DMEM medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin-streptomycin mixture was used. The H9C2 cell (Eoucell, China) were seeded in T25 flasks and grown adherently in a 37 °C, 5% CO₂ incubator (Thermo Fisher Scientific, Waltham, MA, USA), with fresh medium replaced every 2–3 days. When cell density reached 80%–90%, cells were digested with 0.25% trypsin (Solarbio, China) for 1–3 min and passaged at a ratio of 1:3. The freezing medium consisted of 55% base medium, 40% FBS, and 5% DMSO.

Logarithmically growing H9C2 cells were trypsinized, adjusted to a density of 5 × 10³ cells/well, and seeded in 96-well plates with 100 μL of medium per well. After 24 h of pre-incubation, cells were treated with different concentrations of CGA (50 μg/mL, 100 μg/mL, 200 μg/mL) for 24–48 h. Then, 10 μL of CCK-8 reagent (Thermo Fisher Scientific, PA137267) was added to each well, and the plates were incubated in the dark for 1–4 h. Absorbance at 450 nm (OD value) was measured using a microplate reader (Molecular Devices, China). Cell viability was calculated as (OD value of the experimental group - OD value of the blank group)/(OD value of the control group - OD value of the blank group) × 100%.

After removing the medium, cells were washed with PBS and stained with 2 μM acridine orange (AO) (labeling live cells, green fluorescence) and 1 μM propidium iodide (PI, labeling dead cells, red fluorescence) (Solarbio, China) for 15 min in the dark. The percentage of live cells was calculated by fluorescence microscopy.

Cell lysates were collected, and the ability of SOD (Beyotime, China) to scavenge superoxide anions was determined using a WST-8 kit, with activity expressed as U/mg protein.

According to the operating procedures of the MDA assay kit (Beyotime, China), the thiobarbituric acid (TBA) method was used: Cell lysates were mixed with TBA reagent and heated, and the absorbance at 532 nm was measured. MDA content was expressed as nmol/mg protein.

After treatment, cells were stained with 5 μg/mL JC-1 staining solution (Thermo Fisher Scientific, Waltham, MA, USA) for 20 min at 37 C in the dark. The ratio of red to green fluorescence was detected by flow cytometry (normal membrane potential forms red aggregates, while decreased membrane potential deaggregates into green monomers).

Using a ATP assay kit (MedChemExpress, USA, HY-K0314): Cells were lysed, centrifuged to obtain the supernatant, mixed with luciferase reaction solution, and the luminescence intensity was detected by a microplate reader. ATP content was expressed as nmol/mg protein.

In the cell treatment step, one group was treated with the mitochondrial autophagy inhibitor Mdivi-1 (Aladdin, China; 338,967-87-6), while the remaining group was treated normally. The treatment method was the same as before. Cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.1% Triton X-100 (Beyotime, China, ST795) for 10 min, and blocked with 5% BSA for 1 h. Primary antibodies (anti-PINK1 and anti-Parkin, diluted 1:200) (Proteintech Group, China, 23274-1-AP; Bioss, China, bsm-61396R) were incubated overnight at 4 °C, followed by Cy3/FITC-labeled secondary antibodies (diluted 1:500) (Abcam, USA) for 1 h in the dark. Nuclei were counterstained with DAPI (Abcam, USA) before mounting. A fluorescence microscope (Olympus, Japan) was used to observe mitochondrial-localized PINK1/Parkin co-localization signals, and ImageJ was used to quantify fluorescence intensity and co-localization coefficients.

Following a 4-h co-treatment with 200 µM H₂O₂ and 200 μg/mL OLEU, cells were exposed to 100 nM bafilomycin A1 (BafA1) (Abcam, ab120497) for an additional 4 h; control groups received no treatment. After the treatment, both the cells and the supernatant were collected for analysis. Protein extraction was performed using RIPA lysis buffer, followed by heat-denaturation in Laemmli sample buffer at 95 °C for 5 min. The proteins were then resolved by electrophoresis on 12% SDS-polyacrylamide gels.The separated proteins were electrotransferred onto PVDF membranes. After blocking with 5% non-fat milk in TBST, the membranes underwent overnight primary incubation at 4 °C with antibodies specific for P62 (Abcam, ab91526; 1:1,000), LC3 (Biorbyt, orb1495189; 1:2,000), and Parkin (Abcam, ab77924; 1:1,000), all diluted in antibody buffer. Following five washes with TBST (5 min each), the membranes were treated with an HRP-conjugated goat anti-rabbit IgG (H + L) secondary antibody (Sera Care, 5220-0336; 1:5,000 in 5% skim milk/TBST) for 1 h at room temperature and washed again.Protein signal detection was carried out using an ECL substrate, with exposure times adjusted on-the-fly for optimal results. The developed and fixed film was then scanned, and the resulting band intensities were quantified by densitometry using ImageJ software (version 1.53).

To quantitatively measure mitophagic flux, we employed the MT-Keima probe in conjunction with flow cytometry. The MT-Keima reagent (PPL, China, LV01230-2a) was reconstituted as directed by the manufacturer. Cardiomyocytes in the logarithmic growth phase were collected, centrifuged at 1,000 g for 5 min, and washed twice with PBS. They were then incubated with serum-free diluted MT-Keima for 30–60 min at 37 C under 5% CO₂. After incubation, the cells were washed three times with PBS, detached using trypsin, centrifuged again, and resuspended in PBS. The cell density was adjusted to 1–5 × 10⁶ cells/mL for analysis by flow cytometry via the FITC/PE channels, with at least 10,000 events captured under optimized voltage and threshold conditions.

Animal experiments were approved by the Animal Ethics Committee of Jilin University (No. SY202503029). All animal handling and care were performed in accordance with the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health. Sixteen male SD rats (200 ± 20g) underwent ligation of the proximal left anterior descending coronary artery to induce MI, and fractional shortening (FS), ejection fraction (EF), end-diastolic volume (EDV), and end-systolic volume (ESV) were measured, with FS < 30% indicating successful MI modeling. Rats were divided into Sham, MI, and CGA groups.

The CGA group received daily tail vein injections (Q 24 h) of 1.2 mL/kg of 200 μg/mL CGA for 28 days. The sham-operated and MI groups received saline injections at the same volume and frequency as the CGA group. A 27G needle was used to puncture the lower 1/5 of the tail vein at a 45° angle, with an injection rate ≤0.2 mL/min.

To quantify the systemic exposure of CGA under a 200 μg/mL intravenous dosing regimen, a weight-normalized pharmacokinetic simulation was established in rats. The intravenous bolus solution was set at a concentration of 200 μg/mL with an injection volume of 1.0 mL/kg, corresponding to a dose of 0.20 mg/kg. Each group consisted of N = 4 (reflecting individual variations among four rats). Blood sampling time points were: 0, 1, 2, 5, 10, 15, 20, 30, 45, 60, 90, and 120 min.

After 28 days of continuous tail vein injection, H&E and Masson stains (Beyotime, China) were used on frozen sections of MI areas to detect collagen deposition and fibrosis. For H&E staining, myocardial tissue sections (4 μm) were deparaffinized, hydrated, and stained with hematoxylin for 5 min to show nuclei. After rinsing, eosin was applied for 2 min to stain the cytoplasm. For Masson’s trichrome, sections were treated with Weigert’s iron hematoxylin, Biebrich scarlet-acid fuchsin, and aniline blue to distinguish collagen fibers (blue) from muscle fibers (red). Both staining procedures included dehydration and mounting steps. Five random fields per slide were examined under light microscopy (400×) by two blinded investigators to assess tissue morphology and fibrosis.

After 28 days of continuous tail vein injection, echocardiography (Vevo2100, Visualsonic, USA) was used to evaluate left ventricular function in all rats. Rats were anesthetized with isoflurane, and M-mode, short-axis, and long-axis views of the heart were recorded using an M250 probe. EF, FS, left ventricular internal dimension systole (LVIDs) and EDV were measured.

Fresh myocardial tissue (<1 mm³) was quickly harvested after 28 days of tail vein injection, immediately fixed in pre-cooled 2.5% glutaraldehyde (prepared in 0.1 M phosphate buffer, pH 7.4) at 4 C for 2 h, post-fixed with 1% osmium tetroxide for 2 h, dehydrated with acetone, and embedded in epoxy resin. Ultra-thin sections (50–70 nm) were stained with uranyl acetate and lead citrate for electron microscopy observation (Hitachi TEM system, Japan) and identification of mitochondrial autophagosomes.

CD31 is a biomarker for myocardial endothelial cells, while α-smooth muscle actin (α-SMA) is a biomarker for myocardial stromal cells. CD31 and α-SMA staining (Abclonal, China, A19014; Bioss, China, bs-10196R) was used to study vessel density and myocardial fibrosisin infarcted heart tissue, while PINK1 (1:200) and Parkin (1:200) (Proteintech Group, China, 23274-1-AP; Bioss, China, bsm-61396R) were used to detect mitophagy. Tissue sections underwent antigen retrieval in citrate buffer after deparaffinization. Endogenous peroxidase was blocked before incubating with primary antibodies (CD31, α-SMA, PINK1 and Parkin) overnight at 4 C. HRP-conjugated secondary antibody was applied for 1 h at room temperature, followed by 3,3′-diaminobenzidine (DAB) development and hematoxylin counterstaining. Negative controls omitted primary antibodies. Staining intensity was quantified using ImageJ from five random 400× fields per sample, with results verified by two independent observers.

To further demonstrate that CGA promotes mitophagy and activates the PINK1/Parkin pathway, we performed WB analysis of PINK1 and Parkin in the Sham, MI, and CGA-treated groups. The primary antibodies used were as follows: PINK1 (Proteintech Group, China, 23274-1-AP; dilution 1:1,000) and Parkin (Abcam, ab77924; dilution 1:1,000). The experimental procedure was consistent with the method described previously.

Myocardial tissue (<50 mg) from the infarction and border zones of three rats each from the MI and CGA groups was rapidly collected, snap-frozen in liquid nitrogen, and stored at −80 C. Methanol-water (4:1) or acetonitrile-methanol-water (2:2:1) biphasic extraction was performed, followed by ultrasonic disruption and centrifugation (12,000×g, 15 min, 4 C). The LC-MS/MS platform was used for detection, and XCMS or MetaboAnalyst was used for peak alignment, normalization (PQN method), and missing value imputation. Quality control (QC) samples were prepared by pooling sample extracts and used to evaluate the reproducibility of the sample processing procedure under identical conditions. Unsupervised principal component analysis (PCA) was performed using the prcomp statistics function in R (http://www.r-project.org). Data were scaled to unit variance prior to unsupervised PCA. Metabolites with a variable importance in projection (VIP) > 1 and a P-value <0.05 (Student’s t-test) were selected as significantly differential metabolites and visualized using volcano plots. The original relative levels of the identified differential metabolites were row-scaled by unit variance scaling and presented in a heatmap generated with R packages. Identified metabolites were annotated against the KEGG Compound database (http://www.kegg.jp/kegg/compound/), and the annotated metabolites were subsequently mapped to the KEGG Pathway database (http://www.kegg.jp/kegg/pathway.html). Differential metabolites involved in the top-ranked human metabolome database (HMDB) Primary pathways based on P-values were further annotated and visualized using the HMDB database.

By using the HE staining method, the effects of the injected drug on the heart, liver, spleen, lungs, and kidneys were evaluated.

GraphPad Prism seven software was used for statistical analysis. All data are presented as mean ± standard deviation (SD). Student’s t-test was used to determine statistical significance when comparing two groups. For comparisons involving three or more groups, analysis of variance (ANOVA) with post hoc testing was used. Statistical significance was defined as ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

CGA is a condensed phenolic acid formed by caffeic acid and quinic acid, with the chemical formula C16H18O9 (Figure 1A) and CAS number 327-97-9. Supplementary Material confirms the biological properties of CGA (Desite, China, DST221010-021). The CCK assay demonstrates that CGA has no significant cytotoxicity on H9C2 cells within the range of 200 μg/mL (Figure 1B). The IC50 test revealed that 200 μM H2O2 was the most suitable concentration for this experiment (Figure 1C). CGA significantly improved cell viability, with 200 μg/mL CGA exhibiting the greatest effect (Figure 1D). Live/dead staining showed that CGA markedly reduced H2O2-induced cell death (Figures 1E,F).

CGA significantly reduced ROS production induced by H2O2. Within the range of 200 μg/mL (Figures 2A,B), SOD levels in H9C2 cells increased significantly with increasing CGA concentrations (Figure 2C), while MDA levels decreased significantly (Figure 2D).

CGA improved mitochondrial membrane potential levels in H9C2 cells (Figures 3A,B). H2O2 caused a decrease in ATP levels, which was significantly improved by 200 μg/mL CGA (Figure 3C).

CGA upregulated the expression levels of mitophagy markers PINK1 (Figures 4A,B) and Parkin (Figures 4C,D), which were inhibited by the mitophagy inhibitor Mdivi-1. The upregulation of PINK1 and Parkin was concentration-dependent within the range of 200 μg/mL CGA. To specifically inhibit mitophagic flux, the lysosomal inhibitor BafA1 was applied. The results showed that the BafA1-co-treated CGA group displayed markedly increased levels of Parkin, LC3-II, and P62 compared to the CGA-only group, further supporting the promotive role of CGA in mitophagy (Figures 4E–H). Additionally, quantitative evaluation of mitophagic flux using the MT-Keima probe and flow cytometry indicated that CGA significantly enhanced mitophagy relative to cells exposed solely to H₂O₂ (Figures 4I,J).

Following intravenous administration of a 200 μg/mL CGA solution (1.0 mL/kg, dose 0.20 mg/kg), the plasma drug concentration peaked immediately, with an average initial concentration of approximately 0.600 μg/mL. The plasma concentration then declined rapidly, dropping to around 0.220 μg/mL and 0.140 μg/mL at 10 and 15 min, respectively. By 30–45 min, concentrations approached the lower limit of quantification, and became largely undetectable after 60 min. The overall elimination process followed a one-compartment first-order decay model, with an estimated plasma half-life of approximately 7–8 min (Figure 5A). Figure 5B summarizes the pharmacokinetic parameters obtained under the 200 μg/mL solution and 1.0 mL/kg injection volume (dose 0.20 mg/kg) condition. The volume of distribution (V_d) was 0.352 ± 0.081 L/kg, the corresponding initial concentration (C₀) was 0.591 ± 0.131 μg/mL, and the individual half-life (t_1/2) was 7.204 ± 1.309 min. The exposure measured as AUC_0–120 was 6.256 ± 1.404 μg min/mL.

CGA significantly reduced ROS levels in rats with MI (Figure 5C). H&E staining showed improved myocardial tissue after CGA treatment (Figure 5D), and Masson staining revealed a reduced area of myocardial fibrosis (Figure 5E). Echocardiography showed significantly improved cardiac function in rats with MI (Figure 5F), with increased EF, FS, LVIDs, EDV (Figures 5G–J). Immunohistochemical staining of CD31 and SMCs indicated that CGA promotes the expression of myocardial endothelial cells, reduces the expression of myocardial stromal cells, and inhibits myocardial fibrosis and remodeling (Figures 5I–K).

Electron microscopy showed that 200 μg/mL CGA significantly promoted mitophagy and repair of damaged mitochondria in rats (Figure 6A). During MI, the double-membrane structure of the mitochondria is disrupted. With the promotion of mitophagy by CGA, the mitochondrial structure was observed to be restored, accompanied by the emergence of numerous mitophagosomes. Within the yellow circles, typical “mitophagy” is visible, where mitochondria can be clearly identified within the double-membrane structures. Figures 6B–E shows the IF results demonstrating the upregulation of CGA on the expression levels of mitophagy markers PINK1 and Parkin in myocardial tissue. WB analysis further confirmed the regulatory effect of CGA on the expression of mitophagy markers PINK1 and Parkin in cardiac tissue (Figures 6F–H).

The coefficient of variation (CV) is defined as the ratio of the standard deviation to the mean of the raw data, reflecting the degree of data dispersion. More than 85% of the metabolites in the quality control (QC) samples exhibited a CV value below 0.5, indicating relatively stable experimental data. The results presented in Figure 7A further confirm the reliability of the sequencing data. Non-targeted metabolomics analysis of the CGA and MI groups (n = 3) showed distinct metabolic profiles between the two groups (Figure 7B). PCA revealed that 159 metabolites were upregulated and 105 were downregulated in the CGA group compared to the MI group (p < 0.05) (Figure 7C). KEGG enrichment analysis showed that metabolic pathways were the most significantly enriched, with a Rich Factor close to 1.0, indicating their dominance in the samples. Additionally, the enrichment of nucleotide metabolism, thyroid hormone synthesis, and cofactor biosynthesis pathways suggests the reliability of the experimental results, as these pathways collectively regulate ROS levels and mitophagy through energy sensing, metabolic modulation, and redox balance (Figures 7D,E). HMDB enrichment analysis revealed that the following pathways—fatty acid elongation in mitochondria, mitochondrial beta-oxidation of medium chain saturated fatty acids, transfer of acetyl groups into mitochondria, pyruvate dehydrogenase complex deficiency, pyruvate metabolism, and reactive oxygen species metabolism—are all associated with the regulation of mitophagy and/or ROS. These results further confirm the regulatory role of CGA in ROS and mitophagy (Figure 7F).

Immunohistochemical results showed no toxicity of 200 μg/mL CGA on rat liver, spleen, lungs, and kidneys (Figure 8).