Erectile dysfunction (ED) is a worldwide health problem. Although psychogenic factors are a major cause of ED, other factors such as age, diabetes, total prostatectomy and radiation in the pelvis also contribute to the occurrence of ED. These factors cause structural changes as well as functional abnormalities in the cavernous body, and therefore selective phosphodiesterase type 5 inhibitors (PDE5Is) are not so effective in the treatment of these diseases. For the recovery of erectile function in these patients, the regeneration of the cavernous body is necessary. In this regard, much attention has recently been placed on gene therapy and stem cell therapy.
Stem cells are defined as being capable of self-renewal and of differentiation into a variety of phenotypes. There are two categories of stem cells: embryonic stem cells (ESCs) and adult stem cells (ASCs). ESCs were originally isolated from the inner cell mass of blastocysts. ESCs are pluripotent stem cells that can give rise to the three germ layers. However, harvesting ESCs requires the destruction of human embryos and has therefore raised ethical concerns. To overcome this limitation, induced pluripotent stem (iPS) cells have been produced. Adult fibroblasts were reprogrammed by introducing four factors, Oct3/4, Sox2, c-Myc and Klf4 under ESCs culture conditions[3,4]. iPS cells are pluripotent stem cells with very similar characteristics to ESCs. Furthermore, many groups have now succeeded in reprograming somatic cells to create iPS cells by overexpression of variable sets of several transcription factors in cells without employing viruses or vectors[5-7]. Therefore, iPS cells are a promising option for regenerative medicine in the near future. A further option for avoiding the ethical problems of ESCs is the use of ASCs, which are basically multipotent stem cells that reside in various tissues, including the brain, skeletal muscle, bone marrow, adipose tissue, and dental pulp[8-11]. Besides having the potential to differentiate into various cell types, ASCs produce a broad range of cytokines that exert their effects in a paracrine and/or autocrine manner. Among ASCs, bone marrow-derived mesenchymal stem cells (BMMSCs) are the most commonly studied. BMMSCs reportedly have a potential to differentiate into various cell types including bone, cartilage, cardiac muscle, skeletal muscle, vascular endothelial cells (VECs) and vascular smooth muscle cells (VSMCs)[12,13]. Recently, adipose tissue-derived stem cells (ADSCs) have gained much attention because of the simplicity of harvesting hundreds of grams of subcutaneous adipose tissue without using invasive procedures, whereas painful bone marrow aspiration is required to collect just grams of bone marrow. Similar to BMMSCs, ADSCs have the potential to differentiate into various cell types. ADSCs strongly resemble BMMSCs in that they share similar expression patterns of cell surface markers and similar gene expression profiles[15,16].
TREATMENT OF ED USING ASCs
Many studies have demonstrated the efficacy of BMMSCs in the treatment of ED, and their efficacy seems to be reliable[17-27]. However, the mechanism by which BMMSCs restore erectile function remains controversial. BMMSCs were originally believed to home in on damaged tissues efficiently and differentiate into various cells at the target site. Genetically manipulated BMMSCs were also used expecting that these cells remain in the tissues and express a specific gene over a long period. Bivalacqua et al infected BMMSCs with an adenovirus expressing endothelial nitric oxide synthase (eNOS) and injected these cells into the penis of aged rats. The eNOS-modified BMMSCs restored erectile function, with the recovery being associated with increased eNOS protein expression, NOS activity, and cyclic guanosine monophosphate levels. Furthermore, the authors demonstrated that the injected BMMSCs survived for at least 21 d in the cavernous body and expressed markers for VECs and VSMCs. Song et al used immortalized human BMMSCs and injected them into rat penises, where the cells expressed markers for VECs and VSMCs in the cavernous body, although the authors did not examine whether erectile function was restored or not. Qiu et al injected BMMSCs into the penis of streptozotocin (STZ)-induced diabetic rats and found that the BMMSCs injection restored erectile function. The authors also demonstrated that injected cells remained in the cavernous body for at least 4 wk and some expressed markers for VECs and VSMCs. Although the injection of BMMSCs restored erectile function in these studies, there are several problems. First, these studies did not clearly calculate the percentage of injected BMMSCs that remained in the cavernous body and obtained markers for VECs and VSMCs, or the percentage mentioned in these studies was not sufficient to explain the recovery of erectile function. Second, there was a possibility that the injected BMMSCs fused with the residing cells in the cavernous body and acquired the phenotype of the residing cells. Terada et al used BMMSCs obtained from female transgenic mice expressing green fluorescent protein (GFP) and puromycin resistance gene. The BMMSCs were cocultured with a male embryonic stem cell line and then puromycin was added to remove the embryonic stem cells. The remaining cells were GFP positive and puromycin resistant, and morphologically similar to embryonic stem cells. These cells could be induced to differentiate into cardiac myocytes and neuronal cells, suggesting that embryonic stem cell-like pluripotent stem cells were established from BMMSCs. However, following DNA ploidy (the number of DNA copies) analysis using fluorescence-activated cell sorting, the cells were found to be tetraploid (4n) or hexaploid (6n), suggesting that they had developed from spontaneous fusion between the BMMSCs and the embryonic stem cells. The possibility of cell fusion in vivo was also reported. Alvarez-Dolado et al used transgenic mice that contain the lacZ reporter gene downstream of a stop codon flanked by loxP sites (floxed). Therefore, the lacZ reporter gene was only expressed when the loxP-flanked stop codon was excised by Cre recombinase. The authors lethally irradiated these mice and intraperitoneally injected BMMSCs from mice that ubiquitously express Cre recombinase and GFP. If cells from the donor and recipient fused, the Cre enzyme would excise the Lox P-flanked stop codon, thereby allowing expression of the lacZ gene. Results revealed that β-gal+ (fused) and GFP+ cells were found in the brain, heart, and liver of recipients, 2 and 4 mo post-transplantation. Thus, BMMSCs potentially fuse with other cell types in vivo and it appears that BMMSCs are differentiated into other cell types because of this phenomenon. In contrast, the paracrine effects of BMMSCs have been reported. Kendirci et al isolated BMMSCs positive for p75 low-affinity nerve growth factor receptor using magnetic-activated cell sorting, and injected these cells into the penis of rats following bilateral cavernous nerve crush injury. The injection of these cells restored erectile function. The engraftment of these cells in the cavernous tissue occurred very rarely, and the engrafted cells appeared fibroblastic. Furthermore, these cells secreted fibroblast growth factor 2 (FGF2), which suggested that FGF2 might protect the cavernous nerve after crush injury. More direct evidence of the paracrine effects of BMMSCs was reported by Yeghiazarians et al who injected BMMSCs extracts into infarcted hearts and found that the procedure effectively improved cardiac function, suggesting that BMMSCs per se were not required for their tissue protective effects. Although this scenario is attractive, no cytokines that are implicated in the recovery of erectile function have been specifically identified.
The efficacy of ADSCs in the treatment of ED seems to be reliable from the results of many previous reports[31-45]. However, the mechanisms by which ADSCs restore erectile function remain controversial. Ning et al reported that ADSCs could differentiate into VECs. ADSCs injected into the penis obtained a marker for VECs. FGF2 appeared to be necessary for the differentiation of ADSCs into VECs in vitro, although the functional role of FGF2 in the differentiation of ADSCs into VECs in vivo was not studied. Ryu et al demonstrated that the injection of ADSCs into the penis of STZ-induced diabetic mice restored erectile function. They also found that some injected ADSCs became CD31 positive, suggesting that these cells differentiated into VECs. However, ADSCs injected into the penis disappeared within 14 d. Kim et al applied human ADSCs and nerve growth factor-incorporated hyaluronic acid-based hydrogel to the cavernous nerve of rats following bilateral cavernous nerve crush injury. The authors demonstrated that this treatment restored erectile function. They also showed that some ADSCs were engrafted into the cavernous nerve 4 wk after treatment, suggesting that ADSCs could differentiate into nerve tissue. Although these studies showed that ADSCs have the ability to differentiate into the cells located in the cavernous body, the possibility of cell-cell fusion was not excluded in these studies. Alternatively, paracrine effects of ADSCs have been suggested. Several reports suggested the possibility of paracrine effects of ADSCs because ADSCs did not remain in the cavernous body for a long period[32,33]. Albersen et al injected both ADSCs and the lysate of ADSCs into the penis of rats that were subjected to bilateral cavernous nerve crush injury. They demonstrated that the injection of ADSCs lysate significantly restored erectile function. These results suggested that most, if not all, of the effects of ADSCs were mediated through their production of cytokines and/or immune modulators, although the authors did not identify any molecules that are functionally implicated in the restoration of erectile function. Zhang et al reported that ADSCs produced chemokine (C-X-C motif) ligand 5 (CXCL5) and that CXCL5 was implicated in the neurotrophic effects of ADSCs in vitro, although they did not confirm this finding in vivo. We recently reported that adrenomedullin (AM) is implicated in ADSCs-induced restoration of erectile function in diabetic rats. AM was originally isolated from human pheochromocytoma tissue, and has potent vasorelaxant and diuretic effects. In addition, AM is also produced by VECs, VSMCs and macrophages[49-51] and has the ability to stimulate angiogenesis[52-54]. When rat ADSCs were cultured in a medium containing growth factors for VECs, ADSCs produced significant amounts of AM (Figure 1A). The injection of ADSCs into the penis of diabetic rats restored erectile function (Figure1B), the morphology of the cavernous body (Figure 1C), and the expression of vascular endothelial (VE)-cadherin, a marker for VECs. However, when the expression of AM was knocked down using a small interfering RNA for AM, the favorable effects of ADSCs disappeared (Figure 1B and C). Furthermore, when AM was overexpressed in the penis using an adenovirus expressing AM, erectile function and the morphology of the cavernous body were restored in diabetic rats (Figure 1B and C). We also demonstrated that ADSCs produce angiopoietin-1 (Ang-1) and Ang-1 secreted from ADSCs are implicated in ADSCs-induced suppression of neointimal formation and stimulation of reendothelialization in a wire injury model of the rat femoral artery. Furthermore, we reported that overexpression of both AM and Ang-1 using adenoviruses expressing those proteins restored erectile function in diabetic rats to the same level as that observed in age-matched positive control rats (Figure 2). Therefore, it seems obvious that ADSCs produce various cytokines that potentially restore erectile function. The limitation is that ADSCs do not remain in the cavernous body for a long period, usually disappearing within a month[32,33,36]. Where do they go? Do they die or migrate to other tissues? Several interesting papers have been published regarding these points. Lin et al injected rat ADSCs and traced their locations after 2 and 7 d. ADSCs remained not only in the penis but also in the major pelvic ganglia (MPG), spleen and bone marrow. ADSCs preferentially remained in the bone marrow and the number of ADSCs remaining in the bone marrow was much larger than that remaining in the penis. Because ADSCs secrete various cytokines, this result suggests that ADSCs remaining outside the penis may affect the restoration of erectile function, although the survival of ADSCs for long periods was not examined in the study. Several studies have reported that ADSCs injected into the penis or placed around the prostate gland migrated to the MPG[35,39,45]. Fandel et al demonstrated that ADSCs injected into the penis migrated to the MPG, although ADSCs were not engrafted in the nerve tissue. Interestingly, the expression of stromal cell derived factor-1 (SDF-1) increased in MPG, suggesting that ADSCs preferentially migrated to the site of SDF-1 production. Qiu et al also reported that intracavernously injected ADSCs migrated to the MPG, where they remained 17 wk after injection, although the number of ADSCs remaining in the MPG was quite low. You et al showed that periprostatic implantation of ADSCs, but not their intracavernous injection, resulted in the migration of ADSCs to MPG. In summary, some ADSCs injected into the penis migrated to the tissues such as the bone marrow and MPG, and these cells may be implicated in the restoration of erectile function. The possible mechanisms by which ADSCs restore erectile function are summarized in Figure 3.
Figure 1 Production of adrenomedullin by adipose tissue-derived stem cells, and effects of knockdown and overexpression of adrenomedullin on the function and histology of the penis.
A: Adipose tissue-derived stem cells produce adrenomedullin especially when they were cultured in a medium containing growth factors for vascular endothelial cells (VECs); adipose tissue-derived stem cells (ADSCs) were cultured in endothelial basal medium (EBM: white circles) or endothelial growth medium (EGM: black circles) that contains growth factors for VECs. Medium was replaced with serum-free medium and incubated for the indicated periods. Adrenomedullin (AM) accumulated in the medium was measured. aP < 0.05 vs 0 h, bP < 0.01 vs 0 h and dP < 0.01 vs EBM culture at each time point (n = 6 per group); B: Effect of knockdown and overexpression of AM on erectile function. ADSCs were infected with lentivirus expressing negative control siRNA (LV_NCsiRNA) that is predicted not to target any known vertebrate gene or lentivirus expressing AM siRNA (LV_AMsiRNA). ADSCs were cultured in EGM for 1 wk, and those LV_NCsiRNA-infected ADSCs (EGM_NCsiRNA) and LV_AMsiRNA-infected ADSCs (EGM_AMsiRNA) were injected in the cavernous body of STZ-induced diabetic rats. ICP was measured 4 wk after the ADSCs injection. An adenovirus expressing AM (AdAM) or adenovirus expressing green fluorescent protein (AdGFP) was also injected into the cavernous body of STZ-induced diabetic rats, and ICP was measured 4 wk after the infection. Nontreated STZ-injected diabetic rats were used as the negative control (NC). Bar graphs show ICP/MAP (n = 5 per group). aP < 0.001 vs NC, dP < 0.001 vs EGM_NCsiRNA injection and fP < 0.001 vs AdGFP infection; C: Effect of knockdown and overexpression of AM on the morphology of the cavernous body. Experiments were performed in the same way as described in the legend for Figure 1B. The cavernous body was stained by the Elastica van Gieson method. The histology of the root portion of the penis (longitudinal section) is shown. Bars are 200 μm. Note that the size of trabeculae of the cavernous body is smaller when EGM_AMsiRNA or AdGFP was injected into the penis of diabetic rats, compared with when EGM_NCsiRNA or AdAM was injected into the penis. ICP: Intracavernous pressure; STZ: Streptozotocin; MAP: Mean arterial pressure.
Figure 2 Effects of adrenomedullin and angiopoietin-1 on the function and morphology of the penis.
A: Effects of adenovirus expressing adrenomedullin (AdAM), adenovirus expressing angiopoietin-1 (AdAng-1) and AdAM plus AdAng-1 infection on erectile function. These adenoviruses were infected into the cavernous body of STZ-induced diabetic rats. Adenovirus expressing green fluorescent protein (AdGFP) was also infected as the negative control. Age-matched Wistar rats were used as the positive control (PC). Bar graphs demonstrate ICP/MAP (n = 6 per group). bP < 0.001 vs AdGFP infection and dP < 0.001 vs PC; B: Histological analysis of the cavernous body after adenoviral infection. Elastica van Gieson staining of the cavernous body isolated from age-matched Wistar rats (PC), and STZ-induced diabetic rats infected with AdGFP, AdAM, AdAng-1, and AdAM plus AdAng-1. The histology of the root portion of the penis (longitudinal section) is shown. Bars are 300 μm. Note that the size of trabeculae of the cavernous body is small when AdGFP is injected into the penis of diabetic rats, and that the size is restored to a similar level as observed in the age-matched control group when AdAM and/or AdAng-1 are injected. ICP: Intracavernous pressure; STZ: Streptozotocin; MAP: Mean arterial pressure.
Figure 3 Possible mechanisms by which adipose tissue-derived stem cells stimulate the recovery of erectile function.
FGF2: Fibroblast growth factor 2; MPG: Major pelvic ganglia; ADSCs: Adipose tissue-derived stem cells; CXCL5: Chemokine (C-X-C motif) ligand 5.
Endothelial progenitor cells
Endothelial progenitor cells (EPCs) were originally isolated from human peripheral blood by Asahara et al. They isolated CD34-positive mononuclear blood cells and demonstrated that these cells obtained the characteristics of VECs when cultured on fibronectin-coated dishes. They also demonstrated that these cells were incorporated in ischemic tissues in vivo and expressed markers for VECs such as CD31 when introduced into the circulation using a hindlimb ischemia model. Furthermore, the authors showed that Flk-1-positive mononuclear blood cells were also integrated in the capillaries and small arteries when the hindlimb ischemia model was used. These cells were designated as EPCs. EPCs are progenitor cells whose differentiation potential is restricted to one lineage (VECs); therefore, they are not multipotent stem cells. Results of subsequent studies revealed that EPCs express three cell surface markers; CD133 (termed originally AC133), CD34, and Flk-1[58-60]. Premature EPCs either in the bone marrow or immediately after entering into the systemic circulation are positive for CD133/CD34/Flk-1. However, when EPCs become more mature, they lose the expression of CD133 and begin to express CD31 and VE-cadherin. An attempt to use EPCs for the treatment of ED has been reported by Gou et al. They transfected EPCs with the vascular endothelial growth factor gene and injected the transfected cells into the penis of diabetic rats. The authors found that this treatment restored erectile function and the injected cells were integrated into the sites of neovascularization in the cavernous body. However, most researchers in this field regard EPCs as a marker for ED rather than a therapeutic tool for ED. Factors affecting the number of circulating EPCs or their functions have been reported. The number of circulating EPCs and their migratory activity are reportedly reduced in patients with coronary risk factors[62,63]. EPCs isolated from type 2 diabetes patients have a decreased capacity for proliferation and the formation of capillary tubes in vitro. In contrast, the number of circulating EPCs rapidly increases after limb ischemia and acute myocardial infarction[65,66]. Because atherosclerosis and ED share a common feature, i.e., endothelial dysfunction, it has been speculated that the dynamics of EPCs might change in ED. Indeed, several studies have reported that the number of circulating EPCs decreased in ED patients[67-69], suggesting that the decrease in the number of circulating EPCs can predict the presence of ED as well as cardiovascular diseases. Interestingly, the number of circulating EPCs increased when patients were administered statins or PDE5Is[70-73] probably via the mobilization of EPCs from the bone marrow. Therefore, these drugs may improve erectile function via the mobilization of EPCs to the cavernous body.
Muscle-derived stem cells
Muscle-derived stem cells (MDSCs) are ASCs that exist in skeletal muscle. MDSCs can be obtained from autologous muscle biopsies and have been used for the treatment of ED in several studies. Nolazco et al injected MDSCs into the penis of aged rats and found that MDSCs differentiated into VSMCs in the cavernous body, resulting in the recovery of erectile function. Woo et al used a bilateral cavernous nerve injury model in rats and examined the effects of the injection of MDSCs into the penis on erectile function. They demonstrated that MDSCs remained in the cavernous body 4 wk after injection, and erectile function was significantly restored. Kovanecz et al used a bilateral cavernosal nerve resection model of rats and, following the injection of MDSCs into the cavernous body, found that erectile function was restored and α-smooth muscle actin expression was increased. The injection of MDSCs also increased the expression of neural nitric oxide synthase and brain-derived neurotrophic factor. In summary, MDSCs seem to be useful for the treatment of ED. However, because harvesting MDSCs from the skeletal muscle is relatively more invasive than the collection of ADSCs, the beneficial characteristics of MDSCs compared with ADSCs should be clarified before introducing them to clinical application.
Umbilical cord blood stem cells
Umbilical cord blood stem cells (UCBSCs) are an attractive type of stem cells in that they are youngest stem cells among a variety of ASCs. Because they are young, they have less possibility to have DNA damage than other ASCs[77,78]. Bahk et al used human UCBSCs to treat diabetic patients with ED, and demonstrated that erectile function was restored and that blood glucose levels decreased in these patients, although the mechanisms remain unknown.
Brain-derived stem cells
Brain-derived stem cells (BDSCs) reportedly have capacity to differentiate into VSMCs. Song et al isolated fetal BDSCs from embryonal 12-d rats and injected them into the penis. They demonstrated that injected BDSCs obtained characteristics of VSMCs in vivo 6 wk after injection, although they did not examine their effect on erectile function. Considering the source of BDSCs, it will be difficult to use them in clinical settings.
Neural crest stem cells
Neural crest stem cells (NCSCs) are the progenitor cells of several cell types that constitute the peripheral nervous system, including neurons, Schwann cells, adrenal chromaffin cells and smooth muscle cells. Transplantation of NCSCs could reportedly induce the regeneration of connective tissues, VSMCs, skeletal muscle and VECs[82,83]. Song et al injected NCSCs into the penis of rats and demonstrated that they obtained markers for VECs and VSMCs 2 wk after injection, although their effects on erectile function was not analyzed. Clinical application of NCSCs may also be difficult considering the source of these cells.
Treatment of ED using ESCs and iPS cells
Because iPS cells have been created, clinical application of pluripotent stem cells will be intensively explored in the future. To our knowledge, no studies have been published in which iPS cells or iPS cells-derived cells were used to treat ED. Bochinski et al used ESCs that had differentiated into the neural cell line, and injected them into the MPG or cavernous body using a bilateral cavernous nerve injury model. They found that the injection into the both MPG and cavernous body restored erectile function. They also found that neurofilament staining was recovered in the ESCs-injected group. Therefore, ESCs and iPS cells may be useful for treatment of ED. However, these cells may not efficiently home and survive for a long period under persistent inflammation. For example, in diabetic conditions hyperglycemia and adipocytokines induce persistent inflammation in the tissues. Implanted cells may not home and survive under these conditions unless such an inflammation is sufficiently controlled.
As mentioned above, stem cell therapy for ED appears to be a promising strategy. However, several problems should be solved before moving to clinical application.
It is well known that ESCs and iPS cells easily form tumors, because these cells are pluripotent. Although ASCs seem to be less prone to forming tumors, ASCs can form malignant tumors when transplanted in vivo. Jeong et al injected BMMSCs into the peri-infarct area of myocardial infaction (MI) model of mice and hindlimb muscle of mice with diabetic neuropathy. They found sarcoma formation in 30% of hearts in the MI model and in 46% of hindlimbs in the diabetic neuropathy model. Therefore, it will be necessary to sufficiently investigate the malignant potential of stem cells prior to their clinical use and establish methodology to select “healthy” stem cells that will not form tumors.
Fate of injected cells
As described above, some (most) of stem cells injected into the penis do not remain in the penis and migrate to other tissues such as the bone marrow and spleen. Little is known about the fate of these cells that have migrated to non-diseased organs. Detailed examinations will be necessary to detect the fate of these cells before moving to clinical applications.
How to improve homing and survival of stem cells
Most studies suggested that injected stem cells disappeared from the penis in one month. It is crucially important to explore methods to improve homing and survival of stem cells. It was reported that expression of SDF-1 was increased in the MPG and SDF-1 might stimulate migration of ADSCs to the MPG. Therefore, SDF-1 is a candidate that stimulates migration and homing of stem cells into injured sites. Intensive studies will be necessary to identify molecules that are implicated in migration, homing and survival of stem cells.
Activation and mobilization of endogenous stem cells
It is suggested from EPC study that statins or PDE5Is can stimulate mobilization of EPCs from the bone marrow[70-73]. Therefore, it may be possible to activate and/or mobilize tissue-residual endogenous stem cells by some drugs. If endogenous stem cells residing in the penis can be efficiently activated, it will help to regenerate the cavernous body. This possibility should be examined in the future.
Identification of paracrine factors
ASCs produce a variety of paracrine factors that potentially regenerate the cavernous body. However, paracrine factors that are implicated in the regeneration of the cavernous body have not been sufficiently identified. Furthermore, it remains unknown what combinations of these paracrine factors are most suitable to stimulate the regeneration of the cavernous body. If these problems are solved, administration of cytokines cocktail may be more effective to treat ED than ASCs injection.