Gypenosides exert cardioprotective effects by promoting mitophagy and activating PI3K/Akt/GSK-3β/Mcl-1 signaling

Materials and reagents

Rat (Rattus norvegicus) heart/myocardium cells (H9c2; #CRL-1446, ATCC) were purchased from Hunan Fenghui Biotechnology Co., Ltd. (Changsha, China). GYPs (lot number J0423AS, specification ≥ 98%) were purchased from Meilunbio Biotechnology Co., Ltd. (Dalian, China). Polyvinylidene difluoride (PVDF) membrane was purchased from Millipore Corporation (Billerica, MA, USA). The extracellular oxygen consumption rate (OCR) plate assay kit was purchased from Dongren Chemical Technology Co., Ltd. (Shanghai, China). Dox, CCK-8 assay kit, luciferase-based ATP detection kit, adenovirus (Ad)-LC3-GFP, fluorescent dyes including rhodamine 123, dihydroethidium (DHE), MitoTracker Red CMXRos, and LysoTracker Green, as well as antibodies including anti-dynamin-related protein-1 (Drp-1), anti-fission protein 1 (Fis1), anti-mitofusin 1 (MFN1), anti-mitofusin 2 (MFN2), anti-PTEN-induced putative kinase 1 (PINK-1), anti-parkin, anti-myeloid cell leukemia-1 (Mcl-1), and anti-LC3II/I, were purchased from Beyotime Biotechnology Co., Ltd. (Shanghai, China). Anti-phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha (PI3KCA), anti-p-PI3K, anti-Akt, anti-p-Akt (Ser473), and anti-vinculin polyclonal antibodies were purchased from ABclonal Technology Co., Ltd (Wuhan, China). Anti-GSK-3β and anti-p-GSK-3β (Ser9) monoclonal antibodies were purchased from Proteintech Group, Inc. (Wuhan, China). Additionally, anti-α-tubulin monoclonal antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA).

Identifying constituents in GYPs and prediction of targets

Information on the monomeric saponins in GYPs was gathered from the indicated databases. The TCM systems pharmacology database (TCMSP, https://www.tcmsp-e.com/load_intro.php?id=43) was used primarily for screening bioactive compounds, while PubMed (https://pubmed.ncbi.nlm.nih.gov/) and China National Knowledge Infrastructure (CNKI, https://www.cnki.net/) databases were used for manual verification and supplementation. Potential targets of GYPs were predicted by importing the SDF files of the collected components into the SwissTargetPrediction database (http://www.swisstargetprediction.ch/) and limiting the research object to homo sapiens. Targets with a probability of ≥ 0.08 were reserved.

Network construction and enrichment analyses

The term “heart failure and mitochondria” was used to build the search query in two databases, Genecards (https://www.genecards.org/) and Online Mendelian Inheritance in Man (OMIM, https://omim.org/), in order to retrieve mitochondria-related targets in HF. The protein names were converted to the corresponding gene names using the Universal Protein database (UniProt, https://www.uniprot.org/). The disease targets were then crossed with the drug targets to screen out the potential targets of GYPs in regulating mitochondrial homeostasis during HF progression. To better understand the molecular determinants, we used the String 11.0 database (https://cn.string-db.org/) to analyze protein-protein interaction (PPI) relationships. The PPI network was built using Cytoscape 3.8.1.

Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses of the drug-disease targets were conducted using Metascape’s functional annotation tool (https://metascape.org/gp/index.html#/main/step1), with thresholds of P < 0.01. 138 The target-pathway network map was constructed using Cytoscape 3.8. 1.

Cell culture, drug treatment, and sample collection

H9c2 rat cardiomyocytes were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) under 5% CO₂ at 37 °C. After pretreatment with GYPs for 1 h, the cells were treated with 0.5 µmol/L of Dox. The following experiments were conducted after 12 h: ATP measurement, assessment of mitochondrial membrane potential (MMP) and reactive oxygen species (ROS), determination of OCR, staining for mitochondria/lysosomes, and western blot analysis. Cell viability was determined after 24 h using the CCK-8 assay kit and adjusted to the control group.

Quantification of ATP content

ATP concentration was determined using a firefly luciferase-based kit and measured on a multimode microplate reader (PerkinElmer, Waltham, MA, USA). The total ATP level was expressed as nmol/mg protein.

High content screening determination of MMP and intracellular ROS

H9c2 cells were loaded with 0.5 µmol/L of rhodamine 123 at 37 °C for 30 min to indicate the MMP. In addition, the DHE fluorescent dye was used to measure the intracellular ROS levels. After drug treatment, the H9c2 cells were incubated with 10 µmol/L of DHE for 30 min. Images of fluorescently labeled cells were then captured and analyzed using the CELENA^® X High Content Image System (Logos Biosystems, Anyang-Si, Korea).

Measurement of oxygen consumption

H9c2 cells were cultured in a 96-well black bottom plate (10⁴ cells/well). After 1 h of pretreatment with GYPs, the cells were treated with 0.5 µmol/L of Dox for 12 h. Basal oxygen consumption was determined according to the instruction manual of the extracellular OCR plate assay kit (Dojindo E297). The fluorescent signals (Ex: 500 nm, Em: 650 nm) were recorded on a multimode microplate reader every ∼10 min (PerkinElmer, Waltham, MA, USA). OCR was calculated using an analysis of the kinetic profiles acquired from measurements (Saito et al., 2023).

Detection of mitophagy

Co-localization of autophagosomes and mitochondria serves as an indicator of mitophagy. In this study, Ad-LC3-GFP was applied 24 h before the drug administration, MOI = 5. Cells were then stained with 40 nmol/L of MitoTracker Red CMXRos at 37 °C for 30 min. In addition, MitoTracker Red CMXRos and LysoTracker Green (50 nmol/L) were co-loaded into H9c2 cells for 30 min to diagnose mitochondria-lysosome co-localization. The images were taken using a laser scanning confocal microscope (Nikon, Tokyo, Japan).

Western blot analysis

Total protein was extracted from H9c2 cells and quantitated by bicinchoninic acid assay. A fixed amount of the extracted protein was separated by 10–12% SDS-PAGE gel electrophoresis (120 V, 60 min) and then transferred to PVDF membranes (220 mA, 70 min). Following blocking at room temperature with 5% skimmed milk for 1 h, the membranes were incubated with primary antibodies at 4 °C overnight and then labeled with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. Immunoreactive bands were visualized using the enhanced chemiluminescent substrate. Rabbit anti-Drp-1, anti-Fis1, anti-MFN1, anti-MFN2, anti-PINK-1, anti-parkin, anti-Mcl-1, anti-LC3II/I, anti-PI3KCA, anti-p-PI3K, anti-Akt, anti-p-Akt anti-GSK-3β, and anti-p-GSK-3β antibodies were used as primary antibodies. Mouse anti- α-tubulin or anti-vinculin antibody served as the loading control for the whole cell lysis.

Statistical analysis

Data are presented as mean ± SEM. Statistical analysis was performed in SPSS (version 13.0, IBM) using a one-way analysis of variance (ANOVA), followed by the Bonferroni post hoc test. Differences were considered statistically significant at P < 0.05. Graphs were generated using Graph Pad Prism software 9.5.1.

Identifying the bioactive compounds in GYPs

Using the TCMSP, CNKI, and PubMed databases, we identified 230 saponins in G. pentaphyllum. Potential targets of GYPs were predicted using the SwissTargetPrediction database with a probability of ≥ 0.08. Of these, 142 components were excluded when matching targets were not found in the SwissTargetPrediction database. Finally, 88 potential components were selected for further analysis (Table 1). The SwissTargetPrediction database predicted 345 potential targets of GYPs.

GYPs have high molecular weights and polarity due to the presence of one or more monosaccharide molecules, which negatively affect their oral bioavailability (OB) and drug-likeness (DL) scores. Some components, such as Gypnoside V_qt, Gypenoside I_qt, Ginsenoside rb3_qt, and Gypenoside XVIII_qt, shown in Table 1 and Fig. 2A, are derived from native saponins that release monosaccharides through the hydrolysis of glycosidic bonds. These sapogenins exhibit better oral OB and DL than their corresponding saponins while targeting more potential targets.

Identifying drug-disease targets and PPI networks

HF-targeted genes related to mitochondria were retrieved from Genecards and OMIM. Based on the correlation scores, 753 targets were recognized as disease targets. The 753 disease targets were mapped to the 345 potential targets of GYPs to obtain the overlapping targets. As shown in Fig. 2B, 71 targets were confirmed as drug-disease candidates for GYPs.

The PPI network revealed 886 interactions between the 71 drug-disease targets, with 38 nodes in the circle’s center (Fig. 2C). Some targets reflected higher degrees of involvement, including mitogen-activated protein kinase (MAPK), epidermal growth factor receptor (EGFR), PI3KCA, Mcl-1, vascular endothelial growth factor (VEGF), and peroxisome proliferator-activated receptor gamma (PPAR γ), etc. These protein targets may account for the mitochondrial actions of GYPs in treating HF.

GO enrichment and KEGG pathway analyses

The top 15 significant terms of GO enrichment analysis are shown in Fig. 3A, and indicate that the regulation of protein phosphorylation, cellular calcium ion homeostasis, MMP, autophagy, and response to growth factors were involved in the cardioprotective activity of GYPs. In addition, the top 15 signal pathways of KEGG analysis are depicted in Fig. 3B. The majority were involved with PI3K/Akt, MAPK, Ras, and transcription factor forkhead box O (FoxO) signaling pathways, as well as signaling pathways regulating calcium, apoptosis, and mitophagy. Accordingly, the following experiments verify the specific mechanisms of GYPs on HF, mainly from the perspective of the indicated process and pathways. The effects of GYPs on mitochondrial function, membrane potential, dynamics, and mitophagy were also evaluated to identify pivotal processes, followed by an investigation into the associated targets and signaling pathways.

GYPs protect H9c2 rat cardiomyocytes from Dox-induced cytotoxicity

Cell experiments were performed to evaluate the effects of GYPs on cardioprotection and mitochondrial regulation. The drug concentrations of GYPs typically range from five to 75 µg/ml according to previous research (Hu et al., 2022; Tu et al., 2021). Consistently, the CCK8 assay (Fig. 4A) showed that GYPs did not reduce the viability of H9c2 rat cardiomyocytes at concentrations below 80 µg/ml. Therefore, the non-toxic concentrations of GYPs at 10–40 µg/mL were applied in this study.

It is well-known that the anthracycline antitumor agent Dox can induce cardiotoxicity, leading to cardiomyocyte death and HF (Kitakata et al., 2022). Compared to the control group, Dox reduced cell viability by approximately 50%, which was alleviated by GYP pretreatment at 10, 20, and 40 µg/ml, respectively (Fig. 4B).

GYPs ameliorate Dox-induced mitochondrial dysfunction in H9c2 rat cardiomyocytes

Mitochondria integrate fuel metabolism to generate energy in the form of ATP. Mitochondrial dysfunction and oxidative stress have been implicated in Dox cardiomyopathy (Wu, Leung & Poon, 2022). In this study, we observed significant loss of ATP production, basal oxygen consumption, and MMP in Dox-treated H9c2 cells with increased production of ROS. GYPs did not affect mitochondrial function under basal conditions but could improve the cellular ATP, OCR, and MMP in Dox-stimulated H9c2 cells. Interestingly, GYPs showed no effects on Dox-induced ROS overproduction, indicating that GYPs had no anti-oxidative potency in cultured rat cardiomyocytes (Fig. 5).

GYPs alleviate the mitophagy block when facing Dox stimulation

Mitochondria are highly dynamic organelles. Appropriate management of mitochondrial biogenesis, fission/fusion, and mitophagy is essential for preserving mitochondrial integrity and function (Fajardo et al., 2022). In our current work, we detected the expression of critical proteins in mitochondrial quality control using western blot analysis. As shown in Figs. 6A and 6B, no significant changes were observed in MFN 1/2, Drp-1, or Fis-1 after GYP administration, indicating that GYPs do not affect mitochondrial fission and fusion. Notably, pretreatment with GYPs did not affect Dox-induced LC3 II/I, but instead enhanced the protein levels of PINK-1 and parkin, which are central mitophagy priming factors. Therefore, we further detected mitophagy in Dox-stimulated H9c2 cells with and without GYP pretreatment. Fluorescence images show the co-localization of mitochondria with autophagosomes (Fig. 6C) and lysosomes (Fig. 6D). Compared to the control group, co-localizations of both mitochondria/autophagosomes and mitochondria/lysosomes were decreased by Dox administration, whereas pretreatment with GYPs elevated these co-localizations. According to these results, GYPs can alleviate the mitophagy block induced by Dox in H9c2 rat cardiomyocytes.

GYPs increase PI3K/Akt/GSK-3β/Mcl-1 transduction in Dox-stimulated H9c2 rat cardiomyocytes

According to our network pharmacological work, multiple molecular targets and signaling pathways may be implicated in managing the effects of GYPs. The current cell study explored whether GYPs could regulate PI3K/Akt transduction, the corresponding GSK-3β phosphorylation, and Mcl-1 decrease in Dox-stimulated cardiomyocytes. Western blot analysis showed that Dox reduced the p-PI3K and p-Akt levels, which were elevated by GYPs. In addition, GYPs pretreatment also increased p-GSK-3β and the protein level of Mcl-1 compared to the Dox group (Fig. 7). These findings, consistent with the key results from network pharmacology analysis, provide evidence that GYPs exert their effects on mitophagy and cardioprotection, probably through the PI3K/Akt/GSK-3β/Mcl-1 signaling pathway. The current data did not include other potential mechanisms, such as MAPK and EGFR, due to insignificant changes in the cell experiments or lack of research basis.