High-dose SFRP2 attenuates fibrosis and promotes angiogenesis via Wnt signaling modulation in diabetic erectile dysfunction

2.1 Ethics statement and study design

All tunica albuginea (TA) tissues used for the isolation of human corpus cavernosum fibroblasts, as well as all experimental animal procedures, were approved by the Institutional Review Board of Inha University (IRB No. 2007-730). Written informed consent was obtained from all human participants prior to tissue collection in accordance with the Declaration of Helsinki. Primary corpus cavernosum fibroblasts were isolated from tunica albuginea tissue obtained from one male donor undergoing penoplasty for congenital penile curvature (age 21). Non-fibrotic regions of the tissue, carefully dissected during surgery, were extracted for subsequent culture.This study aimed to evaluate the role of SFRP2 in mitigating diabetic fibrosis. First, we examined SFRP2 expression in fibroblasts derived from human corpus cavernosum tissue under normal glucose (NG) and high glucose (HG) conditions. Second, under fibrotic conditions induced by BMP1 and HG, we treated the cells with human recombinant SFRP2 and evaluated its antifibrotic effects by assessing changes in ECM molecules and Wnt signaling activity. Finally, we used HUVECs to investigate the effects of SFRP2 on angiogenesis and cell survival. HUVECs were utilized as a surrogate endothelial model due to the limited availability of primary human penile endothelial cells. HUVECs have been shown to recapitulate key physiological features of penile endothelium, including angiogenic behavior, NO-related signaling, and glucose-induced stress responses, supporting their validity in erectile dysfunction–associated vascular research [13]. A schematic overview illustrating the treatment sequence (NG, HG, BMP1, SFRP2) and corresponding analytical endpoints (fibrosis, angiogenesis, apoptosis) has been included to enhance clarity (Supplementary Figure 1).

Informed consent: Informed consent has been obtained from all individuals included in this study.

Ethical approval: The research related to human use has been complied with all the relevant national regulations, institutional policies and in accordance with the tenets of the Helsinki Declaration, and has been approved by the Institutional Review Board of Inha University (IRB No. 2007-730).

2.2 Primary culture and characterization of human corpus cavernosum (CC) fibroblasts

Primary cultures of human corpus cavernosum (CC) fibroblasts were established from TA tissues as previously described [14], 15]. Briefly, tissues were collected in sterile Hank’s balanced salt solution (Gibco, Carlsbad, CA, USA), washed three times with phosphate-buffered saline (PBS), and sectioned into 1–2 mm fragments. Samples were digested in 12.5 mL Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco) containing 0.06 % collagenase A (Sigma-Aldrich, St. Louis, MO, USA) for 1 h at 37 °C in a humidified incubator with 5 % CO₂. The resulting cell suspension was centrifuged at 400×g for 5 min, washed, and plated onto 100-mm culture dishes (Falcon, Becton Dickinson Labware, Franklin Lakes, NJ, USA) in DMEM supplemented with 10 % fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin. The culture medium was replaced every 2 days, and cells at passages 5 to 8 were used for subsequent experiments.

To confirm cell identity, cultured cells were grown on sterile glass coverslips in 12-well plates until nearly confluent and stained with antibodies against CD90 and vimentin (fibroblast markers; R&D Systems Inc., Minneapolis, MN, USA; 1:100), CD31 (endothelial cell marker; Chemicon, Temecula, CA, USA; 1:50), neural/glial antigen 2 (NG2, pericyte marker; Millipore, San Francisco, CA, USA; 1:50), and 4′,6-diamidino-2-phenylindole (DAPI, nuclear marker; Vector Laboratories Inc., Burlingame, CA, USA), as previously described [16]. Fluorescence images were acquired using a K1-Fluo confocal microscope (Nanoscope Systems, Daejeon, Korea).

2.3 Cell culture and treatment

HUVECs (Lonza, Cohasset, MN, USA) were cultured according to American Type Culture Collection guidelines, then used between passages 2 and 7. Cells were maintained in M199 medium (Gibco) supplemented with 20 % FBS, 0.5 mg/mL heparin (Sigma-Aldrich), 5 ng/mL recombinant human vascular endothelial growth factor (R&D Systems), and 1 % penicillin/streptomycin on 0.2 % gelatin-coated dishes.

To mimic diabetic conditions and induce fibrosis, serum-starved HUVECs were exposed to either NG (5 mM) or HG (30 mM; Sigma-Aldrich) for 72 h, with or without SFRP2 (20 nM or 20 µM). The SFRP2 concentrations used in this study (20 nM and 20 μM) were selected based on prior reports describing the biphasic activity of SFRP2 on collagen regulation and Wnt signaling, where low doses promote collagen maturation while high doses inhibit canonical Wnt signaling and attenuate fibrosis [9], [10], [11]. Preliminary dose–response experiments confirmed these concentrations as effective ranges in human corpus cavernosum fibroblasts. In additional experiments, fibrosis was induced by pre-treating cells with BMP1 (15, 50, or 100 ng/mL) for 24 h, followed by SFRP2 treatment for 48 h [17]. Vehicle-treated cells (PBS) were used as negative controls for all assays, while BMP1 (50 ng/mL) treatment served as a positive fibrotic control to validate collagen responses. For endothelial assays, NG-treated HUVECs served as baseline physiological controls, and HG treatment represented diabetic-like stress conditions.

2.4 Tube formation assay

Endothelial tube formation was evaluated as previously described [18]. Growth factor-reduced Matrigel (50 µL; Becton Dickinson) was added to 96-well plates and allowed to solidify at 37 °C for 10 min. HUVECs pre-treated under NG or HG conditions, with or without SFRP2, were seeded at 5 × 10⁴ cells/well in 150 µL of M199 medium containing 2 % FBS. After 18 h, tube structures were imaged using a CKX41 phase-contrast microscope (Olympus, Tokyo, Japan). The number of master junctions was quantified in a blinded manner from four independent experiments using ImageJ (NIH, v1.34; http://rsbweb.nih.gov/ij/).

2.5 Cell migration assay

Cell migration was assessed using the SPL Scar™ Block system (SPL Life Sciences, Pocheon-si, Gyeonggi-do, Korea) to create uniform scratches, as previously described [19]. HUVECs were seeded to >90 % confluence; after 5 h, the blocks were removed. Cells were then incubated in medium containing 2 % FBS and 2 mM thymidine (Sigma-Aldrich) for 24 h. Images were captured with a phase-contrast microscope, and migration was quantified in a blinded manner using ImageJ.

2.6 Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assay

Apoptosis was detected using a TUNEL assay kit (Chemicon, Temecula, CA, USA), according to the manufacturer’s protocol and a previously described method [20]. Digital images of TUNEL-positive cells were acquired using a confocal microscope (K1-Fluo; Nanoscope Systems) at 200 × magnification. Quantification was performed in four independent experiments using ImageJ in a blinded manner.

2.7 BrdU incorporation assays

Cell proliferation was evaluated by 5-bromo-2′-deoxyuridine (BrdU) incorporation as previously described [21]. HUVECs were incubated with 10 µM BrdU (Sigma-Aldrich) for 1 h at 37 °C, fixed, and subjected to antigen retrieval. BrdU incorporation was visualized with an anti-BrdU antibody (1:200; Bio-Rad, Hercules, CA, USA) and confocal microscopy. BrdU-positive cells were counted in four randomly selected fields at 200 × magnification using ImageJ in a blinded manner.

2.8 Histological examination

For fluorescence histology, cultured cells were fixed with 4 % paraformaldehyde for 15 min at room temperature. Samples were blocked using 1 % bovine serum albumin and incubated overnight at 4 °C with primary antibodies against SFRP2 (Abcam, 1:50), collagen I (Invitrogen, 1:100), and collagen IV (Millipore, 1:100). After samples had been washed with PBS, secondary antibodies were applied: DyLight^® 550-conjugated donkey anti-rabbit (Abcam) and Alexa Fluor^® 488-conjugated donkey anti-mouse and anti-goat (Jackson ImmunoResearch), all at 1:200 dilution. Samples were mounted in a DAPI-containing solution (Vector Laboratories) for nuclear staining [20]. Images were obtained using a confocal microscope (Nanoscope Systems) with the following laser excitation wavelengths: 405 nm for DAPI, 488 nm for FITC, and 561 nm for TRITC. Detector gain and offset settings were kept constant across experimental groups to enable direct comparison and minimize fluorescence variation. Emission ranges were optimized to avoid spectral overlap. Quantification was performed using ImageJ in a blinded manner across four independent biological replicates.

2.9 Western blotting

Protein lysates (50 µg per lane) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis on 4–20 % gradient gels, transferred to polyvinylidene fluoride membranes, and blocked with 5 % nonfat milk. Membranes were probed with primary antibodies against SFRP2 (Abcam, 1:2000), collagen I (Invitrogen, 1:2000), collagen IV (Millipore, 1:2000), Wnt3a and Wnt5a (Cell Signaling Technology, 1:2000), and β-actin (Santa Cruz Biotechnology, 1:4000). β-actin was used as the internal loading control, and its expression stability under NG and HG conditions was verified by comparing band intensities across representative samples. No significant differences in β-actin levels were observed among treatment groups, confirming its suitability for normalization. Detection was performed with an enhanced chemiluminescence system (Amersham Pharmacia Biotech), and band intensities were quantified by densitometry in four independent experiments using ImageJ in a blinded manner.

2.10 Statistical analysis

All quantitative data are expressed as means ± standard errors of the mean from at least four independent experiments. Normality of data distribution was assessed using the Shapiro–Wilk test, and homogeneity of variances was evaluated using Levene’s test prior to applying parametric statistical tests. Unpaired Student’s t-tests were used for comparisons between two groups, and one-way ANOVA followed by Tukey’s post hoc test was applied for multiple group comparisons. Statistical analyses were carried out using GraphPad Prism 8 (GraphPad Software Inc., San Diego, CA, USA). A P-value < 0.05 was considered statistically significant.

3.1 Primary culture and identification of human CC fibroblasts

To investigate the effects of SFRP2 on fibrosis in human CC fibroblasts under HG conditions, primary fibroblasts were successfully isolated from human CC TA tissue. Primary human corpus cavernosum fibroblasts displayed positive immunofluorescence staining for CD90 and vimentin and were negative for CD31 and NG2, confirming fibroblast identity (Figure 1A). Quantification of fluorescence-positive cells revealed markedly higher expression of CD90 and vimentin compared with CD31 and NG2 (Figure 1B). All experiments were performed using cells at passages 3–8. These findings, consistent with a previous report [14], confirmed that the isolated cells were human CC fibroblasts suitable for subsequent analyses.

Figure 1:

SFRP2 expression is increased in human CC fibroblasts under HG conditions. (A) Characterization of human CC fibroblasts by immunofluorescence staining showing positive expression of fibroblast markers CD90 and vimentin, and absence of the endothelial marker CD31 and the pericyte marker NG2. (B) Quantification of the cell marker-positive percentage in immunofluorescence images. (C) Representative western blots for SFRP2 under NG (5 mM) and HG (30 mM) conditions. (D) SFRP2 immunostaining (red) in human CC fibroblasts under NG and HG conditions. Nuclei were labeled with the DNA dye DAPI (blue). (E) Quantification of the SFRP2/β-actin ratio by densitometry. (F) Quantification of the SFRP2-positive (+) area in immunofluorescence images. Data are presented as mean ± SEM (n = 4). The relative ratio in the NG was set to 1. *P < 0.05. Scale bars, 100 µm. CC, corpus cavernosum; NG, normal glucose; HG, high glucose.

3.2 SFRP2 expression increases in human CC fibroblasts under HG conditions

To examine the effect of HG on SFRP2 expression, primary CC fibroblasts were cultured under NG and HG conditions. Exposure to HG (30 mM) for 48 h significantly increased SFRP2 expression compared with NG conditions, as shown by densitometric analysis of Western blots (Figure 1C and 1E) and immunofluorescence staining (Figure 1D and 1F). Quantification demonstrated a ∼2.1-fold elevation in SFRP2 protein levels relative to NG (p = 0.0494). These results indicate that a hyperglycemic, diabetes-like environment can upregulate SFRP2 expression in penile fibroblasts and may initiate early fibrotic signaling.

3.3 BMP1- and HG-induced SFRP2 expression corresponds to the functional range of low-dose SFRP2

The profibrotic effects of SFRP2 are dose-dependent, such that low and high doses produce distinct biological outcomes [11]. To determine whether the HG-induced upregulation of SFRP2 corresponds to low or high functional doses, endogenous SFRP2 levels were compared in fibroblasts treated with BMP1 (a positive control for fibrosis induction), HG, low-dose SFRP2 (20 nM), and high-dose SFRP2 (20 µM). Western blot analysis showed that BMP1- and HG-induced SFRP2 expression was comparable to that in the low-dose SFRP2 group (Figure 2). These results suggest that BMP1- and HG-induced SFRP2 upregulation falls within the low-dose functional range (20 nM) and mimics the profibrotic effects observed under these stimuli, indicating that both treatments functionally recapitulate a low-dose SFRP2 response. Interpretation of low-dose functional activity was based on relative densitometric quantification of Western blot band intensity normalized to β-actin rather than absolute protein levels.

Figure 2:

BMP1 and HG induce SFRP2 expression at levels comparable to low-dose SFRP2. (A) Representative western blots of culture medium showing secreted SFRP2 levels in human CC fibroblasts treated with BMP1 (50 ng/mL), HG (30 mM), low-dose SFRP2 (20 nM), or high-dose SFRP2 (20 μM). Ponceau S staining served as a loading control. (B) Quantification of the SFRP2/Ponceau S ratio by densitometry. Data are presented as mean ± SEM (n = 3–4). The relative ratio in the NG was set to 1.*P < 0.05, ***P < 0.001; ns, not significant. CC, corpus cavernosum; HG, high glucose.

3.4 High-dose SFRP2 attenuates BMP1-induced collagen expression and Wnt signaling activation in human CC fibroblasts

High doses of SFRP2 are known to inhibit the canonical Wnt signaling pathway and suppress collagen-processing enzyme activity, thereby mitigating fibrotic progression [11]. Additionally, SFRP2 exerts pro-angiogenic and cardioprotective effects via modulation of both canonical and non-canonical Wnt signaling pathways [22]. To assess the antifibrotic effect of high-dose SFRP2, human CC fibroblasts were stimulated with BMP1 at various concentrations (15, 50, and 100 ng/mL). As expected, western blot analysis revealed dose-dependent increases in collagen I and collagen IV expression, with maximal induction at 50–100 ng/mL (Figure 3A–C). Based on these results, 50 ng/mL BMP1 was selected for subsequent co-treatment with recombinant SFRP2 (20 µM). High-dose SFRP2 substantially reduced the BMP1-induced expression of both collagen I and collagen IV (Figure 3D–F), with statistically significant differences when compared with BMP1 alone (p < 0.01). Furthermore, SFRP2 suppressed BMP1-induced activation of the Wnt signaling pathway, as indicated by decreased protein levels of Wnt3a and Wnt5a/b (Figure 3D, 3G, 3H). Quantitative densitometry confirmed reductions of approximately 40–66 % relative to BMP1 treatment alone (p < 0.01). Immunofluorescence analysis supported these findings, revealing increased deposition of collagen I and collagen IV in the BMP1 + PBS group, whereas co-treatment with SFRP2 substantially reduced collagen accumulation, in some cases below control levels (Figure 3I–K). These results demonstrate that high-dose SFRP2 effectively inhibits BMP1-induced fibrosis by downregulating collagen production and suppressing canonical Wnt signaling activity in human CC fibroblasts.

Figure 3:

High-dose SFRP2 inhibits BMP1-induced collagen expression and Wnt signaling in human CC fibroblasts. (A) Representative western blot showing dose-dependent induction of collagen I and collagen IV by BMP1 (15, 50, and 100 ng/mL). (B, C) Quantification of collagen I and collagen IV relative to β-actin. (D) Representative western blot showing that co-treatment with SFRP2 (20 μM) suppresses BMP1-induced expression of collagen I, collagen IV, Wnt3a, and Wnt5a/b. (E–H) Quantitative analysis of collagen and Wnt markers. (I) Immunofluorescence images of collagen I and collagen IV (green) in human CC fibroblasts under BMP1 ± SFRP2 treatment. (J, K) Quantification of collagen-positive areas. Data are presented as mean ± SEM (n = 3–5). The relative ratio in the Control was set to 1. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant. Scale bars, 100 µm. CC, corpus cavernosum.

3.5 High-dose SFRP2 inhibits HG-induced fibrosis via the Wnt pathway

Chronic exposure to HG is a well-recognized driver of fibrotic remodeling in multiple tissues, primarily through activation of fibroblasts and their differentiation into myofibroblasts, which increases ECM protein synthesis and collagen deposition via profibrotic signaling pathways such as TGF-β and Wnt [23]. In this study, human CC fibroblasts were exposed to HG (30 mM) for 48 h prior to SFRP2 treatment to maintain consistency with earlier experimental conditions. Accordingly, human CC fibroblasts were cultured under HG (30 mM) conditions with or without recombinant SFRP2 (20 µM). Western blot analysis showed that HG significantly increased collagen I and collagen IV expression compared with NG controls (Figure 4A–C). Co-treatment with SFRP2 considerably reduced HG-induced collagen expression to near or below baseline levels. HG exposure also increased activation of canonical Wnt signaling, as indicated by upregulation of Wnt3a and Wnt5a/b. High-dose SFRP2 treatment effectively downregulated these Wnt ligands, indicating suppression of Wnt signaling activity (Figure 4A, 4D, 4E). Our findings suggest that high-dose SFRP2 mitigates HG-induced fibrotic remodeling by inhibiting Wnt signaling activation and reducing ECM protein overproduction in penile fibroblasts.

Figure 4:

High-dose SFRP2 suppresses HG-induced collagen expression and Wnt signaling in human CC fibroblasts. (A) Representative western blot showing collagen I, collagen IV, and Wnt signaling proteins (Wnt3a and Wnt5a/b) in human CC fibroblasts cultured under NG, HG with PBS, or HG with high-dose SFRP2 (20 μM). (B–E) Quantitative analysis of protein expression relative to β-actin, including collagen I (B), collagen IV (C), Wnt3a (D), and Wnt5a/b (E). Data are presented as mean ± SEM (n = 3–4). The relative ratio in the NG was set to 1. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant. CC, corpus cavernosum; NG, normal glucose; HG, high glucose; PBS, phosphate-buffered saline.

3.6 High-dose SFRP2 promotes angiogenesis and enhances endothelial cell survival under HG conditions

Chronic HG exposure is known to impair angiogenesis and induce apoptosis, thereby compromising endothelial function and contributing to microvascular dysfunction in diabetic complications such as retinopathy, nephropathy, and ED [20]. To assess the pro-angiogenic and cytoprotective potential of high-dose SFRP2 under diabetic-like conditions, HUVECs were treated with HG (30 mM) in the presence or absence of recombinant SFRP2 (20 µM). Tube formation and migration assays demonstrated that SFRP2 significantly enhanced angiogenic and migratory activity compared with the PBS-treated group under HG conditions (Figure 5A–C). To further evaluate whether these beneficial effects were independent of off-target or cytotoxic responses, BrdU incorporation and TUNEL assays were performed to assess cell proliferation and apoptosis under the same treatment conditions. BrdU staining revealed increased endothelial cell proliferation in the high-dose SFRP2 group, whereas TUNEL staining demonstrated reduced apoptosis (Figure 5D–F). The absence of increased apoptosis or impaired cell proliferation indicates that high-dose SFRP2 does not exert detectable cytotoxic effects, supporting the safety of the treatment concentration used. Collectively, these results indicate that high-dose SFRP2 improves endothelial cell function by promoting angiogenesis, increasing proliferation, and reducing apoptosis under hyperglycemic conditions, suggesting its potential to restore vascular integrity in diabetic complications.

Figure 5:

High-dose SFRP2 promotes angiogenesis, migration, and survival of endothelial cells under HG conditions. (A) Representative images from tube formation and migration assays in human umbilical vein endothelial cells (HUVECs) cultured under NG, HG with PBS, or HG with high-dose SFRP2 (20 μM). (B) Quantification of tube formation based on the number of master junctions. (C) Quantification of endothelial cell migration by measuring the migrated area. (D) Representative immunofluorescence images showing BrdU-positive (red) proliferating cells and TUNEL-positive (green) apoptotic cells, with nuclei stained by DAPI (blue). (E) Quantification of BrdU-positive proliferating cells. (F) Quantification of TUNEL-positive apoptotic cells. Data are presented as mean ± SEM (n = 3–9). *P < 0.05, **P < 0.01, ***P < 0.001. Scale bars, 100 µm. NG, normal glucose; HG, high glucose; PBS, phosphate-buffered saline; BrdU, 5-bromo-2′-deoxyuridine; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling.