Rg1-Loaded Tetrahedral DNA Nanostructures Attenuate Myocardial Ischemia–Reperfusion Injury by Suppressing Oxidative and Endoplasmic Reticulum Stress Authors: Fengxia Lin, Jiamei Huang, Zetao Chen, Fanjia Zhong, Luhua Xu, Na Liu, Miao Gan, Haoran Yin, Zhiwen Li, Xiaoduo Zhang, Yinzhi Song, Xi Wang, Hui Zhuo, Hongcai Shang, Sheng Lin, Zhicong Zeng, Rongfeng Yang Published: 2026 ABSTRACT BackgroundMyocardial ischemia–reperfusion injury (MIRI) remains a major clinical challenge in the management of acute myocardial infarction. Ginsenoside Rg1, a bioactive component from Panax ginseng, exhibits cardioprotective properties but suffers from poor bioavailability and limited tissue targeting.MethodsWe designed a novel DNA nanocarrier, Rg1@pTDN, by loading Rg1 onto a tetrahedral DNA nanostructure (TDN) modified with the myocardial-targeting peptide CREKA. The physicochemical characteristics of Rg1@pTDN were evaluated by DLS, zeta potential analysis, AFM, and gel electrophoresis. Biodistribution, biosafety, and cellular uptake were assessed in vitro and in vivo. Cardioprotective efficacy was evaluated in a murine MIRI model and a H/R injury cell model. Mechanistic studies focused on oxidative stress and endoplasmic reticulum (ER) stress pathways.ResultsRg1@pTDN exhibited uniform nanoscale structure, high Rg1 loading efficiency, and good colloidal stability. In vivo imaging revealed preferential accumulation in cardiac tissue following intravenous administration. Rg1@pTDN was well tolerated and improved survival, cardiac function, and myocardial histology in MIRI mice. It significantly reduced serum CK-MB concentrations and oxidative stress markers (MDA), while increasing antioxidant enzyme activities (SOD, GSH-Px). In vitro, Rg1@pTDN suppressed ROS accumulation and H/R–induced apoptosis in H9c2 cells. Furthermore, Rg1@pTDN alleviated H/R–induced ER stress, as shown by decreased GRP78 and CHOP mRNA expression and reduced phosphorylated PERK and CHOP protein levels.ConclusionRg1@pTDN represents a promising nanotherapeutic strategy for myocardial ischemia–reperfusion injury through targeted delivery and dual inhibition of oxidative and ER stress. This DNA-based platform offers a versatile approach for enhancing the efficacy of natural compounds in cardiovascular disease. Introduction Myocardial ischemia–reperfusion injury (MIRI) remains a major clinical challenge in the treatment of acute myocardial infarction.1 In current clinical practice, antiplatelet agents, β-blockers, and lipid-lowering therapies are essential components of acute myocardial infarction management. Current standard therapies and coronary revascularization have substantially improved clinical outcomes, but they do not fully prevent reperfusion–induced cardiomyocyte injury or subsequent adverse ventricular remodeling. While early reperfusion is essential to restore coronary perfusion and rescue ischemic myocardium, it can paradoxically cause further damage due to a cascade of detrimental events, including oxidative stress, calcium overload, inflammatory response, and various types of regulated cell death.2 Among these pathological mechanisms, ER stress plays a pivotal role in driving cardiomyocyte apoptosis. In particular, sustained activation of the PERK (protein kinase RNA-like ER kinase)/CHOP (C/EBP homologous protein) signaling axis promotes maladaptive unfolded protein response (UPR) and contributes significantly to MIRI pathology.3 Ginsenoside Rg1 (Rg1), a principal pharmacologically active component of Panax ginseng, may serve as a complementary cardioprotective agent for reperfusion-induced cellular injury. Rg1 has been reported to exert anti-oxidative, anti-apoptotic, and anti-inflammatory effects, which are closely related to the pathological mechanisms of MIRI.4 Preclinical studies have shown that Rg1 can improve cardiomyocyte survival, reduce apoptosis, and decrease myocardial infarct area in models of ischemic injury.5 Mechanistically, Rg1 mitigates ER stress by downregulating CHOP and p–PERK, thereby preserving cardiomyocyte viability under pathological conditions.6 However, the therapeutic translation of Rg1 remains hampered by its low aqueous solubility, poor stability, rapid metabolism, and inefficient tissue-targeting properties, which severely restrict its bioavailability and efficacy in vivo.7 Therefore, developing a targeted nano-delivery system may help better realize the cardioprotective potential of Rg1 in MIRI. Nanocarrier-based delivery systems have been widely developed to improve the aqueous solubility, stability, biodistribution, and therapeutic efficacy of therapeutic agents.8–11 Several established nanomedicine platforms have achieved broad biomedical or clinical application, including liposomes, lipid nanoparticles, and albumin-bound nanoparticles.12–14 In cardiovascular settings, PLGA-based nanoparticles and liposomes have also been explored for MIRI drug delivery.15,16 Acute MIRI is an early injury process involving myocardial damage, oxidative stress, endothelial dysfunction, and inflammatory activation.1,2 This pathological setting requires timely, injury-oriented, and cell–permeable drug delivery. In recent years, TDNs have emerged as a promising class of self-assembling nucleic acid-based nanocarriers with well-defined geometry, biocompatibility, and sequence-programmable architecture.17,18 More broadly, receptor-mediated cellular internalization has also been demonstrated for other DNA nanostructures.19 Their DNA strands also provide addressable sites for ligand modification, allowing CREKA to be introduced through a defined oligonucleotide anchor. This design was intended to adapt the carrier to the fibrin-enriched injury microenvironment after MIRI.20 In addition, DNA nanostructures have been reported to carry small-molecule drugs through non–covalent or structural association.21 These features support the use of pTDN as a MIRI-oriented scaffold for targeted Rg1 delivery.22 To address the limited myocardial delivery of Rg1 in acute MIRI, we developed a Rg1@pTDN, modified with the cardiac-targeting peptide CREKA to enhance myocardial accumulation. Although DNA nanostructure-based delivery systems have been explored in various disease settings, studies on cardiac injury-targeted DNA nanoplatforms for MIRI remain limited.17,22 This study systematically evaluated the physicochemical characteristics, biosafety, and biodistribution of Rg1@pTDN, and verified its cardioprotective efficacy in both in vivo and in vitro models of ischemia–reperfusion injury. Furthermore, we explored its potential mechanisms of action, focusing on the attenuation of oxidative stress and suppression of ER stress via the PERK/CHOP signaling pathway. Methods Preparation of TDNs All DNA oligonucleotides were synthesized and HPLC-purified by Sangon Biotech (Shanghai, China), and the sequences used for TDN assembly and CREKA anchoring are listed in Table S3. The DNA strands were first dissolved in nuclease-free water and quantified by UV-Vis spectrophotometry using a Hitachi U–3010 spectrometer (Japan). Equimolar concentrations of four DNA strands were mixed at 1 μM final concentration in TM buffer (10 mM Tris–HCl, 50 mM MgCl2, pH 8.0). The four strands were designed to contain complementary segments. During annealing, these segments hybridize through Watson–Crick base pairing and form a TDN with six double-helical edges and four vertices. The mixture was annealed using a thermal cycler by heating at 95°C for 10 min, followed by gradual cooling to 4°C for 20 min. The assembled TDNs were concentrated using Amicon Ultra 30 kDa centrifugal filters (Millipore, USA) and stored at 4°C until use. Modification of TDNs with CREKA Peptide To prepare the targeting-modified nanostructure, a CREKA-functionalized oligonucleotide (CREKA-oligo) was synthesized via click chemistry. An azide-modified CREKA peptide (Cys–Arg–Glu–Lys–Ala–N3) was conjugated to a DBCO-modified single-stranded DNA (5′–DBCO–TTTCGTACGATCATAGATCAAT) complementary to the capture strand of the TDNs (see Figure 1A). Both components were dissolved in phosphate-buffered saline (PBS, pH 7.4) and gently stirred at 25°C overnight. The resulting conjugates were purified by 12% denaturing polyacrylamide gel electrophoresis (PAGE) and stored at −20°C. To obtain CREKA-modified TDNs (pTDNs), the CREKA-oligo (5 μM) was annealed with preassembled TDNs (1 μM) by stepwise cooling from 45°C to 25°C. Excess unbound CREKA-oligo was removed by ultrafiltration using Amicon Ultra–0.5 mL centrifugal filters (Millipore, USA). In this configuration, CREKA was not incorporated into the core assembly of the TDNs. Instead, CREKA was displayed on the outer surface of the TDNs through a complementary DNA anchor, thereby preserving the structural integrity of the nanostructure while enabling potential interaction with fibrin-enriched injured myocardial tissue. Figure 1Construction and physicochemical characterization of Rg1@pTDN. (A) Schematic illustration of TDN self–assembly, CREKA modification, and Rg1 loading. (B) DLS–measured hydrodynamic sizes of TDN, pTDN, and Rg1@pTDN. (C) Zeta potential analysis of the three nanostructures, indicating surface charge and colloidal stability. (D) Atomic force microscopy (AFM) image of TDN, pTDN, and Rg1@pTDN showing dispersed nanoscale structures. The white line is an auxiliary height-analysis path generated by the AFM software during image inspection. (E) Agarose gel electrophoresis verifying the stepwise assembly of pTDN. Lane M, DNA marker; Lane 1, strand a; Lane 2, strands a+b; Lane 3, strands a+b+c; Lane 4, strands a+b+c+d, representing complete TDN; Lane 5, CREKA-modified pTDN. The sequences of strands a–d are listed in Table S3. The gradual retardation of electrophoretic mobility confirmed successful TDN assembly and subsequent CREKA modification. (F) Loading efficiency of Rg1 at varying initial concentrations, measured by LC–MS/MS (n = 3). (G) In vitro release profile of Rg1 from Rg1@pTDN in PBS and in PBS containing 10 μM H2O2 at 37°C, showing gradual release in PBS and accelerated release under oxidative conditions (n = 3).Image A illustrates the workflow: M13mp18 with staple strands forms TDN, CREKA-oligo addition creates pTDN and ginsenoside Rg1 loading results in Rg1@pTDN. Image B′s DLS plot compares sizes of TDN, pTDN and Rg1@pTDN, peaking around 10-20 nm, with Rg1@pTDN slightly larger. Image C′s bar chart shows zeta potential: TDN at -11 mV, pTDN at -7 mV, Rg1@pTDN at -9 mV. Image D presents AFM micrographs of TDN, pTDN and Rg1@pTDN, each with a 200 nm scale bar. Image E displays an agarose gel with lanes M, 1, 2, 3, 4, 5, showing varying band positions. Image F′s plot shows encapsulation rate vs Rg1 concentration, decreasing from 80% at 0 microM to 40% at 150 microM. Image G′s release profile plot shows Rg1 reaching near 100% by 4 h, Rg1@pTDN plus 10 microM H2O2 rising faster than Rg1@pTDN, reaching 80-90% by 12-16 h, while Rg1@pTDN gradually approaches 100% by 40 h.Multi-panel infographic on TDN, pTDN and Rg1@pTDN assembly and characterization. Construction and physicochemical characterization of Rg1@pTDN. (A) Schematic illustration of TDN self–assembly, CREKA modification, and Rg1 loading. (B) DLS–measured hydrodynamic sizes of TDN, pTDN, and Rg1@pTDN. (C) Zeta potential analysis of the three nanostructures, indicating surface charge and colloidal stability. (D) Atomic force microscopy (AFM) image of TDN, pTDN, and Rg1@pTDN showing dispersed nanoscale structures. The white line is an auxiliary height-analysis path generated by the AFM software during image inspection. (E) Agarose gel electrophoresis verifying the stepwise assembly of pTDN. Lane M, DNA marker; Lane 1, strand a; Lane 2, strands a+b; Lane 3, strands a+b+c; Lane 4, strands a+b+c+d, representing complete TDN; Lane 5, CREKA-modified pTDN. The sequences of strands a–d are listed in Table S3. The gradual retardation of electrophoretic mobility confirmed successful TDN assembly and subsequent CREKA modification. (F) Loading efficiency of Rg1 at varying initial concentrations, measured by LC–MS/MS (n = 3). (G) In vitro release profile of Rg1 from Rg1@pTDN in PBS and in PBS containing 10 μM H2O2 at 37°C, showing gradual release in PBS and accelerated release under oxidative conditions (n = 3). Loading of Ginsenoside Rg1 Onto pTDNs (Rg1@pTDN) For Rg1 loading, pTDN (250 nM) was incubated with ginsenoside Rg1 dissolved in 10% methanol at different initial concentrations (20, 40, 80, 120, and 160 μM) for 24 h at room temperature. Excess unbound Rg1 was removed by ultrafiltration using 30 kDa Amicon Ultra–0.5 mL centrifugal filters (Millipore, USA). The resulting nanostructures were designated as Rg1@pTDN. The concentration of unbound Rg1 in the filtrate was quantified by LC-MS/MS. Loading efficiency and loading capacity were calculated to evaluate the Rg1-loading performance of pTDN. Based on the loading profile, the formulation prepared with 40 μM initial Rg1 and 250 nM pTDN was selected for subsequent biological experiments, as it maintained relatively high loading efficiency. Physicochemical Characterization of Rg1@pTDN The structural integrity and physical properties of Rg1@pTDN were evaluated using agarose gel electrophoresis, atomic force microscopy (AFM), dynamic light scattering (DLS), and zeta potential analysis. Agarose gel electrophoresis was performed to assess nanostructure formation and stability. Samples (10 μL) were loaded onto 1% agarose gels prepared in 1× TBE buffer containing 12.5 mM magnesium acetate and stained with GelRed (Biotium, USA). Electrophoresis was conducted at 110 V for 4 hours, and the gels were visualized using a Bio-Rad Gel Doc imaging system. AFM imaging was carried out to examine nanostructure morphology. A 15 μL aliquot of each DNA nanostructure sample (10 nM) was deposited onto freshly cleaved mica and incubated at room temperature for 5 minutes to allow adsorption. Imaging was performed using a Bruker Dimension FastScan AFM equipped with SNL-10 super-sharp probes in PeakForce tapping mode. Images were processed using NanoScope Analysis software. Dynamic light scattering (DLS) and zeta potential measurements were performed using a Zetasizer Nano ZS instrument (Malvern Instruments, UK). For size distribution analysis, Rg1@pTDN samples were diluted to 20 nM in TM buffer (10 mM Tris–HCl, 50 mM MgCl2, pH 8.0). For zeta potential measurements, samples were placed in disposable folded capillary cells and measured in 1 × TAE buffer supplemented with 12.5 mM Mg2⁺. All measurements were conducted at 25°C, and data were analyzed using the manufacturer’s Dispersion Technology Software. Loading and Release Study of Rg1 on pTDN To verify Rg1 loading onto pTDN, Rg1@pTDN samples were analyzed by LC-MS/MS. Briefly, 100 μL of sample was mixed with 300 μL of methanol, vortexed for 60 s, and centrifuged at 17,000 × g for 15 min. The supernatant was collected for analysis. Chromatographic separation was performed using a Waters ACQUITY UPLC system equipped with a C18 column (1.8 μm, 2.1 × 50 mm). The column temperature was maintained at 40°C, the injection volume was 1 μL, and the flow rate was 400 μL/min. The mobile phase consisted of 0.1% formic acid in water as phase A and acetonitrile as phase B. The gradient was as follows: 0–1 min, 95% A; 1–3 min, 95–60% A; 3–5 min, 60–0% A; 5–7 min, 0% A; 7.0–7.1 min, 0–95% A; and 7.1–10 min, 95% A. Mass spectrometric detection was performed on an AB SCIEX 4500 triple quadrupole mass spectrometer in positive electrospray ionization mode using multiple reaction monitoring. The source parameters were set as follows: ion source temperature, 500°C; curtain gas, 25 psi; collision gas, 10 psi; ion spray voltage, 4500 V; and nebulizer temperature, 500°C. Rg1 was quantified using MultiQuant 3.0.3 software based on an external standard calibration method. The concentration of unbound Rg1 in the ultrafiltration filtrate obtained during Rg1@pTDN preparation was quantified by LC-MS/MS. The loading efficiency of Rg1 was calculated as follows: Loading efficiency (%) = (Initial amount of Rg1 added − Amount of free Rg1 in the filtrate) / Initial amount of Rg1 added × 100%. Loading capacity was calculated as the molar amount of loaded Rg1 divided by the molar amount of pTDN and expressed as the Rg1/pTDN molar ratio. For the in vitro release study, free Rg1 or Rg1@pTDN solution was placed in a dialysis bag with a molecular weight cutoff of 30 kDa and immersed in PBS or PBS containing 10 μM H2O2 at 37°C under gentle agitation. PBS containing 10 μM H2O2 was freshly prepared and used as an oxidative release medium to evaluate ROS-responsive Rg1 release. At predetermined time points, the external medium was collected and replaced with an equal volume of fresh corresponding release medium. The released Rg1 was quantified by LC-MS/MS and expressed as a percentage of the total loaded or initial Rg1. Ex vivo Biodistribution and CREKA Targeting Validation To evaluate the contribution of CREKA modification to cardiac targeting, Cy5.5-labeled Rg1@TDN and Cy5.5-labeled Rg1@pTDN were intravenously injected into MIRI mice at the same dose. At 6 h after injection, the mice were euthanized, and the heart, liver, spleen, lung, kidney, and brain were harvested for ex vivo fluorescence imaging. The imaging parameters were kept identical between groups. Regions of interest were drawn for each organ, and the average fluorescence intensity was quantified for semi-quantitative analysis. Establishment of Murine MIRI Model Male C57BL/6 mice (8 weeks old, 20–25 g) were purchased from GemPharmatech (China) and housed under specific pathogen-free (SPF) conditions. All animal procedures were approved by the Institutional Animal Care and Use Committee. MIRI was induced by transient ligation of the left anterior descending (LAD) coronary artery for 30 minutes, followed by reperfusion. Mice were anesthetized with isoflurane (1.5–2%) and mechanically ventilated. After thoracotomy, the LAD was ligated with a 7–0 silk suture approximately 1–2 mm below the left atrial appendage. Ischemia was confirmed by ECG changes and myocardial blanching. After 30 minutes, the ligature was released to allow reperfusion for 72 hours. Sham-operated mice underwent identical procedures without LAD ligation. For therapeutic intervention, Rg1@pTDN was administered via tail vein injection at 10 mg/kg 24 h before surgery and again at the onset of reperfusion. The dose was calculated based on the dry weight of the lyophilized Rg1@pTDN nanomaterial. This dose was selected as a representative working dose within the safety range supported by the biosafety evaluation. Control groups received equivalent volumes of saline, free Rg1, or pTDN. After acclimatization, mice were randomly allocated to different experimental groups using a computer-generated randomization method. Group allocation was completed before model establishment and drug administration. Outcome assessment and data analysis were performed by investigators blinded to group allocation. Survival was recorded during the 72–hour period following reperfusion. At 72 h after reperfusion, mice were deeply anesthetized with sodium pentobarbital (50 mg/kg, i.p.), and the absence of pedal withdrawal reflex was confirmed before terminal sampling. Blood was collected by cardiac puncture under deep anesthesia, followed by transcardial perfusion with PBS and heart tissue harvesting. Death was confirmed by cessation of heartbeat and respiration. All terminal procedures were performed in accordance with institutional animal care guidelines and the AVMA Guidelines for the Euthanasia of Animals. Echocardiographic Assessment of Cardiac Function At 72 hours post-reperfusion, transthoracic echocardiography was performed using a high-resolution ultrasound imaging system (Vevo 2100, VisualSonics, Canada) with a 30 MHz linear transducer. M-mode images were acquired at the papillary muscle level. Left ventricular ejection fraction (LVEF) and fractional shortening (FS) were calculated from parasternal short-axis views using the software provided by the manufacturer. Measurement of Serum CK-MB concentrations Blood samples were collected by cardiac puncture under deep anesthesia immediately. Serum was separated by centrifugation at 3,000 rpm for 10 min and stored at −80°C. Creatine kinase-MB (CK-MB) concentrations were measured using a commercial assay kit (Nanjing Jiancheng Bioengineering Institute, China) following the manufacturer’s instructions. Assessment of Oxidative Stress in Myocardial Tissue To evaluate oxidative stress levels in myocardial tissue after ischemia–reperfusion, the content of malondialdehyde (MDA) and the activities of superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) were measured using commercial kits. Left ventricular tissue from the ischemic region was rapidly harvested, rinsed in ice-cold saline, weighed, and homogenized in extraction buffer provided with the respective kits. The homogenates were centrifuged at 12,000 rpm for 10 minutes at 4°C, and the supernatants were collected for biochemical analysis. MDA levels were determined using the thiobarbituric acid (TBA) method based on colorimetric detection at 532 nm. SOD activity was quantified by the xanthine oxidase method at 550 nm. GSH-Px activity was determined using the corresponding commercial assay kit according to the manufacturer’s instructions. All kits were obtained from Nanjing Jiancheng Bioengineering Institute (China) and used according to the manufacturer’s protocols. Results were normalized to protein concentration and expressed as nmol/mg protein (MDA) or U/mg protein (SOD, GSH-Px). Histological Analysis of Myocardial Tissue Heart tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 5 μm thickness. Sections were stained with hematoxylin and eosin (H&E) for general morphology. Microscopic images were acquired using a Nikon Eclipse Ci light microscope. H/R Injury Model in H9c2 Cells H9c2 rat cardiomyoblasts (ATCC, USA) were cultured in high-glucose DMEM supplemented with 10% fetal bovine serum (Gibco, USA) and 1% penicillin-streptomycin at 37°C in a humidified atmosphere with 5% CO2. To induce H/R injury, cells were seeded in 6-well or 96-well plates and allowed to reach 80% confluence. For hypoxia, the culture medium was replaced with glucose-free DMEM, and cells were incubated in a hypoxic chamber (1% O2, 5% CO2, 94% N2) at 37°C for 6 hours. Reoxygenation was achieved by replacing the medium with complete DMEM and returning the cells to normoxic conditions (21% O2) for 12 hours. The 6 h hypoxia/12 h reoxygenation protocol was selected based on previous H9c2 H/R studies and our preliminary optimization, as this condition induces reproducible cellular injury while retaining sufficient viable cells for subsequent intervention and mechanistic analyses.23–25 Rg1@pTDN or control treatments were applied 2 hours prior to the onset of hypoxia and maintained throughout the reoxygenation phase. For in vitro experiments, cells were seeded under the same conditions, and treatment wells were randomly assigned to different experimental groups to minimize potential positional effects within culture plates. All groups were processed in parallel under identical experimental conditions. Cell Viability Assay Cell viability was assessed using the Cell Counting Kit-8 (CCK-8, Dojindo, Japan). After treatment, 10 μL of CCK-8 reagent was added to each well of 96-well plates and incubated at 37°C for 2 hours. Absorbance was measured at 450 nm using a microplate reader (BioTek, USA). Viability was expressed as a percentage of the normoxic control group. Evaluation of ROS, MDA, and SOD in H9c2 Cells Reactive oxygen species (ROS) levels were detected using the fluorescent probe DCFH-DA (Beyotime, China). After treatment, H9c2 cells were incubated with 10 μM DCFH-DA in serum-free medium at 37°C for 30 minutes in the dark. Cells were then washed three times with PBS and observed under a fluorescence microscope (Leica DMi8, Germany). Fluorescence intensity was quantified using ImageJ software. Cellular oxidative stress was further evaluated by determining MDA content and SOD activity in H9c2 cell lysates. Briefly, after H/R injury and the indicated treatments, cells were rinsed with PBS and lysed using the reagents supplied with the corresponding assay kits. The lysates were centrifuged to remove cellular debris, and the supernatants were used for biochemical measurements. MDA was assessed as an indicator of lipid peroxidation, whereas SOD activity was measured to reflect cellular antioxidant capacity. Both assays were performed using commercial kits from Nanjing Jiancheng Bioengineering Institute (China) according to the manufacturer’s instructions. The values were normalized to total protein concentration and expressed as nmol/mg protein for MDA and U/mg protein for SOD. Apoptosis Analysis by Flow Cytometry Cell apoptosis was quantified using the Annexin V-FITC/PI apoptosis detection kit (BD Biosciences, USA). After treatment, cells were harvested, washed with cold PBS, and resuspended in 1× binding buffer. Annexin V-FITC (5 μL) and propidium iodide (5 μL) were added to 100 μL of cell suspension, followed by incubation in the dark for 15 minutes at room temperature. Samples were analyzed using a BD Accuri C6 flow cytometer, and data were processed with FlowJo software. Transmission Electron Microscopy (TEM) of Endoplasmic Reticulum To evaluate ultrastructural changes of the ER, H9c2 cells were collected after H/R injury with or without Rg1@pTDN treatment. Cells were fixed in 2.5% glutaraldehyde at 4°C overnight, followed by post-fixation in 1% osmium tetroxide for 1 hour. Samples were dehydrated through a graded ethanol series, embedded in epoxy resin, and sectioned using an ultramicrotome. Ultrathin sections (70 nm) were stained with uranyl acetate and lead citrate, and observed under a transmission electron microscope (Hitachi HT7700, Japan) at 80 kV. Representative ER morphologies were imaged for comparison. Quantitative Real-Time PCR for GRP78 and CHOP Expression Total RNA was extracted from H9c2 cells using TRIzol reagent (Invitrogen, USA) following the manufacturer’s protocol. RNA concentration and purity were determined by NanoDrop 2000 (Thermo Fisher Scientific). Reverse transcription was performed using the PrimeScript RT reagent kit (Takara, Japan), and quantitative PCR was conducted using SYBR Green Master Mix (Takara) on a StepOnePlus Real-Time PCR System (Applied Biosystems, USA). Gene-specific primers were listed in Table S1. β-actin was used as the internal reference gene, and relative gene expression was calculated using the 2^–ΔΔCt method. Western Blot Analysis of PERK Phosphorylation and CHOP Protein Levels Total protein was extracted from treated H9c2 cells using RIPA lysis buffer containing protease and phosphatase inhibitors (Beyotime, China). Protein concentration was determined by BCA assay. Equal amounts of protein (30 μg) were separated on 10% SDS-PAGE gels and transferred onto PVDF membranes (Millipore, USA). Membranes were blocked with 5% non-fat milk and incubated overnight at 4°C with the following primary antibodies: Anti-PERK(1:1000, Cell Signaling Technology), Anti-phospho-PERK (1:1000, Cell Signaling Technology), Anti-CHOP (1:1000, Abcam), Anti-α-tubulin (1:5000, Proteintech). After incubation with HRP-conjugated secondary antibodies (1:5000), bands were visualized using enhanced chemiluminescence (ECL) reagent (Thermo Fisher Scientific). Band intensities were quantified using ImageJ software. p-PERK was normalized to total PERK and expressed as the p-PERK/PERK ratio, whereas CHOP was normalized to α-tubulin. Statistical Analysis Experimental data were presented as the mean ± standard deviation and analyzed using GraphPad Prism software (version 9.0). Statistical analyses were performed employing independent sample t-tests for comparisons between two groups and one-way ANOVA for comparisons among three or more groups. A p value of less than 0.05 was considered statistically significant and indicated as “*” in the bar graphs. The level of significance was further denoted as “*” p < 0.05, “**” for p < 0.01, “***” for p < 0.001, and “ns” for non-significant results with p > 0.05. Results Construction and Physicochemical Characterization of Rg1@pTDN To develop a cardiac-targeting nanodelivery system, we designed and constructed a TDN modified with the CREKA peptide and loaded with ginsenoside Rg1 to form Rg1@pTDN. A schematic illustration of the assembly process is presented in Figure 1A. Dynamic light scattering (DLS) revealed that the hydrodynamic size increased sequentially from TDN (5.04 nm) to pTDN (10.21 nm), and further to Rg1@pTDN (12.45 nm), confirming successful peptide modification and drug loading (Figure 1B). Correspondingly, zeta potential analysis showed that all nanostructures retained negative surface charges, with values of −10.96 mV for TDN, −7.396 mV for pTDN, and −8.674 mV for Rg1@pTDN. Although the absolute zeta potential value decreased after CREKA peptide modification and Rg1 loading compared with unmodified TDN, the final Rg1@pTDN formulation still maintained a negatively charged surface. Together with the DLS results, these findings supported the successful construction of Rg1@pTDN and indicated that the formulation remained well dispersed under the measurement conditions (Figure 1B and C). AFM imaging showed dispersed nanoscale particle-like structures, which supported the formation of nanoscale DNA assemblies (Figure 1D). Considering that AFM morphology may be influenced by sample drying, surface adsorption, and tip convolution effects, the successful construction of Rg1@pTDN was further supported by DLS analysis, zeta potential measurement, and agarose gel electrophoresis. Agarose gel electrophoresis further validated stepwise structural assembly: as more strands and CREKA were introduced, migration rates decreased accordingly, confirming successful TDN and pTDN formation (Figure 1E). LC-MS/MS analysis was used to validate and quantify Rg1 loading in pTDN. Rg1 was identified based on its retention time and characteristic MRM transition. The Rg1 standard and Rg1@pTDN sample showed a consistent retention time at approximately 3.9 min, whereas no corresponding Rg1 peak was observed in the blank pTDN sample (Supplementary Figure S1). Quantification using a validated standard curve (Y = 4.283X − 38.07, R2 ≈ 0.999) showed that, as the initial Rg1 concentration increased from 20 to 160 μM, the loading efficiency decreased from 80.76 ± 2.77% to 41.51 ± 3.94%, whereas the loading capacity increased from 64.6 ± 2.2 to 265.7 ± 25.2 Rg1 molecules per pTDN nanostructure and gradually approached a plateau. This inverse trend between loading efficiency and loading capacity was consistent with a concentration-dependent and saturable loading process (Figure 1F and Table S2). In vitro release studies showed that Rg1@pTDN released Rg1 gradually in PBS, whereas exposure to 10 μM H2O2 markedly accelerated Rg1 release, with cumulative release exceeding 80% at 24 h. These results suggest a gradual and ROS–responsive release profile rather than a long-term sustained-release behavior (Figure 1G). In vivo Biosafety Evaluation of Rg1@pTDN To assess systemic biosafety, C57BL/6 mice were intravenously injected with Rg1@pTDN at doses of 0, 2.5, 10, or 40 mg/kg via tail vein. After 72 hours, mice were sacrificed for toxicity evaluation. All mice survived without observable behavioral abnormalities. H&E staining of major organs (heart, liver, spleen, lung, kidney) revealed no pathological alterations such as inflammation or necrosis (Figure 2A). In addition, serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), and creatinine showed no significant differences compared to controls, indicating preserved hepatic and renal function (Figure 2B–D). These results demonstrate that Rg1@pTDN is well-tolerated in vivo and does not induce acute toxicity at doses up to 40 mg/kg, supporting its safety for further therapeutic application. Figure 2In vivo biosafety evaluation of Rg1@pTDN. (A) Representative H&E staining images of major organs (heart, liver, spleen, lung, kidney) from each group, showing no histological abnormalities. (B) Serum levels of ALT. (C) Serum levels of AST. (D) Serum levels of CR. These results indicate preserved liver and kidney function across all treatment groups. Data are presented as mean ± SD. n = 5 mice per group. ns, not significant.Image A shows micrographs of heart, lung, spleen, liver and kidney stained with hematoxylin and eosin for Rg1@pTDN doses of 0, 2.5, 10 and 40 mg/kg. Consistent tissue appearance is noted across doses for each organ. Scale bars are 200 micrometers for heart, spleen, liver and kidney. Image B presents a bar chart of alanine aminotransferase (ALT) levels, with doses on the x-axis (0, 2.5, 10, 40 mg/kg) and ALT (U/L) on the y-axis (0-50). ALT levels cluster between 30-40 U/L across all doses, with comparisons labeled as not significant (ns). Image C shows a bar chart of aspartate aminotransferase (AST) levels, with similar x-axis dosing and y-axis ranging 0-60 U/L. AST levels cluster between 30-45 U/L, also labeled ns. Image D displays a bar chart of creatinine (CR) levels, with doses on the x-axis and CR (micromole/L) on the y-axis (0-100). CR levels cluster between 40-70 micromole/L, with comparisons labeled ns.Histology of H&E organs and dose-related charts for ALT, AST and creatinine. In vivo biosafety evaluation of Rg1@pTDN. (A) Representative H&E staining images of major organs (heart, liver, spleen, lung, kidney) from each group, showing no histological abnormalities. (B) Serum levels of ALT. (C) Serum levels of AST. (D) Serum levels of CR. These results indicate preserved liver and kidney function across all treatment groups. Data are presented as mean ± SD. n = 5 mice per group. ns, not significant. Cardiac Accumulation and Cellular Uptake of Rg1@pTDN To evaluate the contribution of CREKA modification to cardiac accumulation, Cy5.5-labeled Rg1@TDN and Cy5.5-labeled Rg1@pTDN were intravenously injected into MIRI mice, followed by ex vivo fluorescence imaging of major organs 6 h after administration. As shown in Figure 3A, fluorescence signals were detected in the heart and several other organs. ROI-based quantitative analysis showed that CREKA-modified Rg1@pTDN exhibited significantly higher cardiac fluorescence intensity than Rg1@TDN (Figure 3B). Within the Rg1@pTDN group, the heart showed the strongest fluorescence signal among the examined organs, although liver accumulation was also observed (Figure 3C). Confocal fluorescence imaging further showed intracellular red fluorescence mainly in the cytoplasmic region of H9c2 cells after incubation with Cy5.5-labeled Rg1@pTDN (Figure 3D), indicating effective cellular uptake. These results suggest that CREKA modification enhanced the cardiac accumulation of Rg1@pTDN and that Rg1@pTDN could be internalized by H9c2 cardiomyoblasts. Figure 3Cardiac accumulation and cellular uptake of Rg1@pTDN. (A) Ex vivo fluorescence imaging of major organs 6 h after intravenous injection of Cy5.5-labeled Rg1@TDN or Cy5.5-labeled Rg1@pTDN. (B) ROI-based quantitative analysis of fluorescence intensity in major organs from mice treated with non-CREKA-modified Rg1@TDN or CREKA-modified Rg1@pTDN. (C) ROI-based quantitative analysis of organ fluorescence intensity in the Rg1@pTDN group, showing higher fluorescence accumulation in the heart than in other major organs. (D) Confocal fluorescence imaging showing cellular uptake of Cy5.5-labeled Rg1@pTDN in H9c2 cells. Nuclei were stained with DAPI. Scale bar: 40 μm. Data are presented as mean ± SD. n = 3 mice per group for ex vivo fluorescence imaging. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001.The image A shows labeled organs: Brain, Lung, Kidney, Spleen, Heart, Liver, with fluorescence imaging of Cy5.5-Rg1@TDN and Cy5.5-Rg1@pTDN. The image B shows a bar graph comparing average fluorescence intensity (p sec cm superscript -2 sr superscript -1) in organs for Cy5.5-Rg1@TDN and Cy5.5-Rg1@pTDN, with significant differences marked by asterisks. The image C shows a bar graph of average fluorescence intensity in the heart, liver, spleen, lung, kidney and brain, with significant differences marked. The image D shows confocal fluorescence imaging of H9c2 cells with DAPI staining, Cy5.5-Rg1@pTDN uptake and a merged view, with a scale bar of 40 micrometers.Fluorescence imaging and analysis of Rg1@TDN and Rg1@pTDN in organs and cells. Cardiac accumulation and cellular uptake of Rg1@pTDN. (A) Ex vivo fluorescence imaging of major organs 6 h after intravenous injection of Cy5.5-labeled Rg1@TDN or Cy5.5-labeled Rg1@pTDN. (B) ROI-based quantitative analysis of fluorescence intensity in major organs from mice treated with non-CREKA-modified Rg1@TDN or CREKA-modified Rg1@pTDN. (C) ROI-based quantitative analysis of organ fluorescence intensity in the Rg1@pTDN group, showing higher fluorescence accumulation in the heart than in other major organs. (D) Confocal fluorescence imaging showing cellular uptake of Cy5.5-labeled Rg1@pTDN in H9c2 cells. Nuclei were stained with DAPI. Scale bar: 40 μm. Data are presented as mean ± SD. n = 3 mice per group for ex vivo fluorescence imaging. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001. Cardioprotective Effect of Rg1@pTDN in a Murine MIRI Model To investigate the therapeutic potential of Rg1@pTDN, we assessed its protective effect in a mouse model of MIRI. Mice were treated with Rg1@pTDN (10 mg/kg), and cardiac injury was evaluated at 72 hours post-reperfusion. The Rg1@pTDN-treated group showed a higher proportion of surviving mice during the 72-h observation period (Figure 4A). Serum analysis revealed a marked reduction in CK-MB concentrations in the Rg1@pTDN group (Figure 4B), suggesting attenuation of myocardial injury. Echocardiographic imaging (Figure 4C) and corresponding quantitative analysis (Figure 4D) showed significant improvements in LVEF and FS following Rg1@pTDN treatment, reflecting preserved cardiac systolic function. These findings demonstrate that Rg1@pTDN confers effective myocardial protection in vivo, improving both cardiac function and survival after ischemia–reperfusion injury. Figure 4Cardioprotective effects of Rg1@pTDN (A) Kaplan–Meier survival curves of mice during the 72-h observation period following MIRI. Rg1@pTDN treatment showed a trend toward improved survival. (B) Serum CK–MB concentrations measured at 72 hours post–reperfusion. Rg1@pTDN markedly reduced myocardial enzyme release, indicating reduced cardiac injury. (C) Representative M–mode echocardiographic images at 72 hours after reperfusion, showing improved left ventricular motion in the Rg1@pTDN group. (D) Quantification of LVEF and FS based on echocardiography. Rg1@pTDN significantly preserved cardiac systolic function. For survival analysis, n = 12 mice per group at baseline. For CK-MB and echocardiographic analyses, n = 7 mice per group. Data are presented as mean ± SD. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001.The image A shows a line graph depicting the probability of survival over 72 hours for different groups: Sham, MIRI, Rg1, pTDN and Rg1@pTDN+MIRI. The x-axis is labeled ′Hours′ and the y-axis ′Probability of Survival (%)′. The image B shows a bar graph illustrating CK-MB levels in ng/ml for the same groups. The x-axis lists the groups and the y-axis is labeled ′CK-MB (ng/ml)′. Statistical significance is indicated with asterisks and ′ns′ for not significant. The image C shows M-mode echocardiographic images for each group, highlighting differences in left ventricular motion. The image D shows two bar graphs quantifying LVEF (%) and FS (%) for the groups. The x-axis lists the groups and the y-axis is labeled ′LVEF (%)′ and ′FS (%)′. Statistical significance is marked similarly as in image B. These images collectively demonstrate the cardioprotective effects of Rg1@pTDN treatment in a mouse model of MIRI.Rg1@pTDN improves survival, CK-MB levels and heart function after MIRI, shown in graphs and images. Cardioprotective effects of Rg1@pTDN (A) Kaplan–Meier survival curves of mice during the 72-h observation period following MIRI. Rg1@pTDN treatment showed a trend toward improved survival. (B) Serum CK–MB concentrations measured at 72 hours post–reperfusion. Rg1@pTDN markedly reduced myocardial enzyme release, indicating reduced cardiac injury. (C) Representative M–mode echocardiographic images at 72 hours after reperfusion, showing improved left ventricular motion in the Rg1@pTDN group. (D) Quantification of LVEF and FS based on echocardiography. Rg1@pTDN significantly preserved cardiac systolic function. For survival analysis, n = 12 mice per group at baseline. For CK-MB and echocardiographic analyses, n = 7 mice per group. Data are presented as mean ± SD. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001. Antioxidative Effect of Rg1@pTDN in MIRI To investigate the antioxidative effect of Rg1@pTDN, we assessed myocardial tissue 72 hours after ischemia–reperfusion injury by histological and biochemical analyses. H&E staining revealed that myocardial structural damage—characterized by disorganized fibers, edema, and inflammatory infiltration—was alleviated in the Rg1@pTDN-treated group, with notably improved tissue integrity compared to untreated controls (Figure 5A). Biochemical assays showed that MDA levels, an indicator of lipid peroxidation, were significantly reduced following Rg1@pTDN treatment. In parallel, the activities of SOD and GSH-Px were markedly elevated, indicating enhanced antioxidant defense (Figure 5B). These findings suggest that Rg1@pTDN effectively attenuates oxidative stress and contributes to myocardial protection following ischemia–reperfusion injury. Figure 5Antioxidative effect of Rg1@pTDN in MIRI. (A) Representative H&E staining images of cardiac tissue collected 72 hours after reperfusion. Rg1@pTDN treatment alleviated myocardial fiber disruption, edema, and inflammatory infiltration. (B) Quantification of oxidative stress markers in heart tissue: MDA levels were significantly reduced, while SOD and GSH-Px activities were increased following Rg1@pTDN treatment, indicating enhanced antioxidative capacity. Data are presented as mean ± SD. n = 5 mice per group. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001.Image A displays five H&E stained cardiac tissue micrographs labeled Sham, MIRI, Rg1, pTDN and Rg1@pTDN+MIRI, with a 200 micrometer scale bar. Sham shows aligned fibers and sparse nuclei. MIRI reveals disrupted structure and widened spaces. Rg1 has dense cellular areas and irregular fibers. pTDN shows elongated fibers with scattered nuclei. Rg1@pTDN+MIRI presents more continuous fibers than MIRI. Image B features three bar charts titled MDA, SOD and GSH-Px, with groups Sham, MIRI, Rg1, pTDN and Rg1@pTDN+MIRI. MDA chart y-axis is labeled MDA (nmol/mg prot) with ticks at 0, 5, 10, 15. SOD chart y-axis is SOD (U/mg prot) with ticks at 0, 50, 100, 150, 200, 250. GSH-Px chart y-axis is GSH-Px (U/mg prot) with ticks at 0, 50, 100, 150. MIRI shows higher MDA and lower SOD and GSH-Px than Sham. Rg1, pTDN and Rg1@pTDN+MIRI show lower MDA and higher SOD and GSH-Px than MIRI. Significance is marked as ns, *, ** and ***.A multi-panel scientific infographic of cardiac histology and oxidative stress markers across treatments. Antioxidative effect of Rg1@pTDN in MIRI. (A) Representative H&E staining images of cardiac tissue collected 72 hours after reperfusion. Rg1@pTDN treatment alleviated myocardial fiber disruption, edema, and inflammatory infiltration. (B) Quantification of oxidative stress markers in heart tissue: MDA levels were significantly reduced, while SOD and GSH-Px activities were increased following Rg1@pTDN treatment, indicating enhanced antioxidative capacity. Data are presented as mean ± SD. n = 5 mice per group. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001. Rg1@pTDN Attenuates H/R-Induced Oxidative Stress and Apoptosis in H9c2 Cells The protective effects of Rg1@pTDN against H/R-induced oxidative injury were further evaluated in H9c2 cardiomyoblasts. As shown in Figure 6A, H/R exposure markedly decreased cell viability, whereas Rg1@pTDN treatment significantly restored cell viability compared with the H/R group, indicating a cytoprotective effect. Intracellular ROS, detected by DCFH-DA fluorescence staining, were markedly increased after H/R injury (Figure 6B). Quantitative analysis further showed that Rg1@pTDN significantly reduced ROS fluorescence intensity compared with the H/R group (Figure 6C). Consistently, H/R increased MDA levels and reduced SOD activity, while Rg1@pTDN treatment decreased MDA levels and restored SOD activity, suggesting attenuation of oxidative stress (Figure 6D and E). In addition, flow cytometry analysis using annexin V-FITC/PI staining showed that Rg1@pTDN significantly reduced the percentage of apoptotic cells induced by H/R injury (Figure 6F). These findings suggest that Rg1@pTDN protects H9c2 cells against H/R-induced oxidative stress and apoptosis in vitro. Figure 6Rg1@pTDN protects H9c2 cells against H/R injury –induced oxidative stress and apoptosis. (A) Cell viability of H9c2 cells after H/R exposure, measured by CCK-8 assay. Rg1@pTDN significantly improved cell survival compared with the H/R group. (B) Representative fluorescence images of intracellular reactive oxygen species (ROS) levels in H9c2 cells after H/R injury. Scale bar: 200 μm. (C) Quantitative analysis of ROS fluorescence intensity. (D) MDA levels in H9c2 cells after different treatments. (E) SOD activity in H9c2 cells after different treatments. (F) Flow cytometry analysis of apoptotic cell percentages using annexin V–FITC/PI staining. Rg1@pTDN treatment significantly decreased the rate of H/R–induced apoptosis. Data are presented as mean ± SD. n = 3 independent experiments. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001.The infographic on H9c2 cells includes several panels: A bar chart (Image A) shows cell viability, with H/R lower than NC and Rg1 at pTDN plus H/R higher than H/R. Fluorescence micrographs (Image B) are labeled with groups NC, H/R, Rg1 plus H/R, pTDN plus H/R and Rg1 at pTDN plus H/R, each with a 200 µm scale bar. Another bar chart (Image C) displays relative mean fluorescence intensity, where H/R is higher than NC and Rg1 at pTDN plus H/R is lower than H/R. Image D′s bar chart shows MDA levels, with H/R higher than NC and Rg1 at pTDN plus H/R lower than H/R. Image E presents SOD activity, with H/R lower than NC and Rg1 at pTDN plus H/R higher than H/R. Image F features flow cytometry dot plots with FITC-A and PE-A axes, quadrant labels and percentages. A bar chart of apoptotic cell percentage shows H/R higher than NC and Rg1 at pTDN plus H/R lower than H/R, with ns and asterisk markers.Multi-panel infographic of H9c2 assays showing Rg1@pTDN improves H/R outcomes versus H/R alone. Rg1@pTDN protects H9c2 cells against H/R injury –induced oxidative stress and apoptosis. (A) Cell viability of H9c2 cells after H/R exposure, measured by CCK-8 assay. Rg1@pTDN significantly improved cell survival compared with the H/R group. (B) Representative fluorescence images of intracellular reactive oxygen species (ROS) levels in H9c2 cells after H/R injury. Scale bar: 200 μm. (C) Quantitative analysis of ROS fluorescence intensity. (D) MDA levels in H9c2 cells after different treatments. (E) SOD activity in H9c2 cells after different treatments. (F) Flow cytometry analysis of apoptotic cell percentages using annexin V–FITC/PI staining. Rg1@pTDN treatment significantly decreased the rate of H/R–induced apoptosis. Data are presented as mean ± SD. n = 3 independent experiments. ns, not significant; *p < 0.05, **p < 0.01, and ***p < 0.001. Rg1@pTDN Attenuates H/R-Induced ER Stress in H9c2 Cells ER stress is a key contributor to cardiomyocyte injury during ischemia–reperfusion. To determine whether Rg1@pTDN alleviates ER stress, we examined morphological and molecular indicators of ER stress in H9c2 cells following H/R. TEM revealed marked changes in ER ultrastructure after H/R injury. In Figure 7A, the ER regions are indicated by red arrows. In the NC group, the arrow indicates a relatively intact ER-like cisternal structure, consistent with normal ER morphology. In contrast, H/R induced apparent ER dilation and disorganization, indicative of ER stress. These ultrastructural abnormalities were mitigated in cells treated with Rg1@pTDN, as shown by less dilated and more preserved ER morphology compared with the H/R group, suggesting that Rg1@pTDN alleviated H/R-induced ER ultrastructural injury (Figure 7A). At the transcriptional level, qPCR analysis demonstrated that H/R significantly upregulated the expression of ER stress markers GRP78 and CHOP. Rg1@pTDN treatment markedly suppressed this upregulation (Figure 7B), indicating transcriptional inhibition of stress responses. Western blot analysis further showed that phosphorylated PERK and CHOP protein levels were reduced in the Rg1@pTDN group compared with the H/R group (Figure 7C). These findings suggest that Rg1@pTDN attenuates H/R-induced ER stress and is associated with reduced activation of the PERK/CHOP branch of the ER stress response. These results indicate that Rg1@pTDN effectively inhibits ER stress activation under H/R conditions, providing mechanistic insight into its cardioprotective properties. Figure 7Rg1@pTDN attenuates H/R injury–induced endoplasmic reticulum stress in H9c2 cells. (A) TEM images showing endoplasmic reticulum morphology in H9c2 cells. Red arrows indicate representative endoplasmic reticulum regions. H/R induced ER dilation and disorganization, which were alleviated by Rg1@pTDN treatment. (B) Quantitative real–time PCR analysis of ER stress markers GRP78 and CHOP. Rg1@pTDN significantly reduced mRNA expression levels induced by H/R. (C) Western blot analysis of p-PERK and CHOP protein expression. Rg1@pTDN decreased PERK phosphorylation and CHOP levels compared to H/R controls. The symbols “−” and “+” indicate the absence and presence of the indicated intervention, respectively. Data are presented as mean ± SD, n = 3 independent experiments. *p < 0.05, **p < 0.01, and ***p < 0.001.The image A shows TEM images of endoplasmic reticulum morphology in H9c2 cells under three conditions: NC, H/R and Rg1@pTDN+H/R. Red arrows indicate representative ER regions. The NC group shows intact ER-like cisternal structure, while H/R shows ER dilation and disorganization. Rg1@pTDN+H/R shows less dilated ER morphology. The image B shows a bar graph of relative mRNA expression levels of CHOP and GRP78 under NC, H/R and Rg1@pTDN+H/R conditions. The y-axis is labeled ′Relative mRNA expression′ and significant differences are marked with asterisks. The image C shows Western blot analysis of p-PERK, PERK, CHOP and α-tubulin protein expression under the same conditions. The y-axis is labeled ′Relative protein expression′, with significant differences marked by asterisks. The symbols ′−′ and ′+′ indicate absence and presence of interventions, respectively.Images of ER shape, mRNA and protein in H9c2 cells under varied conditions. Rg1@pTDN attenuates H/R injury–induced endoplasmic reticulum stress in H9c2 cells. (A) TEM images showing endoplasmic reticulum morphology in H9c2 cells. Red arrows indicate representative endoplasmic reticulum regions. H/R induced ER dilation and disorganization, which were alleviated by Rg1@pTDN treatment. (B) Quantitative real–time PCR analysis of ER stress markers GRP78 and CHOP. Rg1@pTDN significantly reduced mRNA expression levels induced by H/R. (C) Western blot analysis of p-PERK and CHOP protein expression. Rg1@pTDN decreased PERK phosphorylation and CHOP levels compared to H/R controls. The symbols “−” and “+” indicate the absence and presence of the indicated intervention, respectively. Data are presented as mean ± SD, n = 3 independent experiments. *p < 0.05, **p < 0.01, and ***p < 0.001. Discussion This study demonstrates that Rg1@pTDN, a ginsenoside Rg1-loaded tetrahedral DNA nanocarrier, exerts significant cardioprotective effects in models of MIRI. The nanoplatform improved cardiac function, reduced tissue damage, and attenuated oxidative and ER stress. These findings support the potential of DNA nanostructure-mediated delivery systems in enhancing the therapeutic efficacy of natural compounds for cardiovascular applications. Previous studies have demonstrated that ginsenoside Rg1 exerts cardioprotective effects in various models of myocardial injury, primarily through antioxidative and anti-apoptotic mechanisms.5,6 However, its clinical translation remains limited due to poor solubility, low oral bioavailability, and rapid systemic clearance.7 Liposome-based systems and polymeric nanoparticles have also been explored for MIRI and cardiovascular drug delivery.15,16 These studies support the value of nanodelivery strategies, but the choice of carrier should depend on the therapeutic context. In acute MIRI, myocardial injury, oxidative stress, endothelial dysfunction, and inflammatory activation create a pathological environment that requires injury-oriented and cell-permeable delivery.1,2,10 Therefore, we selected pTDN as the Rg1 delivery scaffold. TDNs have a small and defined tetrahedral structure with reported cellular uptake ability.17,18,21,22 Their DNA strands also provide addressable sites for ligand modification. This allowed CREKA to be introduced through a defined oligonucleotide anchor and adapted the carrier to the fibrin-enriched injury microenvironment.18,26 This design was intended to improve Rg1 delivery to injured myocardium and facilitate cellular entry. In addition, DNA nanostructures have been reported to carry small-molecule drugs through reversible non-covalent or structural association.17,21 Consistently, LC-MS/MS analysis confirmed effective Rg1 loading onto pTDN in this study. In the in vitro release assay, exposure to H2O2 promoted Rg1 release from Rg1@pTDN compared with PBS alone, indicating that oxidative conditions associated with reperfusion injury may facilitate Rg1 liberation from the DNA nanostructure. Thus, Rg1@pTDN provides a MIRI-oriented delivery strategy that combines injury-oriented accumulation, cellular uptake, and small-molecule loading/release. Our findings suggest that the cardioprotective effects of Rg1@pTDN are associated with concurrent attenuation of oxidative stress and ER stress, two interlinked pathological processes in MIRI. Excessive ROS generation during reperfusion not only leads to direct lipid peroxidation and mitochondrial damage, but also triggers ER dysfunction and activation of the unfolded protein response (UPR).27,28 Persistent ER stress can in turn amplify oxidative damage and promote apoptosis during ischemia–reperfusion injury, forming a self-reinforcing loop.29,30 In this study, Rg1@pTDN treatment significantly reduced ROS accumulation, restored antioxidant enzyme activity, and preserved ER ultrastructure in H9c2 cells subjected to H/R injury. These effects were accompanied by downregulation of key ER stress markers, including GRP78, CHOP, and phosphorylated PERK, indicating suppression of the PERK/CHOP branch of the UPR. This signaling axis is known to play a pivotal role in ER stress-induced apoptosis and has been implicated in the progression of ischemic myocardial damage.25 Previous studies have shown that ginsenoside Rg1 can modulate intracellular redox homeostasis and inhibit ER stress in models of myocardial injury.5,6 Our results extend these findings by demonstrating that nanostructure-mediated delivery enhances the molecular effects of Rg1, possibly through improved cellular uptake and subcellular localization. Furthermore, the intrinsic structural stability and negative charge of the DNA nanostructure may contribute to membrane interaction and ER targeting, though this warrants further investigation. Together, these data suggest that Rg1@pTDN may attenuate both oxidative stress and ER stress under ischemia–reperfusion injury conditions. The immunological profile of DNA nanostructures also warrants consideration. TDN-based nanostructures are not entirely new materials, and previous studies have reported favorable biocompatibility and relatively low immunogenicity.31 In addition, the study entitled “Evaluation of Pharmacokinetics, Immunogenicity, and Immunotoxicity of DNA Tetrahedral and DNA Polymeric Nanostructures” further supports the need to evaluate pharmacokinetics, immunogenicity, and immunotoxicity of modified DNA nanostructures in a formulation-specific manner.31 In the present study, our short-term biosafety evaluation did not reveal obvious acute toxicity after Rg1@pTDN administration. Nevertheless, the immune-related responses of the final CREKA-modified and Rg1-loaded formulation remain to be fully defined, particularly under repeated administration and longer observation conditions. Despite the promising results, several limitations of this study should be acknowledged. First, the observation period was limited to 72 hours, focusing primarily on acute-phase myocardial protection. Long-term efficacy, including effects on ventricular remodeling, fibrosis, and functional recovery beyond the acute stage, remains to be evaluated. Second, although Rg1@pTDN demonstrated therapeutic benefit, dose-response relationships and optimal administration schedules were not systematically assessed. These parameters are critical for future translational development. Third, while we verified key markers of oxidative and ER stress, the precise molecular interactions between Rg1 and the PERK/CHOP axis were not elucidated. Whether Rg1 directly modulates PERK activity, or acts through upstream signaling pathways (eg, PI3K/Akt, Nrf2), warrants further investigation using pharmacological inhibitors or gene-silencing approaches. Additionally, the current ex vivo fluorescence imaging suggested cardiac accumulation of Rg1@pTDN at a single time point, but it did not quantitatively determine the myocardial delivery fraction, blood circulation half-life, macrophage uptake, off-target organ distribution, or elimination pathway over time. Therefore, CREKA modification may enhance cardiac accumulation under MIRI conditions, but it does not confer exclusive heart-specific distribution. Signals observed in organs such as the liver and spleen may reflect uptake and clearance by the mononuclear phagocyte system, which could reduce myocardial delivery efficiency and increase off-target organ exposure, especially under repeated administration. Although Rg1@pTDN showed promising acute cardioprotective effects in the present study, its translational application should be interpreted cautiously. The current evidence was mainly derived from mouse MIRI and H9c2 cell models, which may not fully recapitulate the complexity of human myocardial ischemia–reperfusion injury. In addition, more comprehensive pharmacokinetic and clearance profiles, long-term safety evaluation, and quantitative biodistribution analysis would further strengthen the evaluation of this nanoplatform. Future studies using more clinically relevant models, such as large animal models or ex vivo human cardiac tissues, are needed to validate its therapeutic potential. Meanwhile, scalable and reproducible preparation procedures, batch-to-batch consistency, and regulatory-compliant quality control should be established before clinical translation can be considered. Nevertheless, the present findings demonstrate that Rg1@pTDN combines the bioactivity of Rg1 with the structural advantages of TDNs, including biocompatibility, programmable architecture, cellular uptake capacity, and preferential cardiac accumulation. Together with its regulation of oxidative stress and ER stress, this platform may provide a promising nanotherapeutic strategy for ischemic heart disease and may be further optimized through modular modification with additional targeting ligands, therapeutic cargos, or stimuli-responsive elements. Conclusion In summary, we developed a novel Rg1@pTDN, and demonstrated its cardioprotective efficacy in both in vivo and in vitro models of MIRI. Rg1@pTDN exhibited favorable physicochemical properties, excellent biocompatibility, and enhanced cardiac accumulation. Functionally, it improved cardiac performance, reduced myocardial injury, and mitigated oxidative stress and ER stress, accompanied by reduced activation of the PERK/CHOP signaling pathway. These findings support the potential of Rg1@pTDN as a promising nanotherapeutic platform for targeted intervention in ischemic heart disease.