Biomimetic macrophage membrane-coated nanoparticles for targeted cardioprotection against myocardial ischaemia-reperfusion injury

Abstract

Myocardial ischaemia-reperfusion injury (MIRI) exacerbates cardiomyocyte damage through oxidative stress, inflammation, and apoptosis, limiting the efficacy of reperfusion therapies. We developed PB1-Lip@M-CST, a biomimetic nanoparticle comprising procyanidin B1 (PB1)-loaded liposomes coated with macrophage cell membranes and surface-functionalized with the cardiac-specific targeting peptide CSTSMLKAC (CST). The nanoparticles were characterized for hydrodynamic size, zeta potential, core-shell morphology, retention of macrophage surface proteins (CCR2, CD11b, CD36), high encapsulation efficiency, and sustained drug release. In vitro, PB1-Lip@M-CST markedly reduced phagocytic uptake by RAW 264.7 macrophages, enhanced internalization in HL-1 cardiomyocytes in a CST-dependent manner, and provided superior protection in CoCl₂-induced hypoxia-reoxygenation models by suppressing reactive oxygen species, lipid peroxidation, proinflammatory cytokines, and apoptosis. Mechanistic studies using CXCR4-overexpressing HL-1 cells and Plerixafor-treated cells confirmed that these effects were mediated via modulation of the CXCR4-associated PI3K/AKT1 signaling pathway and restoration of the BAX/BCL-2 balance. In a mouse model of MIRI, intravenous administration of PB1-Lip@M-CST prolonged systemic circulation, with markedly enhanced cardiac accumulation (highest heart AUC) and improved therapeutic index, reduced infarct size to 25.58 ± 2.96% (versus 49.22 ± 3.03% in the model group), preserved left ventricular ejection fraction at 50.42 ± 8.74%, attenuated serum injury biomarkers, and inhibited long-term fibrosis and apoptosis, with an excellent biosafety profile. These findings establish PB1-Lip@M-CST as a novel biomimetic platform that integrates macrophage membrane-mediated immune evasion with CST-directed cardiac homing to achieve cardiac-targeted delivery of PB1 and subsequent CXCR4-dependent modulation of the PI3K/AKT1 pathway in MIRI.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-58344-x.

Background

Myocardial ischaemia-reperfusion injury (MIRI) occurs upon restoration of blood flow to ischaemic cardiac tissue following coronary occlusion, paradoxically exacerbating cardiomyocyte damage and contributing substantially to global morbidity, mortality, and healthcare burden¹. Although percutaneous coronary intervention and thrombolytic agents restore vessel patency, these interventions are limited by risks of restenosis, haemorrhage, and a narrow therapeutic window^2,3. Reperfusion itself intensifies injury through oxidative stress, inflammation, and programmed cell death, underscoring the urgent need for targeted cardioprotective therapies^4–6.

The MIRI microenvironment is characterized by excessive reactive oxygen and nitrogen species (ROS/RNS) generated by mitochondria, xanthine oxidase, NADPH oxidase, and uncoupled nitric oxide synthase, resulting in lipid peroxidation, protein modification, DNA damage, and activation of stress pathways. Concomitantly, damage-associated molecular patterns (DAMPs) engage Toll-like receptors (TLRs) and the NLRP3 inflammasome, driving release of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6, IL-18), neutrophil infiltration, and microvascular obstruction⁷. Multiple cell-death modalities-apoptosis (via cytochrome c and caspase activation), necrosis, necroptosis (RIPK1/RIPK3/MLKL), pyroptosis (gasdermin D), ferroptosis (iron-dependent lipid peroxidation), and dysregulated autophagy-further compound tissue loss⁴. Additional contributors include endothelial dysfunction, calcium overload, mitochondrial impairment, and dynamic shifts in innate and adaptive immune responses^8,9. A key molecular player in this pathological cascade is the chemokine receptor CXCR4, which is markedly upregulated on cardiomyocytes and endothelial cells during MIRI. CXCR4 amplifies inflammatory signaling, promotes neutrophil and monocyte recruitment to the injured myocardium, enhances ROS generation, and drives cardiomyocyte apoptosis, at least in part through aberrant activation of downstream PI3K/AKT1 signaling, ultimately exacerbating tissue injury and impairing functional recovery^10,11. Pharmacological or genetic inhibition of CXCR4 has been shown to attenuate these detrimental effects in preclinical MIRI models, highlighting CXCR4 as a promising therapeutic target.

Procyanidin B subtypes, natural dimeric flavonoids abundant in grape seeds and cocoa, are well recognised for their potent antioxidant, anti-inflammatory, and cardioprotective properties¹². To identify the optimal subtype for therapeutic application in MIRI, eight procyanidin B compounds (PB1-PB8) were systematically screened using a CoCl₂-induced chemical hypoxia/reoxygenation (H/R) model in HL-1 cells. Following establishment of the H/R model (200 µM CoCl₂ for 6 h, followed by 24 h reoxygenation), cells were treated with 20 µg/mL of each subtype for 48 h. Intracellular ROS levels were quantified using the DCFH-DA fluorescent probe. As shown in Supplementary Fig. S1, PB1 exhibited the most pronounced inhibitory effect on intracellular ROS production among all eight subtypes (P < 0.0001 versus model group), significantly outperforming PB2, PB3, PB4, and PB6, while PB8 displayed minimal ROS-scavenging activity. PB1 was therefore selected as the therapeutic payload based on its superior ROS-scavenging activity and well-documented antioxidant and anti-inflammatory properties. PB1 has been reported to neutralize ROS, activate Nrf2/ARE signaling, inhibit NF-κB and MAPK cascades, and modulate the BAX/BCL-2 balance, thereby attenuating cardiomyocyte apoptosis¹³. However, PB1’s high water solubility leads to rapid systemic clearance and insufficient cardiac accumulation when administered in free form, severely limiting its therapeutic efficacy in MIRI.

Liposomes provide an established platform for encapsulating hydrophilic agents such as PB1, enhancing stability, prolonging circulation, and potentially enabling controlled release in ischaemic tissues^14,15. Targeted liposomes further advance therapeutic precision by incorporating ligands that promote accumulation in injured cardiac tissues, thereby minimizing systemic exposure, reducing adverse effects, and optimizing efficacy against MIRI¹⁶. Recent biomimetic advances have introduced macrophage membrane coatings that confer immune evasion and inflammation-homing properties. These coatings naturally retain key surface proteins, including the chemokine receptor CCR2, adhesion molecule CD11b, and CD36, which facilitate adhesion to inflamed endothelium, promote recruitment to ischaemic tissue, reduce rapid clearance by the mononuclear phagocyte system, and thereby improve infarct-zone delivery while prolonging circulation^17–20. Recent studies on macrophage membrane-coated or hybrid nanoparticles have confirmed enhanced cardiac accumulation, reduced infarct size, and improved cardioprotection in heart tissue models²¹. Complementing this approach, the cardiac-specific targeting peptide CSTSMLKAC (CST) selectively binds receptors overexpressed on injured cardiomyocytes, thereby promoting precise myocardial accumulation and enhanced cellular internalization²². Importantly, CST itself does not directly bind CXCR4. Instead, CST enhances cardiac homing of the nanoplatform, thereby facilitating the delivery of its therapeutic payload (PB1) to the ischaemic myocardium where CXCR4 is markedly upregulated. While macrophage membrane-coated nanoparticles and CST peptides have been individually explored, their combined application with PB1 for cardiac-targeted MIRI therapy has not been previously reported.

We therefore hypothesized that integrating PB1-loaded liposomes with CST-conjugated macrophage membranes would create a biomimetic nanoplatform (PB1-Lip@M-CST) capable of achieving synergistic immune evasion, inflammation homing, cardiac-specific targeting, and CXCR4-dependent inhibition of the PI3K/AKT1 pathway in the ischaemic myocardium. This integrated strategy is designed to simultaneously attenuate oxidative stress, inflammation, and apoptosis while minimizing off-target effects, with the potential to provide enhanced cardioprotection against MIRI compared with single-modality systems. The present study was undertaken to construct, characterize, and rigorously evaluate this novel platform both in vitro and in vivo.

Materials and methods

Materials

Procyanidin B1 (PB1, Cat. no. P829888), 3,3'-Dioctadecyloxacarbocyanine perchlorate (Dio, Cat. no. D863629), Cobalt(II) chloride (CoCl₂, Cat. no. C804815), Soybean phosphatidylcholine (SPC, Cat: L915126), cholesterol (Cat. no. C6213), and mPEG2000-DSPE (Cat. no. D875871) were obtained from Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China). The cardiac-targeting peptide CSTSMLKAC (CST) and its scrambled control peptide CKSLMATSC (CKS) were synthesized by Anhui Guoping Pharmaceutical Co., Ltd. (Hefei, China). RAW 264.7 macrophage cells were purchased from Wuhan Shang’en Biotechnology Co., Ltd. (Wuhan, China). HL-1 cells were purchased from Shanghai Zhongqiao Xinzhou Biotechnology Co., Ltd. (Shanghai, China). The BCA protein assay kit (Cat. P0009) was purchased from Beyotime Biotechnology Co., Ltd. (Shanghai, China). CD11b Polyclonal Antibody (Cat. no. 31745-1-AP), CD36 Polyclonal Antibody (Cat. no. 32731-1-AP), BAX Polyclonal antibody (Cat. no. 50599-2-Ig), Bcl2 Recombinant monoclonal antibody (Cat. no. 82469-6-RR), AKT1-Specific Recombinant monoclonal antibody (Cat. no. 80457-1-RR), Phospho-AKT1(Ser473) Recombinant monoclonal antibody (Cat. no. 80462-1-RR), PI3K p110 alpha recombinant monoclonal antibody (Cat. no. 82796-4-RR),  and GAPDH Polyclonal Antibody (Cat. no. 10494-1-AP) were obtained from Proteintech (Wuhan, China). CCR2 (PT1511R) PT™ Rabbit mAb (cat. no. YM9353), CXCR4 Rabbit pAb (Cat. YT1800)  and Goat Anti Rabbit IgG (H + L) (HRP) (Cat. no. RS0002) were obtained from Immunoway Biotechnology (Suzhou, China). All other reagents and solvents were of analytical grade and used without further purification.

Cell culture

HL-1 cells were maintained in Claycomb medium supplemented with 10% (v/v) fetal bovine serum (FBS) and 1% (v/v) penicillin-streptomycin, and 0.1 mM norepinephrine, as recommended for HL-1 maintenance. RAW 264.7 macrophages were cultured in high-glucose Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% (v/v) FBS and 1% (v/v) penicillin-streptomycin. Both cell lines were incubated at 37 °C in a humidified atmosphere containing 5% CO₂.

Isolation of macrophage membranes

Macrophage cell membranes were isolated from RAW 264.7 cells using a low-speed mechanical disruption and differential centrifugation protocol adapted from the method described in Gao et al.²³. Cells at a density of 2 × 10⁷ cells/mL were harvested by gentle scraping with a cell scraper, washed twice with ice-cold PBS (pH 7.4), and resuspended in ice-cold TM buffer (10 mM Tris-HCl, pH 7.4, containing 1 mM MgCl₂). The suspension was incubated on ice for 10–15 min to allow hypotonic swelling and lysis, followed by homogenization using a Dounce homogenizer with a tight-fitting pestle (30 passes). The homogenate was then further mechanically disrupted by sequential extrusion through 400-nm and 200-nm polycarbonate membranes using a mini-extruder (at least 15 passes for each pore size) at 4 °C. Sucrose was added to the homogenate to a final concentration of 0.25 M, followed by centrifugation at 2,000 × g for 10 min at 4 °C to remove cell debris and nuclei. The supernatant was collected and centrifuged at 10,000×g for 30 min at 4 °C to pellet the membranes. The membrane pellet was washed twice with ice-cold TM buffer containing 0.25 M sucrose (10,000 × g, 30 min each). Total membrane protein content was quantified using a BCA protein assay kit. Isolated membranes were aliquoted and stored at − 80 °C until use.

Preparation of PB1-loaded liposomes (PB1-Lip)

PB1-loaded liposomes were prepared by the thin-film hydration method followed by extrusion. SPC (20 mg), cholesterol (5 mg), and mPEG2000-DSPE (5 mg) (molar ratio approximately 65:30:5) were dissolved in 5 mL chloroform in a round-bottom flask. The organic solvent was removed under reduced pressure at 40 °C using a rotary evaporator (Gongyi Yuxiang Instrument Co., Ltd., model RE-52 A, Gongyi, China) to form a uniform thin lipid film. The film was further dried under vacuum for 2 h. The lipid film was hydrated with 5 mL PBS (pH 7.4) containing 2 mg PB1 at 50 °C (above the lipid phase transition temperature) with vigorous vortexing for 30 min using a vortex mixer (IKA, model Vortex Genius 3). The resulting multilamellar vesicles were probe-sonicated on ice (100 W, 3 s on/2 s off cycles, total sonication time 5 min) to reduce vesicle size and extruded 10 times sequentially through 400-nm then 200-nm polycarbonate membranes using a mini-extruder (Avanti Polar Lipids, cat. no. 61001-1Ea, Alabaster, AL, USA). Unencapsulated PB1 was removed by ultrafiltration (MWCO 10 kDa) at 4,000×g for 30 min using a high-speed refrigerated centrifuge (Eppendorf, model 5810R, Hamburg, Germany). Blank liposomes were prepared identically without PB1.

CST peptide conjugation and membrane coating (PB1-Lip@M-CST)

The CST-DSPE conjugate was prepared by reacting CST peptide (2 mg) with DSPE-PEG2000-NHS at a molar ratio of 1:1.5 in 1 mL anhydrous DMSO containing 10 µL triethylamine. The mixture was stirred for 12 h at room temperature using a magnetic stirrer (Hangzhou Jingfei Instrument Technology Co., Ltd., model MS2022-171, Hangzhou, China), dialyzed (MWCO 3.5 kDa) against deionized water for 24 h with buffer changes every 6 h and lyophilized using a freeze dryer (Qingdao Yonghe Chuangxin Electronic Technology Co., Ltd., model CTFD-10 S-U, Qingdao, China). The lyophilized CST-DSPE conjugate was stored at − 20 °C until use.

CST-modified macrophage membrane (M-CST) was obtained by incubating the CST-DSPE conjugate (0.2 mg) with isolated macrophage membranes (1 mg total protein) in 1 mL PBS (pH 7.4) at room temperature under gentle shaking for 30 min. After incubation, the mixture was subjected to centrifugal ultrafiltration (10 kDa MWCO) to remove unbound CST-DSPE conjugate. The insertion efficiency was determined by HPLC quantification of unbound CST-DSPE in the filtrate and was 94.23 ± 2.35%. HPLC analysis was performed using a Shimadzu LC-20 series HPLC system (Shimadzu Corporation, Kyoto, Japan) equipped with a reverse-phase C18 column (250 mm × 4.6 mm, 5 μm), a linear gradient of 5–95% acetonitrile in 0.1% trifluoroacetic acid (TFA) over 20 min at a flow rate of 1.0 mL/min, and UV detection at 220 nm.

For membrane coating, PB1-Lip suspension (equivalent to 1 mg total lipid) was mixed with M-CST at a membrane protein-to-lipid weight ratio of 1:100 in PBS (pH 7.4). The mixture was bath-sonicated (42 kHz, 100 W) for 2 min at 4 °C using an ultrasonic cleaner (Shenzhen Yunyi Technology Co., Ltd., model YS0410, Shenzhen, China) and then sequentially extruded 15 times through 400-nm polycarbonate membranes followed by 15 times through 200-nm polycarbonate membranes using a mini-extruder (Avanti Polar Lipids, cat. no. 61001-1Ea, Alabaster, AL, USA) to form PB1-Lip@M-CST. PB1-Lip@M (without CST) was prepared under identical conditions using unmodified macrophage membranes.

Encapsulation efficiency and drug loading

The encapsulation efficiency (EE) and drug loading (DL) of procyanidin B1 (PB1) in PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST were determined as follows. After removal of unencapsulated PB1 by ultrafiltration (MWCO 10 kDa, 4,000 × g, 30 min at 4 °C), 100 µL of the purified nanoparticle suspension was mixed with 900 µL methanol to disrupt the liposomal structure and release the encapsulated PB1. The mixture was vortexed vigorously for 5 min at room temperature and centrifuged at 12,000 × g for 10 min at 4 °C. The supernatant was collected, appropriately diluted, and the PB1 concentration was quantified by LC-MS/MS using the same conditions as those employed for pharmacokinetic and tissue distribution analyses (Shimadzu LCMS-8040 triple quadrupole system operated in negative ion mode; MRM transition m/z 577.10 → 425.15; chromatographic separation on a Shimadzu Inertsil ODS-3 C18 column, 150 mm × 2.1 mm, 5 μm; mobile phase gradient of 5–80% acetonitrile in 0.1% formic acid; flow rate 0.3 mL/min; column temperature 35 °C; external standard calibration curve 3.125–1000 ng/mL).

EE and DL were calculated according to the following equations:

1

2

where M_enc is the mass of PB1 encapsulated in the nanoparticles, M_add is the mass of PB1 initially added during thin-film hydration, and M_NP is the total mass of lyophilized nanoparticles (determined by lyophilizing a known volume of the purified suspension to constant weight and weighing the resulting dry powder). All measurements were performed in triplicate, and results are expressed as mean ± standard deviation (SD).

Characterization of nanoparticles

Hydrodynamic diameter, polydispersity index (PDI), and zeta potential were measured in triplicate by dynamic light scattering (DLS) at 25 °C after 100-fold dilution in deionized water using a Brookhaven NanoBrook Omni (Brookhaven Instruments, Holtsville, NY, USA). Morphology was examined by transmission electron microscopy (TEM, JEM-2100Plus, JEOL Ltd., Tokyo, Japan) after negative staining with 2% (w/v) phosphotungstic acid (pH 7.0) for 30 s followed by air-drying. Attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectra were acquired on a Nicolet iS50 spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) using freeze-dried samples (64 scans at 4 cm⁻¹ resolution). The crystalline state was assessed by X-ray diffraction (XRD) using a Rigaku RINT 2000 diffractometer (Rigaku Corporation, Tokyo, Japan) with Cu Kα radiation over the 2θ range of 5°–80° at a scan rate of 5°/min. Thermal properties were evaluated by differential scanning calorimetry (DSC) on a DSC214 Polyma instrument (NETZSCH-Gerätebau GmbH, Selb, Germany) at a heating rate of 10 °C/min from 30 to 300 °C under nitrogen atmosphere, and by thermogravimetric analysis (TGA) on a TGA2 analyzer (Mettler-Toledo International Inc., Columbus, OH, USA) at 10 °C/min from 30 to 600 °C under nitrogen flow.

In vitro drug release was performed by dialysis (MWCO 3.5 kDa). One milliliter of nanoparticle suspension was placed in a dialysis bag and dialyzed against 200 mL PBS (pH 7.4) containing 0.5% (w/v) Tween-80 at 37 °C with shaking at 100 rpm. At predetermined time points, 1 mL aliquots were withdrawn and replaced with an equal volume of fresh medium to maintain sink conditions. The concentration of released PB1 was quantified by LC-MS/MS using an LCMS-8040 triple quadrupole system (Shimadzu Corporation, Kyoto, Japan).

Successful retention of macrophage membrane proteins on PB1-Lip@M-CST was confirmed by SDS-PAGE and Western blotting. Briefly, samples and raw macrophage membranes (20 µg total protein per lane, quantified by BCA assay) were separated on 10% SDS-PAGE gels, transferred to PVDF membranes, and probed with primary antibodies against CCR2, CD11b, and CD36 (listed in “Materials” section), followed by HRP-conjugated secondary antibodies. Protein bands were visualized by enhanced chemiluminescence (ECL) detection.

In vitro cytocompatibility, intracellular escape, and cellular uptake studies

Cytocompatibility evaluation (CCK-8 assay)

HL-1 cells were seeded in 96-well plates at a density of 5 × 10³ cells per well and allowed to adhere overnight. After cell adhesion, the cells were divided into three treatment groups: free PB1, blank Lip@M-CST, and PB1-Lip@M-CST. Each group was tested across six concentration gradients (0, 5, 10, 20, 40, and 80 µg/mL). Concentrations for the free PB1 and PB1-Lip@M-CST groups refer to the final PB1 concentration, whereas the blank Lip@M-CST carrier was prepared at mass concentrations equivalent to the corresponding PB1 content in the PB1-Lip@M-CST group. Each well received 100 µL of culture medium containing the respective formulation, with four replicate wells per group. Untreated wells (medium + cells only) served as negative controls, and cell-free wells (medium only) served as blanks for background subtraction. After 24 h of incubation at 37 °C, the drug-containing medium was discarded, the cells were washed twice with PBS, and 100 µL of serum-free medium containing 10 µL CCK-8 reagent was added to each well. The plates were incubated at 37 °C in the dark for 35 min. Absorbance was measured at 450 nm using a microplate reader (Molecular Devices, SpectraMax i3x). Cell viability was calculated using the following formula:

3

where A_s is the absorbance of experimental wells (treated with PB1, Lip@M-CST, or PB1-Lip@M-CST), A_c is the absorbance of untreated negative control wells, and A_b is the absorbance of blank wells.

Cellular uptake and targeting specificity in HL-1 cells

HL-1 cells grown to 50–60% confluence were divided into five groups: PB1-Lip, PB1-Lip@M, PB1-Lip@M-CST, CKS blocking (negative control), and CST blocking group. All formulations were labeled with DiO (0.5 wt% of total lipid) by incorporating the dye into the lipid phase prior to film formation and applied at a final concentration of 20 µg/mL (expressed as PB1 equivalent). Each group was tested in triplicate. For the CST blocking group, cells were pretreated with 2 µg/mL free CST peptide for 2 h prior to nanoparticle addition. This concentration was selected based on previous studies demonstrating that CST peptide can effectively achieve competitive receptor saturation and inhibits CST-mediated cellular uptake in cardiomyocytes^22,24. To confirm sequence-specific targeting, parallel wells were pretreated with CKS (CKSLMATSC, possessing the same amino acid composition but randomized sequence; 2 µg/mL) for 2 h at 37 °C before incubation with DiO-labeled PB1-Lip@M-CST. After incubation for 1, 2, or 4 h, the culture supernatant was removed, and cells were washed three times with ice-cold PBS. Cells were fixed with 4% paraformaldehyde for 20 min and washed three times with PBS. DAPI staining solution (1 µg/mL in PBS) was added, and the cells were incubated at room temperature in the dark for 20 min, followed by three PBS washes. Coverslips were mounted cell-side up and immediately examined using a confocal laser scanning microscope under identical gain and exposure settings for all groups. Fluorescence images were acquired at 405 nm (DAPI, blue, nuclei) and 488 nm (DiO, green) excitation wavelengths, with five randomly selected fields captured per sample under identical imaging parameters. All experiments were performed in triplicate.

Intracellular escape analysis in RAW 264.7 macrophages

RAW 264.7 macrophages grown to 50–60% confluence were divided into three groups: PB1-Lip, Lip@M-CST (blank carrier), and PB1-Lip@M-CST. The cells were incubated with the respective DiO-labeled nanoplatforms at a final concentration of 20 µg/mL (expressed as PB1 equivalent) for 1, 2, or 4 h (three replicate wells per group and time point). After incubation, the culture supernatant was aspirated, and the cells were gently washed three times with 2 mL ice-cold PBS. Cells were then fixed with 4% paraformaldehyde for 20 min at room temperature, followed by three additional washes with ice-cold PBS. Subsequently, the cells were stained with DAPI and incubated at room temperature in the dark for 20 min. After staining, the cells were washed three times with PBS to remove unbound dye. Coverslips were carefully removed, placed cell-side up in the center of glass slides, mounted with an appropriate amount of anti-fade mounting medium, and sealed with nail polish. Samples were immediately imaged by confocal laser scanning microscopy under identical acquisition parameters across all samples (laser intensity, exposure time, and gain). DAPI fluorescence was detected using 405 nm excitation, and DiO fluorescence was detected using 488 nm excitation. . All experiments were performed in triplicate.

Evaluation of intracellular ROS levels and early apoptosis in HL-1 cells under hypoxia/reoxygenation

HL-1 cells were seeded in 6-well plates at a density of 5 × 10⁵ cells/well and grown to 70–80% confluence. The cells were then divided into the following groups: (1) Control (normoxic culture without intervention); (2) Model (chemical hypoxia was induced using 200 µM CoCl₂ for 6 h, followed by replacement with fresh complete medium for 24 h to simulate reoxygenation); (3) AC (positive control: hypoxia/reoxygenation model followed by reoxygenation in medium containing 100 µM ascorbic acid for 24 h); (4) free PB1 (hypoxia/reoxygenation model followed by reoxygenation in medium containing 20 µg/mL free PB1 for 24 h); (5–7) PB1-Lip, PB1-Lip@M, or PB1-Lip@M-CST (hypoxia/reoxygenation model followed by reoxygenation in medium containing 20 µg/mL of the respective formulation, all concentrations expressed as PB1 equivalent).

Following the 24 h intervention, medium was discarded and cells were washed three times with ice-cold PBS. For ROS detection, serum-free medium containing 10 µM DCFH-DA was added and cells were incubated at 37 °C in the dark for 30 min. Cells were then washed three times with PBS to remove excess probe prior to imaging or flow cytometry. Intracellular ROS was visualized by fluorescence microscopy (five randomly selected non-overlapping fields per well), and mean fluorescence intensity was quantified using ImageJ software. Parallel samples were analyzed by flow cytometry: cells were harvested by trypsinization, centrifuged at 300 × g for 5 min at room temperature, washed once with PBS, and resuspended at 1 × 10⁶ cells/mL. Aliquots of 100 µL (1 × 10⁵ cells) were analyzed (excitation 488 nm, emission 525 nm), and and mean fluorescence intensity was calculated using FlowJo™ v10 software (BD Life Sciences, Ashland, OR, USA).

Early apoptosis was assessed by Annexin V-FITC/PI staining using a commercial apoptosis detection kit (Abbkine, Cat. no. KTA0006, Wuhan, China). After the same treatment, cells were washed twice with ice-cold PBS, gently detached using EDTA-free trypsin for the minimal time required, with digestion terminated by addition of complete medium. Single-cell suspensions were centrifuged at 300 × g for 5 min, washed twice with PBS (same conditions), resuspended in 500 µL of 1 × Binding Buffer at 1 × 10⁶ cells/mL, and stained with 2 µL Annexin V-FITC plus 2.5 µL PI (25 °C, dark, 15 min). After staining, samples were directly analyzed by flow cytometry within 30 min. The untreated normoxic control group served as the biological negative control, and the early apoptosis rate (Annexin V⁺/PI⁻ population) was quantified from the corresponding double-parameter scatter plots. All experiments were performed in triplicate.

Mechanistic validation of CXCR4/PI3K/AKT signaling in the cardioprotective effects of PB1 and PB1-Lip@M-CST

CXCR4-overexpressing HL-1 Cell Line

The stable CXCR4-overexpressing HL-1 cell line (CXCR4 OE) and the corresponding empty vector control (Vector) were constructed by lentiviral transduction and provided by Guizhou Hejin Biotechnology Co., Ltd. Stable expression was confirmed by reverse transcription quantitative polymerase chain reaction (RT-qPCR) and western blot analysis prior to use.

Experimental grouping and hypoxia/reoxygenation treatment

Logarithmically growing normal HL-1 cells and CXCR4-overexpressing HL-1 cells (CXCR4 OE) were seeded at appropriate densities (12-well plates: 1 × 10⁵ cells/well; 6-well plates: 2 × 10⁵ cells/well) and cultured to 70–80% confluence prior to treatment. All procedures were performed under aseptic conditions. Each experimental group was performed in triplicate and independently repeated three times.

CXCR4 gain-of-function studies (overexpression)

Eight experimental groups were established as follows. CXCR4 OE cells under normoxia were included to assess baseline effects of CXCR4 overexpression: (1) Control (normoxic culture without CoCl₂ treatment); (2) Model (CoCl₂-induced H/R); (3) PB1; (4) PB1-Lip@M-CST; (5) CXCR4 OE; (6) CXCR4 OE + Model; (7) CXCR4 OE + Model + PB1; (8) CXCR4 OE + Model + PB1-Lip@M-CST. Chemical hypoxia was induced using 200 µM CoCl₂ for 6 h, followed by replacement with fresh complete medium for 24 h to simulate reoxygenation. Where indicated, PB1 (20 µg/mL, based on PB1 content) or PB1-Lip@M-CST (20 µg/mL) was added at the beginning of reoxygenation.

CXCR4 loss-of-function studies (pharmacological inhibition)

Plerixafor (Ple) was prepared from a 50 mM stock solution in DMSO and diluted directly into complete culture medium to achieve a final working concentration of 5 µM. This protocol results in a final DMSO concentration of 0.01% (v/v) in all treatment groups. The final DMSO concentration was maintained at 0.01% (v/v), which is generally considered non-cytotoxic in mammalian cell systems²⁵. Equivalent volumes of DMSO (0.01%, v/v) were added to all non-Ple groups as vehicle controls. The 5 µM concentration of Plerixafor was selected on the basis of its established efficacy and specificity as a CXCR4 antagonist in cardiomyocyte hypoxia-reoxygenation models without cytotoxicity¹¹.

Ple pretreatment protocol: In groups receiving Ple, cells were pretreated with 5 µM Ple for 2 h. After the 2 h pretreatment, the medium containing Ple was removed, and cells were washed once with PBS before proceeding with the hypoxia/reoxygenation (H/R) procedure to avoid potential off-target effects of prolonged Ple exposure on reoxygenation-phase signaling. PB1 (20 µg/mL) or PB1-Lip@M-CST (20 µg/mL) was added at the beginning of the reoxygenation phase in the relevant combination groups.

Eight experimental groups were established using normal HL-1 cells as follows:

1. (1) Control: normoxic culture without any treatment;

2. Model: CoCl₂ (200 µM) for 6 h followed by reoxygenation for 24 h;

3. PB1: Model + PB1 (20 µg/mL) added at the beginning of reoxygenation;

4. PB1-Lip@M-CST: Model + PB1-Lip@M-CST (20 µg/mL) added at the beginning of reoxygenation;

5. Ple: Ple (5µM) pretreatment for 2 h without subsequent hypoxia/reoxygenation procedure;

6. Model + Ple: Ple (5µM) pretreatment for 2 h, followed by removal of Ple and subsequent H/R procedure;

7. PB1 + Model + Ple: Ple (5 µM) pretreatment for 2 h, followed by removal of Ple and H/R procedure, with PB1 (20 µg/mL) added at the beginning of reoxygenation;

8. PB1-Lip@M-CST + Model + Ple: Ple (5 µM) pretreatment for 2 h, followed by removal of Ple and H/R procedure, with PB1-Lip@M-CST (20 µg/mL) added at the beginning of reoxygenation.

Chemical hypoxia was induced using 200 µM CoCl₂ for 6 h, followed by replacement with fresh complete medium for 24 h to simulate reoxygenation. All procedures were performed under aseptic conditions. Each experimental group was performed in triplicate.

Molecular and biochemical analyses

RT-qPCR

Total RNA was extracted from treated HL-1 cells using RLT Plus lysis buffer followed by gDNA elimination columns and RNA purification columns. RNA purity and concentration were assessed by spectrophotometry (A260/A280 ratio of 1.8-2.0). One microgram of total RNA was reverse-transcribed into first-strand cDNA using a commercial reverse transcription kit according to the manufacturer’s instructions. Quantitative real-time PCR was performed with 2×SYBR Green PCR Master Mix in a 20 µL reaction volume containing 0.5 µL of each primer (10 µM) using a real-time fluorescence quantitative PCR instrument (Hangzhou Bioer Technology Co., Ltd., model FQD-96 A, China). Thermal cycling conditions were as follows: initial denaturation at 94 °C for 20 s, followed by 40 cycles of 94 °C for 10 s and 60 °C for 20 s. Relative mRNA expression levels of Akt1, Bax, Pi3k, and Bcl-2 were calculated using the 2^−ΔΔCt method, with Gapdh as the internal reference gene. Primer sequences are listed in Table 1.

Table 1.

Primer sequences used for RT-qPCR analysis of key genes in HL-1 cells.

Gene	Forward primer	Reverse primer
Akt1	TCTGAGACTGACACCAGGTA	TAGGAGAACTTGATCAGGCG
Pi3k	GGCACAGACTTGGTGTTTTA	TGTAAGAGTGTAATCGCCGT
Bax	GCTACAGGGTTTCATCCAGG	GCTCCATATTGCTGTCCAGT
Bcl-2	CCTTCTTTGAGTTCGGTGGG	GGGGCCATATAGTTCCACAAA
Gapdh	GGTCCCAGCTTAGGTTCATC	CGTGAGTGGAGTCATACTGG

Western blot analysis

Cells were washed three times with ice-cold PBS and lysed in RIPA buffer supplemented with protease and phosphatase inhibitors on ice for 30 min. Lysates were centrifuged at 12,000 × g for 15 min at 4 °C, and the supernatant was collected. Protein concentration was quantified using a BCA protein assay kit. Equal amounts of protein were mixed with 5 × SDS loading buffer, boiled at 95 °C for 5 min, and separated by 10% SDS-PAGE (80 V for the stacking gel and 120 V for the resolving gel). Proteins were transferred to PVDF membranes (pre-activated in methanol) by wet transfer at a constant current of 300 mA for 60 min at 4 °C. Membranes were blocked with 5% BSA (for phosphorylated proteins) or 5% non-fat milk (for total proteins) in TBST for 1 h at room temperature, incubated with primary antibodies overnight at 4 °C, and then with Goat Anti Rabbit IgG(H + L) (HRP) (Immunoway, Cat. No. RS0002, 1:10000) for 1 h at room temperature. Protein bands were visualized using ECL chemiluminescent substrate and imaged with a chemiluminescence imaging system. Band intensities were quantified using ImageJ software. Total protein levels were normalized to GAPDH, while phosphorylated AKT (p-AKT) levels were normalized to total AKT.

ELISA and oxidative stress marker assays

Following the indicated treatments, cell culture supernatants were collected and centrifuged at 3,000 × g for 10 min at 4 °C to remove cellular debris. The supernatants were analyzed for lactate dehydrogenase (LDH) release, interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and interleukin-10 (IL-10). These measurements were performed the following commercial ELISA kits according to the manufacturers’ instructions: mouse LDH ELISA kit (JL13877-48T, JONLNBIO, Shanghai, China), mouse IL-6 ELISA kit (E-EL-M0044, Elabscience, Wuhan, China), mouse TNF-α ELISA kit (E-EL-M3063, Elabscience, Wuhan, China), and mouse IL-10 ELISA kit (E-EL-M0046, Elabscience, Wuhan, China).

For intracellular oxidative stress markers, the remaining cells were washed three times with ice-cold PBS and lysed in RIPA buffer supplemented with protease inhibitors on ice for 30 min. Lysates were centrifuged at 12,000 × g for 15 min at 4 °C, and the supernatant was collected. Protein concentration was quantified using a BCA protein assay kit. Malondialdehyde (MDA) content and total superoxide dismutase (SOD) activity were then measured using the MDA assay kit (thiobarbituric acid reactive substances (TBARS) method, A003-4-1, Nanjing Jiancheng Bioengineering Institute) and SOD assay kit (A001-1, Nanjing Jiancheng Bioengineering Institute), respectively.

Absorbance was measured at 450 nm (for all ELISA kits), 532 nm (for MDA), and 550 nm (for SOD) using a microplate reader (Molecular Devices, SpectraMax i3x). Concentrations or enzymatic activities were calculated according to standard curves generated for each assay. All assays were performed in triplicate.

Animal model and ethics statement

All animal procedures were approved by the Institutional Animal Care and Use Committee of Chongqing Medical University (Approval No. IACUC-CQMU-20250743). All methods were performed in accordance with the relevant guidelines and regulations. Male C57BL/6J mice (8 weeks, 20–25 g) were housed under standard conditions (12-h light/dark cycle, 22 °C, ad libitum chow/water). MIRI was induced under anesthesia (2% isoflurane in O₂): thoracotomy at the fourth intercostal space, left anterior descending (LAD) coronary artery ligation with 8 − 0 silk suture for 30 min. To guarantee quality control and reproducibility, successful ischaemia was strictly confirmed in every animal by immediate and sustained ST-segment elevation (> 0.1 mV) on continuous lead-II ECG monitoring, accompanied by visible blanching of the anterior left ventricular wall. Reperfusion was verified by prompt restoration of myocardial colour and normalisation of ECG changes upon suture release. These multi-parameter quality control criteria (ECG + visual confirmation) were rigorously applied to all mice prior to randomisation and treatment allocation, consistent with standard practices in mouse MIRI models^26,27. Body temperature was maintained at 37 ± 0.5 °C using a heating pad throughout the surgical procedure. Infarct size was subsequently quantified at 5 days post-reperfusion by TTC staining. Sham underwent thoracotomy without ligation. Mice were monitored post-surgery, and buprenorphine (0.1 mg/kg s.c.) was administered every 12 h for 48 h postoperatively.

Therapeutic efficacy evaluation

One hundred male C57BL/6J mice were randomly divided into five groups (n = 20 per group). For each endpoint analysis, at least five mice per group were randomly selected: sham-operated (Sham), Model (I/R + PBS), I/R + PB1-Lip, I/R + PB1-Lip@M, and I/R + PB1-Lip@M-CST. The Sham and Model groups received PBS via tail vein injection at 0, 1, and 2 days post-reperfusion, while the treatment groups were administered PB1-Lip formulations at an equivalent dose of 10 mg/kg PB1. Infarct size was quantified at 5 days by TTC staining to evaluate subacute infarct remodeling: hearts were excised, rinsed in cold PBS, frozen at-20 °C for 20 min, sliced into 1-mm-thick sections, incubated in 1% TTC solution at 37 °C for 15 min, fixed in 4% paraformaldehyde, and photographed for analysis; infarct area (% of left ventricular area) was quantified using ImageJ software (v1.53; NIH, Bethesda, MD, USA). Serum levels of creatine kinase-MB (CK-MB, Cat. no. JL12422), cardiac troponin I (cTnI, Cat. no. JL11280), and brain natriuretic peptide (BNP, Cat. no. JL12884) obtained from JONLNBIO (Shanghai, China) were measured at 5 days post-reperfusion. Cardiac function was assessed at 7 days post-reperfusion by echocardiography using a Vevo 3100 system (VisualSonics, Toronto, Canada) equipped with a 30 MHz transducer. Measurements of left ventricular ejection fraction (LVEF), left ventricular fractional shortening (LVFS), left ventricular end-diastolic diameter (LVEDD), and left ventricular end-systolic diameter (LVESD) were obtained in M-mode from the parasternal short-axis view. At 7 days post-reperfusion, serum aspartate aminotransferase (AST), lactate dehydrogenase (LDH), superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPX) levels were determined using commercial biochemical assay kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) with a Hitachi 7180 automated biochemistry analyzer (Japan). TUNEL staining was performed on paraffin sections of the left ventricular anterior wall (5 days post-reperfusion) using the One-step TUNEL In Situ Apoptosis Kit (Red, Elab Fluor^® 594; Elabscience, E-CK-A322). At 28 days post-reperfusion, hearts were harvested for histological analysis. Paraffin-embedded Sections  (5 μm thick) were stained with hematoxylin and eosin (H&E) staining kit (Beyotime, C0105S) to assess myocardial morphology and inflammation, and with Masson’s trichrome staining kit (Beyotime, C0189S) to evaluate cardiac fibrosis (collagen fibers stained blue). Images were acquired with an Olympus BX53 microscope and analyzed in a blinded manner using ImageJ.

Immunofluorescence staining for CXCR4 expression

To validate CXCR4 expression in ischemic cardiac tissue, hearts were harvested at 5 days post-reperfusion. Paraffin-embedded sections  (5 μm) were deparaffinized, rehydrated, and subjected to antigen retrieval in 1× Tris-EDTA buffer (pH 9.0) at 95 °C for 20 min. Endogenous peroxidase activity was quenched with 3% H₂0₂ for 15 min prior to HRP-based TSA amplification. After blocking with 3% bovine serum albumin for 30min at room temperature, the sections were incubated overnight at 4 °C with anti-CXCR4 rabbit polyclonal primary antibody (Immunoway, cat. no. YT1800, 1:200 dilution). After washing with PBST, sections were incubated with Goat Anti Rabbit IgG(H + L) (HRP) (Immunoway, cat. no. RS0002, 1:10000) for 30 min at 37 °C. Negative control sections were processed without primary antibody. Subsequently, a Tyramide Signal Amplification (TSA) reaction was performed using a multi-color immunofluorescence kit (Immunoway, cat. no. RS0036) with 594-labeled tyramide (Immunoway, cat. no. YS0010, 1:200 dilution) for 10 min at room temperature protected from light. Nuclei were counterstained with DAPI. Images were acquired using a fluorescence microscope (Olympus BX53) under identical exposure settings for all groups. Fluorescence intensity was quantified using ImageJ software. CXCR4 fluorescence intensity was quantified in the infarct border zone(n = 5 mice per group). Investigators were blinded to treatment allocation during outcome assessment.

Pharmacokinetic and tissue distribution studies

To evaluate the pharmacokinetic profiles and tissue distribution of the PB1 nano-formulations, 195 male C57BL/6J mice subjected to MIRI were randomly divided into three groups (PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST; n = 65 per group). Each formulation was administered intravenously at a dose equivalent to 10 mg/kg PB1 immediately after reperfusion. A composite sampling design was employed, with five mice per group sacrificed at each of 13 predefined time points (0.17, 0.33, 0.5, 0.75, 1, 1.5, 2, 4, 8, 16, 24, 36, and 48 h post-administration).

Mice were anesthetized with 2% isoflurane, and approximately 0.6–0.8 mL of whole blood was collected via cardiac puncture at terminal anesthesia into EDTA-K₂ anticoagulant tubes. After gentle inversion, animals were euthanized by cervical dislocation. Blood was centrifuged at 3,000 × g for 10 min at 4 °C, and the resulting plasma was aliquoted and stored at − 80 °C until analysis. Simultaneously, the heart, liver, spleen, lung, and kidney were excised, rinsed, weighed, and snap-frozen in liquid nitrogen before storage at − 80 °C.

For sample preparation, approximately 20 mg of each tissue was thawed on ice, placed in a microcentrifuge tube with 600 µL phosphate-buffered saline (PBS), and homogenized using a glass homogenizer. The homogenate was centrifuged at 12,000 × g for 15 min at 4 °C. One hundred microliters of tissue homogenate supernatant or 10 µL of plasma were diluted with methanol, followed by further dilution (10-fold for tissue homogenates and 100-fold for plasma) to ensure the concentrations fell within the external standard calibration range of 3.125–1,000 ng/mL. Samples were vortexed for 5 min and centrifuged at 12,000 × g for 10 min at 4 °C. The supernatants were subjected to LC-MS/MS analysis for PB1 quantification as described in “Encapsulation efficiency and drug loading” section. Pharmacokinetic parameters (C_max, T_max, AUC_0−last, t_1/2, CL, and MRT) were calculated by non-compartmental analysis using Phoenix WinNonlin software (version 8.3; Certara, Princeton, NJ, USA).

In vivo localisation of Cy5-labeled PB1-Lip@M-CST nanoparticles

To visualize the myocardial accumulation and distribution of the PB1 nano-formulations, Cy5-labeled variants (Lip-Cy5, PB1-Lip-Cy5, PB1-Lip-Cy5@M, and PB1-Lip-Cy5@M-CST) were prepared and evaluated in C57BL/6J mice subjected to MIRI. Mice were randomly assigned to four groups (n = 5 per group) and received a single intravenous injection of the respective Cy5-labeled formulation at an equivalent nanoparticle dose corresponding to 10 mg/kg PB1 in the therapeutic formulations. Longitudinal in vivo fluorescence imaging was performed at 4, 24, and 48 h post-administration using a PerkinElmer IVIS Spectrum small-animal in vivo imaging system under fixed imaging parameters and detection conditions (excitation/emission: 640/670 nm). Mice were anesthetized with 2% isoflurane during imaging. Regions of interest (ROIs) were manually drawn over the cardiac area, and fluorescence intensity within the myocardial area was quantitatively analyzed using PerkinElmer Living Image software. All images were acquired under identical exposure time, binning, and field-of-view settings.

Animal administration and sample collection

Healthy C57BL/6J mice were randomly divided into four groups (n = 5 per group) and administered PBS or PB1 nano-formulations (PB1-Lip, PB1-Lip@M, or PB1-Lip@M-CST) once daily for three consecutive days via tail-vein injection at a dose equivalent to 10 mg/kg PB1. Seven days after the final injection, blood was collected via orbital venous plexus puncture. Blood samples were allowed to stand at room temperature for 2 h and then centrifuged at 3,500 × g for 15 min. The supernatant (serum) was collected, aliquoted, and stored at − 80 °C until analysis. After blood collection, mice were euthanized, and the heart, liver, spleen, lung, and kidney were harvested and fixed in sufficient 4% paraformaldehyde at room temperature for 24 h.

Hemolysis assay

Fresh EDTA-anticoagulated mouse whole blood (0.5 mL) was diluted 1:1 with normal saline. Twenty microliters of diluted blood was added to 1 mL of each formulation to achieve a final blood volume fraction of approximately 2% (v/v). The mixtures were incubated at 37 °C for 1 h. After incubation, samples were centrifuged at 3,000 × g for 10 min, and the absorbance of the supernatant was measured at 545 nm using a microplate reader (Molecular Devices, SpectraMax i3x). PBS served as the negative control (0% hemolysis) and 0.5% Triton X-100 as the positive control (100% hemolysis). Hemolysis rate (%) was calculated as follows: where Abs_sample, Abs_negative, and Abs_positive represent the absorbance values of the test formulations, the negative control(PBS), and the positive control (0.5% Triton X-100) at 545 nm, respectively.

Serum biochemistry

Serum levels of alanine aminotransferase (ALT; E-BC-K1300-M), aspartate aminotransferase (AST; E-BC-K1301-M), blood urea nitrogen (BUN; E-BC-K183-M), alkaline phosphatase (ALP; E-BC-K1304-M), and creatinine (Cr; E-BC-K188-M) were determined using corresponding colorimetric assay kits (Elabscience, Wuhan, China) on an automated biochemistry analyzer (Hitachi 7180, Japan) according to the manufacturer’s instructions.

Histological examination

Fixed heart, liver, spleen, lung, and kidney tissues were embedded in paraffin, sectioned (5 μm thickness) and stained with hematoxylin and eosin (H&E) for histopathological evaluation under light microscopy.

Statistical analysis

All quantitative data are expressed as the mean ± standard deviation (SD). Normality of distribution was assessed using the Shapiro-Wilk test. For comparisons involving more than two groups, one-way analysis of variance (ANOVA) was performed, followed by Tukey’s multiple-comparison post hoc test. Comparisons between two groups were analyzed using an unpaired two-tailed Student’s t-test. All statistical analyses were conducted with GraphPad Prism software (version 9.0; GraphPad Software, San Diego, CA, USA). A two-sided P value < 0.05 was considered statistically significant. Statistical significance is denoted as follows: ns, P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Results

Preparation and characterization of PB1-Lip@M-CST

PB1-Lip@M-CST nanoparticles were fabricated using a stepwise biomimetic engineering approach. PB1-loaded liposomes were first prepared via the conventional thin-film hydration and extrusion method. These liposomes were subsequently coated with macrophage-derived membranes, after which the cardiac-specific targeting peptide CST was conjugated onto the membrane surface. Dynamic light scattering (DLS) analysis revealed a progressive increase in hydrodynamic diameter from 116.65 ± 2.05 nm (PDI = 0.165 ± 0.006) for PB1-Lip to 149.15 ± 3.47 nm (PDI = 0.192 ± 0.006) for PB1-Lip@M and 152.09 ± 1.96 nm (PDI = 0.192 ± 0.008) for PB1-Lip@M-CST (Fig. 1A; Table 2). This size increment is consistent with successful envelopment of the liposomal core by the macrophage membrane. Concurrently, zeta potential measurements demonstrated a shift toward greater negativity, from-16.97 ± 0.53 mV for PB1-Lip to-28.58 ± 1.68 mV for PB1-Lip@M (P < 0.001), reflecting the anionic character of the native macrophage membrane, which may contribute to improved colloidal stability. Conjugation of the CST peptide partially neutralized the surface charge, resulting in a zeta potential of-20.13 ± 1.53 mV for PB1-Lip@M-CST (Fig. 1B), likely due to electrostatic interactions and steric hindrance introduced by the peptide. All formulations exhibited PDI values within the acceptable range of 0.16–0.20, indicating uniform dispersion without significant aggregation.

Fig. 1.

Preparation and characterization of PB1-Lip@M-CST. (A) Particle size distribution of PB1-Lip@M-CST, PB1-Lip@M, and PB1-Lip. (B) Zeta potential of PB1-Lip@M-CST, PB1-Lip@M, and PB1-Lip (n = 3). (C) Representative TEM images of PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST. (D)–(F) XRD patterns, DSC and TGA of PB1-Lip, PB1 and Blank Lip. (G) Cumulative drug release profiles of PB1 from PB1-Lip and PB1-Lip@M-CST in PBS (pH 7.4) at 37 °C (n = 3). (H) SDS-PAGE bands of Blank Lip, PB1-Lip, Membrane (M), PB1-Lip@M, and PB1-Lip@M-CST. (I) Western blot analysis of PB1-Lip@M-CST for surface characteristic proteins (CCR2, CD11b, CD36).

Table 2.

Particle size, zeta potential, encapsulation efficiency, and drug loading of PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST nanoparticles.

Particle type	Particle size (nm)	PDI	Zeta potential (mV)	EE (%)	DL (%)
PB1-Lip	116.65 ± 2.05	0.165 ± 0.006	-16.97 ± 0.53	86.63 ± 1.00	10.32 ± 0.32
PB1-Lip@M	149.15 ± 3.47	0.192 ± 0.006	-28.58 ± 1.68	83.14 ± 1.23	9.69 ± 0.54
PB1-Lip@M-CST	152.09 ± 1.96	0.192 ± 0.008	-20.13 ± 1.53	81.03 ± 2.73	9.17 ± 0.60

The key physicochemical parameters of the nanoparticles, including particle size, polydispersity index, zeta potential, encapsulation efficiency, and drug loading, are summarized in Table 2. Notably, the encapsulation efficiency remained high across all formulations (86.63 ± 1.00% for PB1-Lip, 83.14 ± 1.23% for PB1-Lip@M, and 81.03 ± 2.73% for PB1-Lip@M-CST), demonstrating that both macrophage membrane coating and CST conjugation caused only minimal loss of PB1. Similarly, the DL was well maintained at 10.32 ± 0.32%, 9.69 ± 0.54%, and 9.17 ± 0.60%, respectively. These results indicate that the biomimetic modification strategy preserved excellent drug-loading capacity while conferring surface properties favorable for prolonged circulation and active targeting.

Transmission electron microscopy (TEM) confirmed the formation of well-defined core-shell nanostructures. PB1-Lip appeared spherical with less distinct boundaries compared with membrane-coated formulations, whereas PB1-Lip@M displayed a core-shell morphology. PB1-Lip@M-CST exhibited a well-defined spherical morphology with sharp, distinct interfaces between the liposomal core and the outer membrane layer (Fig. 1C).

X-ray diffraction (XRD), differential scanning calorimetry (DSC), and thermogravimetric analysis (TGA) further elucidated the physicochemical interactions within the formulations. XRD patterns of PB1-Lip and blank liposomes lacked the characteristic crystalline peaks of free PB1, suggesting that the drug was predominantly present in an amorphous state within the lipid matrix (Fig. 1D). DSC thermograms showed the disappearance of the endothermic peak of crystalline PB1 (approximately 90 °C) in PB1-Lip, indicating molecular dispersion of PB1 with the lipid bilayer (Fig. 1E). TGA curves demonstrated that PB1-Lip possessed a higher onset temperature of thermal decomposition and a slower weight-loss rate compared with free PB1, suggesting a protective effect of lipid encapsulation against thermal degradation (Fig. 1F).

In vitro release studies conducted in PBS (pH 7.4) at 37 °C revealed a controlled diffusion profile. PB1-Lip exhibited a relatively rapid initial release, reaching over 86% cumulative release by 48 h. In contrast, PB1-Lip@M-CST displayed a slower and more sustained release, achieving approximately 80% cumulative release at the same time point without an obvious burst effect (Fig. 1G). Kinetic modeling indicated that both formulations best fit the first-order model (R² = 0.9958 and 0.9947, respectively), while the Korsmeyer-Peppas release exponent (n < 0.43) confirmed a Fickian diffusion mechanism. The additional membrane coating likely contributed to the more controlled release kinetics observed with PB1-Lip@M-CST.

At the molecular level, the biomimetic fidelity of the nanoparticles was verified by protein characterization. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) demonstrated that PB1-Lip@M and PB1-Lip@M-CST retained protein banding patterns highly similar to those of the native macrophage membrane, which were absent in uncoated PB1-Lip and blank liposomes (Fig. 1H). Western blot analysis further confirmed the preservation of key macrophage surface proteins, including CCR2 (42 kDa), CD11b (127 kDa), and CD36 (88 kDa) (Fig. 1I). These findings collectively support the successful construction of a biomimetic nanoplatform that integrates natural cellular camouflage with targeted peptide modification, offering a promising platform for site-specific drug delivery.

In vitro cytotoxicity, targeted cellular uptake and competitive blocking of PB1-Lip@M-CST in HL-1 cells

To determine the optimal working concentration of PB1 and PB1-Lip@M-CST, the cytotoxic effects of free PB1, blank Lip@M-CST carrier, and PB1-loaded PB1-Lip@M-CST were evaluated in normal HL-1 cardiomyocytes using the CCK-8 assay. The blank Lip@M-CST nanocarrier exhibited negligible cytotoxicity across the tested concentration range (0–80 µg/mL, expressed as PB1 equivalent), with cell viability consistently maintained above 90% (Fig. 2B). In contrast, both free PB1 and PB1-Lip@M-CST displayed concentration-dependent cytotoxicity. At 20 µg/mL, cell viability showed no significant difference compared with the 10 µg/mL group for either formulation (P > 0.05). However, a significant reduction in viability was observed at 40 µg/mL relative to 20 µg/mL (free PB1: 68.32 ± 4.12% versus 78.62 ± 2.89%, P < 0.01; PB1-Lip@M-CST: 68.85 ± 6.25% versus 82.41 ± 6.35%, P < 0.05) (Fig. 2A, C). These results indicate that the cytotoxicity at higher concentrations is primarily attributable to PB1 itself. Therefore, 20 µg/mL was selected as the safe working concentration for all subsequent in vitro studies.

Fig. 2.

In vitro cytotoxicity evaluation and cellular uptake of DiO-labeled PB1-loaded nanoplatforms in HL-1 cells. (A) Cell viability of HL-1 cells treated with free PB1 at various concentrations (0, 5, 10, 20, 40, 80 µg/mL) as determined by CCK-8 assay. (B), (C) Cell viability of HL-1 cells after 24 h incubation with (B) blank Lip@M-CST or (C) PB1-Lip@M-CST at matched concentrations. Panel B represents the blank carrier at concentrations equivalent to the PB1 content in panel C (0–80 µg/mL). (n = 4; *P < 0.05, **P < 0.01, ****P < 0.0001). (D) Representative confocal laser scanning microscopy (CLSM) images showing intracellular uptake of DiO-labeled (green) PB1-Lip, PB1-Lip@M, PB1-Lip@M-CST, scrambled control peptide blocking group, and CST blocking group in HL-1 cells after incubation for 1, 2, and 4 h (scale bar = 20 μm). Nuclei were counterstained with DAPI (blue). (E)–(G) Quantitative analysis of mean fluorescence intensity at (E) 1 h, (F) 2 h, and (G) 4 h (n = 3, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant).

To investigate the delivery efficiency and active targeting capacity of the nanoplatforms in HL-1 cardiomyocytes, cellular uptake of DiO-labeled formulations was assessed by confocal laser scanning microscopy following incubation for 1, 2, and 4 h. As shown in Fig. 2D, marked differences in intracellular fluorescence were observed across the different surface-modified nanoplatforms. Compared with the uncoated PB1-Lip group, macrophage membrane coating alone (PB1-Lip@M) produced no significant enhancement in fluorescence intensity at any time point examined (P > 0.05). In contrast, further conjugation with the cardiac-specific targeting peptide CST (PB1-Lip@M-CST) resulted in substantially stronger intracellular fluorescence signals at all incubation times, with uptake increasing progressively in a clear time-dependent manner. Quantitative analysis of mean fluorescence intensity confirmed that the PB1-Lip@M-CST group exhibited the highest cellular internalization among all formulations (Fig. 2E-G). To verify the specificity of CST-mediated targeting, competitive blocking experiments were performed. Pretreatment with excess free CST peptide significantly attenuated the uptake of PB1-Lip@M-CST, reducing the mean fluorescence intensity to levels comparable to those of the non-targeted formulations (P < 0.0001), with the most pronounced inhibition observed at 4 h. In contrast, pretreatment with CKS produced no inhibitory effect. These results demonstrate that CST peptide modification confers potent active targeting capability to the macrophage membrane-coated nanoplatform, suggesting enhanced cellular internalization in HL-1 cells through a sequence-specific mechanism.

Immune-evasion capacity in RAW 264.7 macrophages

To evaluate the immune-evasion capacity of the different nanoplatforms, DiO-labeled PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST were incubated with RAW 264.7 macrophages for 1, 2, and 4 h. Quantitative analysis of mean fluorescence intensity (Fig. 3B-D) revealed that the uncoated PB1-Lip group exhibited markedly higher intracellular fluorescence intensity than both the PB1-Lip@M and PB1-Lip@M-CST groups at all time points examined (P < 0.0001). No significant difference was observed between the PB1-Lip@M and PB1-Lip@M-CST groups at any incubation time (P > 0.05). These findings indicate that macrophage membrane coating effectively reduces nanoparticle uptake by macrophages, thereby conferring robust immune-evasion ability, while additional conjugation of the CST peptide does not compromise this immune-evasive property.

Fig. 3.

Intracellular distribution and escape behavior of different drug-loaded nanoparticles in RAW 264.7 macrophages. (A) Representative confocal laser scanning microscopy (CLSM) images showing intracellular fluorescence of DiO-labeled (green) PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST in RAW 264.7 cells after incubation for 1, 2, and 4 h (scale bar = 20 μm). Nuclei were counterstained with DAPI (blue). (B-D) Quantitative analysis of mean fluorescence intensity at (B) 1 h, (C) 2 h, and (D) 4 h (n = 3, ****P < 0.0001 vs. PB1-Lip group).

PB1-Lip@M-CST exerts the most potent protective effects against hypoxia/reoxygenation-induced ros overproduction and apoptosis in HL-1 cells

The effects of PB1 and its nano-formulations on intracellular reactive oxygen species (ROS) levels were examined in the CoCl₂-induced hypoxia/reoxygenation model in HL-1 cells using the DCFH-DA fluorescent probe, in conjunction with fluorescence microscopy and flow cytometry (Fig. 4A–D). Fluorescence microscopy revealed markedly elevated green fluorescence intensity in the Model group compared with the normoxic Control group (2.47 ± 0.01 versus 1.01 ± 0.13, P < 0.0001), confirming successful induction of oxidative stress (Fig. 4A, B). Treatment with free PB1, PB1-Lip, PB1-Lip@M, or PB1-Lip@M-CST significantly attenuated intracellular ROS accumulation relative to the Model group. Notably, PB1-Lip@M-CST produced the most pronounced reduction, lowering the relative fluorescence intensity to 1.30 ± 0.10, which was significantly lower than that observed in the free PB1 group (2.18 ± 0.16, P < 0.0001) and the PB1-Lip group (1.90 ± 0.21, P < 0.05) (Fig. 4A, B). These results were consistent with flow cytometric analysis, in which the Model group exhibited the highest mean fluorescence intensity (43,729.0 ± 2,198.6) compared with the Control group (24,109.5 ± 2,179.1, P < 0.0001), whereas PB1-Lip@M-CST yielded the lowest value among all treatment groups (Fig. 4C, D).

Fig. 4.

PB1-Lip@M-CST nanoparticles attenuate hypoxia/reoxygenation-induced intracellular reactive oxygen species (ROS) production and apoptosis in HL-1 cells. (A) Representative fluorescence microscopy images of intracellular ROS detected by DCFH-DA probe (green) in HL-1 cells under normoxic control (Control), CoCl₂-induced hypoxia-reoxygenation (Model), or treated with ascorbic acid (AC), free PB1, PB1-Lip, PB1-Lip@M, or PB1-Lip@M-CST. Scale bar = 50 μm. (B) Semi-quantitative analysis of relative DCFH-DA fluorescence intensity from fluorescence microscopy images. (C) Representative flow cytometry histograms of DCFH-DA fluorescence intensity (FITC-H) in the indicated groups. (D) Quantitative analysis of mean fluorescence intensity of DCFH-DA by flow cytometry. (E) Representative flow cytometry scatter plots of Annexin V-FITC/PI double staining showing apoptotic cell populations in each group. (F) Quantification of early apoptosis rate. Data are presented as mean ± SD (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

The effect on early apoptosis was assessed by Annexin V-FITC/PI double staining and flow cytometry (Fig. 4F). The Model group exhibited a substantially elevated early apoptosis rate compared with the Control group (28.17 ± 0.49% vs. 11.85 ± 1.19%, P < 0.0001). Treatment with ascorbic acid, free PB1, PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST reduced the early apoptosis rate to 20.41 ± 1.56% (P < 0.001 vs. Model), 24.08 ± 0.47% (P < 0.05 vs. Model), 20.13 ± 1.86% (P < 0.0001 vs. Model), 18.41 ± 2.02% (P < 0.0001 vs. Model), and 15.19 ± 1.71% (P < 0.0001 vs. Model), respectively. Notably, PB1-Lip@M-CST achieved the lowest early apoptosis rate among all treatment groups, demonstrating significantly superior anti-apoptotic efficacy compared with free PB1 (24.08 ± 0.47%, P < 0.0001) and PB1-Lip (20.13 ± 1.86%, P < 0.05). Taken together, these results demonstrate that PB1-Lip@M-CST provides the most potent protection against both H/R-induced ROS overproduction and early apoptosis in HL-1 cells, outperforming free PB1 and the non-targeted or partially modified nano-formulations.

CXCR4-dependent regulation of PI3K/AKT signaling, apoptosis, and oxidative stress by PB1 and PB1-Lip@M-CST

Role of CXCR4 in the regulation of PI3K/AKT1 signaling by PB1 and PB1-Lip@M-CST

RT-qPCR and western blot demonstrated that hypoxia/reoxygenation (H/R) injury markedly activated the PI3K/AKT1 signaling pathway in HL-1 cells. As shown in Fig. 5A, B, Model group cells exhibited significantly elevated AKT1 and Pi3k mRNA levels compared with the Control group. Consistent with these transcriptional changes, western blot analysis revealed pronounced increases in the p-AKT1/AKT1 protein ratio and PI3K protein expression (Fig. 5C-F). Treatment with PB1 or PB1-Lip@M-CST robustly suppressed both mRNA and protein activation of this pathway, with the nano-formulation exerting a more pronounced inhibitory effect. In gain-of-function experiments, CXCR4 overexpression alone significantly upregulated Akt1 and Pi3k mRNA as well as p-AKT1/AKT1 and PI3K protein levels (Fig. 5A-F). Importantly, PB1 and PB1-Lip@M-CST retained their ability to attenuate this hyperactivation in CXCR4-overexpressing cells, indicating that the compounds may interact with CXCR4-associated signaling. In complementary loss-of-function studies (Fig. 6), pharmacological inhibition of CXCR4 with Ple alone substantially reduced Akt1 and Pi3k mRNA expression (Fig. 6A-D) and lowered p-AKT1/AKT1 and PI3K protein levels (Fig. 6E-F). Under these conditions, neither PB1 nor PB1-Lip@M-CST produced any additional inhibitory effect,  confirming that PB1 acts upstream of or directly on CXCR4 to exert its anti-apoptotic effects. Collectively, these gain- and loss-of-function data suggest that PB1 and, particularly, PB1-Lip@M-CST exert their cardioprotective effects, at least in part, through modulation of CXCR4-associated PI3K/AKT1 signaling induced by H/R injury.

Fig. 5.

CXCR4 overexpression partially counteracts the inhibitory effects of PB1 and PB1-Lip@M-CST on the PI3K/AKT1 signaling pathway in HL-1 cells subjected to hypoxia-reoxygenation. (A), (B) Relative mRNA expression levels of Akt1 (A) and Pi3k (B). (C), (D) Representative western blot bands of p-AKT1, total AKT1, and GAPDH (C) and PI3K and GAPDH (D). (E), (F) Quantitative analysis of the p-AKT1/AKT1 protein ratio (E) and relative PI3K protein expression normalized to GAPDH (F). Data are mean ± SD (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Fig. 6.

CXCR4 inhibition by Ple abolishes the inhibitory effects of PB1 and PB1-Lip@M-CST on the PI3K/AKT1 signaling pathway in HL-1 cells subjected to hypoxia-reoxygenation. (A), (B) Relative mRNA expression levels of Akt1 (A) and Pi3k (B). (C), (D) Representative western blot bands of p-AKT1, total AKT1, and GAPDH (C) and PI3K and GAPDH (D). (E), (F) Quantitative analysis of the p-AKT1/AKT1 protein ratio (E) and relative PI3K protein expression normalized to GAPDH (F). Data are presented as mean ± SD (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns, not significant). .

Regulation of apoptosis-related factors by PB1 and PB1-Lip@M-CST via CXCR4

To determine whether the observed PI3K/AKT1 inhibition translates into anti-apoptotic activity, expression of the pro-apoptotic factor BAX and the anti-apoptotic factor BCL-2 was examined. H/R injury significantly increased mRNA and protein levels of BAX while decreasing BCL-2 expression (Fig. 7A-E). Both PB1 and PB1-Lip@M-CST reversed this imbalance by downregulating BAX protein and upregulating BCL-2 protein, with the nano-delivery formulation demonstrating superior efficacy. CXCR4 overexpression exacerbated the apoptotic shift induced by H/R (Fig. 7A-E), yet subsequent treatment with PB1 or PB1-Lip@M-CST effectively restored the BAX/BCL-2 balance even in the presence of elevated CXCR4. Conversely, pretreatment with the CXCR4 inhibitor Ple alone markedly lowered BAX and elevated BCL-2 expression (Fig. 8A-E). Under CXCR4-inhibited conditions, PB1 and PB1-Lip@M-CST failed to produce any further modulation of BAX or BCL-2 at either the mRNA or protein level. These findings indicate that the anti-apoptotic actions of PB1 and PB1-Lip@M-CST are mediated predominantly through CXCR4-dependent suppression of the BAX/BCL-2 pathway.

Fig. 7.

CXCR4 overexpression attenuates the regulatory effects of PB1 and PB1-Lip@M-CST on BAX and BCL-2 expression in HL-1 cells subjected to hypoxia-reoxygenation injury. (A), (B) Relative mRNA expression levels of Bax (A) and Bcl-2 (B). (C) Representative western blot bands of BAX, BCL-2 and GAPDH. (D), (E) Quantitative analysis of BAX (D) and BCL-2 (E) protein levels normalized to GAPDH. Data are mean ± SD (n = 3, *P < 0.05, **P < 0.01, ****P < 0.0001).

Fig. 8.

CXCR4 inhibition by Ple attenuates the regulatory effects of PB1 and PB1-Lip@M-CST on BAX and BCL-2 expression in HL-1 cells subjected to hypoxia-reoxygenation. (A), (B) Relative mRNA expression levels of Bax (A) and Bcl-2 (B). (C) Representative western blot bands of BAX, BCL-2, and GAPDH. (D), (E) Quantitative analysis of relative BAX (D) and BCL-2 (E) protein expression levels normalized to GAPDH. Data are mean ± SD (n = 3, *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, not significant.).

Modulation of inflammatory and oxidative stress markers by PB1 and PB1-Lip@M-CST through CXCR4

Finally, the impact of PB1 and its nano-formulation on H/R-induced inflammation and oxidative stress was assessed by ELISA quantification of key cytokines and oxidative markers in culture supernatants. As illustrated in Fig. 9A-F, Model group cells displayed significantly elevated levels of the pro-inflammatory cytokines IL-6 and TNF-α, the tissue injury marker LDH, and the lipid peroxidation product MDA, accompanied by reduced anti-inflammatory IL-10 and antioxidant SOD activity. Treatment with PB1 or PB1-Lip@M-CST potently reversed these changes, restoring a protective profile that was more pronounced with the nano-formulation. CXCR4 overexpression further aggravated the inflammatory and oxidative imbalance (Fig. 9A-F), yet PB1 and PB1-Lip@M-CST retained their capacity to normalize all measured parameters. In contrast, CXCR4 inhibition by Ple alone significantly attenuated inflammation and oxidative stress (Fig. 10A-F). Under these conditions, addition of PB1 or PB1-Lip@M-CST elicited no additional beneficial effects on any of the inflammatory or oxidative stress markers. Taken together, these results demonstrate that the anti-inflammatory and antioxidant properties of PB1 and PB1-Lip@M-CST are largely associated with CXCR4 signaling, confirming that CXCR4/PI3K/AKT1 pathway inhibition is an important mechanism contributing to the superior cardioprotective efficacy of the nano-delivery formulation.

Fig. 9.

Effect of CXCR4 overexpression on the modulation of inflammatory and oxidative stress markers by PB1 and PB1-Lip@M-CST in HL-1 cells under hypoxia-reoxygenation. (A)–(B) SOD activity and MDA content measured in cell lysates. (C)–(F) Levels of IL-10, IL-6, TNF-α and LDH measured in culture supernatants by ELISA. Data are mean ± SD (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

Fig. 10.

Effect of the CXCR4 inhibitor Ple on the modulation of inflammatory and oxidative stress markers by PB1 and PB1-Lip@M-CST in HL-1 cells under hypoxia-reoxygenation. (A)–(B) SOD activity and MDA content measured in cell lysates. (C)–(F) Levels of IL-10, IL-6, TNF-α, and LDH measured in culture supernatants by ELISA. Data are mean ± SD (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant).

PB1-Lip@M-CST nanodelivery system confers superior cardioprotection in a mouse model of MIRI

To evaluate the therapeutic efficacy of the PB1-Lip@M-CST nanodelivery system in vivo, a well-established mouse model of MIRI was employed in C57BL/6J mice. The left anterior descending coronary artery was occluded for 30 min followed by reperfusion, and PB1-Lip, PB1-Lip@M, PB1-Lip@M-CST (each at an equivalent dose of 10 mg/kg PB1), or vehicle was administered intravenously at 0, 1, and 2 days post-reperfusion (Fig. 11A).

Fig. 11.

PB1-Lip@M-CST nanodelivery system reduces myocardial infarct size, improves cardiac function, downregulates CXCR4 expression, and attenuates myocardial injury in a mouse model of MIRI. (A) Schematic timeline of the experimental protocol in C57BL/6J mice. (B) Representative photographs of TTC-stained heart sections at 5 days post-reperfusion. (C) Quantification of myocardial infarct size (% of left ventricular area). (D) Representative immunofluorescence staining of CXCR4 (red) with DAPI nuclear counterstaining (blue) in the infarct border zone at 5 days post-reperfusion. Scale bar = 50 μm. (E) Quantitative analysis of CXCR4 mean fluorescence intensity. (F)–(H) Serum concentrations of myocardial injury and heart failure markers at 5 days post-reperfusion: cardiac troponin I (cTnI) (F), creatine kinase-MB (CK-MB) (G), and brain natriuretic peptide (BNP) (H). (I) Representative TUNEL staining images of heart tissue sections at 5 days post-reperfusion. TUNEL-positive nuclei are stained red, and total nuclei are counterstained with DAPI (blue). Scale bar = 50 μm. (J) Quantitative analysis of mean fluorescence intensity in the infarct border zone. (K) Representative M-mode echocardiographic images at 7 days post-reperfusion. (L)–(O) Echocardiographic assessment of cardiac function at 7 days post-reperfusion: left ventricular ejection fraction (LVEF) (L), left ventricular end-systolic diameter (LVESD) (M), left ventricular end-diastolic diameter (LVEDD) (N), and left ventricular fractional shortening (LVFS) (O). (P)–(T) Serum levels of additional myocardial injury and oxidative stress markers at 7 days post-reperfusion: aspartate aminotransferase (AST) (P), lactate dehydrogenase (LDH) (Q), superoxide dismutase (SOD) (R), catalase (CAT) (S), and glutathione peroxidase (GPX) (T). Data are presented as mean ± SD (n = 5 per group). Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

At 5 days post-reperfusion, 2,3,5-triphenyltetrazolium chloride (TTC) staining revealed that brick-red regions represented viable myocardial tissue, while pale unstained areas indicated infarcted myocardial tissue. Compared with the sham-operated (Sham) group, the model group exhibited extensive pale infarct areas due to ischemia-reperfusion (I/R) injury, corresponding to an infarct size of 49.22 ± 3.03% of the left ventricular area, which was significantly higher than 2.82 ± 0.95% in the Sham group (P < 0.0001; Fig. 11B, C). In comparison with the model group, the myocardial infarct sizes in the PB1-Lip group, PB1-Lip@M group, and PB1-Lip@M-CST group decreased to approximately 38.19 ± 2.26%, 32.86 ± 4.65%, and 25.58 ± 2.96%, respectively, with statistically significant differences (P < 0.001, P < 0.0001, and P < 0.0001, respectively). Among these, the PB1-Lip@M-CST group showed the most significant reduction in infarct size (Fig. 11C). Immunofluorescence staining of heart sections at 5 days post-reperfusion demonstrated that PB1-Lip@M-CST was associated with significantly reduced CXCR4 expression in the infarct border zone (Fig. 11D, E), an effect that was more pronounced than that observed with the non-targeted PB1-Lip or the macrophage membrane coated PB1-Lip@M formulations. At 5 days post-reperfusion, serum ELISA analysis revealed that levels of creatine kinase-MB (CK-MB), cardiac troponin I (cTnI), and brain natriuretic peptide (BNP) were significantly elevated in the model group compared with the Sham group (P < 0.0001), indicating successful establishment of the ischemia-reperfusion (I/R) model with substantial myocardial injury. In comparison with the model group, the PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST groups all exhibited significantly reduced serum levels of CK-MB, cTnI, and BNP (P < 0.0001). The reduction effect increased progressively with the degree of nanocarrier modification, with the PB1-Lip@M-CST group demonstrating the most significant decrease (Fig. 11F-H). TUNEL staining of heart sections at 5 days post-reperfusion demonstrated that, compared with the Sham group, the Model group exhibited a significant increase in mean fluorescence intensity in the infarct border zone (P < 0.0001), indicating that I/R injury markedly induced cardiomyocyte apoptosis. In comparison with the Model group, the PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST groups all significantly reduced the mean fluorescence intensity (P < 0.0001 for each group). The level of myocardial apoptosis decreased progressively with increasing degrees of nanocarrier modification, with the PB1-Lip@M-CST group showing the lowest apoptotic activity (Fig. 11I, J). These quantitative findings were consistent with morphological observations. The sham group displayed only scattered TUNEL-positive signals, whereas the Model group showed widespread positive signals with enhanced fluorescence intensity. In contrast, the PB1-Lip@M-CST group exhibited markedly reduced positive signals accompanied by relatively preserved myocardial tissue architecture.

Echocardiographic assessment at 7 days post-reperfusion revealed that, compared with the Sham group, the Model group exhibited markedly impaired left ventricular systolic function, as evidenced by significantly reduced left ventricular ejection fraction (LVEF; 14.12 ± 0.52% versus 67.08 ± 3.69%, P < 0.001) and left ventricular fractional shortening (LVFS; 6.14 ± 0.49% versus 40.78 ± 4.28%, P < 0.001). Concurrently, the Model group displayed significant left ventricular dilation and adverse remodeling, characterized by increased left ventricular end-diastolic diameter (LVEDD; 5.91 ± 0.32 mm versus 3.00 ± 0.30 mm, P < 0.0001) and left ventricular end-systolic diameter (LVESD; 5.52 ± 0.26 mm versus 1.77 ± 0.06 mm, P < 0.0001). In comparison with the Model group, treatment with PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST significantly improved all of these cardiac function parameters. Among the formulations, PB1-Lip@M-CST exerted the most pronounced protective effect, restoring LVEF and LVFS to approximately 50.42 ± 8.74% and 25.62 ± 5.06%, respectively, while reducing LVEDD and LVESD to approximately 3.63 ± 0.33 mm and 2.76 ± 0.41 mm, respectively (P < 0.0001). These improvements were superior to those achieved by the PB1-Lip and PB1-Lip@M groups and brought the functional parameters closer to Sham levels (Fig. 11K-O). To further evaluate the extent of cardiomyocyte injury and oxidative stress imbalance following I/R injury, serum biochemical analyses were performed at 7 days post-reperfusion (Fig. 11P-T). Compared with the Sham group, the Model group exhibited significantly elevated serum levels of aspartate aminotransferase (AST; 146.84 ± 12.66 U/L versus 95.38 ± 7.85 U/L) and lactate dehydrogenase (LDH; 658.84 ± 54.38 U/L versus 315.94 ± 39.16 U/L) (both P < 0.0001). In addition, the activities of the antioxidant enzymes superoxide dismutase (SOD; 55.86 ± 5.83 U/mL versus 158.80 ± 12.92 U/mL), catalase (CAT; 8.48 ± 1.52 U/mL versus 29.52 ± 5.06 U/mL), and glutathione peroxidase (GPX; 38.34 ± 4.67 U/mL versus 54.46 ± 3.96 U/mL) were markedly decreased in the Model group (all P < 0.0001), indicating severe myocardial damage accompanied by pronounced oxidative stress imbalance. In comparison with the Model group, treatment with PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST significantly reduced AST and LDH levels while restoring SOD, CAT, and GPX activities to varying degrees (all P < 0.0001). Among the formulations, PB1-Lip@M-CST conferred the greatest improvement among the tested groups, decreasing AST to 89.76 ± 6.27 U/L, LDH to 329.14 ± 13.23 U/L, SOD to 134.94 ± 10.56 U/mL, CAT to 26.02 ± 1.86 U/mL, and GPX to 52.16 ± 2.10 U/mL (all P < 0.0001 versus Model). These protective effects were superior to those observed with PB1-Lip and PB1-Lip@M.

Collectively, these results demonstrate that PB1-Lip@M-CST confers superior cardioprotective efficacy compared with the other tested formulations in the mouse MIRI model by limiting infarct expansion, preserving systolic function, attenuating cardiomyocyte apoptosis and oxidative stress, and suppressing maladaptive ventricular remodeling. The targeted biomimetic nanocarrier consistently outperformed both free PB1 and the non-CST-modified variants, highlighting the therapeutic advantage of precise myocardial homing.

Attenuation of myocardial fibrosis and adverse ventricular remodeling by PB1-Lip@M-CST

To evaluate the long-term effects of PB1 nano-formulations on ventricular remodeling and fibrosis after MIRI, hearts were harvested at 28 days post-reperfusion for hematoxylin and eosin (H&E) and Masson’s trichrome staining (Fig. 12). In the untreated model group, H&E staining revealed severe disruption of myocardial architecture, including disorganized cardiomyocyte alignment, focal myocyte degeneration and necrosis, and prominent inflammatory cell infiltration (Fig. 12A). Masson’s trichrome staining demonstrated extensive interstitial and replacement fibrosis, with the collagen volume fraction (CVF) in the infarct border zone increasing dramatically from 3.58 ± 0.37% in the sham group to 36.82 ± 1.09% in the model group (P < 0.0001; Fig. 12B, C).

Fig. 12.

Representative hematoxylin and eosin (H&E) and Masson’s trichrome staining of myocardial tissues from different experimental groups at 28 days post-reperfusion. (A) Representative H&E-stained images. Upper panels show whole-heart transverse sections (scale bar = 1 mm); lower panels show magnified views of the infarct border zone (scale bar = 100 μm). (B) Representative Masson’s trichrome-stained images. Upper panels show whole-heart transverse sections (scale bar = 1 mm); lower panels show magnified views of the infarct border zone (scale bar = 100 μm). Collagen fibers are stained blue. (C) Quantification of collagen volume fraction in the infarct border zone. Data are presented as mean ± SD (n = 5, ****P < 0.0001).

Treatment with PB1-Lip@M-CST substantially preserved myocardial structural integrity, as shown by organized cardiomyocyte arrangement and markedly reduced inflammatory infiltration on H&E staining. Correspondingly, Masson’s trichrome staining revealed a pronounced reduction in collagen deposition, lowering the CVF to 7.45 ± 1.63% (P < 0.0001 vs. model group). This anti-fibrotic effect was superior to that observed with PB1-Lip (CVF 20.02 ± 2.79%) and PB1-Lip@M (CVF 11.03 ± 2.24%), suggesting incremental therapeutic improvement with stepwise surface modification of the nanocarrier.

Pharmacokinetic and tissue distribution analysis

Following intravenous administration of PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST to mice subjected to MIRI, pharmacokinetic and biodistribution studies revealed that sequential surface engineering progressively enhanced systemic exposure and cardiac targeting efficiency.

In serum, PB1-Lip displayed rapid clearance, characterized by a short elimination half-life (t_1/2 = 5.87 ± 1.95 h) and limited exposure (AUC_0−last = 47674.55 ± 3,293.73 h·ng/mL) (Fig. 13A; Table 3). Coating with macrophage membranes (PB1-Lip@M) markedly prolonged circulation, resulting in a significantly extended t_1/2 (13.48 ± 1.90 h, P < 0.01), substantially increased AUC_0−last (160301.12 ± 13547.98 h·ng/mL, P < 0.0001), and reduced clearance (CL_obs = 0.0008 ± 0.0001 mL/h/kg, P < 0.0001). Notably, PB1-Lip@M-CST exhibited the most favorable pharmacokinetic profile, with the longest t_1/2 (29.93 ± 3.58 h, P < 0.0001 vs. PB1-Lip and PB1-Lip@M), highest systemic exposure (AUC_0−last = 185557.04 ± 7,777.85 h·ng/mL, P < 0.0001 vs. PB1-Lip, P < 0.01 vs. PB1-Lip@M), and lowest clearance (CL_obs = 0.0006 ± 0.0001 mL/h/kg, P < 0.0001 vs. PB1-Lip, P < 0.001 vs. PB1-Lip@M).

Fig. 13.

In vivo pharmacokinetics and biodistribution of PB1-loaded nanoparticles in MIRI mice. (A) Plasma concentration–time profiles. (B)–(F) Concentration–time profiles in the heart (B), kidney (C), liver (D), lung (E), and spleen (F). Insets show magnified views of the initial 4 h. Data are mean ± SD (n = 5).

Table 3.

Summary of pharmacokinetic parameters of PB1 in serum and major tissues following intravenous administration of PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST in MIRI mice.

Tissue	Parameter	Units	PB1-Lip	PB1-Lip@M	PB1-Lip@M-CST
Serum	t1/2	h	5.87 ± 1.95	13.48 ± 1.90**	29.93 ± 3.58****####
	Cmax	ng/mL	1112.97 ± 51.21	1537.84 ± 54.63****	1755.01 ± 20.12****####
	AUC0−last	h·ng/mL	47674.55 ± 3293.73	160301.12 ± 13547.98****	185557.04 ± 7777.85****##
	Tmax	h	0.50 ± 0.00	0.50 ± 0.00	0.50 ± 0.00
	CLobs	mL/h/kg	0.0022 ± 0.0001	0.0008 ± 0.0001****	0.0006 ± 0.0001****###
	MRTINF_obs	h	11.04 ± 0.24	19.32 ± 0.78***	36.33 ± 3.91****####
Heart	t1/2	h	11.95 ± 2.04	21.18 ± 2.41****	17.43 ± 2.00**#
	Cmax	ng/mL	126.75 ± 12.50	216.88 ± 9.03****	259.84 ± 23.15****##
	AUC0−last	h·ng/mL	11244.40 ± 615.86	30678.02 ± 2851.88****	43001.71 ± 1032.29****####
	Tmax	h	0.50 ± 0.00	0.50 ± 0.00	0.50 ± 0.00
	CLobs	mL/h/kg	0.0116 ± 0.0006	0.0045 ± 0.0003****	0.0033 ± 0.0002****##
	MRTINF_obs	h	17.16 ± 1.01	29.64 ± 2.09****	25.87 ± 1.98****#
Liver	t1/2	h	6.95 ± 0.80	9.28 ± 0.96**	9.48 ± 1.05**
	Cmax	ng/mL	108.51 ± 4.72	166.95 ± 4.79****	173.03 ± 3.57****
	AUC0−last	h·ng/mL	8324.05 ± 905.50	19028.95 ± 1895.64****	21358.43 ± 2340.88****
	Tmax	h	1.00 ± 0.00	1.00 ± 0.00	1.00 ± 0.00
	CLobs	mL/h/kg	0.0130 ± 0.0008	0.0066 ± 0.0005****	0.0059 ± 0.0005****
	MRTINF_obs	h	11.37 ± 0.54	14.54 ± 0.34****	14.80 ± 0.85****
Spleen	t1/2	h	6.96 ± 0.48	16.54 ± 2.66***	14.92 ± 3.66***
	Cmax	ng/mL	102.67 ± 5.23	170.51 ± 1.53****	175.69 ± 1.97****
	AUC0−last	h·ng/mL	9941.35 ± 781.58	27132.35 ± 1596.73****	28719.34 ± 2621.20****
	Tmax	h	1.00 ± 0.00	1.00 ± 0.00	1.00 ± 0.00
	CLobs	mL/h/kg	0.0119 ± 0.0007	0.0051 ± 0.0005****	0.0048 ± 0.0005****
	MRTINF_obs	h	12.64 ± 0.49	22.44 ± 1.94****	21.43 ± 3.50***
Lung	t1/2	h	6.38 ± 0.57	15.91 ± 2.74****	13.39 ± 2.06***
	Cmax	ng/mL	103.92 ± 7.20	173.02 ± 6.91****	180.79 ± 4.80****
	AUC0−last	h·ng/mL	14465.12 ± 1968.01	33813.48 ± 3697.61***	37482.54 ± 5326.77****
	Tmax	h	1.00 ± 0.00	1.00 ± 0.00	1.00 ± 0.00
	CLobs	mL/h/kg	0.0093 ± 0.0008	0.0042 ± 0.0003****	0.0039 ± 0.0003****
	MRTINF_obs	h	14.11 ± 0.60	22.62 ± 2.49****	20.59 ± 1.30***
Kidney	t1/2	h	20.99 ± 0.77	15.90 ± 0.17****	17.39 ± 0.10****
	Cmax	ng/mL	57.93 ± 1.66	68.75 ± 0.73****	69.68 ± 0.74****
	AUC0−last	h·ng/mL	971.22 ± 82.08	3941.04 ± 115.95****	4312.19 ± 115.95****
	Tmax	h	1.00 ± 0.00	1.00 ± 0.00	1.00 ± 0.00
	CLobs	mL/h/kg	0.0474 ± 0.0027	0.0216 ± 0.0005****	0.0203 ± 0.0004****
	MRTINF_obs	h	6.67 ± 0.36	12.46 ± 0.22****	13.79 ± 0.19****

Note: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 versus PB1-Lip; #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001 versus PB1-Lip@M. Statistical analysis: one-way ANOVA followed by Tukey’s post hoc test.

Tissue distribution analysis further demonstrated the superior cardiac targeting of PB1-Lip@M-CST (Fig. 13B; Table 3).

In heart tissue, both membrane-coated formulations significantly enhanced drug accumulation compared with PB1-Lip (Fig. 13B; Table 3). PB1-Lip@M increased C_max to 216.88 ± 9.03 ng/mL (P < 0.0001) and AUC_0−last to 30678.02 ± 2,851.88 h·ng/mL (P < 0.0001). PB1-Lip@M-CST achieved the highest heart exposure, with C_max of 259.84 ± 23.15 ng/mL (P < 0.0001 vs. PB1-Lip, P < 0.01 vs. PB1-Lip@M) and AUC_0−last of 43001.71 ± 1032.29 h·ng/mL (P < 0.0001 vs. both PB1-Lip and PB1-Lip@M). Although the tissue elimination half-life (t_1/2) of PB1-Lip@M-CST in the heart (17.43 ± 2.00 h) was modestly shorter than that of PB1-Lip@M (21.18 ± 2.41 h, P < 0.05), its overall cardiac exposure (AUC_₀−last) remained significantly higher (43,001.71 ± 1,032.29 vs. 30,678.02 ± 2,851.88 h·ng/mL), reflecting superior net drug delivery and targeting efficiency conferred by the CST modification. This pattern is consistent with enhanced cardiomyocyte internalization mediated by the cardiac-specific peptide, resulting in higher initial accumulation (C_max) followed by more rapid intracellular processing and release of PB1. In non-target tissues, both PB1-Lip@M and PB1-Lip@M-CST exhibited significantly higher overall exposure (AUC_₀−last) in the liver, spleen, lung, and kidney compared with PB1-Lip (P < 0.001 for Lung in PB1-Lip@M group; P < 0.0001 for all other groups) (Fig. 13C–F; Table 3). Importantly, CST conjugation did not further increase drug exposure in these organs relative to PB1-Lip@M (no statistically significant difference), demonstrating that the targeting modification enhanced cardiac specificity without exacerbating off-target distribution.

Collectively, these results indicate that macrophage membrane coating effectively prolongs systemic circulation, while CST functionalization provides additional cardiac-specific enrichment. The optimized pharmacokinetic behavior and enhanced cardiac targeting of PB1-Lip@M-CST position it as a promising nanoplatform for targeted cardioprotection in ischemic heart disease.

In vivo injured cardiac tissue-specific localization of Cy5-labeled PB1-Lip@M-CST nanoparticles

To evaluate the myocardial targeting efficiency and in vivo biodistribution of the Cy5-labeled PB1 nanoformulations, Lip-Cy5, PB1-Lip-Cy5, PB1-Lip-Cy5@M, and PB1-Lip-Cy5@M-CST were administered intravenously (10 mg/kg PB1 equivalent) at the onset of reperfusion in C57BL/6J mice subjected to MIRI. Serial non-invasive fluorescence imaging demonstrated markedly enhanced and sustained cardiac accumulation of the macrophage membrane-coated formulations (Fig. 14A). Both PB1-Lip-Cy5@M and PB1-Lip-Cy5@M-CST exhibited strong fluorescence signals in the cardiac region as early as 4 h post-reperfusion, with PB1-Lip-Cy5@M-CST displaying the most intense signal at all examined time points. Quantitative analysis of cardiac radiant efficiency (expressed as [p/s/cm²/sr]/[µW/cm²]) confirmed that PB1-Lip-Cy5@M-CST produced the highest values at 4 h ((8.28 ± 0.55) × 10⁸), 24 h ((6.38 ± 0.18) × 10⁸), and 48 h ((4.26 ± 0.16) × 10⁸), significantly outperforming the other three formulations (P < 0.0001; Fig. 14B). PB1-Lip-Cy5@M also showed substantially higher cardiac signal than the non-targeted groups at all time points (P < 0.0001), whereas no significant difference was observed between Lip-Cy5 and PB1-Lip-Cy5.

Fig. 14.

In vivo and ex vivo fluorescence imaging analysis of myocardial targeting by different Cy5-labeled liposomal nano-formulations in a mouse model of MIRI. (A) Representative in vivo fluorescence images of MIRI mice captured at 4, 24, and 48 h after intravenous injection of the indicated Cy5-labeled formulations (Lip-Cy5, PB1-Lip-Cy5, PB1-Lip-Cy5@M, and PB1-Lip-Cy5@M-CST). (B) Quantitative analysis of cardiac radiant efficiency at 4, 24, and 48 h. (C) Ex vivo fluorescence images of major organs (heart, liver, spleen, lung, and kidney) harvested at 48 h post-reperfusion. Data are presented as mean ± SD (n = 5). Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test (ns, P > 0.05; ***P < 0.001; ****P < 0.0001).

Ex vivo fluorescence imaging of major organs at 48 h post-reperfusion further supported the enhanced cardiac accumulation of PB1-Lip-Cy5@M-CST (Fig. 14C). While fluorescent signals were also observed in the liver and kidneys (as is common for systemically administered nanoparticles^17,21,28, the heart displayed the strongest and most prominent signal among all examined organs in the PB1-Lip-Cy5@M-CST group. In contrast, signals in the spleen and lung remained relatively low. Collectively, these findings indicate that the combination of macrophage membrane coating and CST modification confers markedly enhanced myocardial enrichment and prolonged retention of PB1-Lip-Cy5@M-CST within the ischemic-reperfused myocardium, supporting its superior targeting capability relative to non-modified formulations.

Biosafety evaluation of PB1 nanoformulations in healthy mice

• To evaluate the systemic biosafety of the PB1 nanoformulations, in vitro hemocompatibility and in vivo toxicity assessments were performed in healthy C57BL/6J mice following three consecutive daily intravenous administrations (10 mg/kg PB1 equivalent). In vitro hemolysis assays demonstrated excellent hemocompatibility of all three formulations (Fig. 15A–D). Incubation of mouse erythrocytes with PB1-Lip, PB1-Lip@M, or PB1-Lip@M-CST at concentrations up to 200 µg/mL resulted in hemolysis rates well below the 5% safety threshold, with no statistically significant difference from the PBS negative control (ns, P > 0.05).

• Serum biochemical analysis conducted 7 days after the final injection revealed no significant alterations in liver or kidney function markers (Fig. 15E–I). Levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), blood urea nitrogen (BUN), and creatinine (Cr) remained comparable to those in the control group (ns, P > 0.05) and within normal physiological ranges.

• Consistent with the biochemical results, histopathological examination of major organs (heart, liver, spleen, lung, and kidney) harvested 7 days after the last administration showed normal tissue architecture and cellular morphology in all treatment groups (Fig. 15J). No evidence of inflammatory infiltration, degeneration, necrosis, fibrosis, or other pathological changes was observed.

• Collectively, these findings indicate that PB1-Lip, PB1-Lip@M, and PB1-Lip@M-CST exhibit favorable hemocompatibility and an excellent systemic safety profile at the tested dose and regimen, supporting their potential as safe nanotherapeutic platforms for MIRI.

Fig. 15.

In vitro hemocompatibility and in vivo systemic biosafety evaluation of PB1 nanoformulations following repeated intravenous administration in healthy C57BL/6J mice. (A) Representative photographs of the hemolysis assay after incubation of mouse erythrocytes with PB1-Lip, PB1-Lip@M, or PB1-Lip@M-CST at concentrations of 0.5, 5, 10, 50, 100, and 200 µg/mL. PBS and 0.5% Triton X-100 served as negative and positive controls, respectively. (B)–(D) Quantitative hemolysis rates of PB1-Lip (B), PB1-Lip@M (C), and PB1-Lip@M-CST (D) at the indicated concentrations. (E)–(I) Serum biochemical parameters evaluated 7 days after three consecutive daily intravenous injections: alanine aminotransferase (ALT; E), aspartate aminotransferase (AST; F), blood urea nitrogen (BUN; G), creatinine (Cr; H), and alkaline phosphatase (ALP; I). (J) Representative hematoxylin and eosin (H&E)-stained histological sections of major organs (heart, liver, spleen, lung, and kidney) collected 7 days after the last administration. Scale bars = 100 μm. Data are presented as mean ± SD (n = 5 per group). Statistical significance was determined by one-way ANOVA followed by Tukey’s post hoc test (ns, P > 0.05).

Discussion

The present study demonstrates the feasibility of PB1-Lip@M-CST as a biomimetic nanoplatform that synergistically combines macrophage membrane coating with cardiac-specific CST peptide conjugation for targeted delivery of PB1 in MIRI. By harnessing the inflammation-homing prowess of macrophage membranes and the reported myocardial affinity of CST, this system addresses key challenges in cardiovascular nanomedicine, such as rapid clearance and suboptimal tissue accumulation, to achieve enhanced cardioprotection in both in vitro and in vivo models. Our results demonstrate enhanced biocompatibility, extended circulation, and targeted delivery, culminating in marked reductions in oxidative stress, inflammation, apoptosis, and infarct size, alongside improved cardiac function. This work builds upon established biomimetic strategies by integrating PB1 delivery with macrophage membrane coating and cardiac-specific CST peptide conjugation, creating a biomimetic platform specifically designed for CXCR4-dependent cardioprotection.

Physicochemical characterization of PB1-Lip@M-CST revealed a robust core-shell architecture with hydrodynamic diameter of 152.09 ± 1.96 nm (PDI = 0.192 ± 0.008) and zeta potential of − 20.13 ± 1.53 mV. These parameters, in conjunction with the TEM-confirmed core-shell morphology and retention of key macrophage proteins (CCR2, CD11b, CD36), are likely to contribute to colloidal stability, reduced phagocytic uptake, and preservation of the membrane’s inherent inflammation-homing capability. The observed size increment and surface charge modulation align with the biophysical principles that govern prolonged circulation and reduced opsonization in biomimetic systems, yet the additional CST conjugation introduces a layer of cardiac specificity that distinguishes this platform from single-modality macrophage-membrane coatings reported previously^19–21. Controlled release kinetics (approximately 80% cumulative PB1 release over 48 h without burst effect) further indicate that the hybrid membrane barrier effectively modulates diffusion, addressing the rapid systemic clearance that has historically limited free PB1 translation in cardiovascular settings²⁹. This modification strategy enhances the pharmacokinetic performance of PB1-Lip@M-CST, establishing it as a promising vehicle for sustained PB1 delivery and addressing pharmacokinetic limitations that have hindered PB1’s clinical translation in cardiovascular contexts. This design represents a novel combination of immune-evasive macrophage membrane coating with CST-mediated cardiac targeting.

Hypoxia-reoxygenation injury was induced in HL-1 cells using chemical hypoxia with CoCl₂ as a convenient and well-established mimetic of hypoxia. Cobalt chloride (CoCl₂) was selected as a widely validated chemical hypoxia mimetic because it inhibits prolyl hydroxylases, stabilizes HIF-1α under normoxic conditions, and reproducibly induces oxidative stress, inflammatory responses, and apoptosis in cardiomyocytes³⁰. Compared with chamber-based hypoxia systems, this approach provides superior reproducibility and precise temporal control of hypoxia-reoxygenation conditions³¹. In vitro assessments in CoCl₂-induced hypoxia-reoxygenation models using HL-1 cells demonstrated that PB1-Lip@M-CST significantly suppressed ROS overproduction and reduced early apoptosis from 28.17 ± 0.49% (Model) to 15.19 ± 1.71%, outperforming free PB1 and non-targeted formulations. Surface accessibility of the conjugated CST peptide was functionally validated by competitive blocking with free CST, a standard approach in cardiac-targeting nanomedicine that supports receptor engagement without requiring additional surface-analysis techniques. The sequence-specific uptake, together with time-dependent internalization peaking at 4 h, illustrates how CST-mediated receptor engagement enhances endocytosis in injured cardiomyocytes without compromising the immune-evasion conferred by the macrophage membrane²². Even in CXCR4-overexpressing cells, the nanoparticles partially reversed H/R-induced dysregulation of proinflammatory cytokines (IL-6, TNF-α) while restoring anti-inflammatory IL-10 levels. These findings suggest improved targeting within the inflammatory MIRI microenvironment, characterized by DAMPs, TLR activation, and NLRP3 inflammasome-driven cytokine storms, more effectively than conventional antioxidants, thereby positioning the platform as a CXCR4-dependent therapeutic vehicle^1,8. In H/R models, the formulation modulated CXCR4-associated PI3K/AKT1 signaling and restored BAX/BCL-2 balance; these effects were abolished under pharmacological CXCR4 inhibition or exacerbated by CXCR4 overexpression, supporting the involvement of CXCR4 signaling. Notably, while earlier studies on procyanidins primarily acted through broad NF-κB or MAPK pathways, our platform achieves cardiac-specific delivery via CST conjugation and macrophage membrane coating, followed by PB1-mediated modulation of CXCR4-associated PI3K/AKT1 signaling. This spatially targeted approach may offer advantages over non-targeted antioxidant strategies.

The pharmacokinetic and biodistribution profiles observed in MIRI mice underscore the translational merit of our stepwise biomimetic engineering strategy. PB1-Lip@M-CST exhibited markedly prolonged systemic circulation and substantially enhanced cardiac accumulation (highest C_max and AUC_₀−last in heart tissue) relative to both uncoated and partially modified formulations, without further elevating off-target deposition in reticuloendothelial organs. Notably, the modestly shorter cardiac tissue t_1/2 observed with the CST-modified formulation did not compromise overall exposure, consistent with accelerated cellular uptake in the ischemic myocardium. These findings indicate that macrophage membrane coating primarily extends blood circulation time and reduces rapid clearance, while subsequent CST conjugation provides an additional layer of cardiac enrichment. This synergistic design effectively optimizes the therapeutic index of PB1 by decoupling prolonged systemic exposure from excessive off-target accumulation-a common limitation of conventional nanoparticle platforms and many single-component biomimetic systems reported previously^19,20. In contrast to earlier macrophage membrane-coated nanoparticles, which often achieved cardiac targeting at the expense of significantly increased hepatic and splenic uptake²¹, or ligand-modified liposomes that suffer from rapid RES clearance despite surface functionalization²², our dual-modified formulation simultaneously achieves extended circulation, superior cardiac enrichment, and minimal off-target accumulation. This balanced pharmacokinetic profile represents a clear advancement over existing cardiac-targeted nanoplatforms.

Therapeutic outcomes in the MIRI model further validated the superiority of the fully engineered formulation. PB1-Lip@M-CST achieved the greatest reduction in infarct size and the most pronounced restoration of cardiac function at 7 days post-reperfusion, accompanied by significant downregulation of injury biomarkers. Notably, these acute protective effects translated into long-term benefits, as evidenced by substantially attenuated myocardial fibrosis at 28 days. The progressive improvement observed with each surface modification step demonstrates that the biomimetic strategy not only limits acute necrotic injury but also effectively mitigates adverse ventricular remodeling.This dual benefit distinguishes PB1-Lip@M-CST from previously reported single-component biomimetic systems^18,19. Importantly, comprehensive biosafety evaluations confirmed that all three formulations exhibited excellent hemocompatibility and no detectable systemic toxicity following repeated intravenous administration, with hemolysis rates well below the 5% safety threshold and no abnormalities in serum biochemistry or major organ histopathology. These results collectively establish PB1-Lip@M-CST as a safe and highly effective nanotherapeutic platform, offering a promising strategy for targeted delivery in MIRI with strong potential for clinical translation.

Although the present study demonstrates promising therapeutic efficacy of PB1-Lip@M-CST in a mouse model of MIRI, several limitations should be acknowledged. First, all in vivo experiments were conducted in young male C57BL/6J mice, which may not fully represent the complex clinical scenario of myocardial infarction in elderly patients of both sexes with multiple comorbidities. The C57BL/6J mouse was chosen because it is a well-established, reproducible, and ethically acceptable preclinical platform that recapitulates the core pathophysiological features of human MIRI, including oxidative stress, inflammation, and cardiomyocyte apoptosis. Macrophage membrane-coated liposomes (a form of cell-membrane-hybrid nanocarrier) were evaluated in this model to mimic the immune-evasion and inflammation-homing properties that would be desirable in human therapy. However, translation from mice to humans remains challenging. Species-specific differences in immune system function, cardiac anatomy, pharmacokinetics, and disease comorbidities may influence the performance and safety of such biomimetic nanoplatforms. Therefore, further validation in large-animal models and, ultimately, human clinical trials will be essential to confirm the clinical translatability of this biomimetic strategy. Second, the in vitro hypoxia-reoxygenation model employed chemical hypoxia induced by CoCl₂ rather than physical hypoxia in a controlled oxygen chamber, which may not completely recapitulate the pathophysiological conditions of ischaemia-reperfusion. While CoCl₂ effectively stabilises HIF-1α and induces key downstream responses, it does not fully recapitulate the metabolic consequences of true ischaemia-reperfusion. This limitation should be considered when interpreting the translational relevance of the findings; future studies using gas-controlled hypoxic chambers or ex vivo Langendorff models would provide complementary validation. Third, while we observed beneficial effects up to 28 days post-reperfusion, longer-term studies are warranted to evaluate chronic cardiac remodeling, heart failure development, and long-term safety. Fourth, although CXCR4 dependency was confirmed using both overexpression and antagonist approaches, other potential off-target mechanisms cannot be entirely excluded. Fifth, a direct comparison with free PB1 was not included in the in vivo efficacy evaluation. Although our in vitro data showed that free PB1 had weaker protective effects compared with PB1-Lip@M-CST, we acknowledge that the pharmacokinetics of free intravenous PB1 have not been characterized in this study. Therefore, a direct in vivo comparison between free PB1 and the nano-formulation remains unresolved. Future studies should include a free PB1 group to fully evaluate the therapeutic advantage of PB1-Lip@M-CST in MIRI.

Collectively, this orchestrated interplay of biomimicry and precision targeting provides a promising biomimetic strategy for targeted intervention in MIRI, addressing critical gaps in current reperfusion strategies and warranting advancement to scalable production and humanized models to refine clinical dosing and efficacy.

Conclusions

We have successfully developed a novel biomimetic nanoplatform, PB1-Lip@M-CST, by integrating macrophage membrane coating with the cardiac-specific targeting peptide CST for the targeted delivery of PB1 in MIRI. The engineered nanoparticles exhibited a stable core-shell structure, sustained drug release, and preserved key macrophage surface proteins (CCR2, CD11b, and CD36), which collectively conferred excellent biocompatibility, immune evasion, and enhanced accumulation in ischemic myocardium.

In vitro, PB1-Lip@M-CST effectively protected HL-1 cardiomyocytes against hypoxia-reoxygenation injury through CXCR4-mediated inhibition of the PI3K/AKT1 pathway and restoration of the BAX/BCL-2 balance, demonstrating superior efficacy over uncoated or partially modified formulations. In vivo, the dual-modified system achieved prolonged circulation, preferential cardiac distribution, significant reduction in infarct size, and improved cardiac function, while attenuating systemic injury biomarkers and long-term myocardial fibrosis. Importantly, comprehensive biosafety evaluations confirmed excellent hemocompatibility and the absence of systemic toxicity following repeated administration. These findings establish PB1-Lip@M-CST as an effective and safe biomimetic strategy for CXCR4-dependent cardioprotection in MIRI.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AC

Ascorbic acid

Akt1

Protein kinase B 1

ANP

Atrial natriuretic peptide

AUC_0-last

Area under the curve from time zero to the last measurable concentration

BAX

Bcl-2-associated X protein

Bax

Bcl-2-associated X protein gene

BCL-2

B-cell lymphoma 2

Bcl-2

B-cell lymphoma 2 gene

BNP

Brain natriuretic peptide

CCK-8

Cell Counting Kit-8

CCR2

C-C chemokine receptor type 2

CD11b

Cluster of differentiation 11b

CD36

Cluster of differentiation 36

CK-MB

Creatine kinase-MB

CL

Clearance

CLobs

Observed clearance

CLSM

Confocal laser scanning microscopy

C_max

Maximum concentration

CoCl₂

Cobalt chloride

CPK

Creatine phosphokinase

CRP

C-reactive protein

CST

Cardiac-specific targeting peptide (CSTSMLKAC)

cTnI

Cardiac troponin I

CXCR4

C-X-C chemokine receptor type 4

DAPI

4’,6-Diamidino-2-phenylindole

DCFH-DA

2’,7’-Dichlorofluorescein diacetate

DiO

3,3’-Dioctadecyloxacarbocyanine perchlorate

DLS

Dynamic light scattering

DMEM

Dulbecco’s modified Eagle’s medium

DSPE-PEG-Mal

1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[maleimide(polyethylene glycol)-2000]

DSC

Differential scanning calorimetry

ECG

Electrocardiogram

FBS

Fetal bovine serum

GO

Gene Ontology

H&E

Hematoxylin and eosin

HIF-1α

Hypoxia-inducible factor-1 alpha

HL-1

HL-1 cells

IACUC

Institutional Animal Care and Use Committee

ICAM-1

Intercellular adhesion molecule 1

IL-1β

Interleukin-1 beta

IL-6

Interleukin-6

IL-10

Interleukin-10

IL-18

Interleukin-18

KEGG

Kyoto Encyclopedia of Genes and Genomes

LAD

Left anterior descending

LC–MS/MS

Liquid chromatography-tandem mass spectrometry

LDH

Lactate dehydrogenase

LOQ

Limit of quantification

LVEDD

Left ventricular end-diastolic diameter

LVESD

Left ventricular end-systolic diameter

LVEF

Left ventricular ejection fraction

LVFS

Fractional shortening

MDA

Malondialdehyde

MIRI

Myocardial ischaemia–reperfusion injury

MRT

Mean residence time

MRTINF_obs

Mean residence time extrapolated to infinity observed

MWCO

Molecular weight cut-off

NF-κB

Nuclear factor kappa B

NLRP3

NLR family pyrin domain containing 3

NO

Nitric oxide

NT-proBNP

N-terminal pro-brain natriuretic peptide

OE

Overexpressing

p-AKT1

Phosphorylated AKT1

PB1

Procyanidin B1

PB1-Lip

PB1-loaded liposomes

PB1-Lip@M

PB1-loaded liposomes coated with macrophage membrane

PB1-Lip@M-CST

PB1-loaded liposomes coated with CST-conjugated macrophage membrane

PBS

Phosphate-buffered saline

PDI

Polydispersity index

Pi3k

Phosphoinositide 3-kinase

PPI

Protein-protein interaction

RT-qPCR

Reverse transcription quantitative polymerase chain reaction

RAW 264.7

RAW 264.7 macrophages

RIPA

Radioimmunoprecipitation assay

RNS

Reactive nitrogen species

ROS

Reactive oxygen species

SDS-PAGE

Sodium dodecyl sulfate–polyacrylamide gel electrophoresis

SOD

Superoxide dismutase

SPC

Soybean phosphatidylcholine

t1/2

Half-life

TEM

Transmission electron microscopy

TLRs

Toll-like receptors

T_max

Time to maximum concentration

TNF-α

Tumor necrosis factor-alpha

TTC

Triphenyltetrazolium chloride

TUNEL

Terminal deoxynucleotidyl transferase dUTP nick end labeling

XRD

X-ray diffraction

Author contributions

H.L. and Y.Y. performed formal analysis; H.L. wrote the original draft; Y.Y. and D.Y. validated the data; D.Y. and H.L. developed the methodology; S.L. supervised the project, acquired funding, and revised the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by Standardized Diagnosis and Treatment of Heart Failure (No.2025cyjstg006).

Data availability

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding authors. The human CXCR4 coding sequence was amplified by PCR and subcloned into the pLVX-IRES-Puro vector using EcoRI and BamHI restriction sites. The construct was verified by commercial Sanger sequencing, which confirmed identity with the reference sequence. No novel sequencing data were generated.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the Chongqing Medical University. The animal study protocol was approved by the Ethics Committee for the Welfare of Experimental Animals of the Chongqing Medical University (protocol code: IACUC-CQMU-20250743).

Consent for publication

Not applicable.