Exosomes derived from smooth muscle cells ameliorate diabetes‐induced erectile dysfunction by inhibiting fibrosis and modulating the NO/cGMP pathway

Abstract Erectile dysfunction (ED) is a major health issue among men with diabetes, and ED induced by diabetes mellitus (DMED) is particularly difficult to treat. Therefore, novel therapeutic approaches for the treatment of DMED are urgently needed. Exosomes, nanosized particles involved in many physiological and pathological processes, may become a promising tool for DMED treatment. In this study, we investigated the therapeutic effect of exosomes derived from corpus cavernosum smooth muscle cells (CCSMC‐EXOs) on erectile function in a rat model of diabetes and compared their effect with that of exosomes derived from mesenchymal stem cells (MSC‐EXOs). We incubated labelled CCSMC‐EXOs and MSC‐EXOs with CCSMCs and then observed uptake of the exosomes at different time points using laser confocal microscopy. CCSMC‐EXOs were more easily taken up by CCSMCs. The peak concentration and retention time of labelled CCSMC‐EXOs and MSC‐EXOs in the corpus cavernosum of DMED rats after intracavernous injection were compared by in vivo imaging techniques. Intracavernous injection of CCSMC‐EXOs was associated with a relatively high peak concentration and long retention time. Our data showed that CCSMC‐EXOs could improve erectile function in DMED rats. Meanwhile, CCSMC‐EXOs could exert antifibrotic effects by increasing the smooth muscle content and reducing collagen deposition. CCSMC‐EXOs also increased the expression of eNOS and nNOS, followed by increased levels of NO and cGMP. These findings initially identify the possible role of CCSMC‐EXOs in ameliorating DMED through inhibiting corporal fibrosis and modulating the NO/cGMP signalling pathway, providing a theoretical basis for a breakthrough in the treatment of DMED.


| INTRODUC TI ON
As one of the most common male sexual dysfunctions, erectile dysfunction (ED) is a common disease in diabetic males.
According to epidemiology, the incidence of ED in diabetic patients exceeds 50%, 1 and by 2045, the number of diabetic patients will reach 629 million. 2 Penile erection is a complete hemodynamic process that includes corpus cavernosum smooth muscle relaxation, increased arterial blood flow and blocked venous return. 3 Many studies have shown that important features of ED induced by diabetes mellitus (DMED) are the reduction in corpus cavernosum smooth muscle and decrease in smooth muscle relaxation function under high-glucose conditions, which leads to an inability to obtain or maintain a satisfactory erection during intercourse. [3][4][5] DMED is particularly difficult to treat, and phosphodiesterase type 5 inhibitor, a first-line drug for the treatment of ED, has poor efficacy in DMED. 6 Therefore, a novel therapeutic approach is needed.
Exosomes are nanosized (40- Nevertheless, CCSMCs are unlikely to be directly transferred into the corpus cavernosum. Therefore, we propose a hypothesis about whether DMED can be alleviated by supplementation with exosomes derived from CCSMCs (CCSMC-EXOs). At present, studies have shown that exosomes derived from vascular smooth muscle cells play important roles in intercellular communication and vascular protection. 10,11 In addition, in recent years, studies have reported that exosomes derived from vascular smooth muscle can be targeted for enrichment in vascular smooth muscle. 12 These studies may partly support our hypothesis. Therefore, the aims of this study were to determine the potentially important role of CCSMC-EXOs in erectile function in a rat model of diabetes by their comparison with MSC-EXOs and to develop a novel exosome-based therapy for DMED.

| Animals
Eight-week-old male Sprague-Dawley rats were acquired from the Laboratory Animal Center of Tongji Medical College, Wuhan, China.
This rat model is a classic model used to study DM. 13,14 We used 10 rats for cell isolation and another 60 rats for animal experiments.

| Cell culture and identification
Adipose-derived stem cells (ADSCs) were isolated from inguinal adipose tissue as described in a previous study. 15 Briefly, adipose tissue was isolated, minced into small pieces and incubated with 0.075% collagenase type I (Sigma-Aldrich, St. Louis, USA) for 90 minutes at 37°C. After centrifugation at 1000 × g for 10 minutes, the supernatant was removed. The resulting pellet was treated with erythrocyte lysis buffer for 10 minutes on ice to remove red blood cells.
Subsequently, the remaining cells were suspended in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% foetal bovine serum (FBS), filtered through a 75-μm mesh strainer and cultured in a 5% CO 2 incubator at 37°C.
Bone mesenchymal stem cells (BMSCs) were isolated from rat femora as described in a previous study. 16 Briefly, after the femora were isolated, we used a needle to drill a hole in the distal end of each femur and injected a small volume of serum-containing medium into the bone to flush the marrow into a dish. Subsequently, the cell suspension was centrifuged at 450 × g for 5 min, and the supernatant was removed. The remaining cells were suspended in serum-containing medium and cultured in a 5% CO 2 incubator at 37°C.
CCSMCs were isolated from the rat penises using a standard protocol in our laboratory. 18 Then, the CCSMCs were cultured in DMEM supplemented with 10% FBS. The differential adhesion method was administered to minimize the number of other cells and to ensure that only CCSMCs were present in the culture flask.

| Exosome isolation and characterization
Exosomes derived from cells were isolated from conventional culture medium. Briefly, after cells were cultured in exosome-free medium for 48 hours, the cell supernatants were collected, and exosomes were isolated through multistep centrifugation. Dead cells and debris were eliminated through centrifugation at 300 × g for 10 minutes, 2000 × g for 20 minutes and 10 000 × g for 20 minutes.
Exosomes were isolated using ExoQuick reagent following the manufacturer's recommended protocol. 19,20 To characterize the exosomes, protein levels of CD9 (1:1000; Observations were conducted at 0, 4, 8, 16, 24 and 48 hours after incubation. At the end of each incubation period, the culture medium was discarded, and the cells were fixed with 4% paraformaldehyde. The slides were sealed with DAPI and photographed with a laser confocal microscope (SP8, Leica, Germany).
The same vehicle was injected intraperitoneally into the other 10 control rats (Con group, n = 10). The dose of the drug was based on that used in previous investigations by our laboratory. 5,23 Three and seven days after the intraperitoneal injection, tail-vein blood glucose levels were gauged, and the onset of DM was defined as fasting glucose levels higher than 16.7 mmol/L at both measurements.
Twelve weeks later, 49 of the 50 rats that had been injected with streptozotocin had survived. Subsequently, an apomorphine (APO) test was administered to identify the rats with DMED, and rats with negative results were considered DMED rats according to the protocol in a previous study. 24 Through the APO test, 31 rats were diagnosed with DMED and were further divided into four groups (DMED group, n = 7; DMED + CCSMC-EXO group,

| Evaluation of erectile function in vivo
Four weeks after intracavernous injection, the intracavernous pressure (ICP) was measured as previously described. 5,23 Briefly, under proper anaesthesia (100 mg/kg ketamine and 5 mg/kg midazolam), the major pelvic ganglion and cavernous nerves were exposed via midline laparotomy. When the evaluation of erectile function was over, the rats were killed by intraperitoneal injection of excess anaesthetic. Some corpus cavernosum tissue that had been pricked by the needle was discarded, and the rest was cut into four parts. One part was embedded in paraffin, and the other parts were stored in a −80°C freezer until subsequent experiments.

| Immunohistochemistry and Masson's trichrome staining
Penile tissue sections were processed for immunohistochemical investigations and incubated with antibody against TGF-β1 (1:100; Affinity Biosciences), eNOS (1:100; Affinity Biosciences) or nNOS (1:100; Affinity Biosciences) at 37°C for 1 hour. The sections were then incubated with biotinylated secondary antibodies. Masson's trichrome staining was carried out to determine the ratio between smooth muscle and collagen in the corpus cavernosum as previously described. 26

| Analysis of nitric oxide (NO) and cyclic guanosine monophosphate (cGMP) levels
The NO levels were determined using a nitrate-nitrite assay kit (S0024, Beyotime Biotechnology, China), and the cGMP levels were detected using an enzyme-linked immunosorbent assay kit (F15182, R & D Systems, Minneapolis, MN, USA). All steps were carried out according to the manufacturer's recommended protocol. The NO and cGMP levels were normalized to the protein concentration.

| Statistical analyses
The images were analysed using Image-Pro Plus 6.0 software (Media Cybernetics, Silver Spring, MD, USA), and data were analysed using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA, USA). The results are expressed as the mean plus or minus the standard deviation. Multiple groups were compared using one-way analysis of variance followed by the Tukey-Kramer test for post hoc comparisons.
Statistical significance was set at P < .05.

| Cell identification and exosome characterization
To identify the MSCs (BMSCs and ADSCs) used in this study, we first analysed the expression of cell surface antigens. As shown in Figure 1A, the cells expressed the MSC markers CD29 and CD90 but not the endothelial or hematopoietic markers CD34 and CD45.
Meanwhile, Figure 1B shows that the MSCs were successfully induced into osteoblasts (stained positive with Alizarin Red S) and adipocytes (stained positive with Oil Red O), which indicates the differentiation potential of the MSCs. To identify the CCSMCs used in this study, we subjected cells isolated from the rat penises to immunofluorescence staining for the specific markers of smooth muscle cells α-SMA and desmin. The results in Figure 1C show positive staining for α-SMA and desmin, and the proportion of cells with positive staining is greater than 95%, indicating that the cultured cells were CCSMCs.
The isolated exosomes were visualized and detected using TEM. As shown, the exosomes were approximately 100 nm in diameter ( Figure 1D). Meanwhile, nanoparticle tracking analysis indicated that the diameters of typical particles were mainly approximately 100 nm ( Figure 1D). Furthermore, the Western blot analysis results in Figure 1E revealed that the exosomal markers CD9, CD63 and TSG101 were abundant in our isolated exosomes, while staining for calnexin was negative. All these results indicate that the isolation of exosomes was successful.

| Exosomes derived from CCSMCs were more easily taken up by CCSMCs
To determine the time course of the uptake of different types of exosomes by CCSMCs under different conditions, we labelled CCSMC-EXOs, BMSC-EXOs and ADSC-EXOs with a green fluorescent marker, PKH67, and incubated them with CCSMCs under normal conditions and high-glucose conditions. As shown in Figure 2A

| Exosomes derived from CCSMCs were more easily retained in the corpus cavernosum of DMED rats
To compare the retention time of different types of exosomes in DMED rats after their intracavernous injection, CCSMC-EXOs, BMSC-EXOs and ADSC-EXOs were labelled with DiR, and the signal intensities of the exosomes (used to represent the amount of exosomes) at 0, 1, 2, 4 hours, 1 and 6 days after injection were observed by in vivo fluorescence imaging. As shown in Figure 3A and B, the peak signal intensity for CCSMC-EXOs approximately 4 hours after injection was higher than that for the other two types of exosomes (P < .05). Subsequently, CCSMC-EXOs were maintained at a certain level in the corpus cavernosum at 1 day and 1 week after injection, and the level of CCSMC-EXOs was significantly higher than that of the other two types of exosomes (P < .05). The above results suggest that compared with BMSC-EXOs and ADSC-EXOs, CCSMC-EXOs were more easily retained in the corpus cavernosum of DMED rats.

| Exosomes derived from CCSMCs could improve erectile function in DMED rats
First, as shown in Figure 4E and F, no significant differences in the initial weights and initial fasting glucose levels between rats in each group were observed (P > .05). Nevertheless, at 16 weeks after diabetes induction, the final weight of the Con group was obviously higher than that of other groups (P < .05). In contrast, the final Second, as shown in Figure 4A to D, max ICP/MAP and total ICP were used to evaluate erectile function in the rats. These indicators were significantly reduced in the DMED group and partly increased after administration of the different types of exosomes, although they were still lower than those in the Con group (P < .05). Compared with the injection of BMSC-EXOs and ADSC-EXOs, the injection of CCSMC-EXOs more effectively improved erectile function (P < .05).
The above results indicate that CCSMC-EXOs could effectively improve erectile function in DMED rats.

| Exosomes derived from CCSMCs could inhibit corporal fibrosis in the corpus cavernosum of DMED rats
As shown in Figure 5A Importantly, compared with BMSC-EXOs and ADSC-EXOs, CCSMC-EXOs increased the smooth muscle content to a greater extent (P < .05). TGF-β1, an important profibrotic factor in the corpus cavernosum, was significantly increased in the DMED group. However, the level of TGF-β1 in the CCSMC-EXO group and ADSC-EXO group was significantly lower than that in the DMED group, although it was still higher than that in the Con group (P < .05; Figure 5A,D,F). The above results indicate that CCSMC-EXOs could inhibit corporal fibrosis in the corpus cavernosum of DMED rats.

| Exosomes derived from CCSMCs could improve the NO/cGMP signalling pathway in the corpus cavernosum of DMED rats
Diabetes exerted a significant inhibitory effect on the NO/cGMP signalling pathway, and exosome treatment improved this pathway.
To better understand this process, endothelial nitric oxide synthase (eNOS) and neuronal nitric oxide synthase (nNOS), the two most important factors that produce NO in the corpus cavernosum, were detected using Western blot analysis and immunol staining. As shown in Figure 6A-C, the results of Western blot analysis revealed that the levels of eNOS and nNOS were higher in the Con group than in the other four groups. Although the levels of eNOS and nNOS in the DMED + CCSMC-EXO group were lower than those in the Con group, they were significantly higher than those in the other DMED groups (P < .05). Consistent with the Western blot results, the results of immunol staining for eNOS and nNOS also showed decreased eNOS and nNOS expression in the DMED group, which was partly increased after CCSMC-EXO treatment (P < .05; Figure 6D to H).
In addition, NO and cGMP were directly detected using an assay kit. The NO and cGMP concentrations were obviously reduced in the DMED group compared with the other four groups and partly increased after treatment with exosomes. Furthermore, compared with BMSC-EXOs and ADSC-EXOs, CCSMC-EXOs increased the concentrations of NO and cGMP to a greater extent (P < .05; Figure 6I and J). All the above indicators show that CCSMC-EXOs could improve the NO/cGMP signalling pathway in the corpus cavernosum of DMED rats.

| D ISCUSS I ON
As one of the most common types of ED caused by medical problems, DMED still lack effective and specific treatments. 27 Recent studies have confirmed that paracrine therapy is an important method by which MSCs treat ED. 28  Exosomes can be taken up by cells through unspecific pinocytosis or specific molecular interactions, including membrane-exposed sugars, lipids or proteins. 32 A breakthrough in understanding exosome transmission between cells, exosomes were discovered to express a certain number of parent cell-derived markers. [33][34][35] This means that cells might be more inclined to take up exosomes of the same cell origin. Alvarez-Erviti L et al reported similar conclusions; they found that exosomes have the unique property of homing selectivity because they can express cell type-specific protein markers discovered in the membrane of the parent cell. 36 In our research, we incubated labelled CCSMC-EXOs and MSC-EXOs with CCSMCs and then observed uptake of the exosomes at different time points using laser confocal microscopy. We found that CCSMC-EXOs could be taken up by CCSMCs faster than MSC-EXOs in normal and high-glucose environments. This suggests that CCSMC-EXOs might have a faster onset time as a potential therapeutic drug for DMED.
However, this finding needs to be further confirmed by more studies. Subsequently, consistent with the results of the in vitro uptake study, our in vivo imaging results showed that intracavernous injection of DiR-labelled CCSMC-EXOs was associated with a relatively high peak concentration and long retention time. The exact mechanism by which CCSMC-EXOs were more easily retained in the corpus cavernosum of DMED rats is unclear and needs to be further elucidated in future studies. One possible hypothesis is that CCSMCs can take up CCSMC-EXOs faster, which effectively reduces the loss of CCSMC-EXOs injected into the corpus cavernosum. As reported, because of the anatomical characteristics of the corpus cavernosum, locally injected fluid is not easily retained in the corpus cavernosum for a long time. 37 In summary, our data are the first demonstration of the dynamics of CCSMC-EXOs injected in the corpus cavernosum and thus provide important information regarding the administration of CCSMC-EXOs in exosome-based therapy for DMED.
Erectile dysfunction is typically observed in patients with vasculogenic, neurogenic or endocrinologic diseases and is one of the common early complications of diabetes. 3,4 After successfully establishing a rat model of diabetes, we found that the erectile function of diabetic rats was seriously impaired, which is consistent with observations in diabetic men with ED. Subsequently, we observed that CCSMC-EXOs or MSC-EXOs could preserve the erectile function of DMED rats. The therapeutic effect of MSC-EXOs on DMED has been reported in the relevant literature, 8,9 and our results are consistent with the conclusions reported in these studies. Furthermore, for the first time, our study demonstrates the therapeutic effect of CCSMC-EXOs on DMED.
We found that the specific mechanism of this therapeutic effect might be related to the ability of CCSMC-EXOs to inhibit corporal fibrosis, as shown by the results of immunohistochemistry and Masson's trichrome staining. Our previous research and other studies have shown that corporal fibrosis caused by reduced corpus cavernosum smooth muscle content and increased collagen content is one of the main mechanisms of DMED. 5,38 Here, we observed a dramatic decrease in the smooth muscle content and a significant increase in the collagen content of the corpus cavernosum of DMED rats, and treatment with CCSMC-EXOs or MSC-EXOs effectively F I G U R E 3 Exosome intracavernous injection and in vivo imaging. (A) Representative images of DMED rats that underwent intracavernous injection with DiR-labelled exosomes. (B) The signal intensities of different DiR-labelled exosomes in DMED rats at different time points after injection are presented as a bar graph. Data are expressed as the mean ± standard deviation. * P < .05 compared with the CCSMC-EXO group. CCSMC-EXOs: exosomes derived from corpus cavernosum smooth muscle cells; BMSC-EXOs: exosomes derived from bone marrow stem cells; ADSC-EXOs: exosomes derived from adipose-derived stem cells; DMED: diabetes mellitusinduced erectile dysfunction alleviated these histopathological changes in the corpus cavernosum. On the one hand, our data support previous reports that MSC-EXOs have an inhibitory effect on corporal fibrosis, 28 and on the other hand, they also indicate that CCSMC-EXOs have a more obvious inhibitory effect on corporal fibrosis than other types of exosomes do. We believe that this inhibition is related to the ability of CCSMC-EXOs to effectively reduce the expression of TGF-β1. TGF-β1 is an important profibrotic cytokine that has been recognized as the key factor involved in the formation and development of corporal fibrosis. 39 Kim et al reported that the increased expression of TGF-β1 is closely related to the accumulation and deposition of collagen. 40 We found that compared with BMSC-EXOs, CCSMC-EXOs could significantly reduce the expression of TGF-β1, thereby The cGMP concentration in all five groups was determined with an assay kit. Data are expressed as the mean ± standard deviation. & P < .05 compared with the Con group. # P < .05 compared with the DMED group. * P < .05 compared with the DMED + CCSMC-EXO group. CCSMC-EXOs: exosomes derived from corpus cavernosum smooth muscle cells; BMSC-EXOs: exosomes derived from bone marrow stem cells; ADSC-EXOs: exosomes derived from adipose-derived stem cells; DMED: diabetes mellitus-induced erectile dysfunction; DAPI: 4',6-diamidino-2-phenylindole; eNOS: endothelial nitric oxide synthase; nNOS: neuronal nitric oxide synthase; NO: nitric oxide; cGMP: cyclic guanosine monophosphate; IOD: integral optical density; IOD/area: mean optical density Recently, accumulated evidence has shown that exosomes from an increasing number of cell types, such as Schwann cells and macrophages, can promote nerve regeneration. 46 As a supplement to the above literature, our data provide evidence for the nerve-regeneration effect of CCSMC-EXOs in the corpus cavernosum of diabetic rats. Interestingly, it was reported that MSC-EXOs could increase the expression of nNOS in the corpus cavernosum in a rat model of cavernous nerve injury, 47,48 but no similar phenomenon was observed in our diabetic rat model. We think this may be because of the difference in animal models, which should be confirmed by more studies in the future. In summary, our data indicate that CCSMC-EXOs can ameliorate DMED by inhibiting fibrosis and stimulating NO/cGMP pathway. As we all know, exosomes contain various substances, of which microRNAs and proteins may be important factors that mediate the therapeutic effects of exosomes. 49 Zhu et al reported that exosomes from ADSC can ameliorate DMED by delivering microRNAs, including some proangiogenic microRNAs (miR-126, miR-130a and miR-132) and antifibrotic microRNAs (miR-let7b and miR-let7c). 8 In addition, Wang et al reported that corin in exosomes from ADSC also has the effect of ameliorating DMED. 50 At present, there is no relevant report on CCSMC-EXOs. We speculate that CCSMC-EXOs may also ameliorate DMED through specific microRNAs and proteins, which should be confirmed by more studies in the future.
There are several limitations to our current study. First, we studied the role of CCSMC-EXOs in type 1 diabetes, whereas the effectiveness of CCSMC-EXOs in type 2 diabetes is not yet clear.
Currently, type 2 diabetes is more common than type 1 diabetes.
Although the pathological changes of ED in the corpus cavernosum caused by type 1 diabetes and type 2 diabetes are similar, the role of CCSMC-EXOs in type 2 diabetes remains to be determined. Second, among the several isolation methods for CCSMC-EXOs, we used the ExoQuick method, which is a widely accepted method of exosome isolation. This method can achieve high yield and quality, but its specificity is relatively low. 51 Therefore, additional studies using other isolation methods are necessary to verify our results. Third, we used a single dose (100 μg) of exosomes for intracavernous injection.
Although this is a common dose used in exosome-related in vivo studies, future studies will be conducted to identify the optimum dose. Fourth, it is still a challenge to fully understand the specific composition of exosomes. 52 we found that ADSC-EXOs, BMSC-EXOs and CCSMC-EXOs were different in dynamics and functions.
The main reason for these differences is that the cargo of exosomes from different cells are different. However, we do not yet know the specific differences in the cargo of these exosomes, which needs to be further studied. Finally, we observed that CCSMC-EXOs could effectively remain in the corpus cavernosum and improve erectile function by inhibiting corporal fibrosis and up-regulating the NO/ cGMP signalling pathway, but the exact mechanism has not been fully elucidated. Unknown microRNAs or proteins may participate in this complex process, which our laboratory has recently initiated research to further clarify.
In conclusion, our work initially identified the possible role of CCSMC-EXOs in ameliorating DMED through inhibiting corporal fibrosis and modulating the NO/cGMP signalling pathway. We also uncovered the dynamics of injected exosomes in the corpus cavernosum and compared the therapeutic effects of CCSMC-EXOs and MSC-EXOs on DMED. These findings may provide new insights into the role of exosomes in the treatment of DMED.

ACK N OWLED G EM ENTS
This work was supported by a grant from the National Natural Science Foundations of China (Nos. 81873831 and 81901472).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.