Etiology-oriented rodent models of erectile dysfunction: a narrative review of categorization, modeling strategies, and translational applications

Introduction

Erectile dysfunction (ED), as one of the male sexual dysfunction disorders, impacts not only the quality of life and physical health of men but also causes significant psychological burden to their partners. ED is defined as the persistent or recurrent inability of a male to achieve and maintain an erection adequate for satisfactory sexual intercourse (1). According to statistics, approximately 15% of men worldwide suffer from ED each year. It is predicted that by 2025, ED will affect around 320 million people (2), and its prevalence is increasing due to aging, but there is also a trend of younger men being affected (3). Other risk factors for ED include cardiovascular disease, diabetes, hypertension, and a history of pelvic surgery (4). Although PDE5 inhibitors have offered some hope to ED patients, they simply do not work for roughly 35% of patients (5,6). Moreover, for ED caused by psychological factors such as depression, stress, or anxiety, pharmacotherapy is ineffective, and psychotherapy is necessary to help treat the condition. Much of our current understanding of the physiological and pathological processes underlying sexual activity comes from animal studies. Since Sprague-Dawley rats were first used by Quinlan et al. to electrically stimulate erection of the corpora cavernosa (7), rodents have become more popular. However, there is no comprehensive, etiology-based classification of such models that corresponds to the clinical picture. In particular, rodent models of ED are usually traditionally classified mainly based on the method of induction used (surgical, pharmacological, etc.), as a “how” rather than a “what” type of classification.

Unlike traditional classifications that group models by induction technique (surgical, pharmacological, or chemical), our etiology-oriented framework groups them by underlying pathophysiological drivers—neurogenic injury, metabolic/vascular dysfunction, hormonal deprivation, and psychological stress. This alignment enables researchers to select models based on specific clinical ED subtypes, such as post-prostatectomy ED (neurogenic) or PDE5i-resistant diabetic ED (metabolic/vascular), offering direct translational relevance absent from prior technique‑focused reviews. We present this article in accordance with the Narrative Review reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0460/rc).

Methods

This narrative review was conducted following the established methodological guidelines for narrative reviews, with a transparent and reproducible search strategy to minimize selection bias (Tables 1,2). Two reviewers independently screened titles/abstracts and full texts, with disagreements resolved by a third reviewer.

Core pathophysiological pathways in ED: a framework for etiology-based modeling

Penile erection is fundamentally a neurovascular event governed by the tone of the corpus cavernosum smooth muscle. This tone is regulated by a balance within the autonomic nervous system: parasympathetic dominance initiates erection, while sympathetic activity maintains the flaccid state (Figure 1) (8). ED results from a disruption of this equilibrium, caused by specific etiologies that impair pro-erectile signaling or enhance contra-erectile tone.

Figure 1 The neuromodulation process of erection. This figure was created using BioGDP (www.biogdp.com). NANC, non-adrenergic noncholinergic; NO, nitric oxide.

The pathophysiology across diverse ED models converges on the dysregulation of three principal signaling cascades (Figures 2,3):

• NO-cGMP-PKG pathway (primary relaxation): nitric oxide (NO) is the main neurotransmitter of erection (9). NO is synthesized by NO synthases, namely neuronal (nNOS) and endothelial (eNOS). It activates the sGC-cGMP-PKG pathway to induce smooth muscle relaxation. This pathway is a common endpoint of various injuries: it is decreased in neurogenic models [cavernous nerve (CN) injury] due to loss of nNOS (10); it is decreased in metabolic/vascular models (diabetes, hypertension) due to endothelial dysfunction, eNOS uncoupling, and oxidative stress (11); and it is decreased in smoker models due to oxidative stress (12). Oxidative stress is a unifying driver across etiologies, contributing to eNOS uncoupling and reduced NO bioavailability in the corpora cavernosa (13).
• cAMP-PKA pathway (synergistic relaxation): this pathway is activated by agents such as prostaglandin E1 and works in parallel with the NO/cGMP pathway via cAMP and PKA to induce relaxation (14). Its impairment contributes to ED in conditions such as diabetes and aging.
• RhoA/ROCK pathway (primary contraction): this is a key contra-erectile pathway that enhances Ca²⁺ sensitization to maintain smooth muscle contraction. It is upregulated in models of hypertension, diabetes, and hypogonadism (15).

Figure 2 Molecular mechanisms of smooth muscle contraction. The RhoA/ROCK pathway is the primary contractile signaling cascade in cavernosal smooth muscle. Activation of RhoA by various stimuli (e.g., endothelin-1, angiotensin II, norepinephrine) leads to ROCK-mediated phosphorylation of MYPT1 (myosin phosphatase target subunit 1), which inhibits MLCP (myosin light chain phosphatase). This increases MLC phosphorylation and enhances Ca²⁺ sensitization, promoting actin-myosin cross-bridge formation and sustained contraction. This pathway is upregulated in hypertension, diabetes, and hypogonadism, contributing to the contractile imbalance in these conditions. This figure was created using BioGDP (www.biogdp.com). MLC, myosin light chain.

Figure 3 Molecule mechanism for smooth muscle relaxation. Relaxation is mediated by two synergistic pathways: (I) the NO-cGMP-PKG pathway, initiated by nNOS and eNOS nitric oxide synthase-derived NO, which activates sGC to convert GTP to cGMP; cGMP activates PKG, which reduces intracellular Ca²⁺ and decreases MLC phosphorylation; and (II) the cAMP-PKA pathway, activated by prostaglandin E1 and other agents, which stimulates AC to produce cAMP; cAMP activates PKA, which similarly promotes relaxation. These pathways are differentially impaired across etiologies: the NO-cGMP pathway is predominantly affected in neurogenic and metabolic models, while the cAMP-PKA pathway shows relative preservation in some conditions. This figure was created using BioGDP (www.biogdp.com). AC, adenylyl cyclase; eNOS, endothelial nitric oxide synthases; nNOS, neuronal nitric oxide synthases; PKA, protein kinase A; PKG, protein kinase G; sGC, soluble guanylate cyclase.

Therefore, etiology-based modeling aims to replicate the specific upstream insult that causes a characteristic imbalance among these core pathways. For a model to be valid, it must exhibit molecular alterations in this cascade (e.g., changes in NOS expression, cGMP levels, or ROCK activity) as evidence of a link to human disease.

Etiology‑based rodent models of ED: from clinical subtypes to experimental paradigms

This section reviews major rodent ED models by their underlying etiologies—neurogenic, metabolic/vascular, hormonal, and psychogenic—rather than by induction technique alone. Figure 4 provides a visual overview of how each model category maps onto its corresponding clinical ED subtype. The following subsections detail the modeling strategies, pathophysiological features, and assessment methods for each category.

Figure 4 Etiology-oriented rodent ED models mapped to clinical ED subtypes. CMS, chronic mild stress; CN, cavernous nerve; CUMS, chronic unpredictable mild stress; DM, diabetes mellitus; eNOS, endothelial nitric oxide synthases; ED, erectile dysfunction; GnRH, gonadotropin-releasing hormone; HFD, high-fat diet; HLP, hyperlipidemia; HTN, hypertension; SHR, spontaneously hypertensive rat; SPS, social psychological stress; STZ, streptozotocin; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; VEGF, vascular endothelial growth factor.

Injury models

CN injury models

CN injury models replicate postoperative ED using surgical procedures—compression, transection, resection, or cryoinjury (Table 1). The bilateral CN crush model, mimicking ED after non-nerve-sparing radical prostatectomy, is established by clamping the exposed nerves with a hemostat for 2 minutes (16). This model shows reduced intracavernous pressure (ICP)/mean arterial pressure (MAP) ratios, decreased nNOS expression, and axonal degeneration (17,18), but outcomes are operator-dependent (19).

Transection (simple division) and resection (5-mm excision) cause permanent denervation (20). Resection abolishes regenerative potential, while transection retains some recovery capacity (21). Functional improvement after resection may occur via compensatory innervation from accessory penile nerves (22).

The cryoinjury model, created by freezing the CN at −80 °C for 1 minute, produces profound erectile impairment (23). nNOS-positive fibers decline for over one month, with partial functional recovery by 3 months—possibly due to neurotrophic factor expression (24,25). However, cryoinjury is inconsistent and confounded by tissue damage. The characteristics, limitations, and applicable scenarios of the four cavernous nerve injury models described above are summarized in Table 3.

Radiation induced ED models

Prostate cancer is the most common cancer in men in 112 countries and accounts for approximately 15% of all cancers (28). It has been predicted that the numbers of new cases per year will rise from an estimated 1.4 million for the year 2020 to 2.9 million in 2040 (29) and radiation therapy remains first-line treatment for these patients (30). However, the reported incidence varies widely, ranging from 17% to 90% (31). Although more patients survive prostate cancer, their quality of life is often greatly diminished. Thus it’s meaningful to investigate the mechanisms of RIED via rodent models and then discover treatments. Previously there have been reports on inducing RIED using the method of wide-field irradiation. However, wide-field irradiation has substantial drawbacks, including damage to adjacent organs and excessive off-target interference. Nowadays the current standard is image guided radiation therapy. In this technique, X-ray or computed tomography (CT) imaging is done to find out the prostate gland correctly, aluminum shields or other protective materials are used to protect surrounding tissues from injury. Then researchers administer a certain amount of radiation to only the prostate with special tools. According to one study, after exposure of male Sprague-Dawley rats under 220 kV X-rays and 25 Gy radiation on the prostate locally, there were fewer erections at weeks 4, 9, and 14 post-radiation therapy. Additionally, ICP, MAP, and area under the curve (AUC) were much lower than those in the sham group (32). In another study, a radiation dose of 20 Gy failed to establish a stable RIED model in rats, whereas a dose of 25 Gy successfully induced RIED (33).

The effects of radiotherapy on erectile function primarily involve vascular damage and nerve injury, with the latter appearing to be more critical in the development of RIED. We used transmission electron microscopy (TEM) to analyze the CN in RIED rats. Myelin loss, microvessel damage, and progressive axonal atrophy of CN fibers were observed (34). Radiotherapy can also cause damage to the major pelvic ganglion (MPG) and lead to neuronal apoptosis (35). On the other hand, Schwann cells have been shown to protect against MPG apoptosis (36).

The pathophysiology of RIED involves progressive damage to three tissue compartments. First, radiation directly damages endothelial cells through DNA double-strand breaks and oxidative stress, leading to eNOS downregulation, increased ROS production, and sustained endothelial dysfunction (31). Second, unlike acute surgical injury, radiation-induced neural damage is progressive and delayed, characterized by axonal atrophy, myelin loss, and MPG neuronal apoptosis. Schwann cells have been shown to protect against MPG apoptosis, suggesting that glial dysfunction may contribute to the delayed onset of RIED. Third, radiation activates TGF-β1 signaling, promoting fibroblast-to-myofibroblast transdifferentiation and excessive extracellular matrix deposition, which compromises sinusoidal compliance and veno-occlusive function. This fibrotic response typically peaks at 3–6 months post-irradiation, paralleling the delayed clinical presentation of RIED. The mixed neurovascular and fibrotic pathology explains the poor response to PDE5 inhibitors in this population, underscoring the need for combination therapies targeting neuroprotection, endothelial repair, and anti-fibrotic pathways.

Metabolic and vascular ED models

Diabetic models

Many studies have shown that diabetes mellitus decreases NO expression, impairs cavernous smooth muscle relaxation, and induces ED through endothelial dysfunction, nitrergic nerve damage, increased advanced glycation end products (AGEs), and oxidative stress. In the development of ED associated with type 1 diabetes, non-adrenergic non-cholinergic (NANC) neurotransmission may play a role as a result of diabetic peripheral neuropathy, leading to neuronal degeneration and nNOS downregulation (37). Currently, most studies on diabetic ED use the streptozotocin (STZ)-induced type 1 diabetes animal model, which is generated by intraperitoneal injection of STZ to destroy pancreatic islet β cells. For example, Bivalacqua et al. induced diabetes in rats using STZ and found downregulation of eNOS protein, unchanged NOS activity, decreased cGMP levels in the corpora cavernosa, and a marked reduction in erectile response, indicating impaired erectile function (38).

Given that T2DM accounts for the majority of clinical diabetic ED cases, a more comprehensive discussion of T2DM models is warranted. These models fall into three major categories.

Diet‑induced models: high-fat diet (HFD) feeding in C57BL/6 mice induces insulin resistance, obesity, and metabolic syndrome. However, HFD alone often produces variable hyperglycaemia, and the erectile phenotype may require extended feeding periods to develop fully.

Genetic models: the db/db mouse (leptin receptordeficient) and ob/ob mouse (leptin-deficient) are the most widely used spontaneous T2DM models. Both develop severe obesity, hyperglycaemia, and hyperinsulinaemia by 6–8 weeks of age. In db/db mice, impaired relaxation to acetylcholine has been attributed to reduced KCa2.3, an endothelial calcium-activated potassium channel, leading to endothelial dysfunction in T2DM-associated ED (39). The Otsuka Long-Evans Tokushima Fatty (OLETF) rat is another well-characterized spontaneous T2DM model that develops diabetes around 20–40 weeks of age. Studies have demonstrated significantly reduced eNOS immunofluorescence intensity and protein levels in OLETF rats compared with controls, contributing to endothelial dysfunction in T2DM (40). The Goto-Kakizaki (GK) rat, a non-obese T2DM model, develops mild hyperglycaemia with impaired glucose tolerance and has been used to study ED in the context of non-obese T2DM.

Combined models: a widely used approach combines HFD feeding with a low-dose STZ injection (e.g., 30–35 mg/kg) to induce insulin resistance followed by partial β-cell depletion, creating a T2DM phenotype with stable hyperglycaemia (41). This model offers the advantages of controlled onset timing and relatively high reproducibility, making it suitable for interventional studies.In addition, there are four main pathways by which diabetic hyperglycemia can damage endothelial cells and cause dysfunction: the AGEs pathway, the polyol pathway, the hexosamine pathway, and the PKC pathway. When the PKC pathway is overactivated, it causes increased smooth muscle contraction and decreased endothelium-dependent smooth muscle relaxation. This may be a cause of ED (42). Overall, diabetes-associated ED reflects intertwined endothelial, neurogenic, and metabolic dysfunction, consistent with contemporary clinical and mechanistic syntheses (43).

Hyperlipidemia/hypercholesterolemia models

Hyperlipidemia or hypercholesterolemia itself is not very harmful to the corpora cavernosa. And in the long term, as we know, it leads to atherosclerosis, causing arteries to become blocked and endothelial dysfunction (44). This process has severe consequences for erectile function. This type of animal model is usually created just like type 2 diabetes models; they are mainly caused by a long-term diet with lots of cholesterol. Take a rabbit given a cholesterol-rich 0.5% diet for 16 weeks as an example: there is some plaque deposition within the iliac artery but no great damage done to their erections (45). Then they were fed a diet with about 0.3% cholesterol for the next 80 weeks. This led to some additional atherosclerotic progression in the iliac arteries and remarkable reductions of cavernous smooth muscle relaxation (45).

However, rabbits have long breeding cycles and are relatively resistant to atherosclerosis. So some studies use apolipoprotein E-deficient (ApoE^-/-) mice that were bred in a certain way. Such mice are the best model for atherosclerosis. If fed a normal diet or high-fat food, their blood cholesterol will rise quickly, and they will develop atherosclerosis soon. In some studies, ApoE^-/- mice were placed on a 1.25% cholesterol diet for 2 weeks, 4 weeks, 8 weeks and 12 weeks. The serum cholesterol of ApoE^-/- mice is much higher than that of C57BL/6 mice, and the ICP decreases dramatically (46). They had severe aortic atherosclerosis and weak response amplitudes after CN electrical penile stimulation (47). These models provide a valuable platform for investigating atherosclerosis-related ED.

Hypertension models

Hypertension is a widespread issue among people as it has many causes, and there are different kinds of models available for this disease. Another approach involves implanting an angiotensin II osmotic pump into male rats, and after giving it for twenty-eight days, the systolic blood pressure in the experimental group will stabilize at 220 mmHg: the ICP and AUC were lower in comparison with the control group (48). Furthermore, the MAP ratio went down, eNOS protein content within the penile tissue fell, resulting in impaired erectile function (49).

And then there are also the SHRs, which are genetically bred rats with spontaneous hypertension. This model involves many rats with high blood pressure and a short disease duration. The SHR models can provide some help in understanding the mechanism and treatment of hypertension with ED. And also previous works indicate that high blood pressure can result in sympathetic hyperactivation, vasospasm, vaso-remodeling which will cause imbalance of vasoregulation and diastolic dysfunction. It has been thought that oxidative stress may be involved as well (50). Experiments done with the SHR model showed that levels of BH2, NADPH oxidase, eNOS monomer/dimer ratio and NT in the cavernous tissue of SHR rats were much higher compared to those of the control. These findings suggest that eNOS uncoupling may be a key mechanism linking hypertension to ED. It was also found that icariin could inhibit the uncoupling of eNOS and improve ED of SHR rats (51). Also other works have proved that icariin promoted eNOS post-translational protein-protein (52). Other than the models above, there also exist other classical models of hypertension, like two-kidney, one-clip (2K1C) (53). However, they have very high mortality rates and are not commonly employed in ED research.

Smoking models

Smoking causes vascular damage via nicotine and CO—inducing vasoconstriction, endothelial injury, and atherosclerosis. Mechanistically, smoking reduces eNOS activity, increases superoxide via NADPH oxidase, and activates ROCK (54). Rodent models are limited: TE-2 smoke exposure (5 h/day, 5 days/week, 3 weeks) in C57BL/6 mice reduces erectile responses and increases superoxide (55); similar findings occur in rat second-hand smoke models (56).

Hormone associated ED models

Aging models

Aging-related ED involves oxidative stress, androgen decline, and neurodegeneration (57). Aged rats (24 months) show 73% lower cavernosal testosterone, 30% ICP reduction, and increased venous leakage (58). Nrf2 activators attenuate oxidative stress in aged rats (59). D-galactose-induced aging accelerates senescence, producing ICP/MAP reduction, endothelial dysfunction, and increased SA-β-galactosidase activity within 6–8 weeks (60), offering a practical alternative to natural aging.

Castration models

Bilateral orchiectomy in C57BL/6 mice reduces testosterone, upregulates oxidative stress markers (SOD2, SOD3), and impairs endothelium-dependent relaxation (61). Similar findings in rats show internal pudendal artery changes (62), reduced eNOS/nNOS expression, and CCSMC apoptosis (63). Low androgen levels impair eNOS activity via multiple pathways—mitochondrial-associated membrane proteins (64), endocan (65), TRPV4 (66), GIT1 (67), and galanin (68)—all converging on the eNOS/NO/cGMP pathway.

Non-organic (psychogenic) ED models

Non-organic ED, also known as psychogenic ED (pED), is a type of ED caused by psychological and social factors without any penile injury. Under normal physiological conditions, the paraventricular nucleus (PVN) of the hypothalamus is involved in the regulation of erectile function. The PVN contains various neurotransmitters and neurons. Dopamine binds to D2-type dopamine receptors to activate oxytocinergic neurons. Oxytocin is then released and acts on dopaminergic neurons in the ventral tegmental area (VTA), the lower part of the ventral hippocampus, the posteromedial cortical nucleus of the amygdala, and the thoracolumbar spinal cord (69). Released dopamine activates D2-type dopamine receptors in the nucleus accumbens (NAcc), which in turn releases dopamine to activate the PVN. This pathway regulates erectile function, expected reward, and completion of sexual activity (70). When any of these neural pathways is compromised, psychogenic ED results.

To date, pED studies have primarily focused on depression, and the chronic mild stress (CMS) model is most commonly used. Animals are restrained for several hours daily for approximately 2 weeks. Then, anhedonia is evaluated via the sucrose consumption test (71) to confirm a depressed state. In a study that established a rat model using CMS, sucrose intake in the model group was lower than that in the control group. Sexual behavior tests and apomorphine tests revealed ED. The development of pED was ultimately found to be related to damage to dopamine D2 receptors in the lateral amygdala and nucleus accumbens (72).

In addition, a rat model of chronic unpredictable mild stress has been established using methods such as water deprivation, food deprivation, strobe light, white noise, cage tilting, and social isolation. This model exhibits anhedonia, hypoerectile dysfunction, decreased libido, and absence of ejaculation, making it a suitable model for non-organic ED (73).

Beyond depression: anxiety paradigms and central circuitry in pED. While CMS-based models have dominated pED research, anxiety—another major psychological contributor to clinical ED—has received comparatively less attention in preclinical studies. The restraint stress (RS) model, in which rats are subjected to daily physical restraint for 2–6 hours over 2–3 weeks, has been shown to suppress sexual behavior through modulation of the hypothalamic-pituitary-testicular axis and oxidative balance. Although this model approximates certain aspects of performance anxiety, fully validated “performance anxiety” models specific to erectile function remain scarce in rodents, representing an important gap for future investigation.

At the circuit level, the PVN of the hypothalamus integrates dopaminergic inputs that activate oxytocinergic neurons projecting to the VTA, NAcc, and spinal centers. Dopamine D2 receptors in the NAcc and basolateral amygdala have been identified as critical nodes; their downregulation in CMS models correlates with ED. Emerging techniques such as fiber photometry and chemogenetics now enable real-time monitoring and manipulation of these circuits during stress exposure, offering unprecedented insight into the neural basis of pED.

Cross-etiology convergence on common molecular endpoints

Although rodent ED models are commonly categorized by etiology and induction method, clearer mechanistic value comes from emphasizing their shared functional molecular endpoints in the corpora cavernosa. Erectile function ultimately depends on the balance between relaxation and contraction governed by three core signaling axes—NO-cGMP-PKG, cAMP-PKA, and RhoA/ROCK—and different etiologies determine the specific patterns of dysregulation among these pathways.

In general, neurogenic injury models are primarily characterized by nNOS loss and reduced NO bioavailability, leading to marked impairment of the NO-cGMP-PKG axis and consequently limiting PKG-mediated cavernosal smooth muscle relaxation. Meanwhile, the smooth muscle exhibits relatively enhanced sensitivity to contractile signals following neurogenic injury, resulting in a functional bias toward RhoA/ROCK-mediated contraction. Metabolic and vascular models exhibit a classic “dual disruption” pattern: on one hand, hyperglycaemia, hyperlipidaemia, or hypertension induce endothelial dysfunction and oxidative stress, reducing NO bioavailability and blunting NO/cGMP signaling; on the other hand, vascular remodelling and an altered oxidative microenvironment enhance ROCK-mediated Ca²⁺ sensitization and contraction, shifting the balance toward reduced relaxation and increased contractile tone—providing a mechanistic explanation for the commonly observed PDE5 inhibitor resistance in these patient populations. Hormonal (aging/castration) and non-organic (psychogenic) models, although originating from distinct upstream processes—androgen decline or central dopaminergic dysregulation—ultimately converge on similar functional shifts across the three axes through their effects on autonomic output, endothelial and smooth muscle remodelling, and redox imbalance, whether by reducing eNOS expression and activity to blunt NO/cGMP signaling or by enhancing RhoA/ROCK activity to promote contractile dominance. Thus, despite diverse initiating events across etiologies, their functional endpoints exhibit substantial overlap.

In model selection, researchers should prioritize matching the predominantly impaired signaling axis rather than relying solely on induction labels. Therapeutically, when NO/cGMP deficits dominate, PDE5-related strategies are most appropriate; when RhoA/ROCK‑driven contraction prevails or NO availability is severely compromised, combination approaches incorporating ROCK modulation and endothelial redox repair are likely necessary. Linking upstream etiologies to downstream pathway signatures provides a more mechanistically grounded foundation for preclinical model selection and therapeutic development.

Systematic evaluation and application expansion of animal models

In addition to comparing model characteristics, a systematic overview of assessment methods is essential for experimental design. Table 3 summarizes the functional, behavioral, and histological assessment approaches applicable across all model categories.

Comparative summary of major models

Although various etiological models have significantly advanced ED research, each has its own characteristics and limitations. In this section, we critically compare the main models, which serves as a basis for researchers to choose the most suitable tool for answering specific scientific questions.

A critical gap in the existing literature is the lack of direct comparisons between models within the same etiological category. For instance, a researcher investigating diabetic ED is presented with at least four distinct options—STZ-induced, HFD + STZ, db/db, and OLETF—each with different pathophysiological features, time courses, and translational relevance. Without systematic comparison, model selection becomes arbitrary. To address this, we have synthesized available data to construct a comparative framework (Table 2) that highlights, for each model, its induction time, pathological hallmarks, reproducibility, and recommended applications, enabling evidence-based model selection. A comparative overview is provided in Tables 4,5.

Translational application orientation and advanced optimization strategies for models

Precision intervention in the field of neuroregeneration

CN injury models now integrate bioengineering approaches—hydrogels for localised neurotrophic factor delivery, combined with MSCs or Schwann cells (74). Gene editing (CRISPR-dCas9) enables local upregulation of pro-regenerative genes (75), and engineered ADSCs expressing RXFP-1 show therapeutic promise (76). Cell-free approaches such as stem cell–derived extracellular vesicles are emerging as promising strategies to modulate neuroinflammation and promote functional recovery in ED models (77).

Multidimensional strategies to overcome metabolic PDE5i resistance

Diabetic ED models have revealed complex mechanisms of drug resistance, prompting a shift from monotherapy toward multi-target combination strategies. For example, after it’s been established that overactivation of the RhoA/ROCK pathway is part of the reason why resistance happens (78), then putting PDE5 inhibitors together with Rho-kinase inhibitors like fasudil would be a good option for treatment (79). In addition, while using multi-omics tools, we shall find different new resistance targets in general and use them with a model.

Mechanism elucidation and alternative therapies in aging and hormonal models

Aging model research has progressed beyond antioxidant strategies to focus on targeting fundamental senescence mechanisms. Senolytics such as quercetin can decrease inflammation, delay the aging process and increase testosterone production in the body which then improves erectile function (80). Further study with age model is needed to understand the mechanism of action. In castration models, there is growing interest in androgen-independent protective mechanisms, such as paracrine factors including H₂S and other gaseous signaling molecules (81).

Neural circuit analysis in psychogenic ED models

Models for chronic stress, along with neural circuit analysis models, allow us to observe how neurons in brain regions such as the nucleus accumbens and amygdala change when the subject experiences chronic stress, using fiber photometry (82). Next, we can perform finer neural circuit manipulations using chemogenetics to prove these manipulations to be necessary causes of ED occurrence and thus identify good targets for developing better CNS modulators.

Future directions: developing complex comorbidity models and advancing dynamic physiological monitoring

A core future direction will be more composite models, such as a combination of high-fat diet, CN injury, and mild metabolic disturbance and then to create models like postoperative ED after bariatric surgery to simulate it for more in-depth applications. To achieve this, researchers have employed implantable telemetry to perform continuous, long-term, dynamic measurements in freely moving animals. This method can truly reflect pathological evolution under physiological conditions, which is very important for understanding the mechanism and for determining when to start treatment.

Discussion

Previous reviews have described individual ED models—diabetic, neurogenic, aging-related, and psychogenic—in isolation, focusing primarily on technical aspects of model induction rather than comparative analysis or clinical alignment. Our review fills this gap by: (I) proposing an etiology-oriented classification that directly aligns animal models with clinical ED subtypes; (II) providing systematic comparison of model characteristics, advantages, and limitations; (III) integrating molecular pathway analysis across etiological categories to reveal convergent pathophysiological mechanisms; and (IV) offering practical guidance for model selection based on specific research questions. This combination of features constitutes the incremental value of our framework beyond prior descriptive reviews.

While the etiology-oriented framework offers conceptual clarity, it has inherent limitations. Etiological boundaries are not always discrete—diabetic models have both neuropathic and vascular components, and aging involves hormonal, vascular, and neural changes. The framework does not capture the temporal evolution of ED, as most models represent established disease rather than early pathogenic stages. The classification is also reductionist, whereas clinical ED often involves complex interactions among multiple etiologies, and it does not account for severity gradation. We emphasize that etiology-based classification should be viewed as a conceptual tool for model selection, not a rigid categorization.

Rodent models cannot fully replicate human sexual function. Key differences include: the fibroelastic rodent penis with an os penis differs substantially from the human sinusoidal corpus cavernosum, affecting hemodynamic responses and veno-occlusive mechanisms; rodent erectile responses are primarily driven by olfactory cues and reflexive mechanisms, differing fundamentally from the visual, auditory, and cognitive stimuli that dominate human sexual arousal; anesthesia-based ICP/MAP measurement does not capture the neuropsychiatric and motivational aspects of human sexual function; and rodents lack the psychological complexity contributing to psychogenic ED in humans. These differences must be carefully considered when interpreting preclinical data. Wireless telemetry in freely moving animals and incorporation of behavioral endpoints may partially address these limitations, but inherent species differences remain a fundamental challenge.

To improve translational value, model selection must align with the scientific question. For neural studies, SD rats are suitable, while C57BL/6-nNOSCre mice enable cell-type-specific investigations. For metabolic ED, ZDF rats and db/db mice offer stable phenotypes but require careful interpretation due to systemic complications. For aging studies, natural-aged rats or SAMP8 strains provide practical options. Ethical refinement includes image-guided radiotherapy, dynamic welfare assessments, and exploring 3D organoid alternatives.

Model standardisation requires SOPs for key procedures (nerve crush, diabetic induction). Clinically relevant comorbidity models—”metabolic syndrome + mild nerve injury” or “hypertension + chronic stress”—are needed. Multi-omics “phenotypic anchoring” comparing rodent and human cavernosal tissue can identify conserved disease pathways and biomarkers (83), aiding patient stratification and target validation. These strategies will improve model selection, realism, and clinical translation.

Conclusions

This etiology-oriented framework provides a clinically aligned classification of rodent ED models, facilitating model selection based on underlying pathophysiological drivers rather than induction techniques alone. The convergence of diverse etiologies on common NO-cGMP-PKG, cAMP-PKA, and RhoA/ROCK pathways highlights shared therapeutic targets. However, current models often represent single-cause, end-stage disease, whereas clinical ED is frequently multifactorial and progressive. Future efforts should prioritize developing composite comorbidity models that combine metabolic, neurogenic, and psychological factors, and integrate multi‑omics profiling to refine patient stratification and enhance translational predictability of preclinical findings.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0460/rc

Peer Review File: Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0460/prf

Funding: This work was supported by the National Natural Science Foundation of China (No. 82260141) and the Youth Talent Program of the Technology and Healthcare Joint Program Project (No. GZ2024YLJ124).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0460/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Lewis RW, Fugl-Meyer KS, Corona G, et al. Definitions/epidemiology/risk factors for sexual dysfunction. J Sex Med 2010;7:1598-607. [Crossref] [PubMed]
2. Nicolosi A, Moreira ED Jr, Shirai M, et al. Epidemiology of erectile dysfunction in four countries: cross-national study of the prevalence and correlates of erectile dysfunction. Urology 2003;61:201-6. [Crossref] [PubMed]
3. Shamloul R, Ghanem H. Erectile dysfunction. Lancet 2013;381:153-65. [Crossref] [PubMed]
4. Levine GN. Erectile Dysfunction and Coronary Artery Disease: Unfortunate Bedfellows. J Am Coll Cardiol 2024;83:427-9. [Crossref] [PubMed]
5. Capogrosso P, Albersen M, Burnett AL, et al. Erectile Dysfunction: Update on Clinical Management. Eur Urol 2025;88:388-99. [Crossref] [PubMed]
6. Lee M, Sharifi R. Non-invasive Management Options for Erectile Dysfunction When a Phosphodiesterase Type 5 Inhibitor Fails. Drugs Aging 2018;35:175-87. [Crossref] [PubMed]
7. Quinlan DM, Nelson RJ, Partin AW, et al. The rat as a model for the study of penile erection. J Urol 1989;141:656-61. [Crossref] [PubMed]
8. Dean RC, Lue TF. Physiology of penile erection and pathophysiology of erectile dysfunction. Urol Clin North Am 2005;32:379-95. v. [Crossref] [PubMed]
9. Hurt KJ, Sezen SF, Lagoda GF, et al. Cyclic AMP-dependent phosphorylation of neuronal nitric oxide synthase mediates penile erection. Proc Natl Acad Sci U S A 2012;109:16624-9. [Crossref] [PubMed]
10. Sturny M, Karakus S, Fraga-Silva R, et al. Low-Intensity Electrostimulation Enhances Neuroregeneration and Improves Erectile Function in a Rat Model of Cavernous Nerve Injury. J Sex Med 2022;19:686-96. [Crossref] [PubMed]
11. Wan ZH, Zhang YJ, Chen L, et al. G protein-coupled receptor kinase 2 inhibition improves erectile function through amelioration of endothelial dysfunction and oxidative stress in a rat model of type 2 diabetes. Asian J Androl 2019;21:74-9. [Crossref] [PubMed]
12. Hidaka T, Hata T, Soga J, et al. Increased leukocyte rho kinase (ROCK) activity and endothelial dysfunction in cigarette smokers. Hypertens Res 2010;33:354-9. [Crossref] [PubMed]
13. Zhu D, Pham QM, Wang C, et al. Erectile Dysfunction and Oxidative Stress: A Narrative Review. Int J Mol Sci 2025;26:3073. [Crossref] [PubMed]
14. Lin CS, Lin G, Lue TF. Cyclic nucleotide signaling in cavernous smooth muscle. J Sex Med 2005;2:478-91. [Crossref] [PubMed]
15. Chen K, Huang B, Feng J, et al. Nesfatin-1 regulates the phenotype transition of cavernous smooth muscle cells by activating PI3K/AKT/mTOR signaling pathway to improve diabetic erectile dysfunction. Heliyon 2024;10:e32524. [Crossref] [PubMed]
16. Latcu SC, Novacescu D, Buciu VB, et al. The Cavernous Nerve Injury Rat Model: A Pictorial Essay on Post-Radical Prostatectomy Erectile Dysfunction Research. Life (Basel) 2023;13:2337. [Crossref] [PubMed]
17. Karakus S, Musicki B, La Favor JD, et al. cAMP-dependent post-translational modification of neuronal nitric oxide synthase neuroprotects penile erection in rats. BJU Int 2017;120:861-72. [Crossref] [PubMed]
18. Wu YN, Chen KC, Liao CH, et al. Spontaneous Regeneration of Nerve Fiber and Irreversibility of Corporal Smooth Muscle Fibrosis After Cavernous Nerve Crush Injury: Evidence From Serial Transmission Electron Microscopy and Intracavernous Pressure. Urology 2018;118:98-106. [Crossref] [PubMed]
19. Kaufmann O, Claro J, Cury J, et al. The development of a rat model of erectile dysfunction after radical prostatectomy: preliminary findings. BJU Int 2008;102:1026-8. [Crossref] [PubMed]
20. Bella AJ, Lin G, Garcia MM, et al. Upregulation of penile brain-derived neurotrophic factor (BDNF) and activation of the JAK/STAT signalling pathway in the major pelvic ganglion of the rat after cavernous nerve transection. Eur Urol 2007;52:574-80. [Crossref] [PubMed]
21. User HM, Hairston JH, Zelner DJ, et al. Penile weight and cell subtype specific changes in a post-radical prostatectomy model of erectile dysfunction. J Urol 2003;169:1175-9. [Crossref] [PubMed]
22. Sato Y, Rehman J, Santizo C, et al. Significant physiological roles of ancillary penile nerves on increase in intracavernous pressure in rats: experiments using electrical stimulation of the medial preoptic area. Int J Impot Res 2001;13:82-8. [Crossref] [PubMed]
23. Bochinski D, Hsieh PS, Nunes L, et al. Effect of insulin-like growth factor-1 and insulin-like growth factor binding protein-3 complex in cavernous nerve cryoablation. Int J Impot Res 2004;16:418-23. [Crossref] [PubMed]
24. El-Sakka AI, Hassan MU, Selph C, et al. Effect of cavernous nerve freezing on protein and gene expression of nitric oxide synthase in the rat penis and pelvic ganglia. J Urol 1998;160:2245-52. [Crossref] [PubMed]
25. El-Sakka A, Hassan MU, Bakircioglu ME, et al. Possible molecular mechanisms of cryoablation-induced impotence in a rat model. Urology 1998;52:1144-50. [Crossref] [PubMed]
26. Sezen SF, Blackshaw S, Steiner JP, et al. FK506 binding protein 12 is expressed in rat penile innervation and upregulated after cavernous nerve injury. Int J Impot Res 2002;14:506-12. [Crossref] [PubMed]
27. Ozkara H, Alan C, Atukeren P, et al. Changes of nitric oxide synthase-containing nerve fibers and parameters for oxidative stress after unilateral cavernous nerve resection or manuplation in rat penis. Chin J Physiol 2006;49:160-6.
28. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
29. James ND, Tannock I, N’Dow J, et al. The Lancet Commission on prostate cancer: planning for the surge in cases. Lancet 2024;403:1683-722. [Crossref] [PubMed]
30. Bernal J, Venkatesan K, Martins FE. Erectile Dysfunction in Pelvic Cancer Survivors and Current Management Options. J Clin Med 2023;12:2697. [Crossref] [PubMed]
31. Nukala V, Incrocci L, Hunt AA, et al. Challenges in Reporting the Effect of Radiotherapy on Erectile Function. J Sex Med 2020;17:1053-9. [Crossref] [PubMed]
32. Jackson IL, Pavlovic R, Alexander AA, et al. BIO 300, a Nanosuspension of Genistein, Mitigates Radiation-Induced Erectile Dysfunction and Sensitizes Human Prostate Cancer Xenografts to Radiation Therapy. Int J Radiat Oncol Biol Phys 2019;105:400-9. [Crossref] [PubMed]
33. Mahmood J, Connors CQ, Alexander AA, et al. Cavernous Nerve Injury by Radiation Therapy May Potentiate Erectile Dysfunction in Rats. Int J Radiat Oncol Biol Phys 2017;99:680-8. [Crossref] [PubMed]
34. Mahmood J, Pandita R, Zhang A, et al. RhoA/ROCK pathway inhibitor ameliorates erectile dysfunction induced by radiation therapy in rats. Radiother Oncol 2020;150:174-80. [Crossref] [PubMed]
35. Powers SA, Odom MR, Pak ES, et al. Prostate-Confined Radiation Decreased Pelvic Ganglia Neuronal Survival and Outgrowth. J Sex Med 2019;16:27-41. [Crossref] [PubMed]
36. Randolph JT, Pak ES, McMains JC, et al. Cocultured Schwann Cells Rescue Irradiated Pelvic Neuron Outgrowth and Increase Survival. J Sex Med 2022;19:1333-42. [Crossref] [PubMed]
37. Malavige LS, Levy JC. Erectile dysfunction in diabetes mellitus. J Sex Med 2009;6:1232-47. [Crossref] [PubMed]
38. Bivalacqua TJ, Champion HC, Usta MF, et al. RhoA/Rho-kinase suppresses endothelial nitric oxide synthase in the penis: a mechanism for diabetes-associated erectile dysfunction. Proc Natl Acad Sci U S A 2004;101:9121-6. [Crossref] [PubMed]
39. Comerma-Steffensen S, Prat-Duran J, Mogensen S, et al. Erectile Dysfunction and Altered Contribution of KCa1.1 and KCa2.3 Channels in the Penile Tissue of Type-2 Diabetic db/db Mice. J Sex Med 2022;19:697-710. [Crossref] [PubMed]
40. Jesmin S, Sakuma I, Salah-Eldin A, et al. Diminished penile expression of vascular endothelial growth factor and its receptors at the insulin-resistant stage of a type II diabetic rat model: a possible cause for erectile dysfunction in diabetes. J Mol Endocrinol 2003;31:401-18. [Crossref] [PubMed]
41. Chen GT, Yang BB, Chen JH, et al. Pancreatic kininogenase improves erectile function in streptozotocin-induced type 2 diabetic rats with erectile dysfunction. Asian J Androl 2018;20:448-53. [Crossref] [PubMed]
42. Angulo J, Cuevas P, Fernández A, et al. Enhanced thromboxane receptor-mediated responses and impaired endothelium-dependent relaxation in human corpus cavernosum from diabetic impotent men: role of protein kinase C activity. J Pharmacol Exp Ther 2006;319:783-9. [Crossref] [PubMed]
43. Defeudis G, Mazzilli R, Tenuta M, et al. Erectile dysfunction and diabetes: A melting pot of circumstances and treatments. Diabetes Metab Res Rev 2022;38:e3494. [Crossref] [PubMed]
44. Musicki B, Liu T, Lagoda GA, et al. Hypercholesterolemia-induced erectile dysfunction: endothelial nitric oxide synthase (eNOS) uncoupling in the mouse penis by NAD(P)H oxidase. J Sex Med 2010;7:3023-32. [Crossref] [PubMed]
45. Behr-Roussel D, Bernabe J, Compagnie S, et al. Distinct mechanisms implicated in atherosclerosis-induced erectile dysfunction in rabbits. Atherosclerosis 2002;162:355-62. [Crossref] [PubMed]
46. Xie D, Odronic SI, Wu F, et al. A mouse model of hypercholesterolemia-induced erectile dysfunction. J Sex Med 2007;4:898-907. [Crossref] [PubMed]
47. Behr-Roussel D, Darblade B, Oudot A, et al. Erectile dysfunction in hypercholesterolemic atherosclerotic apolipoprotein E knockout mice. J Sex Med 2006;3:596-603. [Crossref] [PubMed]
48. Jin L, Lagoda G, Leite R, et al. NADPH oxidase activation: a mechanism of hypertension-associated erectile dysfunction. J Sex Med 2008;5:544-51. [Crossref] [PubMed]
49. Chiangsaen P, Maneesai P, Kukongviriyapan U, et al. Tangeretin ameliorates erectile and testicular dysfunction in a rat model of hypertension. Reprod Toxicol 2020;96:1-10. [Crossref] [PubMed]
50. Fratta Pasini A, Garbin U, Nava MC, et al. Nebivolol decreases oxidative stress in essential hypertensive patients and increases nitric oxide by reducing its oxidative inactivation. J Hypertens 2005;23:589-96. [Crossref] [PubMed]
51. Long H, Jiang J, Xia J, et al. Icariin improves SHR erectile function via inhibiting eNOS uncoupling. Andrologia 2018;50:e13084. [Crossref] [PubMed]
52. Liu QW, Yang ZH, Jiang J, et al. Icariin modulates eNOS activity via effect on post-translational protein-protein interactions to improve erectile function of spontaneously hypertensive rats. Andrology 2021;9:342-51. [Crossref] [PubMed]
53. Gonzaga NA, do Vale GT, da Silva CBP, et al. Treatment with nitrite prevents reactive oxygen species generation in the corpora cavernosa and restores intracavernosal pressure in hypertensive rats. Nitric Oxide 2020;94:19-26. [Crossref] [PubMed]
54. Peluffo G, Calcerrada P, Piacenza L, et al. Superoxide-mediated inactivation of nitric oxide and peroxynitrite formation by tobacco smoke in vascular endothelium: studies in cultured cells and smokers. Am J Physiol Heart Circ Physiol 2009;296:H1781-92. [Crossref] [PubMed]
55. Bivalacqua TJ, Sussan TE, Gebska MA, et al. Sildenafil inhibits superoxide formation and prevents endothelial dysfunction in a mouse model of secondhand smoke induced erectile dysfunction. J Urol 2009;181:899-906. [Crossref] [PubMed]
56. Xie Y, Garban H, Ng C, et al. Effect of long-term passive smoking on erectile function and penile nitric oxide synthase in the rat. J Urol 1997;157:1121-6.
57. Tai HC, Tsai WK, Chang ML, et al. Intracavernous injection of platelet-rich plasma reverses erectile dysfunction of chronic cavernous nerve degeneration through reduction of prostate hyperplasia evidence from an aging-induced erectile dysfunction rat model. FASEB J 2023;37:e22826. [Crossref] [PubMed]
58. Davila HH, Rajfer J, Gonzalez-Cadavid NF. Corporal veno-occlusive dysfunction in aging rats: evaluation by cavernosometry and cavernosography. Urology 2004;64:1261-6. [Crossref] [PubMed]
59. Angulo J, El Assar M, Sevilleja-Ortiz A, et al. Short-term pharmacological activation of Nrf2 ameliorates vascular dysfunction in aged rats and in pathological human vasculature. A potential target for therapeutic intervention. Redox Biol 2019;26:101271.
60. de Almeida Rezende MS, Oliveira de Almeida AJP, Gonçalves TAF, et al. D-(+)-Galactose-induced aging: A novel experimental model of erectile dysfunction. PLoS One 2021;16:e0249487. [Crossref] [PubMed]
61. Pierre CJ, Azeez TA, Rossetti ML, et al. Long-term administration of resveratrol and MitoQ stimulates cavernosum antioxidant gene expression in a mouse castration model of erectile dysfunction. Life Sci 2022;310:121082. [Crossref] [PubMed]
62. Alves-Lopes RU, Neves KB, Silva MA, et al. Functional and structural changes in internal pudendal arteries underlie erectile dysfunction induced by androgen deprivation. Asian J Androl 2017;19:526-32. [Crossref] [PubMed]
63. Wang XJ, Xu TY, Xia LL, et al. Castration impairs erectile organ structure and function by inhibiting autophagy and promoting apoptosis of corpus cavernosum smooth muscle cells in rats. Int Urol Nephrol 2015;47:1105-15. [Crossref] [PubMed]
64. Yang HZ, Xiong WJ, Li X, et al. Low androgen status inhibits erectile function by upregulating the expression of proteins of mitochondria-associated membranes in rat corpus cavernosum. Andrology 2022;10:997-1007. [Crossref] [PubMed]
65. Chen Z, Jiang J, Jiang R. A low testosterone level impairs erectile function by increasing endocan expression in rat penile corpus cavernosum. J Sex Med 2024;21:663-70. [Crossref] [PubMed]
66. Liu G, Liu J, Kong X, et al. Effect of hypoandrogenism on expression of transient receptor potential vanilloid channels in rat penile corpus cavernosum and erectile function. J Sex Med 2023;20:1153-60. [Crossref] [PubMed]
67. Gu J, Zhu LK, Zhao X, et al. Low testosterone state inhibits erectile function by downregulating the expression of GIT1 in rat penile corpus cavernosum. Sex Med 2023;11:qfad017. [Crossref] [PubMed]
68. Yuan P, Li X, Xiong WJ, et al. Downregulation of the expression of galanin impairs erectile function in hypoandrogenic rats. Sex Med 2023;11:qfad029. [Crossref] [PubMed]
69. Succu S, Sanna F, Melis T, et al. Stimulation of dopamine receptors in the paraventricular nucleus of the hypothalamus of male rats induces penile erection and increases extra-cellular dopamine in the nucleus accumbens: Involvement of central oxytocin. Neuropharmacology 2007;52:1034-43. [Crossref] [PubMed]
70. Melis MR, Argiolas A. Central control of penile erection: a re-visitation of the role of oxytocin and its interaction with dopamine and glutamic acid in male rats. Neurosci Biobehav Rev 2011;35:939-55. [Crossref] [PubMed]
71. Chen G, Chen J, Yang B, et al. Dopamine D2 receptors in the basolateral amygdala modulate erectile function in a rat model of nonorganic erectile dysfunction. Andrologia 2019;51:e13160. [Crossref] [PubMed]
72. Chen G, Yu D, Wu Y, et al. Dopamine D2 receptors in the nucleus accumbens modulate erectile function in a rat model of nonorganic erectile dysfunction. Andrology 2022;10:808-17. [Crossref] [PubMed]
73. Wang H, Liu X, Zhang Z, et al. Effects of tadalafil on sexual behavior of male rats induced by chronic unpredictable mild stress. Sex Med 2023;11:qfad019. [Crossref] [PubMed]
74. Shao J, Nie P, Yang W, et al. An EPO-loaded multifunctional hydrogel synergizing with adipose-derived stem cells restores neurogenic erectile function via enhancing nerve regeneration and penile rehabilitation. Bioeng Transl Med 2022;7:e10319. [Crossref] [PubMed]
75. Zhang T, Li W, Qiu X, et al. CRISPR/Cas9-mediated TEAD1 knockout induces phenotypic modulation of corpus cavernosum smooth muscle cells in diabetic rats with erectile dysfunction. Nan Fang Yi Ke Da Xue Xue Bao 2021;41:567-73. [Crossref] [PubMed]
76. Sun T, Xu W, Tu B, et al. Engineered Adipose-Derived Stem Cells Overexpressing RXFP1 via CRISPR Activation Ameliorate Erectile Dysfunction in Diabetic Rats. Antioxidants (Basel) 2023;12:171. [Crossref] [PubMed]
77. Ping X, Zhou H, Zhang Y, et al. The New Approach to Treating Erectile Dysfunction: Stem Cell-Derived Extracellular Vesicles. Int J Nanomedicine 2025;20:11249-66. [Crossref] [PubMed]
78. Xu W, Sun T, Wang J, et al. GPX4 Alleviates Diabetes Mellitus-Induced Erectile Dysfunction by Inhibiting Ferroptosis. Antioxidants (Basel) 2022;11:1896. [Crossref] [PubMed]
79. Gur S, Alzweri L, Yilmaz-Oral D, et al. Ivabradine, the hyperpolarization-activated cyclic nucleotide-gated channel blocker, elicits relaxation of the human corpus cavernosum: a potential option for erectile dysfunction treatment. Aging Male 2020;23:1088-97. [Crossref] [PubMed]
80. Huang X, Zhao H, Wu X, et al. Mechanisms of Leydig Cell Aging and Obesity-Related Hypogonadism in Men: A Review. Med Sci Monit 2025;31:e948180. [Crossref] [PubMed]
81. Ihrig CM, Pierre CJ, Azeez TA, et al. Chronic administration of the hydrogen sulfide prodrug SG1002 partially protects against erectile dysfunction resulting from long-term androgen deprivation. Life Sci 2025;380:123976. [Crossref] [PubMed]
82. Muir J, Lorsch ZS, Ramakrishnan C, et al. In Vivo Fiber Photometry Reveals Signature of Future Stress Susceptibility in Nucleus Accumbens. Neuropsychopharmacology 2018;43:255-63. [Crossref] [PubMed]
83. Bright U, Chen Y, Deak JD, et al. Multi-ancestry investigation of the genomics of erectile dysfunction. Nat Commun 2025;16:11602. [Crossref] [PubMed]