Bakh formula ameliorates high-fat diet–induced erectile dysfunction in rats via coordinated restoration of metabolic and tissue homeostasis

Abstract

Introduction

Erectile dysfunction (ED) associated with long-term high-fat diet (HFD) intake is increasingly recognized as a manifestation of systemic metabolic and vascular dysfunction. This study aimed to evaluate the therapeutic effects of the classical Uyghur medicine formula Bakh Formula (BAKHF) on HFD-induced ED in rats and elucidate its underlying mechanisms.

Methods

Male rats with HFD-induced ED were randomly assigned to model, Yimusake (YMSK, positive control), and BAKHF groups. YMSK served as a pharmacological benchmark to validate the model and evaluate BAKHF’s effects on erectile function, histology, and hematology. Erectile function was evaluated by maximal intracavernosal pressure to mean arterial pressure (ICP/MAP), sexual behavior tests, and biochemical assays. The primary outcome was erectile function assessed by ICP/MAP, with secondary outcomes including metabolic, endocrine, molecular, and histopathological parameters. The potential mechanisms of BAKHF were elucidated using integrated transcriptomic, proteomic, and network pharmacology analyses, combined with chemical profiling of serum-absorbed compounds, molecular docking, RT-qPCR, and Western blot.

Results

Both BAKHF and YMSK improved erectile function, characterized by elevated ICP/MAP ratios and improved sexual behavior parameters. Histologically, both formulas restored penile cavernous structure, reduced collagen deposition, and increased smooth muscle content, confirming YMSK’s role as a reliable phenotypic benchmark for the model. However, BAKHF exhibited superior efficacy in restoring systemic metabolic and endocrine homeostasis, including normalization of food intake and serum lipid profiles. Subsequent mechanistic analyses focused on BAKHF and identified its regulatory role in AMPK-associated metabolic and cytoskeletal pathways. Five serum-absorbed compounds, including rosmarinic acid and ferulic acid, were identified as potential bioactive constituents targeting key nodes like GAPDH, PPARγ, and ESR1.

Discussion

These findings suggest that BAKHF acts through multi-component, multi-target, and multi-pathway synergy, potentially involving AMPK signaling to regulate metabolic and tissue homeostasis in HFD-induced ED. The integration of multi-omics analyses with experimentally confirmed serum-absorbed compounds and mechanistic validation strengthens the biological plausibility of the findings. However, limitations include the lack of pathway-specific genetic or pharmacological interventions and unresolved contributions of individual constituents. Targeting systemic metabolic and tissue homeostasis through multi-component herbal formulas may represent a complementary therapeutic strategy for metabolism-related ED.

Graphical Abstract

Graphical Abstract.

Introduction

Erectile dysfunction (ED) is a prevalent male health disorder worldwide.¹ Recent shifts toward high-fat diets (HFDs) have been accompanied by a rising incidence of ED, increasingly affecting younger men.² Accumulating epidemiological and experimental evidence has identified long-term HFD intake as a critical risk factor for ED.^3,4 HFD–induced metabolic disturbances impair penile smooth muscle function and ultimately lead to erectile impairment.^5,6 Importantly, HFD-associated ED is increasingly considered a manifestation of underlying metabolic and vascular dysfunction rather than an isolated penile disorder.^7,8 Previous studies have shown that ED frequently occurs 3-5 years before coronary artery disease, highlighting its potential in early cardiovascular risk stratification.⁹ Therefore, understanding the impact of HFD on ED is important for public health and for developing effective preventive and therapeutic strategies.

Traditional medicine has demonstrated potential benefits in the management of chronic diseases,¹⁰ metabolic dysregulation,¹¹ and reproductive health disorders.¹² In the context of HFD-induced ED, traditional medical systems emphasize a holistic perspective, viewing metabolic, circulatory, and reproductive abnormalities as interconnected pathological processes.¹³ Accordingly, therapeutic strategies aim to restore systemic homeostasis rather than merely alleviating local symptoms.¹⁴ Multi-herbal formulations exert their therapeutic effects through multi-component, multi-target, and multi-pathway mechanisms.^15,16 Growing evidence indicates that such approaches may offer sustained efficacy with fewer adverse effects and improved quality of life.^17,18 These characteristics make traditional multi-herbal therapies particularly attractive for the long-term management of HFD-induced ED, a condition driven by systemic metabolic disturbances.

Bakh Formula (BAKHF), recorded in the Encyclopedia of Chinese Medicine—Uyghur Medicine, is a classical Uyghur prescription widely used for the treatment of ED. Our preliminary studies demonstrated that BAKHF markedly improved erectile function and partially corrected metabolic abnormalities in rats with HFD-induced ED. However, the pharmacological basis and molecular mechanisms underlying the therapeutic effects of BAKHF remain largely unclear, particularly regarding how it modulates metabolic and tissue homeostasis at the molecular level.

Yimusake (YMSK), a traditional Uyghur medicine used for male reproductive disorders, including ED, has been widely applied in clinical practice.¹⁹ Previous studies have shown that YMSK improves erectile function by enhancing endothelial function, regulating nitric oxide signaling, and attenuating oxidative stress and inflammation.^20,21 Given that these processes are central to the pathogenesis of HFD-induced ED, YMSK was included as a positive control to provide a pharmacological benchmark for evaluating the efficacy of BAKHF.

In the present study, we evaluated the therapeutic effects of BAKHF in a rat model of HFD-induced ED and sought to elucidate its underlying mechanisms. Integrated transcriptomic and proteomic analyses were employed to characterize global molecular alterations in penile tissue associated with HFD-induced ED. These approaches were combined with chemical profiling of serum-absorbed compounds and network pharmacology analysis to explore the multi-target and multi-pathway actions of BAKHF and to identify potential key regulatory nodes. Key molecular changes were further validated using targeted molecular biology techniques. Through this integrative strategy, we aimed to systematically elucidate how BAKHF ameliorates HFD-induced ED. Specifically, we asked: Can BAKHF improve erectile function in rats with HFD-induced ED, and are its therapeutic effects mediated through coordinated restoration of metabolic and tissue homeostasis?

Materials and methods

Preparation of BAKHF

The constituent herbs of BAKHF (Table S1) were purchased from Madison Pharmaceutical Co., Ltd. (Xinjiang, China). All botanical drugs were authenticated by a qualified pharmacognosist following standard processing guidelines in accordance with the Pharmacopoeia of the People’s Republic of China (2020 edition). Voucher specimens were deposited in the laboratory (Voucher No. 20230809).

The herbs were mixed according to the prescribed ratios and extracted with purified water at a herb-to-solvent ratio of 1:10 (w/v) under reflux at 100°C for 1 hour using a thermostatic water bath (XMTD-7000, Beijing Yongguangming Medical Instrument Co., Ltd., China). The extracts were filtered under reduced pressure using a vacuum filtration system (Tianjin Jinteng Technology Co., Ltd., China), pooled and concentrated under reduced pressure with a rotary evaporator (OBS-2100, Shanghai Yarong Biochemical Instrument Co., Ltd., China), The concentrated extract was further dried in a vacuum drying oven (DZF-6051, Shanghai Qixin Scientific Instrument Co., Ltd., China) to obtain a dry extract, which was stored at 4°C until use. The rat dosage was calculated from the adult clinical dose using the body surface area conversion method: (1.048 g dry extract/60 kg body weight) × 6.3. Based on preliminary experiments, the medium dose administered by oral gavage was selected for therapeutic evaluation; therefore, no additional dose groups were included in this study.

Reagents

Apomorphine (APO, A4393) and LC–MS–grade solvents, including methanol (≥99.9%, 106035), acetonitrile (≥99.9%, 100029), and formic acid (FX0440), were purchased from Sigma–Aldrich (Shanghai, China). Commercial kits for serum lipid analysis—total cholesterol (TC, 105-000448), triglycerides (TG, 105-000449), high-density lipoprotein cholesterol (HDL-C, 105-000463), low-density lipoprotein cholesterol (LDL-C, 105-000449)—and biochemical quality controls (059323002) were obtained from Sayo Biomedical Technology (Xinjiang, China). Radioimmunoassay kits for follicle-stimulating hormone (FSH, S10950154), luteinizing hormone (LH, S10950161), prolactin (PRL, S10950151), estradiol (E2, S19973002), and testosterone (T, S10940093) were purchased from Beijing North Institute of Biological Technology (Beijing, China).

Hematoxylin–eosin (H&E) staining solution (BL700A; Biosharp Biotechnology, Anhui, China) and a modified Masson’s trichrome staining kit (G1346; Solarbio Life Sciences, Beijing, China) were used for histopathological analysis. Primary antibodies against Cluster of Differentiation 31 (CD31, AF6191), Von Willebrand Factor (VWF, AF3000), Endothelin-1 (ET-1, DF6125), and Alpha-Smooth Muscle Actin (α-SMA, AF1032) were purchased from Affinity Biosciences (USA). Collagen Type I (Collagen I, ab34710, Abcam, UK), β-Tubulin (10094-1-AP), Myosin Light Chain 1 (MYL1, 15 814-1-AP), Tropomyosin 2 (TPM2, 11 038-1-AP), Matrix metalloproteinase 9 (MMP9, 27 306-1-AP), Peroxisome Proliferator-Activated Receptor Gamma (PPARγ,16 643-1-AP), Estrogen Receptor 1 (ESR1, 21 244-1-AP) were purchased from Proteintech Group (Wuhan, China). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH, bsm-33033M, Bioss, Beijing, China). FastKing Reverse Transcription Kit (KR116-02, Tiangen Biotech, Beijing, China), SuperReal Enhanced SYBR Green Premix (FP205-02, Tiangen Biotech, Beijing, China), Beta-Actin (β-actin; B661202, Sangon Biotech, Shanghai, China).

Animals and treatment

Specific pathogen-free (SPF) Sprague–Dawley (SD) rats, including male (6 weeks old, 200 ± 10 g) and female (7 weeks old, 190 ± 10 g), were obtained from an accredited laboratory animal center. Animals were housed under SPF conditions at 22 ± 2°C and 50 ± 10% relative humidity, with a 12 hours light/dark cycle, and had free access to food and water. After one week of acclimatization, male rats were screened to establish baseline sexual proficiency. The specific inclusion criteria (shortcut numbers) for sexually proficient males were defined as achieving ≥2 erections during a 30-minute APO-induced test (90 μg/kg, subcutaneously). To adhere to animal welfare principles, rats failing to meet these thresholds were not sacrificed but reassigned to separate non-behavioral pilot studies for tissue optimization.

ED was induced by feeding male rats an HFD (60% calories from fat, HFD-H10060, Huafu Kang Biotechnology, Beijing, China) for 8 weeks. After model establishment, 30 HFD-induced ED rats were randomly assigned (n = 10 per group) to the model group (M), YMSK group (250 mg/kg, purchased from Hetian Urumqi Pharmaceutical Co., Hetian, China, 20 230 306), or BAKHF group (110 mg/kg). The YMSK and BAKHF doses were calculated based on human clinical dosage using body surface area conversion. Treatments were administered once daily by oral gavage at 17:00 for three consecutive weeks, while normal control (NC) and M groups received an equal volume of physiological saline. All evaluations were performed under blinded conditions. At the end of the experiment, rats were euthanized by exsanguination under deep anesthesia. All animal procedures were approved by the Animal Ethics Committee (IACUC Approval No. 20240228-26, February 28, 2024).

Assessment of erectile function

Erectile function was evaluated via APO-induced erection tests, sexual behavior assessments, and ICP/MAP measurements on separate days. For the APO test, rats received APO (90 μg/kg, subcutaneously) and were placed in a darkened room with 5 minutes of adaptation followed by 30 minutes of observation to record erection frequency and latency. Sexual behavior was assessed in a competitive environment (3:2 male-to-female ratio). Three color-coded males were co-housed with 2 estrous females. Each session consisted of 5 min of adaptation followed by 30 min of observation. Temporal parameters—Mounting Latency, Intromission Latency, and Ejaculation Latency—were recorded as the total time from female introduction to the first occurrence of each event. Frequencies (MF, IF, EF) were recorded as cumulative counts over the 30-minute observation period.^22–24 Detailed definitions are provided in Table S2. To ensure maximum sexual receptivity, stimulus females were bilaterally ovariectomized prior to the experiments. Behavioral estrus was induced by subcutaneous injections of estradiol benzoate (20 μg/rat) at 48 hour and 24 hours pre-test, followed by progesterone (500 μg/rat) 4 hours before testing.²⁵ Estrus was strictly verified via Wright–Giemsa stained vaginal smears; only females exhibiting a field dominated by anucleated cornified epithelial cells were used (Figure S1).

The APO–induced erection test, sexual behavior assessment, and ICP/MAP measurements were conducted on separate days in the same cohort of animals to avoid procedural interference. All rats underwent both behavioral assessments and ICP/MAP measurements. For ICP and MAP measurements, rats were anesthetized with sodium pentobarbital (40 mg/kg, intraperitoneally). The pelvic ganglion and cavernous nerve were exposed via a midline abdominal incision. A 24-gauge needle prefilled with heparinized saline (250 IU/mL) was inserted into the left corpus cavernosum and connected to a PE-50 tube for ICP measurement, while the left carotid artery was cannulated with a PE-50 catheter for MAP recording. Both were connected to a pressure transducer (PT-103, Chengdu Taimeng Software Co., Ltd., China) and recorded using a BL-420F biological signal acquisition system (Chengdu Taimeng Software Co., Ltd., China). The cavernous nerve was stimulated using a bipolar electrode (15 Hz, 7.5 V) for 60 seconds. Erectile function was evaluated as the ratio of maximal ICP to mean MAP, consistent with previously reported methods.²⁶

Assessment of biological parameters

Body weight was recorded once weekly. Surface temperature was measured at a fixed time using an infrared thermal imager (testo 869, Testo SE & Co. KGaA, Germany), and core temperature was measured weekly using a digital rectal thermometer (Kefu Medical Technology Co., Ltd., China) inserted approximately 4 cm into the rectum. Daily food and water intake were recorded using metabolic cages (DX-10, Suzhou Guxiu Laboratory Animal Equipment Co., Ltd., China) and normalized per 100 g body weight. Urine and feces were collected over 24 hours; feces were dried in an oven (DHP-360, Beijing Yongguangming Medical Instrument Co., Ltd., China) at 60°C for 20 minutes and weighed to obtain dry weight, which was also normalized per 100 g body weight. All measurements were performed in triplicate for each animal, and mean values were used for statistical analysis.

Biochemical analysis

Rats were fasted overnight (12 hours) with free access to water, anesthetized with sodium pentobarbital, and subsequently subjected to ICP/MAP measurements. Immediately after completion, blood samples were collected from the abdominal aorta under deep anesthesia, and euthanasia was achieved by exsanguination. Approximately 5 mL was first collected into anticoagulant tubes for hemorheological analysis, including WBV-L, WBV-M, WBV-H, PV, Hct, and ESR-K, using a MEN-C blood rheology analyzer (Shandong Meiyilin Electronic Instrument Co., Ltd., China). Additional blood was collected into procoagulant tubes, allowed to clot at room temperature for 30 minutes, and centrifuged at 2150 × g for 10 minutes at 4°C using a 5810R centrifuge (Eppendorf, Germany). Serum was aliquoted and stored at −20°C for lipid and hormone analyses. Serum lipids (TG, TC, HDL-C, LDL-C) were measured using a BS-240VET automatic biochemical analyzer (Shenzhen Mindray Bio-Medical Electronics Co., Ltd., China) with commercially available enzymatic assay kits under quality control procedures. Pituitary–Gonadal axis hormones (FSH, LH, PRL, E2, T) were quantified using radioimmunoassay kits on an XH6080 radioimmunoassay system (Xi’an Nuclear Instrument Factory, China) according to the manufacturers’ instructions.

Histopathological analysis

At the end of the experiment, the penis, testes, pituitary, and hypothalamus tissues were dissected, rinsed with saline, and blotted. To ensure the precision of subsequent analyses, the superficial skin and urethra were meticulously removed from the penile tissue to isolate the pure bilateral corpus cavernosum. The posterior penis, testes, pituitary, and hypothalamus were fixed in 4% paraformaldehyde, while the remaining cavernous tissue was snap-frozen in liquid nitrogen and stored at −80°C for molecular analyses. Fixed tissues were dehydrated, cleared, embedded in paraffin, and sectioned at 4 μm using a rotary microtome (RM2235, Leica Biosystems, Germany). Sections were stained with H&E,²⁷ and penile sections were further stained with Masson’s trichrome.²⁸ Stained sections were examined under a light microscope (Nikon ECLIPSE Ci, Nikon Corporation, Japan), and representative images were captured. For each section, 5 non-overlapping fields were randomly selected at 200× magnification. Digital image analysis was conducted using ImageJ software (v1.53t, NIH, USA). In Masson’s trichrome staining, the smooth muscle-to-collagen ratio was calculated as the area of red-stained tissue divided by the area of blue-stained tissue. The final value for each rat represented the average of these 5 fields. For transcriptomic and proteomic analyses, isolated cavernous tissues (longitudinal sections) from the NC and M groups (n = 3 per group) were processed as independent biological replicates.

Immunohistochemical assessment of endothelial markers and smooth muscle–collagen balance in penile tissue

Endogenous peroxidase activity and nonspecific binding in penile tissue sections were blocked with 3% H₂O₂ and 5% BSA, followed by overnight incubation at 4°C with primary antibodies (CD31 1:100, VWF 1:100, ET-1 1:50, Collagen I 1:200, α-SMA 1:100). Sections were then incubated with antibody enhancer and secondary antibodies, developed with DAB, counterstained with hematoxylin, dehydrated, cleared, and mounted.²⁹ For each section, 5 non-overlapping fields were randomly selected at 200× magnification using a light microscope (Nikon ECLIPSE Ci, Japan). Images were analyzed using ImageJ software (v1.53t, NIH, USA). The percentage of positive staining area (calculated as positive area/total area × 100%) or the integrated optical density (IOD) was quantified to evaluate the expression levels of endothelial and smooth muscle markers. The final value for each rat was calculated as the average of these 5 fields.

Transcriptomic and proteomic analysis

Total RNA was extracted from penile tissues (NC and M groups) and sequenced using the Illumina HiSeq platform (BGI Genomics, Wuhan, China). Quality control was performed using ShortRead, Biostrings, adapterTrim, and PolyATrimmer. Reads were aligned to the reference genome with Bowtie2, and novel miRNAs were predicted with miRDeep2.³⁰ Differentially expressed genes (DEGs) were identified with FDR ≤ 0.05 and FC ≥ 2.³¹ For proteomic analysis, protein extraction, trypsin digestion, and Tandem Mass Tag labeling were performed following established protocols.³² LC–MS/MS analysis was conducted using an Orbitrap Fusion Lumos mass spectrometer (Thermo Scientific, USA) by BGI Proteomics (Beijing, China). Raw MS data were processed and quantified using Proteome Discoverer (v2.4, Thermo Scientific) against the UniProt rat database. Differentially expressed proteins were defined by P < .05 and fold change >1.19 or <0.84 (logFC ±0.25).³³ Functional enrichment analyses for both datasets were performed using the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases.³⁴

UPLC-QTOF/MS analysis of BAKHF constituents

Chemical profiling of BAKHF was performed using an Acquity H-Class UPLC coupled with a Xevo G2-XS QTof mass spectrometer (Waters, USA) with ESI according to established chemical identification guidelines.³⁵ BAKHF samples were mixed with four volumes of cold methanol, vortexed, and centrifuged (14 000 × g, 20 minutes, 4°C), and the supernatants were analyzed. Separation was achieved on an HSS T3 column (100 × 2.1 mm, 1.8 μm) at 40°C, flow rate 0.3 mL/minute, injection volume 2 μL, using 0.1% formic acid in water (A) and acetonitrile (B) with a gradient program. MS data were acquired in positive and negative ion modes (m/z 50-1200) with MSE acquisition, using leucine-enkephalin as the lock mass. Data were processed with MassLynx v4.2, and constituents were identified based on accurate mass, retention time, and MS/MS fragmentation, compared against MassBank, METLIN, and HMDB (≤10 ppm). Serum samples (0.3 mL) from NC and BAKHF rats were processed similarly to detect serum-absorbed compounds.³⁶

Network pharmacology analysis

To identify the potential targets of BAKHF in the treatment of ED, a systematic network pharmacology approach was employed according to established guidelines.³⁷ Potential ED-related targets were retrieved from GeneCards (https://www.genecards.org) using a relevance score threshold >5. DEGs and proteins from transcriptomic and proteomic analyses were defined as disease-related targets for ED. Targets of absorbed compounds identified in serum by LC–MS analysis were obtained from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) and predicted using SwissTargetPrediction (http://www.swisstargetprediction.ch/), SuperPred (https://prediction.charite.de/), and SEA databases (https://sea.bkslab.org/). Potential key targets were defined as the intersection of predicted, disease-associated, and experimental targets. Protein–protein interaction (PPI) networks were constructed using the STRING database (https://cn.string-db.org). Interaction networks were visualized using Metascape (https://metascape.org) and Cytoscape (v3.7.2).

Molecular docking

Protein structures were prepared using AutoDock Tools (ADT, v1.5.6) by removing water molecules and ions, adding polar hydrogens, and assigning Kollman charges. The chemical structures of the absorbed compounds (eg, rosmarinic acid) were obtained from PubChem database (https://pubchem.ncbi.nlm.nih.gov/) in 2D format, converted to 3D structures using Chem3D software (v16.0, PerkinElmer, USA), and energy-minimized. Molecular docking simulations were performed with AutoDock Vina (v1.1.2; exhaustiveness = 8), with binding affinities ≤ −5.0 kcal/mol considered indicative of stable ligand–protein interactions. Docking poses and interactions were visualized and analyzed using PyMOL (v2.4, Schrödinger, LLC) and Discovery Studio 2019 Client (Dassault Systèmes BIOVIA, San Diego, CA, USA).³⁸

RT-qPCR analysis

Total RNA was extracted from rat penile tissues and assessed for purity and concentration using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA), with acceptable purity defined as A260/A280 ≥ 1.8. First-strand cDNA was synthesized from 50 ng to 2 μg RNA, followed by qPCR using SYBR Green chemistry. Each sample was analyzed in triplicate, with no-template and no–reverse transcription controls included. qPCR was performed on an ABI QuantStudio™ 6 Real-Time PCR system (Applied Biosystems, USA) with the following cycling conditions: 95°C for 15 minutes, then 40 cycles of 95°C for 10 seconds and 60°C for 34 seconds. A melt curve analysis confirmed the specificity of the amplification. ACTB (β-actin) served as the internal reference, and relative expression was calculated using the 2^−ΔΔCt method.³⁹ Amplified products ranged from 80 to 200 bp. Primer sequences are listed in Table S3.

Western blot analysis

Total protein from rat penile tissues was extracted using RIPA buffer with PMSF, quantified by BCA assay, and denatured with 4× loading buffer. Equal amounts (10 μg per lane) were separated by 10% SDS–PAGE using a PowerPac HC Electrophoresis System (Bio-Rad, USA) and transferred to PVDF membranes. Membranes were blocked with 5% non-fat milk and incubated overnight at 4°C with primary antibodies (β-Tubulin 1:2000; GAPDH 1:10000; MYL1, TPM2, MMP9, PPARγ, ESR1 1:1000). After washing, membranes were incubated with HRP–conjugated secondary antibodies, and signals were detected using enhanced chemiluminescence. Chemiluminescence signals were captured and visualized using an Azure 600 Imaging System (Azure Biosystems, USA). Band intensities were quantified with ImageJ software (v1.53t, NIH, USA) and normalized to β-Tubulin.^6,40

Statistical analysis

Data are expressed as mean ± standard deviation (SD). Statistical analyses were performed using SPSS 26.0 (IBM Corp.). Normality and homogeneity of variance were assessed using the Shapiro–Wilk and Levene’s tests, respectively. Normally distributed data with homogeneous variances were analyzed using one-way ANOVA followed by Tukey’s HSD test, whereas non-normally distributed data were analyzed using the Kruskal–Wallis test followed by Dunn’s test with Bonferroni correction. A two-sided P-value < .05 was considered statistically significant. Sample size was determined based on previous studies and pilot experiments.

Results

BAKHF improves sexual function in HFD–induced ED rats

Sexual function was evaluated based on erectile responses and sexual behavior parameters. Compared with the normal control (NC) group, rats in the model (M) group exhibited significantly prolonged latencies and reduced frequencies of sexual behaviors, including erection, mounting, intromission, and ejaculation (P < .001), indicating marked sexual dysfunction (Figure 1A). After 3 weeks of treatment, BAKHF administration significantly shortened the behavioral latencies and increased the behavioral frequencies compared with the M group (P < .001). In contrast, the pharmacological benchmark YMSK failed to show statistically significant improvements in these parameters compared with the M group. Consistent with these findings, ICP/MAP was reduced in the M group compared with NC (P < .001), Following treatment, both BAKHF and YMSK significantly elevated ICP/MAP compared with the M group (P < .001). Notably, BAKHF demonstrated a marked restoration of erectile function, further confirming its pronounced therapeutic efficacy (Figure 1B and C).

Figure 1.

Effects of BAKHF on sexual function in HFD-induced ED rats. (A) Sexual behavior parameters; (B) representative ICP/MAP traces; (C) quantitative analysis of ICP/MAP. Data are presented as mean ± SD (n = 10). Statistical analysis was performed using Tukey’s HSD or Bonferroni-adjusted Dunn’s post hoc tests following normality and homoscedasticity verification. ns P > .05, ^**  P < .01, ^***  P < .001.

BAKHF improves metabolic–related physiological parameters in HFD–induced ED rats

The metabolic–related physiological parameters of rats in each group are summarized in Figure 2. Compared with the NC group, rats in the M group exhibited significantly increased body weight and core temperature (P < .001 for both; Figure 2A and B), along with significantly decreased fecal output, food intake, and surface temperature (P < .01 or P < .001; Figure 2C–E). In addition, urine output and water intake were markedly reduced in the M group (P < .01 or P < .001; Figure 2F and G).

Figure 2.

Effects of BAKHF on metabolic-related physiological parameters in HFD-induced ED rats. (A) Body Weight; (B) Core temperature; (C) Fecal Output; (D) Food Intake; (E) Surface Temperature; (F) Urine Output; (G) Water Intake. Data are presented as mean ± SD (n = 10). Statistical analysis was performed using Tukey’s or Bonferroni-adjusted Dunn’s post hoc tests following normality and homoscedasticity verification. ns P > .05, ^*  P < .05, ^**  P < .01, ^***  P < .001.

Following BAKHF treatment, significant suppression was observed in body weight gain (P < .01; Figure 2A), and a significant recovery was achieved in food intake (P < .05; Figure 2D), with no statistically significant differences compared with the NC group (P > .05). On the other hand, the YMSK group also showed partial improvements in body weight (P < .05); however, the overall effects were less pronounced than those observed in the BAKHF group. These results indicate that BAKHF partially improves metabolic–related physiological parameters in rats with HFD-induced ED.

BAKHF improves lipid metabolism, hemorheological parameters, and pituitary-gonadal axis hormone levels

Serum lipid analysis revealed that compared with the NC group, rats in the M group exhibited significantly elevated levels of TG, TC, and LDL-C, accompanied by a marked reduction in HDL-C (P < .001; Figure 3A). BAKHF treatment significantly reduced TG, TC and LDL-C while elevating HDL-C levels (P < .01 or P < .001), with values not significantly different from those of the NC group (P > .05). The YMSK group also showed partial lipid–lowering effects; however, these changes did not achieve statistical significance directly against the M group. Hemorheological analysis demonstrated that WBV-L, WBV-M, WBV-H, PV, and Hct were all significantly increased in the M group compared with the NC group (P < .05, P < .01 or P < .001; Figure 3B). Both BAKHF treatments significantly reduced WBV-L, WBV-M, and WBV-H (P < .05 or P < .001), whereas the YMSK group showed no statistically significant regulatory effects. Hormonal analysis of the pituitary-gonadal axis showed that compared with the NC group, serum PRL levels were significantly elevated (P < .05; Figure 3C). Following BAKHF treatment, these hormone alterations displayed an observable trend of baseline recovery, though the direct comparisons did not reach statistical significance against the M group.

Figure 3.

Effects of BAKHF on lipid metabolism, hemorheological parameters, and pituitary-gonadal axis hormone levels in HFD–induced ED rats. (A) Serum lipid profiles, including TG, TC, LDL-C, and HDL-C; (B) hemorheological parameters, including WBV-L, WBV-M, WBV-H, PV, Hct, and ESR-K; (C) serum pituitary-gonadal axis–related hormone levels, including FSH, LH, PRL, E2, and T. Data are presented as mean ± SD (n = 8). Statistical analysis was performed using Tukey’s or Bonferroni-adjusted Dunn’s post hoc tests following normality and homoscedasticity verification. ns P > .05, ^*  P < .05, ^**  P < .01, ^***  P < .001.

BAKHF attenuates histopathological damage in pituitary-gonadal axis–related tissues and penile tissue

H&E staining showed no obvious morphological differences in hypothalamic tissue among the groups (Figure 4A). In contrast, compared with the NC group, the M group exhibited pronounced vacuolar degeneration in pituitary cells, disorganized seminiferous tubules with germ cell necrosis and exfoliation in the testes, and marked structural damage in penile cavernous tissue, including sinusoidal dilation. Following treatment, pathological alterations in the pituitary, testes, and penile tissues were alleviated to varying degrees in both the YMSK and BAKHF groups, with the most prominent improvements observed in both the YMSK and BAKHF group, with the most prominent morphological improvements observed in the BAKHF group. Masson’s trichrome staining and quantitative analysis (Figure 4B and C) further revealed that the smooth muscle-to-collagen ratio in penile tissue was significantly decreased in the M group (P < .001), indicating severe fibrosis of the corpus cavernosum. Compared with the M group, both BAKHF and YMSK treatments significantly elevated the smooth muscle/collagen ratio (P < .001). Strikingly, BAKHF treatment demonstrated a superior therapeutic efficacy, successfully restoring the smooth muscle/collagen ratio to baseline levels that exhibited no statistically significant difference compared with those of the NC group (P > .05).

Figure 4.

BAKHF alleviates histopathological damage in pituitary-gonadal axis–related tissues and penile corpus cavernosum in HFD–induced ED rats. (A) H&E staining of the hypothalamus, pituitary gland, testis, and penis (Scale bars = 20 μm); (B) Masson’s trichrome staining of the penile corpus cavernosum (Scale bars = 20 μm); (C) quantitative analysis of the smooth muscle/collagen ratio in penile tissue. Data are presented as mean ± SD (n = 7). Statistical analysis was performed using Tukey’s or Bonferroni-adjusted Dunn’s post hoc tests following normality and homoscedasticity verification. ns P > .05, ^***  P < .001.

Effects of BAKHF on endothelial, fibrotic, and smooth muscle markers in penile tissue

Immunohistochemical analysis demonstrated that compared with the NC group, the positive staining areas of CD31 and VWF in penile tissue were significantly reduced in the M group (P < .001; Figure 5A–C), whereas ET-1 and collagen I expression levels were markedly increased (P < .001; Figure 5D and E). In addition, the α-SMA positive area was significantly decreased (P < .001; Figure 5F), indicating endothelial injury, aggravated fibrosis, and reduced smooth muscle content. Compared with the M group, both YMSK and BAKHF treatments ameliorated these alterations to varying extents. Notably, the BAKHF group exhibited more pronounced and robust therapeutic efficacy than the YMSK group. BAKHF administration achieved highly significant reversals across all tested markers (P < .001), successfully driving the expression levels of CD31, VWF, ET-1, collagen I, and α-SMA back to baseline levels that were not significantly different from those of the NC group (P > .05). In contrast to the limited efficacy of YMSK on VWF and α-SMA, BAKHF comprehensively reversed all pathological alterations, emphasizing its multi-target protective capacity in the penile microenvironment.

Figure 5.

Effects of BAKHF on endothelial injury, fibrosis, and smooth muscle markers in penile tissue of HFD–induced ED rats. (A) Representative IHC staining of CD31, VWF, ET-1, Collagen I, and α-SMA in penile tissue (Scale bars for lower-magnification panels = 20 μm; Scale bars for high-magnification inserts [Enlarge] = 4 μm. Red arrows indicate positively stained structures.); (B–F) quantitative analysis of CD31, VWF, ET-1, Collagen I, and α-SMA positive areas, respectively. Data are presented as mean ± SD (n = 6). Statistical analysis was performed using Tukey’s or Bonferroni-adjusted Dunn’s post hoc tests following normality and homoscedasticity verification. ns P > .05, ^*  P < .05, ^**  P < .01, ^***  P < .001.

Transcriptomic and proteomic analyses reveal AMPK-associated metabolic and cytoskeletal alterations in HFD–induced ED rats

Transcriptomic analysis identified 34 DEGs in the M group compared with the NC group, including both upregulated and downregulated genes (Figure 6A and Table S4). KEGG pathway enrichment analysis indicated that these DEGs were significantly enriched in the AMPK signaling pathway (adjusted P = 0.011435; Figure 6B), while GO analysis revealed significant associations with metabolism–related biological processes (Figure S2). Gene set enrichment analysis further demonstrated upregulation of multiple energy metabolism–related genes in the M group, suggesting marked alterations in AMPK-associated pathways (Figure 6C).

Figure 6.

Transcriptomic and proteomic analyses reveal metabolic and cytoskeletal alterations in penile tissue following HFD. (A) Transcriptomics volcano plot showing differentially expressed genes in M rats compared with NC rats; (B) KEGG enrichment analysis of transcriptomics; (C) GSEA of AMPK signaling pathway; (D) proteomics volcano plot showing differentially expressed proteins in M rats compared with NC rats; (E) KEGG enrichment analysis of proteomics; (F) GSEA of cytoskeleton in muscle cells pathway and motor proteins pathway.

Proteomic analysis identified 169 differentially expressed proteins, including 123 upregulated and 46 downregulated proteins in the M group compared with the NC group (Figure 6D and Table S5). Functional enrichment analysis showed that these proteins were primarily associated with cytoskeletal organization and motor protein–related functions (Figure 6E and Figure S3). Several proteins involved in myofibrillar structure and contractile function exhibited altered expression, indicating pronounced abnormalities in tissue structure and contractility at the protein level in HFD–induced ED rats (Figure 6F).

Validation of transcriptomic and proteomic findings by RT-qPCR and western blot

To validate the reliability of transcriptomic data, RT-qPCR was performed to measure the mRNA expression levels of Lep, Fbp2, and Pck1 in penile tissues from the NC, M, and BAKHF groups (Figure 7A). Compared with the NC group, Lep, Fbp2, and Pck1 expression levels were significantly elevated in the M group, with Fbp2 and Pck1 showing highly significant differences (P < .001) and Lep also significantly increased (P < .01), consistent with transcriptomic results. Western blot analysis was used to assess the protein expression levels of myosin light chain 1 (MYL1) and tropomyosin 2 (TPM2) in penile tissues (Figure 7B and C). Both MYL1 and TPM2 protein levels were significantly elevated in the M group compared with the NC group (P < .001 and P < .01), consistent with proteomic findings. Notably, BAKHF treatment markedly reduced MYL1 protein expression compared with the M group (P < .001).

Figure 7.

Validation of transcriptomic and proteomic alterations in rat penile tissue by RT-qPCR and Western blot. (A) Relative mRNA expression levels of Lep, Fbp2, and Pck1; (B–C) representative Western blot images and quantitative analysis of MYL1 and TPM2 protein expression. Data are presented as mean ± SD (n = 6 for A, n = 4 for C). Statistical analysis was performed using Tukey’s or Bonferroni-adjusted Dunn’s post hoc tests following normality and homoscedasticity verification. ns P > .05, ^*  P < .05, ^**  P < .01, ^***  P < .001.

Characterization of chemical constituents and serum-absorbed compounds of BAKHF

To elucidate the pharmacologically active constituents of BAKHF, its chemical profile was characterized using UPLC-QTOF/MS. Total ion chromatograms were acquired in both negative and positive ion modes (Figure S4), resulting in the identification of 58 chemical constituents sorted by chromatographic peak area (descending) (Table S6). These compounds were classified into 8 major categories, including alkaloids, terpenoids, shikimates and phenylpropanoids, amino acids and peptides, fatty acids, polyketides, carbohydrates, and other components.

To further investigate the in vivo exposure of these constituents, serum samples from both NC and BAKHF rats were analyzed using UPLC-QTOF/MS. No BAKHF-related compounds were detected in the serum of NC rats, whereas 5 compounds identified in vitro were detected in BAKHF rat serum (Table S7), including 2-(3,4-dihydroxyphenyl-d3)ethylamine, ferulic acid, sinapine, rosmarinic acid, and betaine. These 5 serum-absorbed compounds are sorted by peak area (descending), with blank serum background subtracted. These serum-absorbed compounds may represent key bioactive constituents of BAKHF and provide a chemical basis for subsequent pharmacological and mechanistic investigations.

Network pharmacology identifies potential core targets and pathways of BAKHF in ED

A total of 583 potential targets were predicted for the 5 serum-absorbed compounds. Intersection analysis with 726 ED-related genes obtained from the GeneCards database and 202 ED-related targets identified from transcriptomic and proteomic analyses yielded 88 overlapping targets (Figure 8A). A PPI network was constructed using the STRING database (combined score ≥ 0.4) and visualized with Cytoscape, comprising 88 nodes and 954 interaction edges. Several nodes exhibited high connectivity, including GAPDH, ESR1, CASP3, EGFR, MMP9, STAT3, and PPARγ (Figure 8B).

Figure 8.

Network pharmacology analysis of BAKHF in ED. (A) Venn diagram BAKHF-related and ED-related targets; (B) degree–based topology analysis of the PPI network for intersecting targets; (C) top 20 enriched GO terms; (D) top 20 enriched KEGG pathways; (E) drug–absorbed compounds-target-pathways-disease network diagram.

GO enrichment analysis indicated that these targets were mainly involved in biological processes related to circulatory regulation, lipid metabolism, antioxidant defense, and tissue repair (Figure 8C). KEGG pathway analysis further revealed significant enrichment in vascular function–related pathways, the AMPK signaling pathway, estrogen signaling, and multiple metabolism–associated pathways (Figure 8D). Based on these results, an integrated “herbal medicine–serum-absorbed compounds–core targets–signaling pathways” network was constructed to systematically illustrate the multi-target and multi-pathway characteristics of BAKHF in ED (Figure 8E).

Molecular docking analysis of serum-absorbed compounds with core targets

Based on topological parameters including Degree, Betweenness, and maximal clique centrality, 7 core target proteins were selected: GAPDH, ESR1, CASP3, EGFR, MMP9, STAT3, and PPARγ. Molecular docking analysis was performed to evaluate the binding affinities between these targets and the serum-absorbed compounds (Figure S5A).

The docking results demonstrated that 2-(3,4-dihydroxyphenyl-d3)ethylamine, ferulic acid, rosmarinic acid, and sinapine exhibited Vina binding scores lower than −5.0 kcal/mol with all 7 core targets, indicating favorable binding affinities (Figure S5B). Notably, rosmarinic acid displayed the lowest docking scores (below −7.0 kcal/mol) across all targets, suggesting the strongest and most stable binding interactions.

BAKHF regulates the expression of key target genes and proteins in penile tissue

To further validate the regulatory effects of BAKHF on ED–related key targets, RT-qPCR was performed to assess the mRNA expression levels of Gapdh, Pparg, Esr1, and Stat3 in penile tissues (Figure 9A). Compared with the NC group, Gapdh, Pparg, and Stat3 mRNA levels were significantly increased in the M group (P < .01), whereas Esr1 expression was significantly reduced (P < .01). BAKHF treatment significantly downregulated Gapdh mRNA expression and markedly upregulated Esr1 expression (P < .05) compared with the M group.

Figure 9.

Effects of BAKHF on key metabolic- and tissue-related targets in penile tissue of HFD–induced ED rats. (A) RT-qPCR analysis of Gapdh, Pparg, Esr1, and Stat3 mRNA levels; (B–E) representative Western blot images and quantitative analysis of GAPDH, PPARγ, ESR1, and MMP9 protein expressions. Data are presented as mean ± SD (n = 6 for A, n = 4 for C). Statistical analysis was performed using Tukey’s or Bonferroni-adjusted Dunn’s post hoc tests following normality and homoscedasticity verification. ns P > .05, ^*  P < .05, ^**  P < .01, ^***  P < .001.

Western blot analysis further confirmed these findings at the protein level. GAPDH protein expression was significantly elevated in the M group compared with the NC group (P < .01) and was markedly reduced following BAKHF treatment (P < .05), with no significant difference compared with the NC group (P > .05) (Figure 9B and C). PPARγ protein levels were significantly increased in the M group (P < .01) and were substantially reduced by BAKHF treatment (P < .05), with no significant difference compared with the NC group (P > .05) (Figure 9C). ESR1 protein expression was significantly decreased in the M group (P < .001) but was significantly upregulated after BAKHF treatment (P < .01), with no significant difference compared with the NC group (P > .05) (Figure 9C). In addition, MMP9 protein expression was markedly elevated in the M group (P < .001) and was significantly reduced following BAKHF intervention (P < .001), with no significant difference compared with the NC group (P > .05) (Figure 9C).

Discussion

In the present study, rats with HFD-induced ED exhibited pronounced metabolic and endocrine disturbances, including increased body weight, dyslipidemia, elevated blood viscosity, and dysregulation of the pituitary-gonadal axis. These findings further support the growing consensus that systemic metabolic imbalance plays a pivotal role in the initiation and progression of ED, particularly in the context of HFD-related lifestyles.^41–44 Notably, YMSK was employed as a reliable phenotypic benchmark to validate the responsiveness of the HFD–induced ED model.⁴⁵ The therapeutic convergence observed between BAKHF and YMSK, particularly in restoring erectile function and cavernosal histology, likely reflects a shared chemical-pharmacodynamic basis centered on phenolic and organic acid derivatives. For instance, the serum–absorbed ferulic acid and rosmarinic acid identified in BAKHF (Table S7) share structural and functional similarities with key phenolic constituents typically found in YMSK, such as eugenol and galangin.^46,47 Meanwhile, the superior effects of BAKHF on specific metabolic indices, such as the normalization of food intake, may stem from its unique bioactive profile, including constituents like sinapine and betaine, which are not primary markers of YMSK. While we prioritized the mechanistic elucidation of BAKHF’s unique regulatory signature via multi-omics, future head-to-head comparative chemical profiling and plasma analysis between the 2 formulas would provide more granular insights into their synergistic material basis and the molecular rationale behind their nuanced efficacy differences.

At the mechanistic level, our multi-omics analyses suggest that disturbances in energy metabolism and cytoskeletal organization represent key pathological features of HFD-induced ED. AMPK signaling constitutes a central regulatory network governing cellular energy homeostasis and has also been implicated in cytoskeletal organization and actin–myosin dynamics,^48,49 processes that are essential for smooth muscle tone regulation and tissue mechanical properties.⁵⁰ Under conditions of chronic energy surplus induced by HFDs, dysregulation of AMPK-associated signaling networks has been widely reported.⁵¹ Consistent with this notion, KEGG enrichment analyses of both transcriptomic and proteomic datasets revealed that differentially expressed molecules in penile tissue from HFD–induced ED rats were significantly enriched in the AMPK signaling pathway, as well as in pathways related to cytoskeleton organization in muscle cells and motor protein function.

Concurrently, contraction-associated proteins, including MYL1 and TPM2, were aberrantly upregulated in the model group, accompanied by reduced smooth muscle content and increased collagen deposition in penile tissue. These findings suggest that metabolic disturbances induced by HFD may exacerbate ED by disrupting metabolic–cytoskeletal coupling, thereby promoting structural remodeling and functional impairment of penile tissue.⁷ This localized structural decline closely mirrors the pathological milestones characteristic of vascular-driven ED; for instance, Hu et al⁵² demonstrated that hyperlipidemia accelerates corpus cavernosum smooth muscle phenotypic modulation and ECM accumulation, thereby fundamentally impairing sinusoidal elasticity. Importantly, following BAKHF intervention, both histological abnormalities and the expression levels of contraction-related proteins exhibited a clear tendency toward normalization, indicating that its protective effects may involve coordinated modulation of AMPK-associated metabolic–cytoskeletal regulatory networks rather than direct action on a single molecular target.

A key therapeutic advantage of traditional herbal formulas lies in their multi-component, multi-target, and multi–pathway synergistic properties.^53–55 Using a serum-absorbed compound–target–pathway–based network pharmacology strategy, we systematically linked the chemical composition of BAKHF with its potential biological effects. Notably, the circulating compounds identified in vivo—including ferulic acid, rosmarinic acid, sinapine, betaine, and 2-(3,4-dihydroxyphenyl-d3)ethylamine—are supported by substantial pharmacological evidence. Previous studies have demonstrated that ferulic acid and rosmarinic acid can ameliorate metabolic disturbances and suppress inflammatory responses through modulation of AMPK and related signaling pathways,^56,57 whereas sinapine and betaine have been implicated in lipid metabolism regulation and vascular protection.^58,59 In our network analysis, these absorbed compounds exhibited relatively high binding affinities toward several candidate key targets, including GAPDH, PPARγ, ESR1, STAT3, and MMP9, suggesting that they may constitute an important material basis underlying the pharmacological activity of BAKHF.

The observed compositional shift—where certain active compounds like sinapine and 2-(3,4-dihydroxyphenyl-d3)ethylamine achieved robust steady-state exposure despite their relatively low or intermediate chromatographic abundance within the baseline herbal matrix⁶⁰—is highly indicative of botanical matrix synergies. In natural product pharmacokinetics, complex decoction matrices often optimize the systemic availability of organic constituents, which typically suffer from rapid clearance when administered in isolation, by transiently modulating mucosal permeability or metabolic pathways.^61,62 This collective matrix effect supports the developed steady-state circulating levels of these core bioactive ligands at well–tolerated therapeutic doses. Given that the penile corpus cavernosum consists of extensive sinusoidal spaces continuously perfused by systemic blood flow,⁶³ these robust circulating profiles naturally facilitate efficient local tissue availability. Mechanistically, as supported by molecular docking simulations,⁶⁴ core absorbed ingredients display prominent thermodynamic binding energies (≤ –5.0 kcal/mol) toward PPARγ, ESR1, and MMP9. Rather than implying definitive in vivo occupational kinetics, these scores conservatively illustrate a favorable molecular propensity to interact with these candidate targets, theoretically supporting downstream cellular and homeostatic recovery within the cavernous sinusoids.

Crucially, our topology analysis (Figure 8B) identified several high-connectivity nodes that are recognized as “molecular switches” in ED pathology. Integrating our results with established literature provides a more granular understanding of how BAKHF modulates the cavernous microenvironment. For instance, the RhoA/ROCK pathway is a key mediator of hyper-contractility; HFD–induced metabolic syndrome significantly increases RhoA membrane translocation, which impairs nitric oxide (NO)-dependent relaxation.⁶⁵ BAKHF’s ability to modulate RhoA likely restores the delicate balance between pro-erectile and anti-erectile signaling. Furthermore, the preservation of cavernous structural integrity may be attributed to the inhibition of cell death pathways. Hyperglycemia-induced apoptosis, driven by Caspase 3, is a primary cause of endothelial and smooth muscle cell loss in diabetic ED,⁶⁶ a process significantly mitigated by BAKHF.

Additionally, the metabolic stress from HFD activates the mTOR pathway, which fuels NADPH oxidase–derived reactive oxygen species production, further damaging erectile function.⁶⁷ Our findings indicate that BAKHF may act as an upstream modulator to dampen this oxidative stress axis. Finally, the systemic inflammation induced by a Western diet converges on the TLR4/NF-κB1 signaling axis within the penile vasculature. Blockade of TLR4⁶⁸ or downregulation of the NLRP3/NF-κB pathway²⁰ has been shown to alleviate endothelial dysfunction and tissue injury. By addressing these pathways simultaneously, BAKHF facilitates the transition of the cavernous microenvironment from a pro-inflammatory and apoptotic state toward homeostatic recovery.

To better contextualize these findings, the observed network-level modulation of BAKHF was compared with previously reported therapeutic mechanisms in ED. Previous investigations on the positive control, YMSK tablet, have demonstrated that YMSK ameliorates ED through modulation of the NLRP3/NF-κB inflammatory axis and the FXR/FMO3/TMAO lipid metabolism pathway.^69,70 While these mechanisms contribute to functional improvement by targeting specific inflammatory or metabolic pathways, our integrated multi-omics and phenotypic analyses suggest that BAKHF exerts therapeutic effects through broader biological coordination. Unlike the relatively pathway-focused mechanisms reported for YMSK, the pharmacodynamic profile of BAKHF appears to involve coordinated regulation across multiple biological processes. Specifically, BAKHF simultaneously modulated metabolic regulators associated with energy homeostasis (eg, Lep, Fbp2, and Pck1) and structural–cytoskeletal factors related to cavernosal integrity (eg, MYL1 and TPM2), potentially contributing to both metabolic restoration and tissue preservation in penile tissue. These findings suggest that the multi-component nature of BAKHF may offer advantages in addressing the multifactorial pathophysiology of hyperlipidemia-induced ED.

Overall, the present results support a model in which multiple bioactive constituents of BAKHF act synergistically to modulate AMPK-associated metabolic and cytoskeletal signaling networks, thereby promoting coordinated restoration of metabolic and tissue homeostasis at a systemic level. This integrative mode of action may underlie the therapeutic efficacy of BAKHF in HFD-induced ED and highlights the potential advantages of multi-target strategies over conventional single-target interventions for metabolically driven ED.

Nevertheless, several limitations of this study should be acknowledged. The direct regulatory relationships among specific targets require further confirmation through targeted inhibition or genetic intervention approaches. In addition, the relative contributions of individual compounds to the overall efficacy of the formula remain to be elucidated. Future studies combining functional blood flow assessments with pathway-specific interventions may further clarify the synergistic mechanisms underlying the therapeutic effects of BAKHF.

Conclusion

In conclusion, BAKHF ameliorated HFD–induced ED in rats, accompanied by improvements in erectile function, metabolic parameters, endocrine balance, and penile tissue integrity. At the mechanistic level, multi-omics analyses indicated that HFD-induced ED is associated with disturbances in AMPK–associated metabolic pathways and cytoskeletal regulation. BAKHF treatment partially reversed these alterations and modulated the expression of several potential key targets, including GAPDH, PPARγ, ESR1, and MMP9, which are involved in metabolic regulation, vascular function, and tissue remodeling (Figure S6). Collectively, these findings suggest that BAKHF may contribute to the coordinated regulation of metabolic and tissue homeostasis, supporting its potential relevance for the management of metabolism-related ED.

Author contributions

A.Y.: Conceptualization, Funding acquisition, Project administration, Supervision, Writing—review and editing. J.Z.: Methodology, Formal analysis, Data curation, Validation, Visualization, Writing—original draft. J.L.: Data curation, Investigation, Methodology, Resources, Validation. J.Ze.: Funding acquisition, Software, Validation, Visualization, Writing—review and editing. P.H.: Data curation, Investigation, Resources. W.M.: Investigation, Data curation, Supervision. S.A.: Data curation, Investigation, Resources, Supervision. W.L.: Data curation, Investigation, Resources, Supervision.

Funding

This work was supported by the National Natural Science Foundation of China (Grant No. 82260873); the Foundation of the Institute of Traditional Chinese Medicine Health Industry, China Academy of Chinese Medical Sciences (Grant No. CXLH-2025001); the Natural Science Foundation of Jiangxi Province (Grant Nos. 20232BAB216022, 20232BCJ25060, and 20252BAC240473); and the Fundamental Research Funds for the Central Public Welfare Research Institutes (Grant Nos. ZZ16-ND-12, ZZ17-ND-12, and ZZ17-YQ-050).

Conflicts of interest

None declared.

Ethics statement

The animal research protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Xinjiang Medical University (Approval No. 20240228-26, February 28, 2024).

Disclosure

The author(s) report no conflicts of interest in this work.