Photobiomodulation therapy for diabetic erectile dysfunction targeting neuroinflammation and neurovascular regeneration

Abstract

Erectile dysfunction (ED) in diabetes often resists phosphodiesterase type 5 inhibitors due to neuropathy and vasculopathy, both worsened by neuroinflammation. This study evaluated light-emitting diode (LED) therapy’s effects on diabetes-induced neurovascular damage using a diabetic mouse model. Diabetes was induced in C57BL/6 mice with streptozotocin, followed by treatment with RED (660 nm) and near-infrared (NIR; 830 nm) LED light, separately and combined, for ten days over two weeks. Functional and molecular analyses assessed neurovascular regeneration. LED therapy significantly improved intra-cavernous pressure (ICP), with combined RED and NIR wavelengths restoring ICP to 90% of normal levels, indicating enhanced nerve and vascular function. Histological analyses showed increased endothelial cell density, angiogenesis, pericyte recruitment, and neural regeneration. Molecular findings revealed upregulation of neurotrophic factors (NGF, NT-3, BDNF), angiogenic markers (VEGF, eNOS), and phosphorylated PI3K, alongside reduced apoptosis and increased cell proliferation. These results demonstrate that LED therapy mitigates diabetes-induced neuropathy and vasculopathy by enhancing neurovascular repair and modulating neuroinflammatory pathways. The study highlights the potential of combined RED and NIR LED therapy as a non-invasive treatment for diabetic ED and related neurovascular complications, offering a promising approach to improving patient outcomes.

Introduction

Erectile dysfunction (ED) is a common and debilitating disorder that profoundly impacts men’s quality of life, particularly those with diabetes mellitus (DM), and compared with that in non-diabetic individuals, diabetic-induced erectile dysfunction (DIED) develops earlier and is notably more severe. The pathophysiology of DIED is complex, and is often attributed to the vascular and neuropathic damage induced by chronic hyperglycemia¹. However, despite the availability of pharmacological therapeutic options, such as phosphodiesterase type 5 inhibitors (PDE5-i), a substantial subset of diabetic men shows poor responses to these treatments. Indeed, it has been reported that almost 50% of such patients do not obtain satisfactory outcomes², thereby highlighting the significant therapeutic challenge presented by this condition.

The conventional approach to addressing the problem of ED tends to focus primarily on symptomatic relief rather than addressing underlying pathologies, thus emphasizing need for effective non-invasive approaches for the treatment of this population. In this regard, photobiomodulation (PBM) therapy using LEDs is considered to offer a promising alternative. PBM has been shown to have a positive influence on cellular functions by enhancing mitochondrial metabolism and promoting significant antioxidant effects, which are crucial in the management of disorders like DIED, where oxidative stress is significantly involved³.

Recent advances in PBM have highlighted its potential utility in modulating mitochondrial dysfunction and enhancing the antioxidant capacity of diabetic tissues and in cases of nerve injury, thereby providing evidence for a novel mechanism whereby PBM could ameliorate ED^4,5. PBM has been observed to activate mitochondrial respiratory chain components, thus promoting increases in ATP production and reducing the concentrations of reactive oxygen species (ROS), the latter of which is assumed to mitigate cellular damage and promote neurovascular regeneration. Additionally, PBM has been demonstrated to induce the upregulated expression of growth factors including vascular endothelial growth factor (VEGF) as well as nerve growth factor (NGF) which are essential for tissue repair and rejuvenation⁴. Furthermore, evidence suggests that PBM regulates oxidative stress response genes’ expression, inflammation, and apoptosis, including Nrf2, HO-1, and Bcl-2, thereby contributing to an enhancement of cellular resilience and survival⁵.

In our previous study, in a model of nerve injury, we showed that PBM could restore erectile function, thereby indicating its potential application in the treatment of DIED⁴. Indeed, the non-invasive nature of PBM and its capacity to target causal factors at the cellular level make this approach an attractive area for further research. Given the limitations of current pharmacological therapies and the innovative approach of LED therapy in enhancing mitochondrial function and reducing oxidative damage, in this study, we aimed to assess the efficacy and safety of PBM as a non-pharmacological treatment intervention for DIED. In doing so, we hope to provide a new perspective on non-invasive treatment strategies that could meet the unmet needs of those with ED, thus potentially transforming the therapeutic landscape for this disorder.

Results

DM adversely affects erectile tissue cell morphology, thereby causing functional alterations, as exemplified by the impaired erectile function observed in DM mice with prolonged hyperglycemia. To investigate such DM-mediated changes in erectile tissue, we monitored dynamic alterations in the composition of endothelial cells, smooth muscle cells, nerves, and pericytes within the corpus cavernosum. Figure 1 shows a conceptual framework of the PBM therapy system we employed for treating mice with modeled DM. The system includes a low-light LED based delivery device designed for the administration of NIR and RED light to the abdominal region of mice that can be applied in both in vivo and in vitro studies. The LED module board is strategically positioned beneath a transparent, non-toxic acrylic cage, allowing for the unhindered movement of the animals throughout the treatment process. Light is directed perpendicularly to the abdominal region and is administered in a continuous mode for a period of 30 min.

Fig. 1.

PBM device and treatment. Schematic overview of PBM in the STZ diabetic mouse model. PBM treatment commenced 10 weeks post diabetes induction. The irradiation was applied perpendicularly to the dorsal region of the mice, as well as to the upper surfaces of the culture plates in the in vitro study. STZ streptozotocin, PBM photobiomodulation.

PBM therapy improves erectile function by enhancing cavernous angiogenesis and nerve regeneration in DM mice

The impact of PBM on erectile function was evaluated through the application of electrical stimulation to the cavernous nerve, conducted 10 weeks post DM induction by STZ. In comparison to the age-matched controls, the heat-treated diabetic mice exhibited a notable reduction in the ratio of maximal and total ICP to mean MSBP (Fig. 2A, E, F). In diabetic mice subjected to PBM utilizing RED, NIR, and a combination of both wavelengths, we detected a significant enhancement in total and max ICP, resulting in a restoration to 90% of normal levels. We observed no notable disparities between the experimental and control groups with respect to MSBP.

Fig. 2.

Restoration of endothelial, pericyte, and neuronal cells in diabetic mice in response to PBM enhances the recovery of erectile function. (A) Representative ICP responses for age-matched control mice with normal glucose levels (C) and diabetic (DM) mice, assessed 2 weeks after PBM irradiation therapy with Heat (H), RED (R), NIR (N), or a combination of RED and NIR (R + N). The solid bar indicates the stimulus period. (B) Immunostaining of endothelial cells (PECAM-1; green, p-eNOS; red), pericytes (NG-2; red), and smooth muscle cells (α-SMA; green) in cavernous tissue obtained from age-matched control mice with normal glucose levels (C) and diabetic (DM) mice, assessed 2 weeks after irradiation with Heat (H), RED(R), NIR (N), or a combination of RED and NIR (R + N). Scale bar = 100 μm. (C) Immunostaining for neuronal cells (NF; green) in dorsal nerve bundles from age-matched control mice (C) and diabetic (DM) mice, assessed 2 weeks after irradiation with Heat (H), RED(R), NIR (N), or a combination of RED and NIR (R + N). Nuclei were labeled with DAPI. Scale bar = 100 μm. (D) Representative western blots for neurotrophic factors (BDNF, NGF, and NT-3) and angiogenic factors (VEGF, FGF-2, and Ang-1) in cavernous tissue from age-matched control mice (C) and diabetic (DM) mice, assessed 2 weeks after irradiation with Heat (H), RED(R), NIR (N), or a combination of RED and NIR (R + N). (E and F) Ratios of the mean maximum and total ICP to the mean systolic blood pressure (area under the curve) were determined for each group and are presented as the mean values with standard error of the mean (SEM; n = 10; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). (g-j) Quantitative analyses of PECAM-1, NF, NG-2, and α-SMA expression in cavernous tissue using Image J, with data being presented as the means ± SEM (n = 8; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). (K-P) Band intensity values of NGF (K) NT-3 (L), BDNF (M), Ang-1 (N), VEGF (O), and FGF-2 (P), normalized to the density of GAPDH, quantified using Image J, with data being presented as the means ± SEM (n = 5; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). The relative ratio in the control group was defined as 1. ICP, intracavernous pressure; PBM, photobiomodulation; DM, diabetes mellitus; PECAM-1 platelet endothelial cell adhesion molecule; p-eNOS, phosphorylated endothelial nitric oxide synthase; NG-2, nerve/glial antigen 2; α-SMA, alpha-smooth muscle actin; NF, neurofilament; BDNF, brain-derived neurotrophic factor; NGF, nerve growth factor; NT-3, neurotrophin-3; VEGF, vascular endothelial growth factor; Ang-1, angiopoietin-1; FGF-2, Fibroblast growth factor 2; SEM, standard error of the mean; DAPI, 4,6-diamidino-2-phenylindole.

Immunofluorescence staining revealed increased expression of endothelial cells (PECAM-1), smooth muscle cells (α-SMA), and pericytes (NG-2) in the cavernosum tissue of diabetic mice treated with PBM (Fig. 2B, G–J). Quantitative analysis of PECAM-1 expression using ANOVA revealed a significant effect of PBM treatment on endothelial cell regeneration (F = 20.37, p < 0.0001, R² = 0.6995), confirming that PBM significantly enhances PECAM-1 expression. Similarly, α-SMA expression, a marker of smooth muscle cells, showed significant group differences (F = 44.86, p < 0.0001, R² = 0.8568), indicating improved smooth muscle content following PBM treatment. Quantitative analysis of NG-2 expression further revealed a significant effect of PBM on pericyte regeneration (F = 27.65, p < 0.0001, R² = 0.7980), supporting its role in vascular restoration. Diabetic ED develops as a consequence of a combination of endothelial dysfunction, autonomic neuropathy, and vascular complications^6,7, and we have previously demonstrated that PBM can contribute to the regeneration of neurons in a model of cavernous injury⁴, on the basis of which we accordingly postulated that PBM would might also have similar effects on neuronal cells and peripheral nerves in the penile tissues of mice with STZ-induced diabetes. To verify this assumption, we examined the expression of Neurofilament-2000 (NF) in the dorsal nerve bundles of cavernous tissues obtained from control, STZ-induced diabetic, and PBM-treated DM mice.

Immunofluorescent labeling revealed compared with control mice, there was a pronounced increase in expression of NF in the dorsal nerve bundles of STZ-induced diabetics mice treated with PBM (Fig. 2C). Consistent with these findings, Western blot analysis confirmed increased expression of NGF, NT-3, and BDNF in the PBM-treated diabetic mice compared to untreated diabetic controls (Fig. 2D, K-M). ANOVA analysis demonstrated significant differences in NGF (F = 10.37, p < 0.0001, R² = 0.5970), NT-3 (F = 10.57, p < 0.0001, R² = 0.6789), and BDNF (F = 12.71, p < 0.0001, R² = 0.6212), confirming that PBM effectively upregulates these neurotrophic factors.

PBM was shown to stimulate the production of angiogenic factors, including VEGF, Ang-1, and FGF-2, as demonstrated by the band intensity values normalized to GAPDH (Fig. 2N-P). Statistical analysis revealed significant differences in VEGF (F = 3.31, p = 0.0244, R² = 0.3209), Ang-1 (F = 8.92, p = 0.0003, R² = 0.6408), and FGF-2 (F = 6.62, p = 0.0011, R² = 0.5352), confirming PBM’s angiogenic effects. In contrast, these expressions were markedly reduced in STZ-induced diabetic mice. These results suggest that PBM can enhance the regrowth of penile nerves in diabetic mice induced by STZ. Supporting previous findings, PBM was also shown to increase the expression of neurotrophic factors such as BDNF, NT-3, and NGF (Fig. 2D). Furthermore, PBM significantly stimulated the production of angiogenic factors, including Ang-1 and VEGF (Fig. 2D). Analysis of phosphorylated eNOS (p-eNOS), a critical regulator of endothelial function, showed a significant increase in expression in PBM-treated diabetic mice compared to controls (ANOVA: F = 29.43, p < 0.0001, R² = 0.7325), indicating improved endothelial health.

Collectively, our findings indicate that PBM enhances the process of neurovascular regeneration in DM mice by activating angiogenic (VEGF, Ang-1, and FGF-2) and neurotropic (BDNF, NT-3, and NGF) factors. While diabetic animals exhibited significantly elevated fasting and postprandial blood glucose levels in comparison to the control group mice, there were no notable differences in body weight or blood glucose levels among the diabetic mice, irrespective of the treatment administered (Supplementary Table S1).

In summary, our experimental results provide convincing evidence to indicate that PBM utilizing RED, NIR, and their combination can effectively replenish the levels of cavernous endothelial cells, smooth muscle cells, and neurons, thereby contributing to a marked amelioration of erectile dysfunction in diabetic mice.

PBM enhances mitochondrial ATP production and reduces ROS production in vivo and in vitro

To assess the effectiveness of PBM in enhancing ATP production and addressing the compromised mitochondrial function resulting from elevated ROS levels in DM, we immunostained penile tissue using ox-LDL (Fig. 3A) and nitrotyrosine (Fig. 3B). The findings indicated increased levels of ROS in the penile tissues of diabetic mice, which played a role in the onset of erectile dysfunction. Following PBM treatment, we detected a notable reduction in ROS levels, with RED, NIR, and RED + NIR irradiation being found to have prominent effects. Additionally, nitrotyrosine levels were significantly reduced following PBM treatment (ANOVA: F = 7.883, p < 0.0001, R² = 0.3916), indicating that PBM effectively mitigates oxidative stress in penile tissues (Fig. 3D, E). Excessive levels of ROS are associated with reductions in mitochondrial membrane potential and general mitochondrial malfunction, as revealed by a reduction in the fluorescence of TMRM (Fig. 3C). The mitochondrial dysfunction was effectively reduced in response to PBM therapy using RED, NIR, and their combination, as confirmed by a recovery of TMRM fluorescence (ANOVA: F = 11.24, p < 0.0001, R² = 0.2939) at Fig. 3C, the ratio of TMRM-positive to total mitochondria was significantly increased following PBM treatment, further supporting mitochondrial recovery (ANOVA: F = 19.12, p < 0.0001, R² = 0.5016) (Fig. 3I-K).

Fig. 3.

PBM enhances ATP production in mitochondria and ameliorate ROS production in vivo and in vitro. (A and B) Immunostaining for reactive oxygen species (ROS), ox-LDL, and nitrotyrosine in cavernous tissue from age-matched control mice (C) and diabetic (DM) mice, assessed 2 weeks irradiation with Heat (H), RED (R), NIR (N), or a combination of RED and NIR (R + N). (C) Staining to detect functional mitochondrial membrane potential activity using tetramethylrhodamine methyl ester (TMRM, red), and MitoTracker (green) after irradiation with Heat (H), RED(R), NIR (N), or a combination of RED and NIR (R + N) for three consecutive days under high-glucose (HG) conditions. Nuclei were labeled with DAPI. Scale bar = 100 μm. (D-H) Quantitative analysis of cavernous endothelial cell staining for ox-LDL and nitrotyrosine using Image J, with data being presented as the means ± SEM (n = 10; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). (I-K) Quantitative analysis of active endothelial mitochondrial cells stained with TMRM and MitoTracker per high-power field (HPF) using Image J, with data being presented as the means ± SEM (n = 20; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). ROS, reactive oxygen species; ox-LDL, oxidized low-density lipoprotein; DM, diabetes mellitus; ATP, adenosine triphosphate; HG, high glucose; DAPI, 4ʹ,6-diamidino-2-phenylindole; TMRM, tetramethylrhodamine.

Quantitative analysis of ox-LDL/PECAM-1 ratios provided further evidence of PBM’s efficacy in reducing oxidative stress while preserving endothelial function. PBM treatment significantly reduced the ox-LDL/PECAM-1 ratio (ANOVA: F = 11.19, p < 0.0001, R² = 0.5280). Similarly, nitrotyrosine/PECAM-1 ratios demonstrated a significant reduction with PBM treatment (ANOVA: F = 4.951, p = 0.0020, R² = 0.2921), highlighting its protective effects on endothelial cells under oxidative stress conditions (Fig. 3F-H).

Finally, quantitative analysis of mitochondrial markers revealed significant improvements in mitochondrial abundance and activity following PBM treatment. Active endothelial mitochondrial cells stained with TMRM and MitoTracker showed a marked increase per high-power field (HPF) (ANOVA: F = 4.864, p = 0.0015, R² = 0.1997), further confirming the mitochondrial protective effects of PBM (Fig. 3I-K).

Collectively, these findings demonstrate that PBM significantly reduces ROS levels and oxidative stress markers while improving mitochondrial function in diabetic mice. By reducing oxidative damage and restoring mitochondrial activity, PBM holds promise as a therapeutic strategy to mitigate erectile dysfunction associated with DM.

PBM induces vascular sprouting and improves cellular connectivity, mitigating diabetic damage in endothelial and neural tissues

The reduction in apoptosis in the DM mice cavernous tissue as well as high glucose environment of endothelial cell suggests that PBM effectively mitigates the apoptotic signaling cascades (Fig. 4A-C). This is likely achieved through stabilization of mitochondrial function and reduction in oxidative stress, preserving cellular viability. Quantitative analysis of TUNEL-positive cells revealed a significant reduction in apoptotic cells following PBM treatment (Fig. 4F, H), as confirmed by ANOVA (F = 14.01, p < 0.0001, R² = 0.5897 for cavernous tissue; F = 15.57, p < 0.0001, R² = 0.7218 for endothelial cells). This finding is crucial for tissue repair, as excessive cell death can limit regenerative potential.

Fig. 4.

PBM suppresses apoptosis and promotes endothelial cell proliferation in vivo and invitro. (A and B) Immunostaining for the proliferation of endothelial cells (phospho-histone H3; PHH3, red) and apoptosis (TUNEL; green) in cavernous tissue from age-matched control mice (C) and diabetic (DM) mice, assessed 2 weeks after irradiation with Heat (H), RED (R), NIR (N), or a combination of RED and NIR (R + N). Nuclei were labeled with DAPI. Scale bars = 100 μm. (C) Immunofluorescence staining of MCECs with anti-BrdU antibody (red) and TUNEL assay (green) in cells irradiated with Heat (H), RED(R), NIR (N), or a combination of RED and NIR (R + N) for three consecutive days under NG or HG conditions. Nuclei were labeled with DAPI (blue). Scale bars = 50 μm. (D) Immunofluorescence staining of VE-cadherin and ZO-1 in MCECs irradiated with Heat (H), RED (R), NIR (N), or a combination of RED and NIR (R + N) under NG or HG conditions for 3 days. Scale bars = 50 μm. (E and F) Quantitative analysis of cavernous proliferation (E) and apoptosis (F) using Image J, with data being presented as the means ± SEM (n = 8; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). (G and H) Quantitative analysis of BrdU positive (G) and TUNEL-positive (H) endothelial cells per high-power field (HPF) using Image J, with data being presented as the means ± SEM (n = 5; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). (I and J) Quantitative analysis of VE-cadherin (I) and ZO-1(J) immunopositive staining in MCECs using Image J, with data being presented as the means ± SEM (n = 7; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). DAPI, 4ʹ,6-diamidino-2-phenylindole; PHH3, phospho-histone H3; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling; BrdU, 5ʹ-bromo-2ʹ-deoxyuridine; ZO-1, zonula occludens-1.

In addition to reducing apoptosis, PBM treatment was associated with a marked increase in cellular proliferation. BrdU (Bromodeoxyuridine) and pH3 (Phospho-Histone H3) expression data demonstrate a marked increase in cellular proliferation within the cavernous tissue (Fig. 4A, E) and endothelial cell (Fig. 4C, G) after PBM treatment. BrdU is a thymidine analog that incorporates into newly synthesized DNA, which marks cells in the S-phase of the cell cycle, showed a significant increase after PBM treatment (ANOVA: F = 5.902, p = 0.0014, R² = 0.4575). Similarly, pH3 staining, indicative of mitotic activity, revealed significantly higher proliferation in PBM-treated groups (ANOVA: F = 8.302, p < 0.0001, R² = 0.3765). This dual effect of reduced apoptosis and enhanced proliferation is essential for tissue regeneration, as it allows for replacement of damaged or lost cells in the diabetic milieu.

ZO-1, a tight junction protein critical for maintaining endothelial barrier integrity (Fig. 4D), while VE-cadherin (Fig. 4D) is a key component of adherent junctions, essential for endothelial cell–cell adhesion and vessel stability were also evaluated. PBM treatment significantly increased the expression of ZO-1 (ANOVA: F = 12.26, p < 0.0001, R² = 0.5768) and VE-cadherin (ANOVA: F = 3.708, p = 0.0167, R² = 0.3724) in endothelial cells (Fig.4I, J). These findings suggest that PBM contributes to the preservation and stabilization of endothelial junctional integrity, which is critical for maintaining vascular homeostasis in diabetic conditions.

Collectively, these results demonstrate that PBM exerts dual effects in diabetic tissue: it reduces apoptosis and promotes cellular proliferation, while enhancing the expression of key junctional proteins that support endothelial integrity. These effects contribute to the restoration of tissue homeostasis and regeneration in diabetic mice. All analyses confirm PBM’s potential as a therapeutic intervention for mitigating endothelial and cavernous dysfunction associated with diabetes.

The upregulation of these proteins post-PBM treatment suggests that PBM enhances endothelial barrier function and stability, counteracting the vascular dysfunction typically observed in diabetic conditions. This improved endothelial integrity is crucial for promoting angiogenesis, as stable endothelial junctions are necessary for the formation of new blood vessels. This is further evidenced by the results of assays assessing MCEC tube formation (Fig. 5A, F), aortic ring sprouting (Fig. 5B, G), and endothelial migration (Fig. 5C, H). Quantitative analysis of master junctions in tube formation assays showed significant increases following PBM treatment (ANOVA: F = 8.956, p = 0.0001, R² = 0.5890), while micro vessel sprouting from aortic rings demonstrated even more pronounced effects (ANOVA: F = 30.81, p < 0.0001, R² = 0.7550). Similarly, PBM significantly enhanced endothelial cell migration, as reflected by increased migrated cells in wound healing assays (ANOVA: F = 13.30, p < 0.0001, R² = 0.6803).

Fig. 5.

PBM induces vascular sprouting and improves cellular connectivity, mitigating diabetic damage in endothelial and neural tissues. (A) Tube-formation assay: mouse cavernous endothelial cells (MCECs) were irradiated with Heat (H), RED (R), NIR (N), or a combination of RED and NIR (R + N) for five consecutive days under normal glucose (NG) or high glucose (HG) conditions. Representative images were acquired at 24 h. Scale bar = 100 μm. (B) Ex vivo mice aortic ring micro vessels outgrowth assay: aorta rings were irradiated with Heat (H), RED (R), NIR (N), or a combination of RED and NIR (R + N) for five consecutive days under NG or HG conditions. Representative images of sprouting micro vessels were acquired at 5 days. Scale bar = 100 μm. (C) Migration assay: MCECs were irradiated with Heat (H), RED(R), NIR (N), or a combination of RED and NIR (R + N) for three consecutive days under NG or HG conditions. Representative images of migrated cells were acquired at 24 h. Scale bar = 100 μm. (D) βIII tubulin staining of mouse major pelvic ganglion (MPG) and dorsal root ganglion (DRG) tissue that irradiated with Heat (H), RED (R), NIR (N), or a combination of RED and NIR (R + N) for five consecutive days under normal glucose (NG) or high glucose (HG) conditions. Representative images were acquired at 72 h. Scale bar = 25 μm. (E) Representative western blots for neurotrophic factors (NGF, NT-3, and BDNF), angiogenic factors (VEGF, FGF-2, Ang-1), eNOS, and AKT of MCECs irradiated with heat (H), RED (R), NIR (N), or RED + NIR (R + N) for five consecutive days under NG or HG conditions. (F) Quantitative analysis of master junctions using Image J, with data being presented as the means ± SEM (n = 6; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). (G) Quantitative analysis of micro vessel sprouting area from aortic rings, using Image J, with data presented as the means ± SEM (n = 9; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). The relative ratio in the NG group was defined as 1. (H) Quantitative analysis of migrated cells using Image J, with data being presented as the means ± SEM (n = 5; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). The relative ratio in the NG group was defined as 1. (I and J) Quantitative analysis of βIII-tubulin–immunopositive neurite length in MPG or DRG tissue using Image J, with data being presented as the means ± SEM (n = 7; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). The relative ratio in the NG group was defined as 1. (K–O) Band intensity values of each factor normalized to the density of GAPDH, quantified using Image J, with data being presented as the means ± SEM (n = 6; ^*P < 0.01; ^**P < 0.05; ^#P < 0.001). The relative ratio in the NG group was defined as 1. DAPI, 4',6-diamidino-2-phenylindole; PHH3, phospho-histone H3; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling; PBM, photobiomodulation; PI3K, phosphoinositide 3-kinase; VEGF, vascular endothelial growth factor; Ang-1, angiopoietin-1; FGF-2, Fibroblast growth factor 2; BDNF, brain-derived neurotrophic factor; NGF, nerve growth factor; NT-3, neurotrophin-3; SEM, standard error of the mean.

To evaluate PBM’s neuroprotective effects and its role in promoting neurogenesis, MPG and DRG explants were exposed to a high-glucose environment and immunolabeled for βIII-tubulin. PBM-treated tissues exhibited significant increases in neurite sprouting compared to untreated controls (Fig. 5D, I, J). Quantitative analysis confirmed that PBM markedly enhanced neurite length in MPG (ANOVA: F = 26.90, p < 0.0001, R² = 0.7820) and DRG tissues (ANOVA: F = 23.06, p < 0.0001, R² = 0.7250). These results highlight PBM’s ability to foster neurovascular regeneration and facilitate nerve regeneration in high-glucose conditions, supporting functional recovery in the cavernous tissue.

The regenerative capacity of PBM is further enhanced by its stimulation of key growth factors involved in angiogenesis and tissue repair, including VEGF, FGF-2, and Ang-1 (Fig. 5K-O). Western blot analysis revealed a significant upregulation of VEGF (ANOVA: F = 8.804, p < 0.0001, R² = 0.5571), FGF-2 (ANOVA: F = 7.303, p = 0.0004, R² = 0.5197), and Ang-1 (ANOVA: F = 25.30, p < 0.0001, R² = 0.7430) in response to PBM.

This regenerative capacity is further enhanced by PBM’s known effects on stimulating growth factor release, such as VEGF and Fibroblast growth factor-2 (FGF-2) (Fig. 5E) as the are key regulators of angiogenic factors that drives angiogenesis and tissue repair, while Ang-1 shown by western blot in Fig. 5B promotes vessel maturation and stability. The upregulation of these factors, along with enhanced aortic sprouting and tube formation assays in endothelial cells exposed to PBM under high-glucose conditions, reflects PBM’s broad role in enhancing vascular repair mechanisms.

Additionally, PBM-treated diabetic mice demonstrated significant increases in the phosphorylation of endothelial NO synthase (p-eNOS) and AKT (p-AKT), as confirmed by western blot analysis (Fig. 5E). Quantitative analysis of p-eNOS expression (Fig. 5K) revealed a significant increase in PBM-treated groups (ANOVA: F = 15.81, p < 0.0001, R² = 0.6855), while p-AKT/AKT (Fig. 5L) ratios were also significantly elevated (ANOVA: F = 11.84, p < 0.0001, R² = 0.6121). These findings suggest that PBM enhances nitric oxide signaling and AKT activation, both of which are critical for endothelial cell survival and angiogenesis.

Collectively, these findings provide evidence significant therapeutic potential in addressing diabetes-induced cellular and tissue dysfunction through a combination of proliferative, angiogenic, neurogenic, and anti-apoptotic effects shown at Fig. 6 PBM enhanced mitochondrial function by optimizing mitochondrial membrane potential, activating cytochrome c oxidase (CcOx), increased ATP production, and mitigating oxidative stress, and elevated nitric oxide (NO) levels that acted synergistically to stimulate cellular proliferation, angiogenesis, and neurogenesis while reducing apoptosis protected cells, preserving their viability and function. These effects highlight PBM’s ability to comprehensively restore cellular and tissue health in diabetic conditions, offering a novel, non-invasive therapeutic strategy for managing neuropathy, angiopathy, and related complications.

Fig. 6.

Schematic representation of the proposed molecular mechanism by which PBM preserves erectile function in diabetic mice. PBM enhances mitochondrial function, activating the PI3K/AKT signaling pathway and promoting AKT phosphorylation. This activation upregulates proangiogenic factors, including VEGF, eNOS, Ang-1, and FGF-2, supporting endothelial repair and vascular regeneration while inhibiting pro-apoptotic Caspase-3 signaling. Activated AKT also facilitates the synthesis of nucleotides, proteins, and lipids essential for tissue regeneration. PBM increases NO levels, which contribute to pro-survival signaling and vasodilation and induces neurotrophic factors such as BDNF, NGF, and NT-3 promoting neurogenesis and neural repair. These combined effects highlight PBM’s potential in restoring cellular, vascular, and neuronal functions in diabetic conditions. PBM, photobiomodulation; PI3K, phosphoinositide 3-kinase; VEGF, vascular endothelial growth factor; eNOS, endothelial nitric oxide synthase; Ang-1, angiopoietin-1; FGF-2, Fibroblast growth factor 2; NO, nitric oxide; BDNF, brain-derived neurotrophic factor; NGF, nerve growth factor; NT-3, neurotrophin-3.

Discussion

In this study, we demonstrate the therapeutic potential of PBM in mitigating erectile dysfunction in a diabetic mouse model. The superiority of PBM compared with conventional laser-based therapies can be ascribed to its broader wavelength spectrum and deeper tissue penetration without causing thermal damage³. We specifically observed that RED, NIR, and a combination of both these wavelengths effectively restored erectile function, as indicated by increases in the ratios of maximal and total ICP to MSBP. These improvements are consistent with the findings of earlier studies that have demonstrated to capacity of PBM to enhance cavernous angiogenesis and nerve regeneration in different animal models of ED^3,8,9.

At the molecular and cellular level, our findings reveal that PBM can enhance the proliferation and reduce the apoptosis of pericytes, endothelial, as well as smooth muscle cells as indicated by increases in the expression of PECAM-1, α-SMA, and NG2, thereby providing evidence of enhanced tissue regeneration and vascularization. The results align with earlier observations regarding the angiogenic properties of PBM, highlighting its ability to alleviate endothelial dysfunction and promote vascular health in diabetic environments⁵. Moreover, the stimulation of angiogenesis and enhancement of endothelial function in ischemic tissues further highlights the therapeutic potential^10–12. In addition, PBM was found to promote increases in the expression of NF in dorsal nerve bundles, thereby providing evidence of enhanced neuronal regeneration. This is also consistent with the findings of previous studies that have demonstrated the neuroprotective effects PBM mediated via activation of the PI3K/AKT pathway and the subsequent upregulation of neurotrophic factors^4,8,13,14.

Elevated levels of ROS have been established to impair mitochondrial membrane potential and function, thereby contributing to diabetic complications. Our findings in the present study indicate that PBM can contribute to modulating mitochondrial function and reduce oxidative stress in penile tissues. This is evidenced by reductions in ROS levels and an increase in ATP production in response to PBM treatment, as shown by the recovery of TMRM fluorescence. These findings are consistent earlier reports indicating the efficacy of PBM’in enhancing mitochondrial biogenesis and antioxidant defenses in different tissues^15–17.

Moreover, we also demonstrated the effects of PBM in promoting neurovascular regeneration in high-glucose environments. Specifically, PBM was observed to enhance neurite sprouting in MPG and DRG explants subjected to elevated high glucose levels, thus providing evidence indicating that this treatment may counteract the detrimental effects of hyperglycemia on neuronal growth and function. Furthermore, it was observed that PBM increased the expression of cell–cell junction proteins (ZO-1 and VE-cadherin) in endothelial cells thereby indicating improved vascular integrity. This aligns with earlier findings that demonstrate the potential of PBM to strengthen endothelial barrier function and reduce vascular leakage^16,18. The enhanced phosphorylation of endothelial nitric oxide synthase (p-eNOS) and AKT (p-AKT) further indicates improved endothelial function and nitric oxide production, which are essential for vasodilation and cell survival^5,7,12. The PBM-promoted upregulation of these signaling pathways enhances vasodilation and blood flow, which may contribute to improved erectile function by promoting nitric oxide production and inhibiting apoptosis in the corpus cavernosum^19,20.

In addition to these beneficial effects, a particularly important aspect of this study, is that we utilized LED-based PBM, which offers several advantages over laser-based PBM, including the provision of non-invasive broad-spectrum light treatment that effectively penetrates deep tissues and promotes cellular regeneration without causing thermal damage. The broader range of wavelengths associated with LED irradiation is conducive to more tailored treatment approaches, thereby enhancing the applicability of this modality for clinical use. The consistent improvements observed in this study serve to emphasize the efficacy of PBM, particularly LED-based modalities, in mitigating the deleterious effects of diabetes on erectile tissue. This in turn highlights the potential for integrating PBM into clinical practices for managing diabetic-induced erectile dysfunction, and plausibly other related vascular and neurological complications.

Our findings in this study provide convincing evidence in support of the therapeutic potential of LED-based PBM in mitigating erectile dysfunction in diabetic mice. The multi-faceted effects of PBM on multiple molecular pathways, as well as its capacity to promote neuro-vascular regeneration, highlight its potential as a novel treatment approach for this debilitating condition. The superior performance of LED-based PBM compared with traditional laser-based therapies highlights its potential as an effective non-invasive treatment for diabetic-induced erectile dysfunction. Consequently, additional studies and clinical trials are necessary to evaluate the relevance and effectiveness of LED-based PBM in treating human subjects.

Materials and methods

Ethics statement and animal study design

The research included the examination of 8-week-old male C57BL/6 J strain mice, totaling 110 specimens, acquired from Orient Bio (Seongnam-si, Gyeonggi-do, Republic of Korea). All procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Inha University (approval number: 210114-744) and were conducted in accordance with the National Institutes of Health guidelines for animal research. Additionally, we confirm that all experimental procedures have been reported in compliance with the ARRIVE guidelines for reporting animal experiments.

In order to induce DM, the dosage of streptozotocin (STZ) was 50 mg/kg body weight in 0.1 M citrate buffer (pH 4.5) administered intraperitoneally for five days in a row, following the method previously reported²¹. After eight weeks of induction, the Accu-Chek blood glucose meter (Roche Diagnostics) was used to monitor blood glucose levels. A level more than 300 mg/dL was considered to be an indication of DM. To ensure consistency and confirm sustained hyperglycemia, random blood glucose levels were measured in all diabetic animals at both Week 8 (post-STZ induction) and again at Week 10, immediately prior to intracavernosal pressure (ICP) measurement. These repeated assessments verified that hyperglycemia was maintained in all diabetic groups throughout the experimental period, minimizing inter-group variability and confirming the validity of the diabetic model prior to downstream functional analyses.

All mice were housed in plastic cages with ad libitum access to food and water in a climate-controlled environment under a 12-h light/dark cycle. Five age-matched groups categorized according to the kind of LED used for PBM treatment: (1) a normal glucose control group subjected to heat treatment, (2) a DM group subjected to heat treatment (3) a DM group that received RED LED PBM treatment, (4) a DM group that received near-infrared (NIR) LED PBM treatment, and (5) a DM group that received a combination of RED and NIR LED PBM treatments. PBM treatment was administered for 30 min daily, commencing 8 weeks after STZ injection and DM confirmation, and was continued until 10 weeks post-injection. All study procedures were performed in a double-blinded manner.

Evaluation of erectile function

Erectile function was assessed by monitoring intracavernous pressure (ICP), as outlined in prior research^21,22. Anesthesia was induced in mice by intraperitoneal administration of xylazine (5 mg/kg) and ketamine (100 mg/kg). We used a tail-cuff blood pressure monitoring device (Visitech Systems, Apex, NC, USA) to measure systemic blood pressure. The lower midline of the abdomen was incised using an incision that was 2 to 3 cm wide, cutting through the peritoneum and skin. The bladder and prostate were gently moved aside to expose the cavernous nerve, which was electrically stimulated using a 25G bipolar platinum electrode (Biopac Systems Inc., Santa Barbara, CA, USA) positioned under the nerve, and a 25G needle was used to introduce a pressure transducer into the exposed penile crus (Biopac Systems Inc., Santa Barbara, CA, USA).

To prevent the needle tip from being blocked by blood clots, approximately 30 µL of a solution containing heparin (250 U/mL) was injected into the cavernous area. The stimulation settings established to 5 V, 12 Hz, a 1-ms pulse width, and a 1 min duration, and ICP was quantified using Biopac Student Lab software (BSL 3.7.7; Biopac Systems Inc., Santa Barbara, CA, USA). Maximal intracavernossal pressure (maxICP) and total ICP shown by under the curve area, were assessed during tumescence. To normalize systemic blood pressure variations, the ratios of the area under the curve (total ICP) or maximum ICP (cm H2O) to the mean systolic blood pressure (MSBP) were calculated.

Cell culture

The mouse cavernous originated endothelial cells (MCEC) were primarily cultivated from mouse cavernous tissues and endothelial cells were initially isolated and subsequently cultured in M199 medium supplemented with necessary complements, following previously described methods^22–24. Surgically removed penile tissues were carefully deposited in sterile vials that contained Hanks’ Balanced Salt Solution (Cat# 14025092; Gibco, Carlsbad, California, USA). The tissue was rinsed twice with phosphate-buffered saline (PBS) prior to excising the penis glans, dorsal neurovascular bundle, and urethra, retaining solely the corpus cavernosum tissue for MCEC cultivation. Experiments were performed using cells passaged two to four times.

To simulate diabetic vasculopathy and neuropathy, Cells were deprived of serum overnight and subsequently exposed to high glucose (HG, 30 mM glucose) or normal glucose (NG, 5 mM glucose; Sigma-Aldrich) conditions for 3 days (for MCECs) or 5 days for major pelvic ganglia (MPG), dorsal root ganglia (DRG), and aortic rings maintained at 37 °C in a humidified environment with 5% CO₂²⁵.

Photobiomodulation treatment

PBM was administered using an array of 162 LED bulbs. In order to expose the ventral region of the mice directly to light, the experimental design included placing an LED device at the base of the cage enclosure at a fixed distance of 4 cm. The mice underwent daily light therapy while maintaining unrestricted mobility during the PBM treatment. Administering sedation and confinement to animals during treatment may have improved PBM dosimetry; however, this approach was considered unethical due to the extended treatment schedule and preliminary studies suggesting that direct contact PBM therapy under anesthesia led to increased mortality rates⁴.

Mice in each experimental group were subjected to either heat treatment, RED light, NIR light, or a combination of RED and NIR light for 30 min daily over a span of 14 consecutive days, starting 8 weeks post-STZ injection and DM confirmation. The RED light used in the in vivo experiments was emitted at a wavelength of 660 nm ± 3%, with an intensity of 46.8 mW/cm² (voltage: 16.00 V, current: 0.3 A, total energy: 4.68 J/cm²). NIR light was emitted at a wavelength of 830 nm ± 2%, with an intensity of 85.3 mW/cm² (voltage: 12.00 V, current: 0.75 A, total energy: 8.5 J/cm²).

For in vitro experiments, the RED light was set at 660 nm ± 3%, with an intensity of 36.0 mW/cm² (voltage: 12.10 V, current: 0.3 A, total energy: 3.6 J/cm²). The NIR light used for treatment of the in vitro control and heat-treated DM groups was set at 830 nm ± 2%, with an intensity of 81 mW/cm² (voltage: 9.0 V, current: 0.91 A, total energy: 81 J/cm²). To mitigate radiation and light scattering effects during heat treatment the LED bulbs were enclosed within aluminum foil throughout the procedures.

TUNEL assay

MCECs cultured under NG or HG conditions for three days were subsequently subjected to treatment with either heat or PBM. Cell death was evaluated using a TUNEL assay, adhering to the protocol supplied with the ApopTag Fluorescein in Situ Apoptosis Detection Kit (Catalog # S7111; Chemicon, Temecula, CA, USA). Apoptotic cells were quantified across different fields of view at ×200 magnification utilizing a confocal fluorescence microscope (K1-Fluo) and quantitative analysis of the fluorescence signals was performed using ImageJ software (version 1.34, NIH, http://rsbweb.nih.gov/ij/).

Analysis of neurite outgrowth

MPG and DRG tissues were collected and preserved according to an established protocol^4,26. DRG tissues attached to the Lumbar 3 (L3) to Lumbar 5 (L5) regions and MPG tissues tightly attached to the ventral prostate area were removed from 8-week-old C57BL/6 male mice under a dissecting microscope. These tissues were placed in Hanks’ Balanced Salt Solution, rinsed, and washed twice with PBS. The tissues were subsequently sectioned into 1–2 mm fragments, placed on poly-D-lysine hydrobromide-coated six-well glass plates, and overlaid with Matrigel (BD Biosciences, Franklin Lakes, NJ, USA), and incubated at 37 °C for 10 min using 1 ml of Gibco complete neurobasal media, serum-free B-27 (Gibco), and (0.5 nM) GlutaMAX™-I 2% in a 5% CO2 setting subjected to NG or HG conditions and treated with heat or PBM.

Neurite outgrowth was evaluated after 5 days of PBM treatment. The tissues were preserved in 4% paraformaldehyde for 1 h, washed multiple times with PBS. Immunofluorescent labeling was performed using an anti-βIII tubulin antibody at a dilution of 1:100 (Abcam). The area of neurite outgrowth was quantified using ImageJ software (version 1.34), by measuring the area of βIII tubulin-positive labeling, with the images thus obtained being converted to 8-bit graphics to facilitate the selection of the region of interest (ROI). The region of optimal neurite sprouting was determined by adjusting the threshold and selection tools, and any residual noise not eliminated by thresholding was removed manually.

Aortic ring assay

Abdominal and thoracic aorta of 8-week-old C57BL/6 mice were placed in the wells of an eight-well Nunc Lab-Tek Slide System (Sigma-Aldrich) then overlaid with with 50 μL of Matrigel. Aortic rings were cultured in full M199 media for five days under NG or HG conditions, with subsequent treatment using either heat or PBM as specified. Aortic segments and outgrowth were assessed via phase-contrast microscopy and quantified with ImageJ software.

Histological examinations

Penile tissues designated for immunofluorescence were fixed in 4% paraformaldehyde overnight. The tissue was rapidly frozen in OCT gel and sectioned into 10-μm thickness. Before the analysis, the slide underwent multiple washes with PBS (Gibco, Carlsbad, CA, USA) to eliminate the OCT. Tissue sections were incubated overnight at 4 °C with primary antibodies against, anti-βIII tubulin antibody (1:100; Abcam, Cambridge, MA, USA), neuron-glial antigen-2 (NG-2, 1:100; Millipore, Temecula, CA, USA), neurofilaments (NF, 1:100; Sigma-Aldrich, St. Louis, MO, USA), platelet endothelial cell adhesion molecule-1 (PECAM-1, 1:100; Millipore), oxidized low-density lipoprotein (Ox-LDL, 1:100; Abcam), nitrotyrosine (1:100; Millipore), vascular endothelial growth factor (VEGF, 1:100; Novus Biologicals, Littleton, CO, USA), phospho-endothelial nitric oxide synthase (p-eNOS, 1:100; Cell Signaling, Beverly, MA, USA), phosphohistone H3 (pHH3, 1:100; Millipore), zonula occludens-1 (ZO-1, 1:100; Millipore), vascular endothelial cadherin (VE-cadherin, 1:100; Santa cruz), anti-BrdU antibody (1:200; Bio-Rad, Hercules, CA, USA), tetramethylrhodamine-methyl ester-perchlorate (TMRM, 1:200 nM; Sigma-Aldrich), mitotracker (Green, FM, 1:200 nM; Invitrogen, Carlsbad, CA, USA). After a 10-min rinse with PBS, the slides were incubated for 2 h at room temperature with species-specific secondary antibodies conjugated to either tetramethyl-rhodamine-isothiocyanate (TRITC) or fluorescein-isothiocyanate (FITC) at a dilution of 1:300 (Zymed Laboratories, South San Francisco, CA, USA). The slides were then mounted using a DAPI-containing nuclear stain (Vector Laboratories, Inc., Burlingame, CA, USA). Fluorescence signals were visualized using a confocal microscope (K1-Fluo; Nanoscope Systems, Inc., Yuseong-gu, Daejeon-si, Korea). Quantitative analysis was performed using ImageJ (NIH version 1.34), available at http://rsbweb.nih.gov/ij/.

The quantification of immunofluorescent images involved adjusting threshold parameters using the “Image–Adjust–Threshold” function to accurately identify areas of interest, with a dark background setting employed to enhance fluorescence contrast. Once the threshold was established, the “Analyze–Set Measurements” and “Limit to Threshold” functions were applied. Regions of Interest (ROIs) were then manually refined using tools like freehand or ellipse selection. The“ROI Manager”was utilized to define and restrict quantification parameters to the selected areas. Finally, the “Analyze–Measure” command in the ROI Manager was used to obtain measurement results, while the “Analyze Particles” function facilitated cell analysis and intensity evaluations within the designated ROIs. Ensuring consistent staining, image acquisition, processing methods, and settings, as well as using a uniform background, was critical for accurate comparisons across multiple specimens.

Western blotting

After assessing erectile function, the penile tissue was immediately stored in a container with liquid nitrogen for preservation. A small volume of liquid nitrogen was used to crush the tissue. Lysates of the tissue preparations were obtained by treating with RIPA buffer (Sigma-Aldrich) containing protease and phosphatase inhibitors (1:100; Sigma-Aldrich), with the protein content of these lysates being determined using an ELx800G Universal Microplate Reader (BioTek Instruments Inc., Winooski, VT, USA). Protein samples (30 µg) were separated electrophoretically on 8–12% sodium dodecyl sulfate–polyacrylamide gels and then transferred to polyvinylidene fluoride membranes.

After subsequently blocked the membranes with 5% non-fat milk for 1.5 h at room temperature, membranes were treated overnight at 4 °C with primary antibodies targeting phospho-endothelial nitric oxide synthase (p-eNOS, 1:3000; Cell Signaling), endothelial nitric oxide synthase (eNOS, 1:3000; BD Biosciences), p-AKT (1:3000, Cell Signaling), AKT (1:3000, Cell Signaling), angiopoietin-1 (ab8451, 1:3000; Abcam), vascular endothelial growth factor (VEGF, 1:3000; Santa Cruz), fibroblast growth factor 2 (FGF-2, 1:3000; Santa Cruz Biotechnology Inc, Dallas, TX, USA), nerve growth factor (NGF, 1:3000; Santa Cruz), neurotrophin-3 (NT-3, 1:3000; Santa Cruz), phospho-phosphoinositide-3-kinase (p-PI3K, 1:3000, Cell Signaling), brain-derived neurotrophic factor (BDNF, 1:3000; Santa Cruz), and glyceraldehyde-3-phosphate-dehydrogenase (GAPDH, 1:5000, ABclonal, Woburn, MA, USA), after which, they were washed five times with PBS-T. Signals were visualized using an ECL detection system (Amersham Pharmacia Biotech, Piscataway, NJ, USA).

After primary incubation, membranes were washed five times with PBS-T, and signals were visualized using an ECL detection system (Amersham Pharmacia Biotech, Piscataway, NJ, USA). To accommodate multiple targets with varying molecular weights, membranes were physically spliced after protein transfer and blocking but prior to antibody incubation. This approach was used to optimize reagent use and probe multiple proteins efficiently. We confirm that no part of the gels was manipulated or artificially spliced—all protein bands presented originated from a continuous gel run. Image cropping was performed solely to enhance clarity and focus on bands of interest. The corresponding full, unedited membrane images, including molecular weight markers and marked regions of interest, have been provided in the supplementary data for transparency.

Statistical analysis and image post-processing

Immunofluorescence and western blot band densities were quantitatively evaluated using ImageJ image analysis software. The data are expressed as means ± standard error of the mean (SEM). Statistical analyses were conducted utilizing GraphPad Prism version 10.2.3 (GraphPad Software, San Diego, CA, USA). The Newman–Keuls post hoc test was conducted following ANOVA analysis. A p-value below 0.05 was deemed to indicate statistical significance.

Author contributions

Conceptualization: LA, CMJ, JKS, JKR; Data curation: LA, CMJ; Formal Analysis: LA, CMJ; Funding acquisition: JKR, LA; Investigation: LA, CMJ, GNY, OJY, MHK; Methodology: GNY, OJY, MHK; Project administration: JKS, JKR; Resources: LA, CMJ; Software: GNY, MHK, OJY; Supervision: JKS, JKR, SSH, JHK; Validation: LA, JKS, JKR, SSH, JHK; Visualization: JKS, JKR, SSH, JHK; Writing-original draft: LA; Writing-review and editing: JKS, JKR.

Data availability

The datasets utilized and/or analyzed in this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-04873-w.