The Role of Autophagy in Erectile Dysfunction

Generally speaking, a complete sexual response cycle is composed of four stages, namely desire, arousal, orgasm, and resolution. The arousal disorder is the key to erectile dysfunction (ED), which is defined as consistent or recurrent inability to achieve and/or maintain an adequate penile erection for satisfactory sexual performance [1]. ED is a common male disease. It is predicted that the global number of ED patients will reach 322 million by 2025 [2]. Nowadays, in addition to the aging factor, a large number of risk factors, such as chronic tobacco or alcohol consumption, obesity, diabetes mellitus, hypertension, hyperlipidemia, and depression, et al, are all not conducive to penile erection. Indeed, penile erection is a complex interplay of psychogenic, physiologic, neuroendocrine and cardiovascular system. But briefly, the essence of erection is a neurovascular response. Therefore, nerve injury, endothelial dysfunction, smooth muscle cell apoptosis, and fibrosis of corpus cavernosum have been extensively investigated in various forms of ED. Besides, with the storm of autophagy research, the role of autophagy has also begun to be studied in ED. In the past decade, a growing number of studies have shown that autophagy plays an important role in ED resulted from hyperlipidemia, diabetes, castration, cavernous nerve injury, and aging [3, 4, 5, 6, 7], but the exact mechanism has not been completely elucidated. Previous studies have revealed that autophagy plays a significant role in clearing oxidative stress, DNA damage, and dysfunctional proteins and organelles. And autophagy can not only regulate intracellular homeostasis and determine cell survival or death, but also regulate the secretion of exosomes and participate in tissue fibrosis [8, 9, 10]. Given that ED involves multiple mechanisms, and the role of autophagy varies in different cells. This review aims to elucidate autophagy and its possible role in various EDs from a cellular perspective, and expecting to provide a new perspective for the pathological research and clinical treatment of ED in the future.

Autophagy is a lysosome dependent and highly conservative material degradation process. Autophagy exists in all eukaryotic cells, and with an astonishing number of connections to human pathophysiology process, such as cancer, neurodegeneration, microbial infection, embryonic development and aging, et al [11, 12]. In 1963, the famous Belgian cell biologist Christian de Duve first coined the concept of “autophagy” and won the Nobel Prize in physiology or medicine in 1974 because of his seminal work in the discovery of the lysosomes [13]. In 2004, Aaron Ciechanover, et al was awarded the Nobel Prize in Chemistry for their research on the degradation mechanism of ubiquitinated proteins. Afterwards, the Japanese biologist Yoshinori Ohsumi, who devoted his life to the study of autophagy, achieved the Nobel Prize in physiology or medicine in 2016 due to his discovery of the autophagy mechanism of protein degradation and utilization in organisms. Which reignited the passion of autophagy research [14].

When understand autophagy from different perspective, autophagy will be classified by different ways. For example, according to the different routes of autophagy substrates transport to lysosomes, autophagy is divided into macroautophagy (it is usually referred to as autophagy), microautophagy, and chaperone-mediated autophagy (CMA), as shown in Fig. 1 [15, 16]. The traditional autophagy includes four processes, (1) the initiation of autophagy. In general, the autophagy initiator unc-51 like autophagy activating kinase 1 (ULK1) is suppressed by mammalian target of rapamycin complex 1 (mTORC1), which is a sensor of extracellular stimuli, such as starvation, drugs, hormones, and so on. In most cases, the suppressed ULK1 maintains a low level of autophagy in tissues or organs [17, 18]. But when the upstream mTORC1 is suppressed, for example using rapamycin (a classical inhibitor of mTORC1), autophagy will be activated significantly. (2) The formation of autophagosome. First of all, the isolation membrane is formed in cytoplasm, and it will further extend and wrap intracellular useless materials to develop a special organelle with the double-membrane structure, namely ‘autophagosome’ [19]. (3) The formation of autolysosome. Autophagosomes are transported along the intracellular microtubules to lysosomes and converted into autolysosomes [20]. (4) Digestion. Finally, all packaged materials are degraded by lysosomal enzymes. Compared with macroautophagy, microautophagy is much simpler. Lysosomes directly engulf cytosolic substrates through the inward invagination of their membranes [15]. The CMA mainly selectively degrades some soluble proteins, which contain a special ‘Lys-Phe-Glu-Arg-Gln (KFERQ)’-like motif. This structure will be recognized by the heat shock cognate 70 (HSC 70), and transported to lysosomes with the help of lysosome-associated membrane protein type 2A (LAMP 2A) [16, 21].

Fig. 1
The process and classification of autophagy. According to the different routes of autophagy substrates transport to lysosomes, autophagy is divided into macroautophagy, chaperone-mediated autophagy and microautophagy. Macroautophagy is often referred to as autophagy. It is more complex and involves four processes and six categories of core ATG protein complexes. First, when the mTORC1 is inhibited by intracellular or extracellular factors, ULK1, the autophagy promoter, will be activated. Autophagy is initiated, and isolation membranes are formed. Second, under the synergistic action of multiple complexes including PI3K III-Beclin1, ATG5–ATG12–ATG16, ATG2–ATG18, ATG9 and LC3-PE, autophagosomes are formed. They are specific markers of autophagy. Third, autophagosomes are transported along microtubules to lysosomes. Here, they fuse with each other to form autophagosomes. Finally, all of substrates are degraded by lysosomal enzymes to generate new products and energy. The chaperone-mediated autophagy mainly selectively degrades substrates that have ‘Lys-Phe-Glu-Arg-Gln (KFERQ)’-like motif. These special substrates can be recognized by HSC 70, and transported into the lysosomal lumen by a receptor or translocon like lysosome-associated membrane protein type 2A (LAMP 2A), and ultimately digested. Microautophagy refers that lysosomes directly engulf cytosolic substrates through the inward invagination of their limiting membranes. It involves none-selective microautophagy and selective microautophagy. The former has not well understood until now. But the selective microautophagy has been extensively reported, such as mitophagy, reticulophagy, endosomal microautophagy, et al. Which contributes to intracellular homeostasis by clearing dysfunctional organelles. ATG: autophagy-related gene, mTORC1: mammalian target of rapamycin complex 1, ULK1: unc-51 like autophagy activating kinase 1, PI3K: phosphatidylinositol 3-kinase, LC: microtubule-associated protein 1 light chain 3, PE: phosphatidyl ethanolamine, HSC: heat shock cognate.

• Click for larger image
• Download as PowerPoint slide

Besides, depending on the degree of autophagy, basal autophagy and inducible autophagy are put forward. Basal autophagy is mild, but continuously occurs in most cells, which is significant for the renewal of cellular components and the maintenance of intracellular homeostasis under physiological condition. Whereas, induced autophagy is severe. It can protect cells from damage to a certain extent. While uncontrolled overactivation of autophagy will lead to cell apoptosis, which is identified as ‘autophagic cell death’ [22, 23]. Besides, in recent years, autophagy has also been studied based on its target organelles. Therefore, mitophagy, endoplasmic reticulum autophagy (reticulophagy), ribosomal autophagy (ribophagy) and peroxisomal autophagy (pexophagy) are formed [24, 25].

From yeast genetic studies, up to now, 42 autophagy-related genes (Atgs) have been identified. And 16 of them have been identified as core Atg genes, which are highly conserved and essential for autophagosome formation in eukaryotes [26, 27]. Furthermore, these genes are categorized 6 functional units, including (1) Atg1/ULK1 (yeast Atg1 is homologous to mammalian ULK1) complex, a serine/threonine kinase, which is regulated by upstream mammalian target of rapamycin (mTOR). (2) The phosphatidylinositol 3-kinase (PI3K)-Atg6/Beclin 1 (yeast Atg6 is homologous to mammalian Beclin 1) complex, which can promote the formation of autophagosome. (3) Atg5-Atg12-Atg16 conjugation. (4) Atg8/microtubule-associated protein 1 light chain 3 (LC3, yeast Atg8 is homologous to mammalian LC3) conjugation system, which is linked to the lipid phosphatidylethanolamine in an Atg12 system-dependent manner, plays a vital role in the closure, trafficking or fusion. (5) Atg9/mAtg9, which promote the separation of mature autophagosome and autophagy-related proteins, and transporting and reutilizing transmembrane proteins. (6) Atg2-Atg18 complex, which cooperates with Atg9 by binding to phosphatidylinositol 3-phosphate, is very important for membrane elongation [12, 28]. Extensive investigations have evaluated autophagic activity or explored the role of autophagy through dynamically monitoring autophagic flux, autophagosome marker LC3, or depleting of some core Atg genes [12, 29].

Previous studies have summarized autophagy mainly performs two functions, generation new products through degradation, and clearance of intracellular defective macromolecules and organelles [30]. The former is to produce nutrients to promote cell survival in case of nutrient deficiency or increased demand for cell growth [26]. The second function of autophagy is often achieved at the basal level or under induction conditions (damaged organelles or invasive bacteria). Which is of great significance to control cell quality and benefit for long-lived cells and organisms [26, 31].

Despite a tremendous increase in knowledge about the molecular mechanisms and functions of autophagy, there are still many problems that need to be addressed urgently. Firstly, the mechanisms of autophagy may be not as simple as previously thought. Because autophagy dysfunction (or core Atg genes knockout) mice manifest various abnormalities in different organs and in different ways, but the underlying mechanisms are not elucidated [12, 32]. Secondly, extensive investigations have suggested that the number of autophagosomes is insufficient to assess autophagic activity. Because an increase in the number of autophagosomes may manifest as an enhancement of autophagy, or inhibition of autophagy consumption or lysosomal degradation. So the monitoring of ‘autophagic flux’ is recommended by a large number of studies [12, 26, 33]. Nevertheless, though satisfactory autophagy detection results have been achieved in cultured cells by monitoring ‘autophagic flux’, it remains difficult to practice in animals and humans. Thus, it is significant to identify other typical autophagy biomarkers. Thirdly, the knowledge about autophagic regulation, especially under the complicated influence of multiple stimulatory and inhibitory signals, is poorly understood [15]. Finally, in order to manipulate autophagy for human health, it is essential to have a more comprehensive understanding of autophagy, including its types, their different significance and roles in physiology, disease and medicine [34, 35, 36].

CCECs play an essential role in smooth muscle contraction as well as vascular barrier structure. CCECs injury is a major cause of ED, which is particularly common in diabetes mellitus-induced ED (DMED) [37]. With sexual arousal, neuronal and endothelial-derived nitric oxide synthase (nNOS and eNOS) rapidly convert L-arginine and oxygen to nitric oxide (NO). Then, NO enters into corpus cavernosum smooth muscle cells (CCSMCs), promoting soluble guanylate cyclase and guanosine triphosphate to produce cyclic guanosine monophosphate (cGMP). Further cGMP activates proteinase kinase G to trigger the relaxation of CCSMCs by decreasing intracellular calcium concentration [38]. So the NO/cGMP pathway is accepted as the major relaxation pathway of penile erection, whose abnormality is closely related to various EDs [39].

Many risk factors including changes of blood component, viscosity and shear stress, et al, are all competent to impair CCECs and lead to endothelial dysfunction. Meanwhile, autophagy has been found to play an important role in regulating the homeostasis of CCECs and deciding their survival and death. Numerous studies have revealed that the epigallocatechin gallate and analogs of resveratrol (pterostilbene and dimethoxystilbene) can increase catabolism, and clear the excess lipid substrates in endothelial cells by inducing autophagy [40, 41]. Rapamycin may activate autophagy to promote the survival of endothelial cell under hypoxia and starvation [42]. Moreover, it has been reported that increased autophagy can protect endothelium from cytotoxicity caused by cardiovascular risk factors, such as advanced glycation end-products, reactive oxygen species, oxidized low-density lipoprotein, et al [43, 44].

Shear stress, which is produced from the sliding of blood on the surface of endothelium, is extensively investigated in cardiovascular disease and endothelial autophagy. However, the results of studies are controversial. Yang et al [45] showed that the expression of autophagy-related molecules LC3-II and Beclin-1 decreased under low shear stress (5 dyn/cm²) in comparison to 15 dyn/cm². And the decreased autophagy was presumed to associate with the downregulated eNOS and upregulated endothelin-1. But Ding et al [46] addressed that low shear stress (3 dyn/cm²) promoted endothelial autophagy and attenuated the cytotoxicity of lipopolysaccharide, and autophagy would gradually decrease with the increasing magnitude of shear stress. Despite the role of magnitude of shear stress in autophagy is not clear. As it has been summarized by Jiang [42], in most cases, induced autophagy has cytoprotective effects. Combining with these reports and a previous study of us [6], we considered one of the mechanisms by which vacuum therapy increases autophagy in bilateral cavernous nerve crush (BCNC, a classic animal model simulates ED after radical prostatectomy) rat penis may be related to the changes of shear stress. Because it has well known that loss of spontaneous erection and continuously insufficient blood supply to the penis caused by cavernous nerve injury is the main pathogen of ED after radical prostatectomy [47]. In other words, the shear stress of CCECs may be decreased after cavernous nerve injury. Meanwhile, applying a vacuum device to distend the corporal sinusoids and increase the blood inflow to the penis, which would certainly increase the shear stress on CCECs. Our study found that the activated autophagy is beneficial to regulate the homeostasis of CCECs by improving mitochondrial function, which manifested as the increased eNOS and erectile ability. More interestingly, a similar conclusion was later drawn in the study of Ye et al [48]. Which presented that rapamycin as well as salidroside could improve ED in BCNC rats by promoting protective autophagy to attenuate the loss of nerve contents, CCECs and CCSMCs [48]. Meanwhile, Zhang et al [49] found that impaired autophagy is associated with the cavernosal endothelial dysfunction and DMED. Human urine-derived stem cells (USCs), which are isolated from the urine of healthy adult male, have the same proliferation and differentiation properties as other stem cells. Intracavernous injection of human USCs could ameliorate DMED and endothelial function by increasing autophagic activity in the CCECs. In summary, an appropriate increase of autophagy in CCECs with medicines, vacuum erectile device, or human USCs, et al has a good prospect for the treatment of ED.

The contraction and relaxation of CCSMCs is the key to deciding the amount of penile blood flow and the degree of penile erection [50]. In recent years, autophagy and apoptosis of CCSMCs have been proven to play an important role in various forms of ED, including aging-related ED, diabetes mellitus-related ED, cavernous nerve injury-induced ED, obstructive sleep apneaassociated ED, and androgen deprivation-induced ED.

In fact, the crosstalk of autophagy and apoptosis has been widely described in physiological and pathological conditions [51, 52]. Briefly, proper autophagy is to remove unwanted cellular components, thereby maintaining intracellular homeostasis and cell survival. But once the intensity or duration of stress exceeds the limit of the cell, cell apoptosis instead of autophagy would be activated. Furthermore, the activated caspase will digest some essential autophagy-related proteins such as Atg3 and Beclin 1. Conversely, in some cases, excessive autophagy can lead to cell apoptosis or necrosis. But the boundary of autophagy escalating to apoptosis remains unclear nowadays. Therefore, autophagy and apoptosis are often evaluated simultaneously in many studies.

Extensive investigations have shown that CCSMCs apoptosis plays a vital role in organic ED [53, 54]. But the measurement of autophagy in ED is inconsistent, as shown in Table 1. Most studies have shown that autophagy is increased in penis or CCSMCs of diabetic rats because of abnormal glucose and energy metabolism. A further increase in autophagy can make full use of intracellular substances to maintain energy balance and stability of CCSMCs. For example, applying rapamycin, simvastatin, liraglutide, nicotinamide adenine dinucleotide phosphate (NADPH) oxidase 1/4 inhibitor, and mesenchymal stem cell with or without low-energy shock waves, et al to enhance autophagy can improve ED in diabetes rats by inhibiting CCSMCs apoptosis [4, 55, 56, 57, 58]. But the opposite results were subsequently presented in the studies of Zhang et al [59, 60] and Zhang et al [61]. They showed that inhibiting autophagy with icariside II or/and metformin can ameliorate ED in diabetes rats. In their opinion, autophagic upregulation is bad for the proliferation and survival of CCSMCs. Promoting the expression of B-cell lymphoma-2 (Bcl 2, a well-known suppressor of apoptosis) and interacting with Beclin 1 (an essential regulator of autophagy), can significantly prevent CCSMCs from autophagic damage. In addition to diabetic rats, autophagy in BCNC rats is also controversial. Our study found that Beclin 1 and LC3B were decreased in BCNC rats, and autophagosomes were reduced in CCSMCs. Increasing autophagy by vacuum therapy helped to improve ED of BCNC rats [6]. Therefore, we proposed a hypothesis that continuous pathology stimuli resulted from cavernous nerve injury may break up the balance between autophagy and apoptosis. In this situation, apoptosis prevails over autophagy, and excessive apoptosis may impair autophagy by degrading Beclin 1 and LC3 proteins through activated caspases. But with the improvement of penile microenvironment after treatment, autophagy may gradually dominate over apoptosis. Interestingly, the same conclusions of CCSMCs apoptosis in BCNC rats were drawn in both Ye et al [48] and our study. But a slight difference in autophagy was presented by Ye et al [48] who showed that autophagy was increased in BCNC rats, further boosting autophagy with salidroside could improve ED. In addition, it has been proven that autophagy was decreased in ED caused by aging, castration, and hyperlipidemia. Upregulating autophagy in penis or CCSMCs through (tankyrase 1 and human tissue kallikrein 1) gene therapy and testosterone supplementation may protect erectile function [5, 7, 62]. In conclusion, the specific roles of autophagy in CCSMCs apoptosis and ED have not been fully understood to date. And more studies are needed to explore the boundary of autophagy escalating to apoptosis, thereby playing a greater protective role of autophagy in the treatment of ED.

Table 1
Changes of apoptosis and autophagy before and after the treatment of ED

• Click for larger image
• Click for full table
• Download as Excel file

In recent years, in addition to apoptosis, autophagy has been reported to associate with the phenotype modulation of CCSMCs. Phenotype modulation is a new term and widely studied in vascular smooth muscle cells (VSMCs). It is defined as an increase in cell proliferation, migration, and extracellular matrix, and a decrease in smooth muscle contractile molecules [63]. Currently, the phenotype modulation of CCSMCs has been investigated in several studies [64]. They showed CCSMCs under hypoxia for 48 hours may significantly increase apoptosis, fibrosis and phenotype modulation in vitro, which was speculated to be an important mechanism of hypoxia-induced ED. Indeed, Yang et al [65] and Qiang et al [66] have successively proven that the phenotype modulation of CCSMCs is related to ED in BCNC rats. It is manifested as a decrease in contractile phenotype marker including smooth muscle myosin heavy chain, α-smooth muscle actin (α-SMA), desmin, and calpoponin-1, and a significant increase in synthetic phenotype marker including osteopontin and vimentin. But despite the significance of autophagy and phenotype modulation is gradually understood, their complex interactions in ED remains undefined. Increasing lines of evidence suggests that hypoxia may regulate multiple microenvironmental factors to regulate autophagy and CCSMCs phenotype modulation, such as hypoxia inducible factor-1 alpha (HIF-1α), platelet derived growth factor (PDGF), transforming growth factor-beta 1 (TGF-β1) and endothelin-1 [67, 68]. Specially, on the one hand, hypoxia may promote the accumulation of HIF-1α, and which can activate adenovirus E1B 19 kDa protein-interacting protein 3 (BNIP3) to compete Bcl-2, thereby disrupting intracellular Beclin 1-Bcl 2 complex and inducing autophagy due to the release of Beclin1 [64]. On the other hand, hypoxia may upregulate TGF-β1, which is an important differentiation factor. TGF-β1 may maintain the contractile phenotype of CCSMCs and overexpress Bim to combine with Beclin1 and LC3, thereby suppressing autophagy [69, 70]. Besides, previous studies have found that induced autophagy by PDGF-BB plays an essential role in attaining the synthetic phenotype and proliferation of VSMCs under hyperoxidative stress caused by vascular lesions [71]. And the similar effect of PDGF-BB in CCSMCs has been subsequently confirmed by Luo et al [67]. Meanwhile, vasoactive factors like endothelin-1 and Angiotensin II have also been indicated to regulate the phenotype modulation of CCSMCs under hypoxia by mediating autophagy [64].

Numerous studies have revealed that fibrosis is a significant pathological mechanism of ED. Penile fibrosis exists in ED caused by a variety of etiologies, such as aging, hypertension, diabetes mellitus, castration, radical prostatectomy, chronic smoking and alcoholism, et al [72]. Under the influence of these etiologies, the overexpression of fibrogenic factors such as TGF-β, cytokines and vasoactive peptides can induce fibroblast-to-myofibroblast differentiation or phenotype modulation of CCSMCs. Which will directly lead to the loss of CCSMCs and excessive deposition or disorganization of extracellular matrix in the corpus cavernosum, and ultimately result in penile fibrosis and ED [73]. However, though numerous studies on fibrosis in ED have been reported, the specific molecular mechanism is not fully understood until now.

Fibrotic factors including growth factors (TGF-β, connective tissue growth factor, and PDGF), vasoactive peptides and reactive oxygen species, et al, have been widely investigated in penile fibrosis, but little evidence about antifibrotic factors has been reported. Intriguingly, in recent years, autophagy as a cellular pathway devotes to degrade and recycle cellular components, has been shown to play a significant role in preventing penile fibrosis. The related studies are listed in Table 2. According to the study of Li et al [3], the increased apoptosis and decreased autophagy of corpora cavernosum cells are import causes of penile fibrosis, which impaired the penile structure and function in hyperlipidemia rats. Consistent with this study, Wang et al [5] demonstrated that promoting autophagy with testosterone supplementation could inhibit penile fibrosis and CCSMCs apoptosis in castration rats. And the underlying mechanism has been speculated that androgen affects the counter-regulation of autophagy and apoptosis by regulating the interaction of Beclin1–Bcl-2.

Table 2
Autophagy in penile fibrosis after treatment of ED

• Click for larger image
• Click for full table
• Download as Excel file

However, though current studies have indicated that autophagy plays an important role in penile fibrosis, few literatures have shown the role of autophagy in detail. According to the report of Lv et al [74], autophagy activity of fibroblasts generally maintains at a high level in most tissues. Once autophagy dysfunction caused by knockdown of LC3B or Beclin1 or ATG5 can promote the fibroblasts differentiation into myofibroblasts, and release a large amount of collagen. Interestingly, during the recovery of fibrosis, macromolecular collagen is first decomposed into collagen monomers or small fragments by proteases like extracellular matrix metalloproteinases. Then macrophages and fibroblasts phagocytize and degrade these products through autophagy. Because autophagy can non selectively degrade long-lived proteins, organelles, and provide cellular nutrition and energy. But if the dysregulation of autophagy fails to clear these abnormal macromolecular, their deposition can increase intracellular reactive oxygen species, endoplasmic reticulum stress, apoptosis, and tissue inflammatory, thereby exacerbating fibrosis [74]. This hypothesis was similarly verified by Lin et al [4], who revealed that rapamycin could ameliorate DMED in rats by increasing autophagy, and inhibiting endothelial dysfunction, apoptosis and fibrosis. This study showed that rapamycin could inhibit corpora fibrosis in diabetic rats through reducing TGF-β1, α-SMA, interstitial fibroblast collagen III, and basement membrane collagen IV. Followed by Ding et al [55] who further indicated that autophagy plays an important role in anti-fibrosis in DMED rats. Their study revealed that autophagy-related pathway, 5′-AMP activated protein kinase (AMPK)-dependent forkhead box O3 (FoxO3a) pathway is dysfunctional in DMED. AMPK agonist (simvastatin) could improve DMED in rats by upregulating FoxO3a phosphorylation, and inhibiting the transcription of S-phase kinase-associated protein 2 (SKP2), thereby increasing the expression of coactivator-associated arginine methyltransferase 1 (CARM1) and autophagy [75]. Furthermore, the increased autophagy was assumed to alleviate corpora fibrosis, as showed that after treatment, the contents of α-SMA, collagen I and collagen III in the cavernosum of diabetic rats were similar to those of normal rats.

Extracellular vesicles including exosomes, are important components of paracrine and play important roles in various pathophysiological processes. Currently, stem cells including bone marrow-derived mesenchymal stem cells (BMSCs), adipose-derived stem cells (ADSCs) and USCs, these have all been confirmed to benefit for the improvement of ED [76, 77], but the exact mechanism has not been fully elucidated. With the rapid development of molecular biology techniques, plenty of microRNAs and proteins are identified in stem cells-derived extracellular vesicles, which make more researchers believe that it is paracrine rather than transdifferentiation of stem cells that improves ED. For example, Zhu et al [78] revealed that ADSC-derived exosomes could restore DMED by inducing endothelial cell proliferation and decreasing penile fibrosis, and it was explained that the ADSC-derived exosomes contains some proangiogenic microRNAs (miR-126, miR-130a, and miR-132) and antifibrotic microRNAs (miR-let7b and miR-let7c). Furthermore, Song et al [79] revealed that compared to BMSCs and ADSCs, CCSMC-derived exosomes have greater potential to ameliorate DMED by inhibiting fibrosis and regulating the NO/cGMP pathway. Harrell et al [80] revealed that extracellular vesicles may activate and/or inhibit apoptosis, necrosis, and oxidative stress in multiple cells through transporting mRNAs and miRNAs. However, although significant effects of exosomes have shown in treating ED, the roles of exosomes are actually complicated and multidimensional. It has been proven by Gao et al [10] that overactivation of mTOR by PDGF may lead to inhibition of autophagy and activation of ras homolog family member (Rho)-associated protein kinase 1 (ROCK1), which can mediate hepatic stellate cells releasing fibrogenic extracellular vesicles to promote liver fibrosis. Similarly, Gan et al [81] has also reported that Ang II-induced vascular remodel depends on the biogenesis, secretion, and internalization of senescence associated small extracellular vesicles, which leads to mitophagy dysfunction.

Nerve injury is another important cause in organic ED. Nerve injury is common in spinal cord injury, cavernous nerve injury caused by radical prostatectomy or other diseases, like diabetes. Nerve injury usually leads to a series of pathological changes in corpora cavernosum, including downregulating nNOS expression and reducing cavernosal oxygen, upregulating TGF-β1 and collagen expression, activating NADPH oxidase and RhoA/ROCK contractile pathway, and triggering inflammatory damage, et al. Recent years, autophagy has been suggested playing a substantial neuroprotection role by eliminating abnormal aggregation of proteins and toxic substances in nerve cells [82].

A previous study reported that rapamycin, as an immunophilin ligands, after binding to its receptors such as FK506 binding proteins (FKBPs), FKBP 38 and FKBP 65, could improve ED in BCNC rats through its neuroprotection and nerve-regenerative function [83]. Consistent with this study, Pan et al [84] demonstrated that the neuroprotective effect of rapamycin is partially related to the increased autophagy. Because autophagy contributes to degrade misfolded and aggregated proteins in neurons, which can significantly reduce the neuronal loss under pathological stimulation. Similarly, as Lin et al [4], who showed that nerve injury caused by diabetes would reduce the nNOS expression, rapamycin could significantly reverse its reduction and improve DMED through the neuroprotection effect of autophagy. Moreover, apart from rapamycin, other methods like salidroside and vacuum erectile device, have also been reported to ameliorate ED by mediating autophagy to improve nerve injury [6, 48]. Similar evidence was subsequently presented in the study of Lin et al [85] that activated autophagy after zinc treatment could protect nerve cells in spinal cord injury by inhibiting pyrin domain containing protein 3 (NLRP3) expression and ubiquitination. Meanwhile, Gao et al [86] suggested that melatonin could protect nerve cells in spinal cord injury by mediating Sirtuin 1/AMPK signaling pathway to promote autophagy. Nowadays, a large number of studies have revealed the role of autophagy in neuroprotection, but in addition to the degradation function of autophagy, more details about autophagy or core Atg genes in neuroprotection remain poorly understood.

Growing studies suggest autophagy plays a significant role in ED. Autophagy deficiency or overactivation can lead to dysfunction of CCECs and CCSMCs, and disorder of the cavernous microenvironment. Autophagic activity varies in ED with different causes. For example, autophagy is increased in diabetic rats, but decreased in aging rats and castrated rats. Enhancing autophagy in these rats can significantly improve erectile function. Nevertheless, the exact mechanism of autophagy in ED has not been completely elucidated until now. Based on previous literature, we try to clarify the possible effect of autophagy in endothelial dysfunction, CCSMCs apoptosis, penile fibrosis and cavernous nerve injury, which are the main causes of ED. And according to previous literature, we learn that other than blood components, shear stress on the endothelial cells may also play an important role in regulating autophagy, proper adjusting shear stress may improve ED. Broken balance of autophagy and apoptosis in CCSMCs is involved in various EDs, increasing autophagy and inhibiting apoptosis may improve most EDs. Inducing autophagy may degrade collagen macromolecules to improve penile fibrosis. Autophagy may also perform neuroprotection in the recovery of ED.

Previous studies have enriched our standing of autophagy in ED to some extent, but it is believed that the role of autophagy in ED is not as simple as currently reported. Moreover, so far, our understanding of autophagy regulation is still very limited. Most studies have discussed the classic autophagy regulation pathway, PI3K protein kinase B (Akt)-mTOR or mTOR in ED, but autophagy is generally regulated by multiple signals. And besides, it has been suggested that autophagy varies in different organs and in different ways [12]. This may be similar to its roles in cavernosal cells. Interestingly, a recent study provides a very detailed introduction to human corpus cavernosum cells. In this study, seven major types of cavernous cells and all subclusters of fibroblast, smooth muscle cell, and endothelial cell were identified. More preciously, the heterogeneity of each cell subgroup, including the spatial location, biological function, and regulatory pathway, has also been identified [87]. Furthermore, this study reveals that different cell subclusters play different roles in psychological and pathological process. Therefore, the study of autophagy in ED may need to classify not only the causes of ED, but also the subclusters of corpus cavernousm cells.