunosuppression before or after surgery.

Erectile Function Evaluation

Erectile function was evaluated using total ICP and the ratio of ICP to MAP 4 weeks after intracavernous injection, as previously described^11,14. Briefly, rats were anesthetized with pentobarbital sodium (40 mg/kg) via an IP injection. The left carotid artery was subsequently isolated and cannulated with a PE-50 catheter filled with heparinized saline (250 IU/ml) to monitor the MAP. For the ICP measurement, a 25-gauge needle was inserted into one side of the penile crus and connected to another pressure transducer. The CN was identified and isolated with a midline laparotomy. A bipolar hook electrode attached to a signal generator (BL-420F; Taimeng, Chengdu, P.R. China) was placed to the left CN distal to the nerve injury for stimulation. Monophasic rectangular pulses (stimulus parameter settings of 0.2-ms width, 1.5 mA, 20-Hz frequency, and 60-s duration) were recorded and analyzed with BL New Century 2.1 software (Taimeng). The erectile function was evaluated as the ratio of ICP (mmHg)/MAP (mmHg) to normalize for variations in systemic blood pressure. Total ICP was measured by the area under the ICP curve (AUC; mmHg × second) according to the method described in the literature²⁷. The penis was then harvested for Western blot and histological analysis.

Western Blot Analysis

Whole proteins extracted from the homogenized penis or USC lysate were analyzed by Western blot on the Odyssey infrared imaging system (LI-COR Biosciences, Lincoln, NE, USA). Briefly, protein samples were collected, and the concentrations were determined with a bicinchoninic acid (BCA) protein assay kit (Beyotime Institute of Biotechnology, Haimen, P.R. China). Thereafter, equal amounts of protein (10–20 μg per lane) were loaded and separated on a 6–12% sodium dodecyl sulfate (SDS) polyacrylamide gel. Primary antibodies were applied after the proteins were transferred to nitrocellulose membranes; glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the loading control. Signals were obtained in the linear range of detection and quantified with the Odyssey infrared imaging system. The data were analyzed and presented as the relative density of each protein relative to GAPDH or β-actin. The primary antibodies used in this study were as follows: rabbit anti-PEDF polyclonal antibody (sc-25594; dilution 1:500; Santa Cruz Biotechnology); mouse anti-CD31 monoclonal antibody (ab24590; dilution 1:1,000; Abcam); rabbit anti-eNOS polyclonal antibody (ab66127; dilution 1:200; Abcam); mouse anti-smoothelin (ab204305; dilution 1:500; Abcam); mouse anti-transforming growth factor-β1 (TGF-β1) monoclonal antibody (ab27969; dilution 1:1,000; Abcam); and mouse anti-GAPDH monoclonal antibody (60004-1-Ig; dilution 1:10,000; Proteintech, Rosemont, IL, USA).

Histological Analysis

The fresh tissues were fixed in cold 2% formaldehyde, followed by overnight immersion in a buffer solution containing 30% sucrose. The tissues were then embedded in optimum cutting temperature compound (Sakura Finetek, Torrance, CA, USA) and stored at −80°C. For immunohistochemical staining, the sections were cut at 6-μm thickness and adhered to charged slides. After rehydration with PBS, the tissue sections were rinsed with hydrogen peroxide and methanol to block endogenous peroxidase. The sections were washed three times in PBS and incubated with 3% goat serum/PBS/0.3% Triton X-100 (Sigma-Aldrich) for 1 h; the tissues were subsequently incubated overnight at 4°C with a mouse anti-nNOS (sc-5302; dilution 1:200; Santa Cruz Biotechnology), mouse anti-nestin (ab6142; dilution 1:500; Abcam), mouse anti-S100 (ab526642; dilution 1:1,000; Abcam), mouse anti-neurofilament (NF)200 (ab6142; dilution 1:500; Abcam), mouse anti-TGF-β1 (ab27969; dilution 1:100; Abcam), or mouse anti-CD68 antibody (ab955; dilution 1:200; Abcam). The slides were washed with PBS and immunostained using the Polink-1 HRP 3,3′-diaminobenzidine (DAB) Detection System (PV-6002; Golden Bridge International, City of Industry, CA, USA). The sections were then stained with DAB followed by hematoxylin (Golden Bridge International, City of Industry) counterstaining.

For immunofluorescence, the sections were incubated with 3% bovine serum albumin (BSA) for 1 h and then with a primary antibody targeting an endothelial marker (CD31; ab24590; dilution 1:1,000; Abcam; or eNOS; ab66127; dilution 1:100; Abcam) or SMC marker (smoothelin; ab204305; dilution 1:3,000; Abcam) overnight at 4°C. The samples were treated by immersion in the secondary antibody conjugated with Alexa 594 (goat anti-rabbit or goat anti-mouse; Invitrogen) at a 1:500 dilution for 1 h. Finally, 4′,6-diamidino-2-phenylindole (DAPI; Invitrogen) staining was perform to visualize the nuclei. For Masson's trichrome staining, the tissues were fixed in 4% paraformaldehyde (PFA; Sigma-Aldrich), followed by ethanol gradient dehydration and fixation in paraffin. The sections were cut at 5-μm thickness and stained according to the Masson's trichrome-staining protocol for connective tissue and smooth muscle. Quantitative image analysis was performed using Image-Pro Plus 6.0 software (Media Cybernetics, Rockville, MD, USA). Three penile tissue sections per rat were chosen for statistical analysis. The sections were photographed and recorded with a digital camera coupled to a Leica microscope (Leica Microsystems). The ratio of the nNOS-positive fibers over the total area of the nerve in pixels was calculated at a 400× magnification, which included all branches of the penile dorsal nerves per section. For the Masson's trichrome staining analysis, three sections per animal were analyzed using the Image-Pro Plus 6.0 software (Media Cybernetics). Magnified images (100×) were analyzed for smooth muscle (stained in red) and collagen (stained in blue) in terms of pixels and expressed as the ratio of smooth muscle to collagen.

TUNEL Assay

The TUNEL assay was performed with the TUNEL detection kit (Roche, Basel, Switzerland) according to the manufacturer's instructions. Briefly, after rinsing three times with PBS, the frozen sections were blotted in 3% hydrogen peroxide at RT for 10 min. The sections were immersed in 1% Triton X-100 (Sigma-Aldrich) on ice for 5 min and washed twice with PBS. The sections were then incubated with the TUNEL reaction mixture in the dark for 60 min at 37°C. The cell nuclei were stained with DAPI. The sections were analyzed by fluorescence microscopy, and the percentage of TUNEL-positive cells (shown with green fluorescence) to the total cells (shown with blue fluorescence) was presented as the apoptotic index (AI).

Statistical Analysis

Comparisons between groups were analyzed using the GraphPad Prism v.6.0 software (GraphPad Software, La Jolla, CA, USA). Continuous variables were expressed as the mean ± standard deviation. Multiple groups were compared using one-way analysis of variance (ANOVA) followed by the Student–Newman–Keuls post hoc test. Values of p < 0.05 were considered statistically significant.

Results

Characterization of Cultured USCs

The USC colonies appeared 1 week after plating. These cells exhibited the typical rice-shaped appearance (Fig. 1A). Flow cytometry analysis demonstrated that the cells used in this study were strongly positive for the mesenchymal stem cell (MSC) markers CD24, CD44, CD73, CD90, CD133, and CD146 and negative for the hematopoietic stem cell (HSC) markers CD31, CD34, and CD45, which was in accordance with our previous studies (Fig. 1B). The adipogenic- or osteogenic-induced differentiation of USCs (stained with Alizarin red solution or Oil red O, respectively) confirmed the multipotent differentiation ability of the USCs (Fig. 1A).

Figure 1.

Characterization of human urine-derived stem cells (USCs) in vitro. (A) The typical “rice-shaped” appearance of USCs (p1 and p6), and osteogenic- and adipogenic-induced USCs with Alizarin red S staining or Oil red O staining, respectively. Scale bar: 100 μm. (B) Flow cytometric analysis showed that the USCs (p3) were positive for mesenchymal stem cell (MSC) surface markers CD24, CD44, CD73, CD133, and CD146, but were negative for hematopoietic stem cell markers CD31, CD34, and CD45. (C) Representative images of USCs transfected with GFP or GFP/PEDF and screened by puromycin and cultured for 30 days in vitro. Scale bars: 200 μm. (D) Western blot analysis confirmed the significantly increased expression of PEDF by USCs^GFP/PEDF+ in vitro compared to USCs or USCs^GFP+. (E) The ELISA assay showed that PEDF secretion by USCs^GFP/PEDF+ into the culture medium peaked on day 14 postinfection and remained stable on day 30.

PEDF Secretion by USCs In Vitro

After transduction of the PEDF/GFP lentivirus or GFP lentivirus into USCs and screening by puromycin (Fig. 1C), the transduction efficiency (GFP-positive cells to total cells) was approximately 95%. Western blot confirmed PEDF expression in the USCs, USCs^GFP+, and USCs^GFP/PEDF+ in vitro (Fig. 1D). The ELISA assay showed that the PEDF protein levels in the culture medium of the USCs^GFP/PEDF+ reached the peak at 14 days (at passage 7) and remained stable for more than 30 days (at passage 12) (Fig. 1E), and the transfection rate remained 95% 30 days after the transduction.

Expression of PEDF in the Cavernous Tissue

Western blot analysis demonstrated that the expression level of PEDF in the BCNI group was increased on day 14 and decreased to basal level on day 28, just the same as in our previous study²⁴. Moreover, PEDF was significantly increased in the USC, USC^GFP+, and USC^GFP/PEDF+ groups compared to the BCNI group on day 28. The USC^GFP/PEDF+ group manifested the highest expression level out of the five groups (Fig. 2A and B). Furthermore, greater expression levels of PEDF in the USC, USC^GFP+, and USC^GFP/PEDF+ groups were shown on day 28 day than on day 14.

Figure 2.

The expression of PEDF in the cavernous tissue in each experimental group. (A) Western blot and (B) quantitative analysis identified increased expression of PEDF within the cavernous tissue in the USC, USC^GFP+, and USC^GFP/PEDF+ groups compared with the sham group 28 days after cell injection. Expression was higher in the USC^GFP/PEDF+-treated group than with USC or USC^GFP+ treatment, but no difference was observed between USC and USC^GFP+ treatment. Greater expression levels of PEDF in the USC, USC^GFP+, and USC^GFP/PEDF+ groups were shown on day 28 day than on day 14. *p < 0.05. n = 8 per group. GAPDH, glyceraldehyde 3-phosphate dehydrogenase.

Erectile Function Measurement

To evaluate the erectile function, ICP tracing responses to electronic stimulation of the CN were assessed in all groups 4 weeks after USC intracavernous injection (Fig. 3A). The total ICP and the ICP/MAP ratio of the BCNI group were 32 ± 6 mmHg and 26 ± 5%, respectively, which were significantly lower than the values in the sham group (p < 0.01). The USC and USC^GFP+ groups showed greatly increased ICP and ICP/MAP ratios compared to the BCNI group, but were still significantly lower than the sham group (p < 0.05). The USC^GFP/PEDF+ group displayed a greatly increased protective effect of erectile function. The total ICP and the ICP/MAP ratio of the USC^GFP/PEDF+ group were 93 ± 10 mmHg and 75 ± 13%, respectively, which were higher than the values in the USC and USC^GFP+ groups and represented 84% and 87% of the values in the sham group (Fig. 3B and C).

Figure 3.

USCs and USCs^GFP/PEDF+ improved erectile function in bilateral cavernous nerve injury rats. (A) Representative ICP tracing responses to the electronic stimulation of the cavernous nerve (5 V, 20 Hz, and 60-s duration) in the sham rats and BCNI rats 4 weeks after intracavernous injection of PBS, USCs, USCs^GFP+, or USCs^GFP/PEDF+ (n = 8 rats per group). (B) The effects of USCs, USCs^GFP+, or USCs^GFP/PEDF+ on the increase of ICP in the BCNI rats. (C) The ratio of total ICP to MAP was calculated for all five groups. *p < 0.05, **p < 0.01. ICP, intracavernous pressure; MAP, mean arterial pressure.

Neural Markers Expression in the Penile Dorsal Nerves

Intracavernously injected USCs had a neuroprotective effect on the penile dorsal nerves. The nNOS-, NF200-, nestin-, and S100-positive fiber ratios were significantly increased in the USC, USC^GFP+, and USC^GFP/PEDF+ groups compared with the BCNI group (p < 0.05), although the number was still lower than the sham group. Moreover, the nNOS-, nestin-, and NF200-positive fiber ratio in the USC^GFP/PEDF+ was higher than in the USC and USC^GFP+ groups (p < 0.05), while there was no significant difference of the S100-positive fiber ratios among these three groups (p > 0.05) (Fig. 4A and B).

Figure 4.

USCs, USCs^GFP+, and USCs^GFP/PEDF+ prevented the destruction of penile dorsal nerves. (A) Representative IHC images are presented for the nNOS-, NF200-, nestin-, and S100-positive fibers in the penile dorsal nerves in each group. n = 6 per group. Scale bars: 20 μm. (B) Quantitative analysis of nNOS-, NF200-, nestin-, and S100-positive fibers to the total area of penile dorsal nerves in pixels recorded by the Image-Pro Plus 6.0 software. *p < 0.05. nNOS, neuronal nitric oxide synthase; NF200, neurofilament heavy polypeptide; IHC, immunohistochemistry.

Endothelial Marker Expression in the Cavernous Tissue

The endothelial marker (CD31 and eNOS) expression levels were determined by immunofluorescence staining and Western blot analysis (Fig. 5A and D).

Figure 5.

USCs and USCs^GFP/PEDF+ increased the expression of endothelial cell markers (CD31 and eNOS) in the cavernous tissue. (A) Immunofluorescence staining and (B, C) semiquantitative analysis revealed that the expression levels of CD31 and eNOS were increased in the USC-, USC^GFP+-, and USC^GFP/PEDF+-treated groups 28 days after cell transplantation. Scale bars: 200 μm. (D) Western blot analysis and (E, F) quantitative analysis for Western blot confirmed the significantly higher expression of CD31 and eNOS in the cavernous tissue. n = 6 per group. *p < 0.05. eNOS, endothelial nitric oxide synthase.

The result showed a large decrease in CD31 and eNOS expression in the BCNI rats compared with the sham rats (p < 0.01). Intracavernous injection of USCs, USCs^GFP+, and USCs^GFP/PEDF+ partially restored the endothelial content in the penile tissues (p < 0.05). However, there were no significant differences among these three SC-treated groups (Fig. 5B, C, E, and F).

SMC Marker Expression in the Cavernous Tissue

The mean SMC/collagen ratios significantly decreased from 14.9 ± 1.2% in the sham rats to 5.1 ± 0.8% in the BCNI rats (p < 0.05). Intracavernous injection of USCs or USCs^GFP+ improved the smooth muscle content, with the ratios reaching 8.2 ± 1.6% and 8.4 ± 2.0%, respectively. Furthermore, the ratio in the USC^GFP/PEDF+-treated rats was 12.9 ± 2.2%, which was significantly higher than that in the USC and USC^GFP+ groups (p < 0.05) and just slightly lower than in the sham group (Fig. 6A and B). This finding indicated the significant protective effect of USCs^GFP/PEDF+. Immunofluorescence staining and Western blot analysis indicated that the expression of the SMC marker smoothelin was significantly decreased within the cavernous tissue in BCNI rats compared to the sham rats (p < 0.05). In contrast, the USC, USC^GFP+, and USC^GFP/PEDF+ groups exhibited a significant increase in smoothelin expression compared to the BCNI group. Immunohistochemistry staining and Western blot analysis showed a much higher TGF-β1 expression level within the cavernous tissue in the BCNI group compared to the sham group (Fig. 6F–H). USCs, USCs^GFP+, and USCs^GFP/PEDF+ resulted in a significantly lower TGF-β1 expression level compared to the BCNI group (p < 0.05). The TGF-β1 expression level in the USC^GFP/PEDF+ group was slightly higher than that in the sham group.

Figure 6.

USCs and USCs^GFP/PEDF+ increased the expression of the smooth muscle cell marker smoothelin and the smooth muscle-to-collagen ratio by decreasing TGF-β1 expression and cell apoptosis in the cavernous tissue. (A) Representative images of the penile tissue stained with Masson's trichrome in all groups. Smooth muscle was stained red, whereas collagen and connective tissue were stained blue. n = 6 per group. Scale bar: 100 μm. (B) The smooth muscle-to-collagen ratio in the cavernous tissue was calculated in each experimental group. (C) Immunofluorescence staining and (D, E) Western blot indicated the increased expression of smoothelin (with arrow) in the USC-, USC^GFP+-, and USC^GFP/PEDF+-treated groups 28 days after cell transplantation compared to the BCNI group. Scale bar: 200 μm. (F) IHC and (G, H) Western blot analysis demonstrated the decreased expression of TGF-β1 in the cavernous tissue in the USC- and USC^GFP/PEDF+-treated rats compared to the BCNI rats. Scale bar: 100 μm. (I) Representative images of the apoptotic cells in the cavernous tissue via the TUNEL assay. The apoptotic cells are indicated with green fluorescence. Scale bar: 100 μm. (J) The apoptosis index (AI; apoptosis-positive cells to total cells) was calculated in each group. USCs and USCs^GFP/PEDF+ significantly decreased the AI in the corpora cavernosa. n = 6 per group. *p < 0.05, **p < 0.01. TGF-β1, transforming growth factor-β1; TUNEL, transferase-mediated deoxyuridine triphosphate.

Cell Apoptosis in the Cavernous Tissues

TUNEL assay analysis demonstrated that the mean AI in the BCNI group was 9.2 ± 0.8%, which was significantly higher than the mean AI in the sham group (Fig. 6G and H). In the USC and USC^GFP+ groups, the mean AIs were decreased to 3.0 ± 0.5% and 2.8 ± 0.7%, respectively. PEDF demonstrated a large effect on the prevention of cell apoptosis; the mean AI in the USC^GFP/PEDF+ group was 1.8 ± 0.3%, which was not significantly different from the sham group (p > 0.05).

Macrophage Marker Expression in the Cavernous Tissue

No macrophages infiltrated the cavernous tissue 14 or 28 days after cell transplantation, as no CD68-positive cells could be observed in the cavernous tissue of rats injected with PBS, USCs, USCs^GFP+, or USCs^GFP/PEDF+ via immunohistochemistry (IHC) (Fig 7).

Figure 7.

Cavernous tissue stained with macrophage marker (CD68). No CD68-positive cells could be observed in the cavernous tissue in the USC-, USC^GFP+- and USC^GFP/PEDF+-treated groups 14 and 28 days after cell transplantation via IHC. Scale bars: 50 μm.

Discussion

Structurally, the key constituents of erections are the endothelium, smooth muscle, and CN (specifically the nNOS-positive nerves). Functionally, precise interactions among these three components are critical. The pathophysiology of CNIED is closely associated with CN denervation and then subsequently with endothelial dysfunction, apoptosis of corporal SMCs, and accumulation of collagen I and III²⁸. In the present study, we demonstrated that intracavernous injection of USCs and USCs^GFP/PEDF+ significantly protected the number of nNOS-positive penile dorsal nerves, increased the expression of endothelial cell markers (CD31 and eNOS) and the smooth muscle content, and improved the ICP, thereby enhancing erectile function in the CNIED rodent model.

PDE5 inhibitors, which can protect endothelial functions, mobilize endothelial progenitor cells, and even exhibit neuromodulatory effects to some extent^29,30, are the first-line treatment for patients suffering from ED following RP. A previous study strongly supported the use of PDE5 inhibitors for rehabilitation, but numerous clinical studies reported controversial results³. To date, several studies have utilized different SCs to treat CNIED (e.g., BM-MSCs, ADSCs, and MDSCs) and have shown a significant restoration in erectile function and the cavernous structure. However, these SCs must be obtained by invasive methods, thus limiting their use in clinical practice. In the present study, we utilized a new type of adult SCs (USCs) that can be noninvasively obtained from human voided urine and possess a high proliferation rate and multipotent differentiation capacity^17,18,20. The results indicated therapeutic effects of USCs for the protection of erectile function and structure in CNIED rats.

Calenda et al. found that the PEDF gene was immediately upregulated (threefold compared to the normal control) 48 h after CNI but recovered to the basal level at 14 days in MPG tissue based on whole-genome microarray analysis³¹. Our previous study has verified the increased expression of the PEDF protein in the cavernous tissue until 14 days after CNI and recovery to the normal level by day 28²⁴. PEDF has been shown to possess neurotrophic and neuroprotective properties via the activation of nuclear factor κB (NF-κB), which in turn induces the expression of several neurotrophic factors [glial cell line-derived neurotrophic factor (GDNF), BDNF, and NGF] and antiapoptotic genes (Bcl-x and Bcl2) in cerebellar granule cells³². The temporary and insufficient expression of PEDF in the cavernous tissue may be involved in the CNIED rat model. In our previous study, we discovered that USCs could secret detectable PEDF in vitro. In this study, USCs and USCs^GFP/PEDF+ displayed a large neuroprotective effect because the numbers of neural marker (nNOS and NF200)-positive fibers within the penile dorsal nerve in the USC- and USC^GFP/PEDF+-treated rats were significantly increased when compared to the number of fibers in the BCNI group. Furthermore, the USCs^GFP/PEDF+ displayed a greater therapeutic effect than the USCs, which should be attributed to the overexpression of PEDF in vivo. Recently, several studies have shown the important role of PEDF in maintaining the pluripotency and self-renewal of adult SCs. Liang et al. reported that the downregulation of PEDF inhibited the proliferation of human cardiac SCs and induced cell differentiation via the Notch signaling pathway³³. In the present study, we found that the neural stem cell (NSC) marker (nestin)-positive fibers were significantly increased after USC^GFP/PEDF+ treatment, which demonstrated the NSC-protective effect of PEDF. Although the Schwann cell marker (S100)-positive fibers all increased in the cell-treated groups compared to the BCNI group, no significant difference was found among them, which indicated that the protective effect of Schwann cells was not attributable to PEDF.

Animal experiments have demonstrated that chronic hypoxia in the cavernous tissue followed by neuropraxia of the CN can lead to increased apoptosis of SMCs and exaggerated deposition of collagen^3,34–36. A reduced SMC-to-collagen ratio is another important etiology of CNIED. In our present study, the SMC-to-collagen ratio in the cavernous tissue was reduced by 14.9% in the sham group and to approximately 5.1% in the BCNI group. After intracavernous injection of USCs or USCs^GFP+, the ratio increased to approximately 8.2% or 8.4%, respectively, and with USC^GFP/PEDF+ administration it increased to 12.9%. Research has indicated that PEDF deficiency increases the risk of fibrosis in the pancreas because PEDF-deficient pancreases show significantly enhanced TGF-β1 expression³⁷. Moreland demonstrated that hypoxia could induce TGF-β1 expression, which in turn increased the synthesis of collagen in the cavernous tissue³⁸. Leungwattanakij et al. also reported that the TGF-β1 mRNA and protein levels were significantly increased in the cavernous tissues of CNIED rats³. Herein, we found that the TGF-β1 protein expression level was greatly decreased after USC^GFP/PEDF+ injection. This result indicated another mechanism underlying the effect of PEDF on the restoration of the cavernous structure in CNIED rats.

The innervation of the penile tissue by CNI increases the apoptosis of SMCs and thus plays an important role in the pathogenesis of ED. Mulhall et al. demonstrated that the mean AI within the cavernous tissues of CNIED rats was 63 ± 4%, compared with the AI of 10 ± 7% in the sham group³⁹. Furthermore, administration of sildenafil citrate significantly decreased the AI within the cavernous tissue after CNI. Wang et al. reported that PEDF and its derived 44-mer peptide protected cultured primary cardiomyocytes and H9c2 cells from hypoxia-induced apoptosis and necrosis by an antioxidative mechanism⁴⁰. Our present study showed that USCs or USCs^GFP/PEDF+ significantly reduced apoptosis in the cavernous tissue. Furthermore, the AI of the USC^GFP/PEDF+-treated group decreased to 1.8%, which was nearly the same level as the sham group.

In adult tissues, PEDF inhibits pathological neovascularization and tumor angiogenesis^41,42. Interestingly, PEDF appeared to have an opposite effect on endothelial cells of different phenotypes. Although PEDF promoted the apoptosis of endothelial cells, it exerted a synergistic proliferative effect on endothelial cells with VEGF⁴³. In our present study, both USCs and USCs^GFP/PEDF+ increased the expression levels of endothelial cell markers in the cavernous tissues of the CNIED rats; indeed, no differences were detected between them. One possible explanation is the paracrine response of proangiogenic growth factors by USCs, such as VEGF, bFGF, and platelet-derived growth factor (PDGF). Our previous studies have shown that intracavernous injection of USCs restored the endothelial function in a diabetic-induced ED rat model via the paracrine effect of USCs.

USCs have multidifferentiation potential in vitro and can differentiate into endothelial cell and myogenic lineages in vivo in the presence of specific growth factors^20,44. In our previous study, intracavernous injection of USCs genetically modified with FGF2 significantly improved erectile function and restored the cavernous structure in diabetic-induced ED rats14. However, no labeled USCs could be tracked in the penile tissue 28 days after cell transplantation, which was similar to most studies that utilized other MSCs to treat ED rat models. In this study, no GFP-positive USCs were observed in the cavernous tissue 28 days after cell injection in the USC^GFP+- and USC^GFP/PEDF+-treated groups. The USCs improved the erectile function and protected the cavernous structure by paracrine effects rather than direct differentiation into endothelial cells, SMCs, or nerve cells.

Inevitably some limitations existed in our study. First, no GFP-positive USCs could be tracked in the cavernous tissue 28 days after cell transplantation. Thus, the fate of the injected USCs should be monitored by other techniques, such as in vivo fluorescence imaging. Second, we demonstrated that USCs^GFP/PEDF+ could protect the erectile function and cavernous structure of the CNIED rat model, but the exact pathway involved needs further study. Third, the paracrine effect of USCs is complicated, and PEDF is only one of many paracrine growth factors. Other growth factors may also participate in the therapeutic effect and thus should be taken into consideration.

Conclusion

In summary, we demonstrated that intracavernous injection of USCs genetically modified with PEDF can protect erectile function in the CNIED rat model by preventing destruction of the nerves, improving endothelial cell function, increasing the SMC-to-collagen ratio, and decreasing fibrosis and cell apoptosis in the cavernous tissues.

Footnotes

Acknowledgments

The study was supported by the following grants: the National Natural Science Foundation of China (Nos. 81471449, 81270696, 81302223, and 81401197) and the Natural Science Foundation of Guangdong Province (Nos. 2014A030313072, 201300000094, and 2015A030310027). The authors would like to thank Professor Xenong Zou and his laboratory (Guangdong Provincial Key Laboratory of Orthopedics and Traumatology) for providing technical support and guidance to the study. The authors declare no conflicts of interest.

References

Hatzimouratidis

, Burnett

A.L.

, Hatzichristou

, McCullough

A.R.

, Montorsi

, Mulhall

J.P.

Phosphodiesterase type 5 inhibitors in postprostatectomy erectile dysfunction: A critical analysis of the basic science rationale and clinical application. Eur Urol. 2009; 55(2): 334–47.

Fode

, Ohl

D.A.

, Ralph

, Sonksen

Penile rehabilitation after radical prostatectomy: What the evidence really says. BJU Int. 2013; 112(7): 998–1008.

Leungwattanakij

, Bivalacqua

T.J.

, Usta

M.F.

, Yang

D.Y.

, Hyun

J.S.

, Champion

H.C.

, Abdel-Mageed

A.B.

, Hellstrom

W.J.

Cavernous neurotomy causes hypoxia and fibrosis in rat corpus cavernosum. J Androl. 2003; 24(2): 239–45.

Ficarra

, Novara

, Ahlering

T.E.

, Costello

, Eastham

J.A.

, Graefen

, Guazzoni

, Menon

, Mottrie

, Patel

V.R.

, Van der Poel

, Rosen

R.C.

, Tewari

A.K.

, Wilson

T.G.

, Zattoni

, Montorsi

Systematic review and meta-analysis of studies reporting potency rates after robot-assisted radical prostatectomy. Eur Urol. 2012; 62(3): 418–30.

Rassweiler

, Hruza

, Teber

, Su

L.M.

Laparoscopic and robotic assisted radical prostatectomy—Critical analysis of the results. Eur Urol. 2006; 49(4): 612–24.

Hatzimouratidis

, Amar

, Eardley

, Giuliano

, Hatzichristou

, Montorsi

, Vardi

, Wespes

Guidelines on male sexual dysfunction: Erectile dysfunction and premature ejaculation. Eur Urol. 2010; 57(5): 804–14.

Dorsey

, Keel

, Klavens

, Hellstrom

W.J.

Phosphodiesterase type 5 (PDE5) inhibitors for the treatment of erectile dysfunction. Expert Opin Pharmacother. 2010; 11(7): 1109–22.

Albersen

, Lin

C.S.

, Lue

Stem-cell therapy for erectile dysfunction. Arab J Urol. 2013; 11(3): 237–44.

Bochinski

, Lin

G.T.

, Nunes

, Carrion

, Rahman

, Lin

C.S.

, Lue

T.F.

The effect of neural embryonic stem cell therapy in a rat model of cavernosal nerve injury. BJU Int. 2004; 94(6): 904–9.

10.

Kendirci

, Trost

, Bakondi

, Whitney

M.J.

, Hellstrom

W.J.

, Spees

J.L.

Transplantation of nonhematopoietic adult bone marrow stem/progenitor cells isolated by p75 nerve growth factor receptor into the penis rescues erectile function in a rat model of cavernous nerve injury. J Urol. 2010; 184(4): 1560–6.

11.

Liu

, Sun

, Dai

, Zheng

, Wang

, Huang

, Bian

, Deng

Chronic administration of sildenafil modified the impaired VEGF system and improved the erectile function in rats with diabetic erectile dysfunction. J Sex Med. 2010; 7(12): 3868–78.

12.

Woo

J.C.

, Bae

W.J.

, Kim

S.J.

, Kim

S.D.

, Sohn

D.W.

, Hong

S.H.

, Lee

J.Y.

, Hwang

T.K.

, Sung

Y.C.

, Kim

S.W.

Transplantation of muscle-derived stem cells into the corpus cavernosum restores erectile function in a rat model of cavernous nerve injury. Korean J Urol. 2011; 52(5): 359–63.

13.

Liu

, Sun

, Bian

, Wu

, Guan

, Ouyang

, Huang

, Xiao

, Luo

, Atala

, Zhang

, Deng

Correction of diabetic erectile dysfunction with adipose derived stem cells modified with the vascular endothelial growth factor gene in a rodent diabetic model. PLoS One 2013; 8(8): e72790.

14.

Ouyang

, Sun

, Han

, Chen

, Yao

, Gao

, Bian

, Huang

, Zhang

, Wan

, Yang

, Xiao

, Songyang

, Liu

, Zhang

, Deng

Human urine-derived stem cells alone or genetically-modified with FGF2 Improve type 2 diabetic erectile dysfunction in a rat model. PLoS One 2014; 9(3): e92825.

15.

Lin

C.S.

, Xin

Z.C.

, Wang

, Deng

, Huang

Y.C.

, Lin

, Lue

T.F.

Stem cell therapy for erectile dysfunction: A critical review. Stem Cells Dev. 2012; 21(3): 343–51.

16.

Bharadwaj

, Liu

, Shi

, Markert

, Andersson

K.E.

, Atala

, Zhang

Characterization of urine-derived stem cells obtained from upper urinary tract for use in cell-based urological tissue engineering. Tissue Eng Part A 2011; 17(15-16): 2123–32.

17.

Lang

, Liu

, Shi

, Bharadwaj

, Leng

, Zhou

, Liu

, Atala

, Zhang

Self-renewal and differentiation capacity of urine-derived stem cells after urine preservation for 24 hours. PLoS One 2013; 8(1): e53980.

18.

Zhang

, McNeill

, Tian

, Soker

, Andersson

K.E.

, Yoo

J.J.

, Atala

Urine derived cells are a potential source for urological tissue reconstruction. J Urol. 2008; 180(5): 2226–33.

19.

Bharadwaj

, Liu

, Shi

, Wu

, Yang

, He

, Fan

, Lu

, Zhou

, Liu

, Atala

, Rohozinski

, Zhang

Multipotential differentiation of human urine-derived stem cells: Potential for therapeutic applications in urology. Stem Cells 2013; 31(9): 1840–56.

20.

Liu

, Wang

, Sun

, Deng

, Atala

, Zhang

The effect of urine-derived stem cells expressing VEGF loaded in collagen hydrogels on myogenesis and innervation following after subcutaneous implantation in nude mice. Biomaterials 2013; 34(34): 8617–29.

21.

Filleur

, Nelius

, de Riese

, Kennedy

R.C.

Characterization of PEDF: A multi-functional serpin family protein. J Cell Biochem. 2009; 106(5): 769–75.

22.

Tombran-Tink

, Johnson

L.V.

Neuronal differentiation of retinoblastoma cells induced by medium conditioned by human RPE cells. Invest Ophthalmol Vis Sci. 1989; 30(8): 1700–7.

23.

Minkevich

N.I.

, Lipkin

V.M.

, Kostanyan

I.A.

PEDF—A noninhibitory serpin with neurotrophic activity. Acta Naturae 2010; 2(3): 62–71.

24.

Chen

, Yang

, Zheng

, Bian

, Sun

, Shi

, Liang

, Gao

, Liu

, Deng

Neurotrophic effect of adipose tissue-derived stem cells on erectile function recovery by pigment epithelium-derived factor secretion in a rat model of cavernous nerve injury. Stem Cells Int. 2016; 2016: 5161248.

25.

Sun

X.Z.

, Liu

G.H.

, Wang

Z.Q.

, Zheng

F.F.

, Bian

, Huang

Y.P.

, Gao

, Zhang

Y.D.

, Deng

C.H.

Over-expression of VEGF165 in the adipose tissue-derived stem cells via the lentiviral vector. Chin Med J. (Engl) 2011; 124(19): 3093–7.

26.

Bian

, Dai

Y.P.

, Sun

X.Z.

, Liu

G.H.

, Deng

C.H.

, Ye

Y.L.

[Establishment of a rat model of erectile dysfunction after radical prostatectomy by crush injury or resection of the cavernous nerve]. Nan Fang Yi Ke Da Xue Xue Bao 2011; 31(2): 230–3.

27.

Hannan

J.L.

, Kutlu

, Stopak

B.L.

, Liu

, Castiglione

, Hedlund

, Burnett

A.L.

, Bivalacqua

T.J.

Valproic acid prevents penile fibrosis and erectile dysfunction in cavernous nerve-injured rats. J Sex Med. 2014; 11(6): 1442–51.

28.

Mulhall

J.P.

, Bella

A.J.

, Briganti

, McCullough

, Brock

Erectile function rehabilitation in the radical prostatectomy patient. J Sex Med. 2010; 7(4 Pt 2): 1687–98.

29.

Zhang

, Zhang

R.L.

, Cui

, LaPointe

M.C.

, Silver

, Chopp

Tadalafil, a long-acting type 5 phosphodiesterase isoenzyme inhibitor, improves neurological functional recovery in a rat model of embolic stroke. Brain Res. 2006; 1118(1): 192–8.

30.

Zhang

, Wang

, Zhang

, Tsang

, Lu

, Chopp

Sildenafil (Viagra) induces neurogenesis and promotes functional recovery after stroke in rats. Stroke 2002; 33(11): 2675–80.

31.

Calenda

, Strong

T.D.

, Pavlovich

C.P.

, Schaeffer

E.M.

, Burnett

A.L.

, Yu

, Davies

K.P.

, Bivalacqua

T.J.

Whole genome microarray of the major pelvic ganglion after cavernous nerve injury: New insights into molecular profile changes after nerve injury. BJU Int. 2012; 109(10): 1552–64.

32.

Yabe

, Wilson

, Schwartz

J.P.

NFkappaB activation is required for the neuroprotective effects of pigment epithelium- derived factor (PEDF) on cerebellar granule neurons. J Biol Chem. 2001; 276(46): 43313–9.

33.

Liang

, Duan

, Hou

, Yi

, Zhang

, Jin

, Zhai

, Yu

, Liu

, Yi

The roles of nanocarriers on pigment epithelium-derived factor in the differentiation of human cardiac stem cells. Cell Tissue Res. 2015; 362(3): 611–21.

34.

Iacono

, Giannella

, Somma

, Manno

, Fusco

, Mirone

Histological alterations in cavernous tissue after radical prostatectomy. J Urol. 2005; 173(5): 1673–6.

35.

Jin

H.R.

, Chung

Y.G.

, Kim

W.J.

, Zhang

L.W.

, Piao

, Tuvshintur

, Yin

G.N.

, Shin

S.H.

, Tumurbaatar

, Han

J.Y.

, Ryu

J.K.

, Suh

J.K.

A mouse model of cavernous nerve injury-induced erectile dysfunction: Functional and morphological characterization of the corpus cavernosum. J Sex Med. 2010; 7(10): 3351–64.

36.

User

H.M.

, Hairston

J.H.

, Zelner

D.J.

, McKenna

K.E.

, McVary

K.T.

Penile weight and cell subtype specific changes in a post-radical prostatectomy model of erectile dysfunction. J Urol. 2003; 169(3): 1175–9.

37.

Schmitz

J.C.

, Protiva

, Gattu

A.K.

, Utsumi

, Iwakiri

, Neto

A.G.

, Quinn

, Cornwell

M.L.

, Fitchev

, Lugea

, Crawford

S.E.

, Chung

Pigment epithelium-derived factor regulates early pancreatic fibrotic responses and suppresses the profibrotic cytokine thrombospondin-1. Am J Pathol. 2011; 179(6): 2990–9.

38.

Moreland

R.B.

Is there a role of hypoxemia in penile fibrosis: A viewpoint presented to the Society for the Study of Impotence. Int J Impot Res. 1998; 10(2): 113–20.

39.

Mulhall

J.P.

, Muller

, Donohue

J.F.

, Mullerad

, Kobylarz

, Paduch

D.A.

, Tal

, Li

P.S.

, Cohen-Gould

, Scardino

P.T.

The functional and structural consequences of cavernous nerve injury are ameliorated by sildenafil citrate. J Sex Med. 2008; 5(5): 1126–36.

40.

Wang

, Zhang

, Lu

, Zhang

, Li

, Dong

, Zhang

PEDF attenuates hypoxia-induced apoptosis and necrosis in H9c2 cells by inhibiting p53 mitochondrial translocation via PEDF-R. Biochem Biophys Res Commun. 2015; 465(3): 394–401.

41.

Barnstable

C.J.

, Tombran-Tink

Neuroprotective and antiangiogenic actions of PEDF in the eye: Molecular targets and therapeutic potential. Prog Retin Eye Res. 2004; 23(5): 561–77.

42.

Tombran-Tink

, Barnstable

C.J.

PEDF: A multifaceted neurotrophic factor. Nat Rev Neurosci. 2003; 4(8): 628–36.

43.

Hutchings

, Maitre-Boube

, Tombran-Tink

, Plouet

Pigment epithelium-derived factor exerts opposite effects on endothelial cells of different phenotypes. Biochem Biophys Res Commun. 2002; 294(4): 764–9.

44.

Liu

, Pareta

R.A.

, Wu

, Shi

, Zhou

, Liu

, Deng

, Sun

, Atala

, Opara

E.C.

, Zhang

Skeletal myogenic differentiation of urine-derived stem cells and angiogenesis using microbeads loaded with growth factors. Biomaterials 2013; 34(4): 1311–26.