Rg1-R1 attenuates cardiac ischemia/reperfusion-induced endothelial cell injury through activating the ULK1/PGAM5-FUNDC1-mitophagy pathway

Abstract

Background

Mitochondrial dysfunction has been recognized as a pivotal pathological mechanism underlying myocardial ischemia/reperfusion injury (MIRI).Ginsenoside Rg1 and notoginsenoside R1 exhibits cardioprotective effects against MIRI. However, their molecular mechanisms remain unclear. This study aims to investigate the therapeutic potential of Rg1 and R1 in ameliorating cardiomyocyte injury through mitophagy regulation, with a focus on elucidating the molecular crosstalk between these compounds and key mitophagy-related signaling pathways.

Methods

Cardiac injury in mice was induced by subjecting the heart to 45 min of ischemia followed by 6 h of reperfusion. Post-injury, the mice were treated with intraperitoneal injections of Rg1-R1. The effects of Rg1-R1 on MIRI were assessed through electrocardiography, echocardiography, HE/Masson staining, and Transmission Electron Microscope. The impact of Rg1-R1 on biochemical markers of myocardial injury was also analyzed. Cardiac microvascular endothelial cells (CMECs) were pretreated with Rg1-R1 prior to being exposed to hypoxia/reoxygenation (H/R). Subsequently, cellular function and mitochondrial function were evaluated.

Results

Our results indicated that in vivo, Rg1-R1 improved MIRI-induced cardiac dysfunction; in vitro, exposure of CMECs to Rg1-R1 reduced H/R injury severity and protected mitochondria. Further studies illustrated the protective effect of Rg1-R1 achieved via the regulation of FUNDC1-mediated mitophagy. In addition, we found that Rg1-R1 exerted these protective effects by activating FUNDC1-dependent mitophagy through the ULK1/PGAM5 pathway.

Conclusions

Our results indicated that Rg1-R1 attenuates MIRI-induced endothelial cell injury through activating the ULK1/PGAM5-FUNDC1-mitophagy pathway, and may represent a novel therapeutic target in the context of MIRI.

Graphical abstract

1. Introduction

Acute myocardial infarction (AMI) represents a leading cause of global cardiovascular mortality due to its substantial disease burden and poor clinical outcomes [1,2]. While timely reperfusion therapy remains the cornerstone intervention for minimizing ischemic myocardial damage and salvaging viable myocardium, accumulating evidence reveals that reperfusion itself paradoxically exacerbates tissue injury—a phenomenon termed myocardial ischemia-reperfusion injury (MIRI) [3,4]. At the core of this pathophysiological cascade lies mitochondrial dysfunction, characterized by diminished ATP synthesis, calcium overload, reactive oxygen species (ROS) overproduction, and mitochondrial permeability transition pore (mPTP) dysregulation, collectively culminating in apoptotic cell death and aggravated myocardial injury [[5], [6], [7], [8]].Emerging research highlights mitophagy—a selective autophagy mechanism for eliminating damaged mitochondria—as a critical regulatory node in MIRI pathophysiology [9]. Under physiological conditions, mitophagy maintains mitochondrial quality control by clearing defective organelles. However, MIRI disrupts this homeostatic balance, leading to pathological mitochondrial accumulation and dysfunction [10]. These findings position mitophagy modulation as a promising therapeutic strategy for cardiovascular diseases [9].

Mitophagy, an important pathway for cellular clearance of damaged mitochondria, involves the receptor protein FUN14 domain-containing 1 (FUNDC1) as one of the key mediators of mitophagy [11]. Clinical studies implicate FUNDC1 in maintaining cardiovascular homeostasis under pathological stress [12,13]. Mechanistically, ULK1 kinase activation during hypoxia induces FUNDC1 phosphorylation at Ser17 to initiate mitophagy, while phosphoglycerate mutase 5 (PGAM5)-mediated dephosphorylation at Ser13 further amplifies this process [14,15]. This ULK1/PGAM5-FUNDC1 signaling axis thus represents a potential therapeutic target for MIRI intervention.

Notably, traditional Chinese medicine (TCM) compounds exhibit mitochondrial protective effects through multi-target modulation of mitophagy-related pathways [[16], [17], [18]]. Our previous investigations identified ginsenoside Rg1 (from Panax ginseng Meyer) and notoginsenoside R1 (from Panax notoginseng (Burk.) F. H. Chen) as bioactive components of the classic Ginseng-Notoginseng herb pair, demonstrating cardioprotective properties through autophagy regulation [19].

Building upon these findings, we hypothesize that the Rg1-R1 combination may attenuate MIRI by enhancing FUNDC1-dependent mitophagy.To test this hypothesis, we established both in vivo murine MIRI models and in vitro myocardial endothelial cell hypoxia/reoxygenation (H/R) systems. This integrated approach aims to elucidate the pharmacological mechanisms underlying Rg1-R1's cardioprotective effects, potentially unveiling novel therapeutic strategies for MIRI management and drug development.

2. Materials and methods

2.1. Animal and MIRI in vivo

A total of 40 C57BL/6J mice, aged 7 weeks with an equal number of males and females, were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. All experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals issued by the US National Institutes of Health and were approved by the Institutional Animal Care and Research Advisory Committee of Shandong University of Traditional Chinese Medicine. To establish the MIRI model, the mice were anesthetized and secured on a surgical table. The left pectoral muscle was gently separated, and the chest cavity was opened to expose the heart. The left anterior descending coronary artery was ligated for 45 min, followed by 6 h of reperfusion. Throughout the procedure, electrocardiogram (ECG) was monitored in real-time using a multichannel physiological recorder. Subsequently, the ejection fraction of the mice was measured by ultrasound to evaluate the successful establishment of the MIRI model. Drug administration was initiated only after the stabilization of vital signs. Using standardized random number table method to randomly divide mice into five groups (n = 8 per group) as follows: Sham operation group: Underwent sham surgery (thoracotomy without coronary artery ligation) and received an intraperitoneal injection of an equivalent volume of 0.9 % sodium chloride. Model group: MIRI-induced mice received an intraperitoneal injection of an equivalent volume of 0.9 % sodium chloride. Rg1 group: MIRI-induced mice were administered Rg1 (10 mg/kg) via intraperitoneal injection. R1 group: MIRI-induced mice received R1 (10 mg/kg) via intraperitoneal injection. Rg1-R1 group: MIRI-induced mice were co-administered Rg1 (10 mg/kg) and R1 (10 mg/kg) intraperitoneally [4,20]. After 4 weeks of drug administration, tissue samples were collected from the mice for subsequent experiments.

2.2. Cell extraction/modelling and drug preparation

Suckling mouse hearts were first isolated into single-cell suspensions using the Neonatal Heart Dissociation Kit (Miltenyi Biotec, Germany) in accordance with the producer's instructions, followed by sorting of cardiac microvascular endothelial cells using CD31 magnetic beads (Miltenyi Biotec, Germany) for future use.

The cells were positioned inside a hypoxia chamber incubator for 45 min while the medium was substituted with D-Hanks solution. Cell culture medium was substituted with typical high glucose ECM and shifted to a standard CO2 incubator for 6 h.

The drug monomers (Standard Biotech, China) were dissolved in DMEM medium at stored concentration, safeguarded from light and preserved at −20 °C for subsequent experimental dilution.

2.3. Echocardiography

We performed echocardiography by using the Experimental animal ultrasonic instrument(SigmaVET, China). Ejection fraction (EF), and fractional shortening (FS) were measured to assess cardiac function.

2.4. H&E and Masson

Heart tissue was embedded in paraffin, sectioned and stained with H&E using standard procedures. Specimens were dehydrated in xylene, made transparent and finally sealed.

Mouse heart tissues were fixed, followed by dehydration with gradient alcohol and embedding in paraffin. Sections were subjected to Masson staining using Weigert's iron hematoxylin and Ponceau S acid fuchsin stain (Servicebio, China).

2.5. Transmission electron microscope

The heart samples were pre-fixed with 3 % glutaraldehyde and then post-fixed with 1 % osmium tetroxide. After dehydration, infiltration, and embedding, semi-thin positioning and ultra-thin sectioning were performed. Finally, the samples were stained with uranyl acetate for 10–15 min, followed by staining with lead citrate for 1–2 min at room temperature.

2.6. Immunofluorescence staining

Heart tissue paraffin sections were deparaffinized and rehydrated. Following antigen retrieval, the sections were blocked with bovine serum albumin (BSA) (GC305010, 3 %, Servicebio) for 30 min at room temperature. Subsequently, the sections were incubated with a primary antibody against CD31 (ab182981, 1:200, Abcam) overnight at 4 °C in a humidified chamber. After washing, a horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG secondary antibody (GB23303, 1:100, Servicebio) was applied and incubated for 50 min at room temperature. A second round of antigen retrieval was then performed to enable sequential staining. Next, the sections were incubated with the second primary antibody, eNOS (#32027, 1:100, CST), overnight at 4 °C. Following this, a Cy3-conjugated goat anti-rabbit secondary antibody (GB21303, 1:100, Servicebio) was added and incubated for 50 min at room temperature. Finally, the cell nuclei were counterstained with DAPI (G1012, Servicebio) for 10 min, and the sections were mounted with an anti-fade mounting medium.

2.7. Proliferation assay

Lysate was added to dissolve each group of MTT purple crystalline. Absorbance was measured at 570 nm to assess the metabolic viability of the cells.

2.8. Migration assay

Tolerant or non-tolerant H/R cells were added to the upper chamber for culture and then stained with H&E. CMECs migrated to its lower face were counted in 3 views per membrane under a microscope (ZEISS, Germany).

2.9. Tube formation assay

Tube formation facilitated by Matrigel (BD, USA) was monitored using an inverted microscope (PerkinElmer-Opera Phenix, USA). Image J was used to calculate the total number of branching points.

2.10. Scratch assay

Cells were cultured on 6-well plates in ECM medium until confluent. The monolayers were wounded by scraping with a 10 μL pipette tip. Imaging was performed at 0/6/12 h with a microscope.

2.11. Flow cytometry

The harvested cells were washed twice and fluorescent dye (Meilunbio, China) was added for incubation. Next, allow the flow cytometer (Beckman, USA) to scan and record the fluorescence signals emitted by the FITC and 7-AAD.

2.12. Transmission electron microscopy

Treated cells were pre-embedded in agar and fixed with osmium tetroxide. After osmotic embedding, the mixture was polymerised into blocks for ultrathin sectioning at 60–80 nm. Negative dye was used to enhance the contrast. Images were acquired on the transmission electron microscope (Hitachi, Japan).

2.13. Seahorse

The treated cells were seeded into Seahorse XFe96 (Agilent Technologies, USA) microtiter plates at a density of 8000 cells per well and allowed to adhere overnight. The cells were then washed with assay medium (non-buffered DMEM supplemented with 10 mM glucose, pH 7.4) and incubated in the same medium at 37 °C for 30 min in a non-CO₂ incubator.

OCR and ECAR, which reflect oxidative phosphorylation and glycolysis, respectively, were measured using the Seahorse XFe96 analyzer. Basal oxidative phosphorylation and glycolysis were calculated by averaging three baseline OCR and ECAR measurements taken prior to the injection of specific metabolic inhibitors: oligomycin (2 μg/mL), FCCP (5 μM), and antimycin A (2 μM). Maximum glycolysis and ATP-linked respiration were assessed following oligomycin injection. ATP-linked respiration was determined by subtracting the residual OCR after oligomycin injection from the baseline OCR. FCCP was injected to evaluate the maximum respiratory capacity, which was calculated as the average of three measurements after injection. The respiratory reserve was determined by subtracting the baseline OCR from the FCCP-induced maximum OCR.

2.14. ROS, JC-1 and mPTP

Cells were treated using DCFH-DA (Beyotime, China), Calcein AM staining solution (Beyotime, China), Mitosox™Red (ThermoFisher, USA), and mitochondrial membrane potential (Solarbio, China) according to the manufacturer's procedure before image capture using the microscope.

2.15. Western blotting and co-immunoprecipitation

Protein samples that have been prepared were run on precast SDS-PAGE gels (Vazyme, China) and electrophoretically converted onto the polyvinylidene difluoride membrane. After blockade treatment then incubated overnight with specific primary antibodies against ET-1(ab2786, Abcam, USA), e-NOS(#32027, CST, USA), Mitofusin-2 (#9482, CST, USA), OPA1 (#80471, CST, USA), Fis1 (#32525, CST, USA), p-DRP1(#4494, CST, USA), DRP1(#8570, CST, USA), FUNDC1(#49240, CST, USA), P-FUNDC1(PA5-114576, Invitrogen, China), P62 (ab109012, Abcam, USA), LC3B (#3868, CST, USA), ULK1(#8054, CST, USA), PGAM5(#63454, CST, USA), BNIP3(68091-1-Ig, Proteintech, China), BNIP3L(12986-1-AP, Proteintech, China), GAPDH (10494-1-AP, Proteintech, China). After three rinses with Tris-buffered saline with Tween, the membrane was incubated with corresponding secondary antibody (#7074, CST, USA). ECL (Vazyme, China) was utilised for exposure, followed by image acquisition and analysis.

2.16. mt-keima

Cells were transfected by Mito-Keima (Hanbio, China) upon reaching confluency. Transfection lasted for 48 h, and a multiplicity of infection of 10 was used. Imaging was carried out at the end of the experimental cycle using a confocal microscope (Zeiss LSM880+Fast Airyscan, Germany).

2.17. siRNA transfection

ULK1 siRNA (100 nmol/l) and PGAM5 siRNA (75 nmol/l) were diluted to the working concentrations as per the RNAFit transfection reagent instructions (Hanbio, China). Followed by incubation for 48 h with the medium containing RNAFit and siRNA transfection complexes for induction of siRNA entry and function within cells.

2.18. Tunel staining

After cell processing, 4 % paraformaldehyde fixation was performed for permeabilization. Apoptosis signaling were captured under a microscope after sequential use of TdT enzyme reaction solution, streptavidin-fluorescein conjugate labelling as directed by the manufacturer (KeyGEN BioTECH, China).

2.19. Statistical analysis

The outcomes were calculated by means of GraphPad Prism V.9.0 software. All the data were expressed as mean ± SD. Multiple-group comparisons were analyzed using Student's t-test followed by a one-way analysis of variance. Values of P < 0.05 was considered to indicate the difference was statistically significant.

3. Results

3.1. Rg1-R1 ameliorates MIRI-induced cardiac function damage

As illustrated in Fig. 1A, compared to the control group of mice subjected to sham surgery, those undergoing MIRI exhibited significant myocardial injury, as evidenced by alterations in the ST segment and T waves. Treatment with Rg1-R1 effectively mitigated these electrocardiogram abnormalities. Subsequent echocardiogram analysis (Fig. 1B) revealed a reduction in both the ejection fraction (EF) and left ventricular short-axis fractional shortening (FS) in the MIRI model group. However, compared to the model group, the Rg1-R1 treatment group showed varying degrees of improvement in EF and FS (Fig. 1C and D). To further assess myocardial injury, we measured the levels of cardiac biomarkers, including cardiac troponin T (cTnT), creatine kinase-MB (CK-MB), and lactate dehydrogenase (LDH). While these markers were elevated in the MIRI model group compared to the sham group, their levels were significantly reduced following Rg1-R1 treatment (Fig. 1E–G).Histopathological evaluation using H&E and Masson staining provided additional insights. H&E staining (Fig. 1H) revealed myocardial cell disarray, fragmentation of myocardial fibers, interstitial edema, extensive inflammatory cell infiltration, and partial nuclear pyknosis in the MIRI group. In contrast, Rg1-R1 treatment resulted in a more orderly arrangement of myocardial cells, reduced myocardial fiber rupture, mild interstitial edema, decreased inflammatory cell infiltration, and only occasional nuclear condensation. Masson staining (Fig. 1I) further demonstrated that Rg1-R1 alleviated the structural disorganization of myocardial cells and reduced collagen deposition induced by MIRI. Moreover, electron microscopy analysis indicated that MIRI led to the widening of tight junction gaps between endothelial cells, a phenomenon that was ameliorated by Rg1-R1 treatment (Fig. 1J). To assess the state and function of cardiac microvascular endothelial cells, we performed dual immunofluorescence staining for e-NOS and CD31. The results revealed comparable levels of e-NOS expression across all groups (Fig. 1K). Western blot analysis (Fig. 1L–Q) revealed that MIRI significantly suppressed the expression of phosphorylated endothelial nitric oxide synthase (p-eNOS) while increasing endothelin-1 (ET-1) levels. Notably, Rg1-R1 treatment reversed these effects, enhancing p-eNOS expression and reducing ET-1 levels. These findings suggest that Rg1-R1 exerts a protective effect on cardiac function under MIRI conditions. Additionally, we observed that MIRI significantly inhibited the expression of phosphorylated FUN14 domain-containing protein 1 (p-FUNDC1) and microtubule-associated protein 1 light chain 3B (LC3B), whereas Rg1-R1 treatment upregulated the levels of these proteins. Collectively, these results underscore the potential of Rg1-R1 in mitigating MIRI-induced cardiac damage and improving cardiac function. Results are expressed as mean ± SD, ∗∗∗p < 0.0001, ∗∗p < 0.01, ∗p < 0.05.

Fig. 1.

Rg1-R1 ameliorates MIRI-induced cardiac function damage. (A) The typical results of the ECG. (B–D) Echocardiography was used to observe the structural and functional status of the heart. (E–G) Assessment of myocardial injury using Elisa assay kit to detect levels of myocardial markers. (H) Cardiac samples were dyed with hematoxylin-eosin to observe myocardial histopathologic changes.The red arrow indicates the site of myocardial injury. (I) Masson staining to distinguish the degree of myocardial fibrosis. (J) Electron microscopy detection of vascular endothelial structural changes.The blue arrow represents the tight junction structure. (K) CD31 and e-NOS staining of cardiac tissue (Scale bar: 20 μm). (L–Q) Proteins extracted from cardiac samples were analyzed for e-NOS、FUNDC1,LC3B and ET-1 levels by Western blotting. Experiments were repeated at least three times and data are shown as mean ± SD (n = 8 mice per group). ∗∗∗p < 0.001, ∗∗p < 0.01, ∗p < 0.05.

3.2. Rg1-R1 suppresses H/R-induced endothelial cell damage

Firstly, Transwell experiments showed that H/R inhibited the migratory response of CMECs, and these changes were markedly ameliorated in cells supplemented with Rg1-R1 (Fig. 2A and B). Secondly, using the MTT assay, endothelial cell viability was reversed by Rg1-R1 (Supplemental Fig. 1) supplementation compared with the model group (Fig. 2C). We used cell tube formation assays (Fig. 2D and E) and cell scratch (Fig. 2F and G) and identified Rg1-R1 as a potent guardian of migration and tube formation in H/R injury. Flow cytometry showed that Rg1-R1 significantly reduced apoptosis induced by H/R (Fig. 2H and I) in competition with model group. Results are expressed as mean ± SD, ∗∗p < 0.01, ∗p < 0.05.

Fig. 2.

Rg1-R1 attenuates H/R-induced CMEC injury. (A, B) Cell migration was determined by means of transwell assay. (C) Cell viability was estimated using the MTT. (D, E) The ability of endothelial cells to divide and migrate rapidly was detected using the tube-forming assay. (F, G) The cell healing capacity was measured by means of the scratch assay. (H, I) Flow cytometry with membrane-linked protein V-FITC/7-AAD to detect apoptotic cells. Experiments were repeated at least three times and data are shown as mean ± SD (Three independent cellular alleles per group). ∗∗p < 0.01, ∗p < 0.05.

3.3. Rg1-R1 alleviates H/R-induced endothelial mitochondrial dysfunction

To investigate the impact of Rg1-R1 on endothelial mitochondrial dysfunction induced by H/R, we initially examined the changes in endothelial cells after treatment with Rg1-R1 using electron microscopy. Following H/R injury, we observed disorganized arrangement of organelles, mitochondrial swelling, disruption and dissolution of cristae, and rupture of the outer membrane. In contrast, the Rg1-R1 group reversed mitochondrial swelling, mitochondrial membrane incompletion, and promoted selective autophagy of mitochondria by autophagosomes (Fig. 3A). To further explore the influence of H/R on mitochondrial damage, we conducted endothelial OCR detection in CMECs (Fig. 3B). Importantly, treatment with Rg1-R1 not only preserved mitochondrial morphology but also protected mitochondrial respiratory function and the production of ATP. CMECs displayed an enrichment of mitochondrial and cytoplasmic ROS (Fig. 3C) as a consequence of H/R-induced respiratory chain damage. Rg1-R1 reduced mitochondrial ROS generation, leading to a reduction in cytoplasmic ROS levels in CMEC after H/R lesion. As a result of the elevation in oxidative stress, the potential of the mitochondrial membrane dropped and the velocity of mPTP opening elevated. However, Rg1-R1 balanced the mitochondrial membrane potential and diminished the velocity of mPTP aperture in CMECs after H/R injury (Fig. 3D and E). Finally, Western blot analysis revealed that H/R injury stimulated mitochondria fission in CMECs but blocked mitochondria fusion (Fig. 3F). Treatment with Rg1-R1 inhibited mitochondrial fission and strengthened mitochondrial fusion, thus restoring the normal mitochondrial morphology of CMECs subjected to H/R treatment, which is consistent with the aforementioned results observed through electron microscopy. It is evident that Rg1-R1 can ameliorate the pathological changes in CMECs mitochondria induced by H/R-induced injury. Results are expressed as mean ± SD, ∗∗p < 0.01, ∗p < 0.05.

Fig. 3.

Rg1-R1 alleviates H/R-induced endothelial cell mitochondrial dysfunction by culturing and isolating CMECs. (A) Changes in mitochondria and autophagic vesicles of endothelial cells in each group were observed using transmission electron microscopy. Red arrows indicate mitochondria and yellow arrows indicate mitochondria wrapped by autophagic lysosomes. (B) OCR measurement was used to observe cellular respiratory function and the production of ATP with or without Rg1-R1 treatment. (C) Mitochondria ROS were gauged by Mitosox™Red and cytoplasmic ROS were gauged by H2DCF-DA. (D) JC-1 was utilised to monitor the membrane potential of the mitochondria. (E) The openness of mPTP was determined based on the intensity of green fluorescence of calcein acetoxymethyl esterin mitochondria. (F) Expression of p-Drp1, Mff, Fis1, Mfn2, and Opa1 was examined by Western blot. Experiments were repeated at least three times and data are shown asmean ± SD (Three independent cellular alleles per group). ∗∗p < 0.01, ∗p < 0.05.

3.4. Rg1-R1 alleviates H/R-induced mitochondrial injury in CMECs by activating mitophagy

To evaluate the potentially beneficial impact of Rg1-R1 on mitochondria and CMECs during the process of H/R by mediating mitophagy, we employed a PH-dependent fluorescent protein called mt-Keima to visualize the occurrence of mitophagy. The results indicated that H/R injury significantly suppressed mitophagy in CMECs (Fig. 4A) as evidenced by reduced formation of acidic autolysosomes, indicative of impaired mitochondrial flux. Nevertheless, treatment with Rg1-R1 ultimately enhanced the interaction between mitochondria and lysosomes in CMECs following H/R injury. Furthermore, immunofluorescence (Fig. 4B) results revealed that treatment with Rg1-R1 elevated the level of LC3B in mitochondria in comparison with the control group. Collectively, these outcomes imply that Rg1-R1 induces mitophagy in the context of H/R injury. Western blot analysis of mitophagy markers demonstrated a considerable reduction in the LC3II/LC3I ratio in H/R-treated CMECs, which indicates impaired mitophagy (Fig. 4C and D). However, treatment with Rg1-R1 reversed the decrease in the LC3II/LC3I ratio. Moreover, while H/R inhibited the degradation of the ubiquitin-binding protein p62, its degradation was enhanced in cells treated with Rg1-R1 (Fig. 4C–E). It is worth noting that H/R injury downregulated the manifestation of p-FUNDC1 in CMECs, whereas treatment with Rg1-R1 ameliorated these effects (Fig. 4C–F). To provide further evidence for mitophagy activation, we employed the lysosomal inhibitor Bafilomycin A1 (Baf-A1) and monitored the autophagic markers LC3B and p62. The immunoblotting results demonstrated that the pro-autophagic effect of Rg1-R1 was abolished by Baf-A1 treatment (Fig. 4G). Notably, the protein expression of the mitophagy receptors BNIP3 and BNIP3L was not significantly modulated by either Rg1-R1 or Baf-A1 (Fig. 4G), suggesting that Rg1-R1 activates mitophagy independently of altering the abundance of these key receptor proteins. Taken together, our results collectively suggest that Rg1-R1 promotes mitophagy in CMECs under I/R conditions, an effect that is at least partially mediated through the activation of the FUNDC1 pathway. Results are expressed as mean ± SD, ∗∗p < 0.01, ∗p < 0.05.

Fig. 4.

Rg1-R1 activates FUNDC1-dependent mitochondrial autophagy. (A) Cellular autophagy in CMECs was analyzed by the mt-keima method. (B) Representative photos of co-localization of LC3B and Mito-Tracker in separate groups of CMECs. (C–F) Representative immunoblots and quantitative analysis of mitochondrial autophagy-associated proteins LC3B, P62, and FUNDC1. (G) Representative immunoblots and quantitative analysis of mitochondrial autophagy-associated proteins LC3B and P62. Experiments were repeated at least three times, and data are shown as mean ± SD (Three independent cellular alleles per group). ∗∗p < 0.01, ∗p < 0.05.

3.5. The abolition of ULK1 or PGAM5 abrogates the protective effect of Rg1-R1 against H/R-induced endothelial cell impairment

Knockdown experiments targeting ULK1 and PGAM5 were performed to investigate whether Rg1-R1 attenuates H/R-induced endothelial cell injury through the induction of the ULK1/PGAM5-FUNDC1 pathway. OCR analysis revealed that H/R injury suppressed mitochondrial respiration function in CMECs, and Rg1-R1 ameliorated these effects (Fig. 5A and B). Importantly, the beneficial effects of Rg1-R1 were abolished when ULK1 siRNA and PGAM5 siRNA were used to inhibit the activation of ULK1 and PGAM5 in CMECs supplemented with Rg1-R1, highlighting the vital part played by ULK1 and PGAM5 in holding against endothelial injury (Supplemental Fig. 2). In addition to evaluating mitochondrial respiration function, we also assessed endothelial cell migration ability. The supplementation of Rg1-R1 significantly raised the proportion of migrated cells triggered by H/R compared to the model group (Fig. 5C and D). However, the knockdown of ULK1/PGAM5 genes reduced endothelial cell migration despite Rg1-R1 supplementation. Similarly, TUNEL staining (Fig. 5E) results showed that Rg1-R1 prevented H/R-induced cell apoptosis but had no effect on the up-regulation of apoptosis in CMECs with down-regulated ULK1 or PGAM5. Furthermore, we assessed mitochondrial autophagy using the dual-excitation ratiometric fluorescent protein mt-Keima. Our results (Fig. 5F and G) demonstrated that endothelial cells with ULK1 or PGAM5 Knockdown were more susceptible to H/R injury compared to normal endothelial cells, indicating the essential role of ULK1 and PGAM5 in protecting endothelial cells. Ultimately, Rg1-R1 treatment significantly increased the overall red signal, indicating the restoration of mitochondrial flux. Moreover, Rg1-R1 treatment significantly enhanced the collaboration in FUNDC1 and LC3B compared to the H/R model group, as confirmed by immunoprecipitation results (Fig. 5H). Demolition of ULK1 and PGAM5 separately weakened the implications of Rg1-R1 on mitochondrial autophagy flux and FUNDC1-LC3B interaction, suggesting the synergistic activation of mitochondrial autophagy by Rg1-R1 through the ULK1-FUNDC1 and PGAM5-FUNDC1 signaling pathways. Results are expressed as mean ± SD, ∗∗p < 0.01, ∗p < 0 0.05 (see Fig. 6).

Fig. 5.

Demolition of ULK1 or PGAM5 abrogates the sheltering effect of Rg1-R1 against H/R-induced CMEC injury. (A) Cellular respiratory function was detected using the seahorse cell energy metabolism analyzer. (B, C) Transwell assay was conducted to determine endothelial cell migration (D, E) TUNEL assay labeled apoptotic endothelial cells and counted apoptosis rate. (F, G) Representative images of mt-keima puncta. (H) Representative immunoblot and quantitative analysis of LC3B binding to FUNDC1. Input, direct detection of FUNDC1 and LC3B in cell-extracted protein samples. IgG, negative control. Experiments were repeated at least three times, and data are shown as mean ± SD (Three independent cellular alleles per group). ∗∗p < 0.01, ∗p < 0.05.

Fig. 6.

Rg1-R1 promoted FUNDC1 and LC3B binding and improved FUNDC1-mediated mitophagy by synergistically regulating ULK1 and PGAM5, which in turn ameliorated MIRI.

4. Discussion

In this study, we utilised endothelial injury following myocardial MIRI as an entry point to investigate the modulation of FUNDC1-mediated mitochondrial autophagy through the pharmacological modulation of Rg1-R1. In current project, we confirmed the great capacity of the ULK1/PGAM5-FUNDC1 mitochondrial autophagy signal transduction pathway to protect the vascular endothelium and antagonise myocardial MIRI. We additionally identified Rg1-R1 as an endothelial-targeted protectant that augments vascular endothelial endurance against MIRI by preserving endothelial functional and structural integrity. Furthermore, Rg1-R1 prevented MIRI-induced endothelial dysfunction by enhancing mitochondrial autophagy and thereby improving mitochondrial condition. In summary, our study elucidated that Rg1-R1 exerts its protective effects against endothelial damage induced by myocardial MIRI through FUNDC1-dependent mitochondrial autophagy by activating the ULK1/PGAM5 pathway.

The endothelial protectant role of R1/Rg1 has been characterized in multiple comprehensive studies. In ischaemic stroke resulting from middle cerebral artery occlusion, R1 has been reported to restore cerebral blood flow, improve mitochondrial energy metabolism and stimulate antiangiogenic activity by activating the NAMPT-NAD + -SIRT1 cascade [21]. Similarly, R1 has been found to inhibit oxLDL-mediated endothelial injury during the formation of atherosclerosis by suppressing the generation of inflammatory cytokines [22]. R1 attenuated high glycosylation-induced endothelial impairment in endothelial cells by regulating the intracellular redox state [23]. With the exception of attenuating endothelial dysfunction, Rg1 has been observed to alleviate the senescence of human endothelial progenitor cells cultured in isolation by increasing the proliferative phase of endothelial progenitor cells and decreasing the resting phase [24]. In the mouse model of stroke, Rg1 was shown to shrink the area of infarction by upregulating CD31 in the peri-infarct cortex, bromodeoxyuridine-positive/CD31+ microvessels, and glial fibrillary acidic protein-positive vessels [25]. These studies indicate that R1/Rg1 enhances mitochondrial antioxidant activity, strengthens inflammation inhibition properties, and induces the pro-survival program in endothelial cells by activating specific signaling pathways. Based on these historical observations, combined with our investigations, Rg1-R1 exhibited multi-functional and structural benefits of endothelial cells. Rg1-R1 ameliorated cardiac dysfunction by maintaining the integrity of the endothelial barrier and the secretion of its relaxing factors. At the cellular and molecular level, Rg1-R1 also exerted the comprehensive blockade of H/R-induced endothelial cell injury. Our research represents the first conclusive demonstration that Rg1-R1 is an effective agent for the treatment of acute vascular endothelial dysfunction.

Autophagy, as a "double-edged sword", is instrumental in MIRI. During the ischemic phase, mitochondrial autophagy functions to remove damaged mitochondria. However, during the reperfusion phase, mitochondrial autophagy is suppressed, leading to the disruption of mitochondrial homeostasis in cardiac myocytes. This disruption is characterized by decreased glucose oxidation, opening of the mPTP, decreased mitochondrial membrane potential (MMP), inhibition of the tricarboxylic acid cycle and oxidative phosphorylation, a shift in cellular energy metabolism from aerobic respiration to glycolysis, and reduced ATP synthesis efficiency. Consequently, these events result in cellular damage and ultimately lead to cardiac dysfunction [26,27]. Interestingly, LC3 aggregation on CMECs mitochondria, which is involved in H/R processing, is inhibited, while Rg1-R1 treatment increases the recruitment of LC3 in mitochondria. FUNDC1, a tertiary transmembrane protein located on the exterior mitochondrial membrane, stably exists in the outer membrane under normal conditions without mediating mitophagy. Concentrations of FUNDC1 and LC3 rise when mitochondria are dysfunctional or impaired. Dephosphorylation of FUNDC1 induces mitophagy, a process of selective degradation of mitochondria [28]. The phosphorylation/dephosphorylation of the FUNDC1 protein serves as an important component in mitophagy, triggering our exploration of how Rg1-R1 modulates upstream signalling to trigger activation of FUNDC1-induced mitophagy.

The ULK1/FUNDC1 pathway has been engaged in protecting cells against hypoxia-induced damage. Activation of ULK1 and subsequent phosphorylation of FUNDC1 at Ser17 promote the interaction between FUNDC1 and LC3, leading to the selective removal of damaged mitochondria through mitophagy. This process helps maintain cellular homeostasis and prevents the clustering of dysfunctional mitochondria that can contribute to a broad spectrum of diseases, including neurodegenerative and cardiovascular diseases [28,29]. Similarly, the PGAM5/FUNDC1 pathway is also involved in cellular protection. Dephosphorylation of FUNDC1 at Ser13 by PGAM5 enhances its binding to LC3 and promotes mitophagy. This pathway has been shown to play a protective role in conditions such as oxidative stress, mitochondrial dysfunction, and neuronal injury. Dysregulation of PGAM5/FUNDC1-mediated mitophagy has been identified in the pathogenesis of neurodegenerative diseases, ischemic stroke, and cardiac disorders [15]. Overall, both the ULK1/FUNDC1 and PGAM5/FUNDC1 pathways serve as protective mechanisms by regulating mitophagy and maintaining mitochondrial quality control under pathological conditions. Understanding these pathways can provide insights into the development of therapeutic strategies targeting mitochondrial dysfunction-related diseases. Our experiments have demonstrated that Rg1-R1 attenuate mitochondrial injury in CMECs by inducing restoration of mitophagy via the ULK1/PGAM5-FUNDC1 pathway in response to H/R injury. If ULK1 or PGAM5 is ablated, it can impede the beneficial influence of Rg1-R1 on mitochondrial respiration and inhibit mito-apoptosis in CMECs. These findings suggest that the primary mechanism by which Rg1-R1 maintains endothelial mitochondrial functionality in response to H/R insult is through the preservation of mitophagy.

We have found the earliest support for a linkage involving Rg1-R1 and the ULK1/PGAM5-FUNDC1 mitophagy pathway. Of course, there are constraints involved in this research. Firstly, despite the fact that we have identified Rg1-R1 against endothelial injury, there is no direct evidence of an effect of Rg1-R1 on vascular endothelial function in vivo. Coronary angiography is considered the gold standard for evaluating coronary endothelial function and serves as a reference. Therefore, Further experiments are needed to confirm the influence of Rg1-R1 exerted on vascular endothelial physiology. Secondly, we have identified the modulating ability of Rg1-R1 on mitochondrial fission/fusion and mitophagy in H/R-induced cardiac endothelial impairment, but further work is warranted to elucidate the complicated interplay between Rg1-R1 and mitochondrial fission/fusion and mitophagy in H/R-induced cardiac endothelial lesion. Finally, we did not further distinguish the phosphorylation status of the Ser17 and Ser13 sites of p-FUNDC1, which represents a significant limitation of current mechanism research. To address this issue, future research should construct FUNDC1 Ser17/Ser13 site specific mutants and combine immunoprecipitation with phosphorylation site specific antibody detection to clarify whether Rg1 regulates FUNDC1 activity through the ULK1-FUNDC1-Ser17 phosphorylation axis or inhibits the PGAM5-FUNDC1-Ser13 dephosphorylation pathway.

5. Conclusions

In conclusion, we unveiled the efficacy of the Rg1-R1 combination on MIRI for the first time. In parallel, our study revealed the mechanism by which Rg1-R1 inhibits reperfusion-induced cardiac impairment: the enhancement of FUNDC1-mediated mitophagy through the synergistic regulation of ULK1 and PGAM5. The discovery of the relationship between Rg1-R1, ULK1/PGAM5-FUNDC1-mitophagy and endothelial protection will help to explore new therapeutic targets and drugs for myocardial MIRI and other cardiovascular diseases share the same underlying biology.

CRediT authorship contribution statement

Xiayinan Song and Jinlan Deng: Writing - original draft, Data curation. Danyang Wang: Conceptualization. Zhenzhen Zheng: Visualization. Chao Li: Supervision. Jie Li: Writing - review &editing.

Funding

This work was supported by the National Natural Science Foundation of China (82474441,82004277). The Outstanding Youth Innovation Team of Shandong Institutions of Higher Learning (2023KJ189). Jinan City-School Integration Development Strategic Project (JNSX2024018).

Declaration of competing interest

The authors have declared that they have no conflicts of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Availability of data and materials

All data generated or analyzed during this study are included in this published article.