Heme cascade-triggered nanovesicles for myocardial ischemic core delivery

Abstract

Myocardial ischemia-reperfusion injury (MIRI) remains a significant therapeutic challenge due to insufficient targeted drug delivery. Recognizing heme as a key endogenous signal that has yet to be utilized in delivery systems, we designed the first heme-responsive nanovesicle (DecAS-PC@NM) for lesion-specific therapy through a cascade reaction involving two functionalized phospholipids. Specifically, heme activates the artemisinin-modified phospholipid (A-PC) to generate reactive oxygen species, which oxidize the thioether structure in another phospholipid (S-PC), triggering a hydrophilic-to-hydrophobic transition and subsequent vesicle disassembly. Co-assembly with neutrophil membranes further enhances chemotaxis toward inflammatory sites. In vitro studies confirm the unique heme responsiveness of DecAS-PC@NM, while in vivo data highlight its ischemic core targeting and significant therapeutic improvement. This innovative heme-triggered phospholipid cascade offers a promising strategy for MIRI treatment.

Graphical abstract

Highlights

• •Heme as a specific trigger for the ischemic core.
• •Heme-cascade response based on phospholipids.
• •Heme cascade-triggered vesicles for MIRI treatment.

1. Introduction

Myocardial ischemia-reperfusion injury (MIRI) is a frequent pathophysiological complication following myocardial infarction (MI) interventions. Central to its pathogenesis is the ischemic core, a region of significant microvascular dysfunction that triggers and exacerbates myocardial damage during reperfusion [[1], [2], [3], [4]]. This ultimately leads to secondary myocardial damage, triggering major adverse cardiovascular events and mortality. Microcirculatory dysfunction exhibits distinct pathological features and microenvironmental alterations predominantly within the MIRI ischemic core. Nevertheless, targeted therapeutic strategies for this critical region remain virtually non-existent. Particularly in drug delivery, no effective approach enables precise targeting of the ischemic core [5,6]. Current drug delivery systems exploit triggers such as inflammatory chemotaxis, hypoxia, and acidic pH for stimuli-responsive release. However, these triggers lack adequate specificity for the complex environment of the ischemic core, resulting in off-target deposition within the peri-ischemic border zone and restricting therapeutic efficacy [[7], [8], [9], [10]]. Thus, it is urgently necessary to develop innovative release mechanisms and delivery systems specifically designed for the unique microenvironment of the ischemic core. This progress is essential for tackling the substantial therapeutic challenge posed by MIRI.

Ferroheme (heme) is an endogenous active molecule containing an iron-centered porphyrin ring. As the core carrier for oxygen transport in hemoglobin, its reversible Fe²⁺-O₂ binding plays vital physiological roles in cellular aerobic metabolism [[11], [12], [13]]. However, during ischemia-reperfusion, heme detaches from normal metabolic function and abnormally accumulates [[14], [15], [16]]. This accumulation interacts with the microenvironment of the MIRI ischemic core characterized by hypoxia, acidity, high reactive oxygen species (ROS), and inflammation ultimately becoming a contributing pathological factor that worsens ischemic core injury [17]. Quantitative analysis shows that heme levels rise significantly in the ischemic core post-MIRI. This triggers endoplasmic reticulum stress, causes intracellular iron overload in cardiomyocytes, and intensifies lipid peroxidation, ultimately inducing ferroptosis. Ferroptosis, a non-apoptotic cell death pathway dependent on iron and lipid peroxidation, has been shown to drive cardiomyocyte death within the ischemic core [[18], [19], [20], [21], [22]]. These findings indicate that altered heme levels serve as an endogenous, specific molecular marker of ischemic core damage. Consequently, research on the dynamic metabolism and spatial distribution of heme within the ischemic core remains crucial. Developing heme-responsive delivery strategies is particularly important. Intelligent delivery systems capable of precisely recognizing zones of abnormal heme accumulation could significantly improve drug enrichment efficiency in the ischemic core, thereby enhancing therapeutic outcomes.

The development of heme-responsive delivery systems represents a significant yet largely unexplored therapeutic avenue. As a unique iron-porphyrin molecule with redox activity, heme can react with responsive chemical groups like peroxide bonds, keto/aldehyde groups, and pyridine/imidazole moieties [23,24]. Among these, peroxide bridges (O-O bonds) show outstanding reactivity and specificity toward heme [[25], [26], [27]]. This property was validated in antimalarial therapy: parasite heme rapidly cleaves artemisinin peroxide bridge to generate ROS, killing parasites [23,[28], [29], [30]]. In the previous studies [[31], [32], [33], [34]], we synthesized an artemisinin-phosphatidylcholine prodrug (A-PC) based on artemisinin structure. By replacing hydrophobic fatty acid chains at sn-1/sn-2 positions of phosphatidylcholine with artemisinin-succinate, A-PC retains phospholipid properties while enabling direct preparation of nanovesicles (e.g., liposomes). Extensive in vitro and in vivo tests confirmed A-PC is efficiently activated by heme. However, heme-responsive peroxide cleavage alone, although it generates ROS, is not sufficient to effectively disrupt self-assembled structures and achieve substantial drug release. In vitro degradation assays, A-PC–based nanovesicles showed only slow structural disintegration [31]. Fortunately, ROS-responsive chemistries are well-established in biological systems. For example, thioether groups convert from hydrophobic to hydrophilic states under ROS exposure, thereby disrupting hydrophobic domains and disassembling structures [35,36]. We previously synthesized a ROS-sensitive thioether phosphatidylcholine (S-PC) [37]. S-PC assemblies demonstrate precise ROS-triggered drug release in high-ROS environments. By integrating A-PC and S-PC into one platform, we achieve cascade responsiveness to heme and ROS signals. This dual-targeting approach enables precise drug delivery to areas of pathological heme accumulation.

To combat heme-induced cardiomyocyte ferroptosis, we previously identified that Decursin (a natural plant-derived compound) inhibits heme-triggered ferroptosis in MIRI [38,39]. This intervention reduces iron overload and lipid peroxidation in the ischemic core, significantly alleviating myocardial injury [[40], [41], [42], [43], [44]]. However, the extremely short half-life and lack of targeting specificity severely limit its in vivo bioavailability [[45], [46], [47], [48]]. Building on this finding, our heme-responsive delivery system significantly increases drug concentration in heme-rich tissues and may substantially enhance the therapeutic efficacy of Decursin in improving MIRI.

This work focuses on surmounting the significant barrier of efficient drug delivery to the ischemic core in MIRI. For the first time, we demonstrate for the first time that heme shows gradient accumulation within the ischemic core and can act as a specific trigger. Based on this, we developed a heme-responsive drug delivery system that significantly improves targeted drug delivery and therapeutic effects in the ischemic core. This system includes two functional phospholipid molecules, A-PC and S-PC. These lipids self-assemble into vesicles loaded with Decursin. The vesicles are then coated with neutrophil membranes using an extrusion method to form hybrid nanoparticles (DecAS-PC@NM) (Fig. 1a). After intravenous injection, DecAS-PC@NM uses the natural inflammation-targeting ability of neutrophils to accumulate preferentially at ischemic sites [49,50]. Within the ischemic core, high heme levels trigger a cascade response: A-PC undergoes cleavage of its peroxide bridge, and S-PC undergoes thioether oxidation. This dual activation greatly enhances the controlled release of Decursin. The released free Decursin subsequently diffuses and accumulates further within the ischemic core (Fig. 1b). This site-specific concentration enables precise delivery to the damaged tissue, thereby enhancing overall therapeutic efficacy (Fig. 1c). Our optimized DecAS-PC@NM combines the natural targeting by neutrophils with a dual heme/ROS-responsive cascade mechanism. This achieves drug delivery to the ischemic core and provides a new strategy for the precise treatment of MIRI.

Fig. 1.

Design of DecAS-PC@NM for targeting and effect on ischemia-reperfusion process. (a) Schematic diagram of the design and synthesis of DecAS-PC@NM. (b) The process of targeting myocardial ischemic regions by DecAS-PC@NM and the heme-responsive degradation release of Decursin. (c) Mechanistic pathways underlying Decursin-mediated protection against myocardial ischemia-reperfusion injury.

2. Results

2.1. Potential threptic ability of Decursin on MIRI

The effect of Decursin on MIRI had been verified in this study, and the detail of animal experiment protocol was outlined in Fig. S1a. To ensure high success rates in surgical modeling, each mice underwent electrocardiogram monitoring both before and after surgery. ST-segment elevation was used to confirm precise ligation of the anterior descending branch (Fig. S1b), thereby minimizing modeling deviations. Decursin significantly reduced myocardial infarct size in mice undergoing ischemia/reperfusion (I/R) surgery, as shown by TTC staining compared to vehicle-treated controls (Fig. S1c and S1d). PET-CT analysis of myocardial viability indicated a notable decrease in the mean standardized uptake value (SUV) in the I/R surgery group, while the Decursin-treated group exhibited a significant increase in mean SUV (Fig. S1e and S1f). Furthermore, 2D M-mode echocardiography, demonstrated that Decursin treatment effectively reversed the reductions in left ventricular ejection fraction (LVEF) and left ventricular shortening fraction (LVFS) caused by I/R, thereby enhancing contractile function (Fig. S1g–1i). These findings suggested that Decursin alleviated the ischemic region, promoting improved myocardial glucose uptake and cardiac function.

To elucidate the mechanism by which Decursin mitigates ischemia/reperfusion (I/R) injury, differentially expressed genes (DEGs) were assessed using transcriptome sequencing. Principal component analysis (PCA) of the RNA-Seq data for the Sham, vehicle-treated I/R, and Decursin-treated I/R groups was presented in Fig. S2, while DEGs comparisons among the three groups were shown in Fig. S1j [51]. Co-expression trends and pathway enrichment analyses revealed that DEGs were segregated into three expression sets, indicating that Decursin negatively regulated DEGs in opposition to I/R. Pathway enrichment analysis revealed associations with ferroptosis, inflammation, and apoptotic signaling, suggesting that Decursin exerts protective effects against myocardial ischemia-reperfusion injury by modulating these pathways (Fig. S3).

2.2. Feature of DecAS-PC@NM response to Heme

Ensuring the stability of Decursin and enhancing its targeted release to the infarcted region are central to this study [52]. Therefore, we first aimed to characterize the microenvironment of the infarcted area, particularly the distribution characteristics of heme. By assessing heme concentrations in various pathological regions, the distribution profile of heme in the infarcted zone was elucidated. Hematoxylin and Eosin (H&E) staining analysis revealed erythrocyte aggregation in the infarct zone on the first day post-IR surgery, with erythrocytes exhibiting uniform biconcave morphology and bright red coloration, indicative of microvascular obstruction (MVO). On the third day post-IR surgery, erythrocytes displayed heterogeneous shapes and sizes, suggesting microvascular damage and widespread erythrocyte lysis, characteristic of intramyocardial hemorrhage (IMH) (Fig. 2a). Fe²⁺ levels, partially indicative of heme concentrations, were assessed before heme quantification. Fe²⁺ assays showed significantly elevated concentrations in myocardial tissue homogenates from the I/R group compared to the sham group (Fig. 2b). The ischemic myocardium was divided into three zones based on injury severity: remote zone, border zone, and lesion core areas (Fig. 2c). Equal weights from these regions were analyzed, and heme content was quantitatively assessed after homogenization. Results demonstrated a gradient distribution of heme, with concentrations progressively increasing from remote zone to lesion core tissues. The difference in heme levels between the Sham group and the ischemic core region was highly significant (P < 0.0001) (Fig. 2d). This observation suggests a significant correlation between heme accumulation and the severity of myocardial injury. Moreover, the localization of heme in the core of the lesion indicates its potential role as a biomarker for targeting MIRI. Based on this characteristic, responsive nanovesicles were designed to achieve controlled drug release.

Fig. 2.

Quantification of heme and heme-response of DecAS-PC@NM. (a) Representative images of cardiac sections of H&E-stained on the first- and third-day post-IR surgery. (b) Quantitative analysis of Fe²⁺ in heart homogenate using HPLC-ICPMS coupling after I/R surgery (n = 5 biologically independent animals). (c, d) Schematic diagram (c) of myocardial ischemic zone division and quantitative analysis of heme (d) in each area (n = 5 biologically independent animals). (e) Representative TEM images of AS-PCs and DecAS-PC@NM. (f) Representative TEM image showing surface cracks of DecAS-PC@NM. (g, h) Diameter and zeta potentials of AS-PCs and DecAS-PC@NM after conventional storage (g) and 24-h room temperature storage (h) (n = 5 biologically independent samples). (i, j) The evaluation of the Z-average diameter and polydispersity index of DecAS-PC@NM through dynamic light scattering analysis after conventional storage (i) and 24-h room temperature storage (j). (k, l) Spectral analysis (k) and quantitative evaluation (l) of Decursin loading in DecAS-PC@NM nanovesicles. (m) Determination of Decursin release using HPLC. (n) Release curves of DecAS-PC@NM treated with PBS and heme. (n = 3 biologically independent samples). The data are expressed as mean ± s.d. NS indicates not significant. ∗∗P < 0.01 and ∗∗∗∗P < 0.0001. Data in b were analyzed by two-tailed unpaired Student's t-test. Data in d were analyzed by one-way ANOVA followed by Bonferroni's post hoc test. Data in n were analyzed by Welch ANOVA followed by Bonferroni's post hoc test.

Considering the heme distribution characteristics and the requirements for targeted release, this study engineered composite nanovesicles DecAS-PC@NM. Transmission electron microscopy (TEM) was used to confirm the spherical or elliptical morphology of the fabricated AS-PC and DecAS-PC@NM with complete encapsulation. Unlike traditional liposome-like membrane vesicles, the DecAS-PC@NM was observed to have a thicker outer membrane of higher density (Fig. 2e). To ensure the purity of neutrophil membranes used for nanoparticle coating, we evaluated the purity of neutrophil isolates by flow cytometry. The proportion of neutrophils co-expressing the specific markers CD45, CD11b, and Ly6G was consistently high, with purity greater than 90% across independent preparations (Fig. S4). Further investigation into heme-triggered structural changes in DecAS-PC@NM was conducted using TEM, which vividly captured the degradation process. Exposure to 50 μM heme for 6 h, consistent with the physiological concentration found in lesion cores during I/R injury, led to the formation of discernible cracks on the nanocarrier surface (Fig. 2f).

Dynamic light scattering (DLS) measurements indicated that the size of DecAS-PC@NM was approximately 170 nm, slightly larger than that of AS-PC, with a zeta potential of approximately −22 mV (Fig. 2g). The physical stability of particle size and zeta potential remained unaffected after standing at room temperature for 24 h (Fig. 2h). Likewise, the diameter and zeta potential of AS-PCs and DecAS-PC@NM remained unchanged after storage at room temperature for 24 h in serum-containing culture medium (Fig. S17). These results collectively confirmed the in vivo stability of DecAS-PC@NM. DLS was further employed to systematically characterize the particle size distribution profile and polydispersity index (PDI) of DecAS-PC@NM. The results showed that DecAS-PC@NM exhibited a narrowly distributed, unimodal size distribution, with a Z-average hydrodynamic diameter of 170 nm and a PDI of 0.181, indicating good size uniformity and relatively low polydispersity of the formulation (Fig. 2i). After storage at room temperature under static conditions for 24 h, the Z-average diameter was determined to be 175 nm with a PDI of 0.175, without any significant change compared with the initial values (Fig. 2j). The intensity-weighted DLS size distribution curves did not display any new peaks at larger sizes or pronounced shoulders, and no detectable aggregation or obvious multimodal distribution was observed. Together with the zeta potential measurements, these data indicate that the particle size and surface charge of DecAS-PC@NM remain stable over 24 h at room temperature, demonstrating good short-term physical stability of the nanovesicle system. Such controlled dispersity is expected to be beneficial for the reproducibility of its in vivo targeting and release behavior. As shown in Fig. 2k, the UV-Vis absorption spectra of Decursin at different concentrations, revealing concentration-dependent spectral features. UV-Vis quantitative analysis confirmed the successful loading of Decursin into DecAS-PC@NM nanovesicles. The encapsulation efficiency (EE) and drug loading content (DL) were calculated using an external calibration curve of absorbance versus concentration (Fig. 2l), yielding an EE of 78% and a DL of 13% (determined after a 20-fold dilution of the formulation), indicating robust encapsulation and effective retention of Decursin within the biomimetic vesicle system.

DecAS-PC@NM was subsequently incubated with PBS or with heme solutions at varying concentrations (35 μM, 40 μM, and 50 μM) for 24 h, followed by high-performance liquid chromatography (HPLC) analysis. The results showed that under PBS conditions, only minimal vesicle disintegration occurred, leading to the release of merely trace amounts of resveratrol. In contrast, heme exposure induced substantial disruption of the nanocarriers, accompanied by a clear concentration-dependent pattern. As the heme concentration increased, the vesicle rupture and corresponding drug-release efficiency exhibited a pronounced positive correlation. These findings validate the concentration-dependent relationship between heme levels and vesicle release efficiency within this biomimetic nanodelivery system (Fig. 2m). Quantitative analyses using HPLC revealed that the percentage release curves of decursin from DecAS-PC@NM, measured at intervals of 1 h, 6 h, 12 h, 24 h, 48 h, and 72 h after treatment with PBS or with heme solutions at varying concentrations (35 μM, 40 μM, and 50 μM), showed a logarithmic pattern (Fig. 2n). These results confirm that heme effectively induces the disintegration of DecAS-PC@NM, facilitating the controlled release of the encapsulated therapeutic agent. The results demonstrate that DecAS-PC@NM exhibits excellent stability as nanovesicles and enables controlled release of decursin in response to heme, thus prolonging the sustained stability of intravenous decursin administration.

2.3. Targeting and safety properties of DecAS-PC@NM

Previous studies have shown that neutrophil membranes not only enhance the accumulation of nanoparticles in inflamed myocardial regions, but also exhibit biomembrane-specific homing behavior. Therefore, mimicking neutrophil migration is considered a potential nanostrategy for targeting myocardial injury [49,50]. To simulate ischemia-reperfusion (I/R) injury in myocardial cells, H9c2 cells were subjected to oxygen-glucose deprivation/reperfusion (OGD/R) in vitro. DecAS-PC@NM and AS-PC were fluorescently labeled with DiIC18(5) dye (DiD) to assess their targeting efficacy on injured cells. Confocal Laser Scanning Microscopy (CLSM) was used to visualize the retention of DiD-labeled nanocarriers around H9c2 cells. After incubation and subsequent PBS washing to remove unbound nanocarriers, DecAS-PC@NM nanocarriers showed selective adherence to injured cell, while AS-PC aggregated in cell-free regions (Fig. 3a and c; Fig. S5). This aggregation of AS-PC in the acellular regions is nonspecific and occurs at a low frequency, which may be caused by the washing pathway and hydrodynamic distribution, but it does not affect the quantitative comparison of the targeting performance of the material. These results indicate that DecAS-PC@NM exhibits exceptional targeting and retention on myocardial cells damaged by I/R injury.

Fig. 3.

Efficiency evaluation of targeted delivery in vivo and in vitro. (a) Confocal Microscope images showing the binding of DiD-labeled AS-PCs and DecAS-PC@NM to H9c2 cells after incubation. Scale bars, 10 μm. (b) Fluorescence imaging of organ-specific distribution in vivo. (c) Quantitation of DiD-labeled AS-PCs and DecAS-PC@NM remaining on H9c2 cells after elution (n = 3 biologically independent samples). (d) Quantification of radiant efficiency for organ-specific fluorescence distribution in vivo. (e, f) Fluorescence imaging (e) and quantitation of radiant efficiency (f) by these mice after intravenous injection at 1 h, 12 h, 24 h, and 48 h (n = 3 biologically independent animals). (g, h) Comparative analysis of the accumulated levels of Decursin in various organs (g) and different regions of myocardial ischemia (h) using HPLC. The data are expressed as mean ± s.d. NS indicates not significant. ∗∗∗P < 0.001. Data in c, d were analyzed by one-way ANOVA followed by Bonferroni's post hoc test. Data in f were analyzed by Welch ANOVA followed by Bonferroni's post hoc test.

To further delineate the biodistribution profile suggested by in vivo imaging, ex vivo fluorescence imaging of excised major organs (heart, liver, spleen, lung, and kidney) was performed at 48 h after intravenous administration. As shown in Fig. 3b, representative images and corresponding quantification, the DecAS-PC@NM group displayed a pronounced increase in fluorescence intensity in the heart compared with the AS-PC group, confirming enhanced cardiac accumulation/retention. In contrast, fluorescence signals in the liver and spleen were not elevated in the DecAS-PC@NM group, several off-target organs exhibited unchanged or reduced fluorescence relative to AS-PC (Fig. 3d). To thoroughly evaluate the targeting effectiveness of DecAS-PC@NM for ischemic myocardial tissue and its specificity to the lesion core in vivo, mice undergoing myocardial ischemia-reperfusion surgery were randomly assigned to receive tail vein injections of DiIC18(7)-labeled AS-PC or DecAS-PC@NM on the third day post-IR surgery. Using in vivo imaging technology, both AS-PC and DecAS-PC@NM groups showed significant fluorescence enhancements in the cardiac region compared to controls, with the DecAS-PC@NM group displaying the highest fluorescence intensity and prolonged signal duration (Fig. 3e and f). Liver accumulation in systemic delivery remains a significant challenge for nanoparticles. After the administration of DecASP@NM, Decursin reaches its highest concentration in the liver, which is consistent with the typical distribution characteristics of nano-delivery systems in vivo. Additionally, HPLC revealed that decursin concentrations in the heart were second only to those in the liver following DecAS-PC@NM administration (Fig. 3g; Fig. S6), highlighting effective targeting and accumulation in the ischemic myocardium. To further clarify the localization within the ischemic area, the heart was divided into Remote zone, Border zone, and Lesion core, and myocardial tissues from each region were analyzed via HPLC. Decursin levels demonstrated a gradient distribution from the remote zone to the lesion core, mirroring heme distribution patterns (Fig. 3h; Figs. S7 and S16), indicating that the nanocarriers respond to heme-mediated signals to release the drug precisely at the target site. Collectively, these results confirm that DecAS-PC@NM selectively and sustainably accumulate in ischemic myocardial tissue and effectively target the lesion core.

In vitro cytotoxicity assessments were conducted on H9c2 and RAW264.7 cells using AS-PC, decursin, and DecAS-PC@NM. The cells were treated with the respective drugs or nanocarriers at a concentration of 50 μM, while PBS served as the control, over a 24 h incubation period. Subsequent analysis using the Cell Counting Kit-8 (CCK-8) assay revealed no statistically significant differences in cell viability among the treatment groups (Fig. 4a–b). Additionally, H&E staining was conducted on the major non-cardiac organs of mice in each experimental group to assess the systemic toxicity of AS-PC, decursin, and DecAS-PC@NM administered via tail vein injections. The lung, kidney, liver, and spleen were collected for histopathological analysis and results showed no significant pathological changes across the different treatment groups (Fig. 4c). Furthermore, blood compatibility was quantitatively evaluated using a hemolysis assay based on hemoglobin release. After incubation with Vehicle or DecAS-PC@NM, the supernatants were collected and the absorbance at 540 nm was measured to determine the hemolysis rate. Both Vehicle and DecAS-PC@NM resulted in hemolysis rates comparable to the negative control, with no appreciable increase in hemoglobin release, whereas the lysis buffer produced near-complete hemolysis (Fig. 4d and e). At day 7 after ischemia-reperfusion (I/R), we performed statistical comparisons of key liver function markers (ALT and AST) as well as renal function indicators (UACR and CRE). The results demonstrated no significant differences between the DecAS-PC@NM-treated group and the control groups (Fig. S18). Collectively, these findings suggest that AS-PC, Decursin, and DecAS-PC@NM exhibit favorable in vitro or in vivo biocompatibility and do not present toxic effects following intravenous administration.

Fig. 4.

Biocompatibility evaluation of DecAS-PC@NM in vitro and in vivo. (a, b) Proliferation of H9c2 (a) and RAW264.7 (b) cells detected by CCK-8 assay after co-culture on the third day (n = 5 biologically independent samples). (c) Each organ (heart, lung, spleen, liver, kidney) was performed by H&E staining after intravenous injection. (d) Photographs of the in vitro hemocompatibility test of DecAS-PC@NM (n = 5 biologically independent animals). (e) Quantitative analysis of in vitro hemolysis for blood compatibility evaluation of DecAS-PC@NM. Data were analyzed by one-way ANOVA followed by Bonferroni's post hoc test.

2.4. Assessment of DecAS-PC@NM therapy on MIRI

To assess whether intravenous administration of DecAS-PC@NM improves the therapeutic efficacy of decursin in I/R injury by targeting delivery to the ischemic myocardial lesion core, a series of detailed animal experiments were conducted as shown in Fig. S8. Consistent with previous methodologies, precise ligation of the anterior descending branch was confirmed through ST-segment elevation (Fig. 5a), thereby minimizing modeling deviations. Cardiac function assessments via echocardiography showed that both decursin and DecAS-PC@NM significantly improved LVEF and LVFS post-I/R, with DecAS-PC@NM exhibiting particularly pronounced effects Fig. 5b, g and 5h). During the entire murine echocardiographic assessment, the heart rate was closely monitored and maintained within 550 ± 50 beats·min⁻¹ to minimize bias caused by fluctuations in anesthesia (Fig. S9) [53,54]. Supporting these findings, PET-CT evaluation of myocardial viability revealed superior improvement in SUV with DecAS-PC@NM therapy (Fig. 5c and i; Fig. S10). Histopathological analyses, including H&E, Masson's trichrome, and Sirius Red staining, consistently demonstrated that DecAS-PC@NM treatment markedly reduced typical manifestations of myocardial injury, such as muscle degeneration, fibrosis, and collagen deposition (Fig. 5d, e and 5j; Fig. S11). Moreover, quantification of myocardial infarct size using TTC staining further substantiated the superior cardioprotective effects of DecAS-PC@NM compared decursin alone (Fig. 5f and k). Collectively, these findings strongly suggest that DecAS-PC@NM enhances the therapeutic efficacy of decursin in mitigating post-I/R myocardial injury, offering a promising approach for targeted cardiac therapy.

Fig. 5.

Efficacy evaluation in vivo. (a) Representative electrocardiograms of mice before and after LAD artery occlusion. (b) Representative M-mode images of echocardiographic. (c) Representative PET imaging with ¹⁸F-FDG on the seventh day post-IR surgery. (d) Representative images of H&E-stained mice cardiac sections harvested on the seventh day post-IR surgery. Scale bars, 1 mm. (e) Representative images of whole and local amplified MT-stained mice cardiac sections. Scale bars, 1 mm. (f) Representative images of TTC-stained cardiac sections. Scale bars, 1 mm. (g, h) LVEF (g) and LVFS (h) assessed by M-mode echocardiography (n = 15 biologically independent animals). Scale bars, 1 mm. (i) Quantitative analysis of cardiomyocyte viability based on SUV of ¹⁸F-FDG (n = 5 biologically independent animals). (j) Quantitative analysis of fibrotic tissue based on the MT staining on the seventh day post-IR surgery (n = 3 biologically independent animals). (k) Quantitative analysis of infarct size from TTC-stained cardiac sections (n = 5 biologically independent animals). The data are expressed as mean ± s.d. NS indicates not significant. ∗∗∗∗P < 0.0001. Data in g, i, j, k were analyzed by one-way ANOVA followed by Bonferroni's post hoc test. Data in h were analyzed by Welch ANOVA followed by Bonferroni's post hoc test.

2.5. DecAS-PC@NM mitigates ferroptosis and oxidative stress in ischemic myocardial tissue

As previously described, the RNA-seq results demonstrated that decursin effectively inhibits I/R-induced ferroptosis, inflammation, and apoptosis. In the cellular studies, intracellular ferrous iron (Fe²⁺) was first assessed using FerroOrange staining. Compared with the OGDR group, DecAS-PC@NM treatment markedly reduced FerroOrange fluorescence intensity, indicating an effective attenuation of intracellular Fe²⁺ accumulation (Fig. S12a and d). Consistently, lipid peroxidation staining showed a significant decrease in the lipid peroxidation–positive area in the DecAS-PC@NM group, suggesting that DecAS-PC@NM suppresses OGDR-induced membrane lipid peroxidation (Fig. S12b and e). Finally, Acrolein Red staining, commonly used to detect acrolein, a reactive aldehyde end-product generated during lipid peroxidation and closely associated with ferroptotic damage, was substantially diminished upon DecAS-PC@NM treatment, further supporting its inhibitory effect on OGDR-triggered ferroptosis (Fig. S12c and f).

Further validation is planned under I/R conditions using DecAS-PC@NM. On the seventh day post-IR surgery in murine models, Prussian blue staining showed a significant reduction in iron-positive areas within the myocardial tissue of subjects treated with DecAS-PC@NM, compared to the I/R group (Fig. 6a and e). This observation indicated a marked decrease in iron accumulation. Concurrently, 4-hydroxynonenal (4-HNE) immunohistochemistry demonstrated a substantial reduction in positively stained areas within the cardiomyocyte cytoplasm in the DecAS-PC@NM cohort (Fig. 6b and f), suggesting significant attenuation of lipid peroxidation in ischemic myocardium. Ultrastructural analysis via transmission electron microscopy on the seventh day post-IR surgery revealed aberrant mitochondrial morphology and disorganized arrangement in the I/R group, which was notably ameliorated by DecAS-PC@NM administration (Fig. 6c; Fig. S13a and S14b and S13c). Dihydroethidium (DHE) immunofluorescence staining of ischemic myocardial tissue demonstrated that DecAS-PC@NM significantly mitigated ROS production (Fig. 6d and g). Furthermore, Western blot analysis of ferroptosis-related markers in H9c2 cells subjected to OGD/R and treated with DecAS-PC@NM showed that the levels of molecules that promote ferroptosis, such as transferrin receptor 1 (TFR1) and acyl-CoA synthetase long-chain family member 4 (ACSL4), were decreased. In contrast, the levels of molecules that inhibit ferroptosis, including solute carrier family 7 member 11 (SLC7A11), glutathione peroxidase 4 (GPX4), and ferroptosis suppressor protein 1 (FSP1), were increased (Fig. 6h and i). After treatment with DecAS-PC@NM, the quantitative analysis of the GSH/GSSG ratio in mouse myocardial tissue showed a significant restoration of the GSH/GSSG balance. This provides rigorous quantitative biochemical evidence supporting the inhibition of ferroptosis (Fig. S19). Corroborating these findings, H2DCFDA immunofluorescence assays in H9c2 cells yielded concordant results (Fig. S14a and S14b). Collectively, these findings suggest that DecAS-PC@NM effectively inhibits ferroptosis in ischemic myocardial tissue by decreasing iron accumulation, reducing lipid peroxidation, enhancing mitochondrial integrity, and lowering ROS levels.

Fig. 6.

Suppression of cardiomyocyte ferroptosis. (a) Representative images of Prussian blue-stained cardiac sections. (b) Representative images of 4HNE-stained cardiac sections. (c) TEM images at were utilized to evaluate the mitochondrial structure of myocardial cell. (d) H&E staining and DHE fluorescent co-staining of the same cardiac section. (e, f) Percentages of Prussian blue or 4HNE positive areas in the cardiac section (n = 3 biologically independent animals). (g) Quantitative analysis of DHE Fluorescent Staining (n = 3 biologically independent animals). (h, i) Western Blot Detection and Quantitative Analysis of Ferroptosis Marker Proteins in H9c2 Cells after OGD/R Treatment. (n = 5 biologically independent samples). The data are expressed as mean ± s.d. NS indicates not significant. ∗∗P < 0.01, ∗∗∗P < 0.001 and ∗∗∗∗P < 0.0001. Data were analyzed by one-way ANOVA followed by Bonferroni's post hoc test.

2.6. DecAS-PC@NM mitigates oxidative DNA Damage, apoptosis, and improves survival in ischemic myocardial injury

Immunofluorescence studies were conducted using CD86 and CD163 as markers for pro-inflammatory (M1) and anti-inflammatory (M2) macrophages, respectively, in murine ischemic myocardial tissue. The results demonstrated that both decursin and DecAS-PC@NM promoted the M1-to-M2 transition, with DecAS-PC@NM showing particularly pronounced effects (Fig. 7a, c and 7d). Supporting evidence was obtained in vitro using RAW264.7 cells subjected to OGD/R, where DecAS-PC@NM significantly enhanced the shift towards the M2 phenotype (Fig. 7b, e and 7f). To functionally validate the M1/M2 skewing inferred from surface markers, macrophages were FACS-sorted and restimulated ex vivo, followed by quantitative cytokine analysis. Compared with the AS-PC group, macrophages from the DecAS-PC@NM–treated produced significantly lower levels of the pro-inflammatory cytokines IL-12 and TNF-α, while exhibiting increased secretion of the anti-inflammatory cytokine IL-10 (Fig. 7g). This cytokine profile is consistent with the observed shift in M1/M2 proportions (reduced M1 and enhanced M2 macrophages), supporting that DecAS-PC@NM not only alters marker expression but also functionally reprograms cardiac macrophages toward an anti-inflammatory phenotype. Furthermore, comprehensive cytokine profiling of serum from post-I/R mice and supernatants from OGD/R-treated RAW264.7 cells revealed that DecAS-PC@NM treatment resulted in a marked reduction of pro-inflammatory cytokines and a concomitant increase in anti-inflammatory mediators (Fig. 8f). Flow-cytometric profiling of macrophage polarization yielded results consistent with our in vivo and in vitro immunofluorescence data. Compared with the OGDR group, DecAS-PC@NM significantly decreased the proportion of CD86⁺ (M1-like, proinflammatory) macrophages (Fig. S15a and c) while increasing the proportion of CD206⁺ (M2-like, anti-inflammatory) macrophages (Fig. S15b and d), indicating that DecAS-PC@NM promotes a shift from M1 to M2 polarization under OGDR conditions. These findings collectively indicate that DecAS-PC@NM facilitates macrophage polarization towards the anti-inflammatory M2 type, inhibits the secretion of pro-inflammatory cytokines, and thereby exerts significant anti-inflammatory effects in the context of post-I/R injury.

Fig. 7.

Anti-inflammation of myocardial cells in vivo and in vitro. (a, b) Representative images of CD86⁺ macrophages (M1) and CD163⁺ macrophages (M2) in the RAW264.7 macrophage cells (a) and cardiac tissue sections (b) for identifying the inflammatory phenotype. (c-f) Proportions of M1 and M2 macrophages (n = 3 biologically independent animals and biologically independent samples). (g) Quantitative analysis of macrophage inflammatory cytokines IL-12, TNF-α, and IL-10 via ELISA. The data are expressed as mean ± s.d. NS indicates not significant. ∗∗∗∗P < 0.0001. Data were analyzed by one-way ANOVA followed by Bonferroni's post hoc test.

Fig. 8.

Prevention of cardiomyocyte apoptosis in vivo and in vitro. (a) Representative images of 8-OHDG-stained cardiac sections. (b) Representative TUNEL staining images in the Cardiac Sections to assess cardiomyocyte apoptosis. (c) Cells death was detected by Live/Dead cells assay. (d) Percentages of 8-OHDG positive areas in the cardiac section (n = 3 biologically independent animals). (e) Quantitative analysis of cells death. (n = 5 biologically independent animals). (f) Quantitative analysis of inflammatory cytokines in RAW264.7 cell culture supernatant and mice plasma (n = 5 biologically independent samples and biologically independent animals). (g) Survival curve analysis was conducted seven days post-surgery in mice. The data are expressed as mean ± s.d. NS indicates not significant. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001 and ∗∗∗∗P < 0.0001. Data in d, e were analyzed by one-way ANOVA followed by Bonferroni's post hoc test. Survival distributions were estimated by the Kaplan–Meier method and compared by the log-rank test (g).

Histochemical analysis of 8-OHDG showed a significant increase in nuclear positive rates on the seventh day post-IR surgery in mice compared to the sham group, indicating heightened DNA oxidative damage. Administration of DecAS-PC@NM markedly reduced nuclear 8-OHDG positivity, suggesting a potent inhibition of DNA damage pathways (Fig. 8a and d). Furthermore, TUNEL staining of myocardial tissues demonstrated a substantial increase in TUNEL-positive cells following I/R injury, reflecting extensive chromosomal DNA double-strand breaks and heightened apoptosis. Treatment with DecAS-PC@NM significantly reduced TUNEL positivity, thereby decreasing apoptosis (Fig. 8b; Fig. S16). Live-dead cell assays supported these findings, showing that OGD/R markedly induced cell death, as evidenced by increased red fluorescence. However, DecAS-PC@NM treatment resulted in a significant reduction in dead cells and a concomitant increase in viable cells (Fig. 8c and e). Additionally, survival analysis further substantiated the therapeutic advantage of the biomimetic vesicle system. With the sample size are 30 mice per group, mortality by day 7 was 46.7% in the IR + Vehicle group, 43.3% in the IR + AS-PC group, 36.7% in the IR + Decursin group, and only 13.3% in the IR + DecAS-PC@NM group. Statistical evaluation of the Kaplan–Meier curves revealed a significant survival benefit in the IR + DecAS-PC@NM group compared with both IR + Vehicle (P = 0.008 < 0.01) and IR + Decursin (P = 0.04 < 0.05). Collectively, these findings demonstrate that DecAS-PC@NM exerts substantial cardioprotective effects by mitigating DNA damage and apoptosis in cardiomyocytes, thereby enhancing survival outcomes following I/R injury.

This work is the first to report a heme-responsive drug delivery strategy for MIRI treatment. The introduction of neutrophil membranes significantly enhanced the inflammation-targeting ability of the carriers, leading to their preferential accumulation in ischemic myocardial tissue. This was confirmed by in vivo imaging and biodistribution analysis. Mechanistically, the sequential activation of A-PC and S-PC ensured that drug release was strictly confined to heme-rich areas. This maximized the anti-ferroptosis and anti-inflammatory effects of decursin while effectively protecting normal tissues. This “trigger-release” strategy allowed the therapeutic action to closely match the spatiotemporal dynamics of MIRI pathology, outperforming traditional nanocarriers. Importantly, the inherent biocompatibility of the neutrophil membranes significantly reduced the immune clearance of the carriers, a characteristic crucial for subsequent translational applications.

Moreover, this work establishes a novel microenvironment-responsive therapy that links nanomedicine with cardiac pathophysiology. By utilizing endogenous heme as both a biomarker and a trigger, our engineered nanoparticles, DecAS-PC@NM, demonstrate a precision therapeutic strategy developed using principles of synthetic biology. Our findings not only confirm the significant value of heme as a potential drug target but also provide a basis for extending this platform to other ischemia-related diseases. Future research should focus on translational validation, including the optimization of pharmacokinetics and conducting large animal studies, to accelerate the process of clinical translation. In summary, this study significantly advances the field of targeted cardiac therapy and offers a valuable reference for reprogramming treatment strategies for myocardial ischemia-reperfusion injury and other related conditions by harnessing pathological cues.

3. Conclusion

MIRI remains a significant clinical challenge due to the lack of therapeutic strategies that can achieve precise drug delivery to the ischemic area while minimizing systemic off-target effects. Traditional drug delivery systems often fail to effectively utilize the complex and dynamic pathological microenvironment of MIRI, where heme accumulation has been identified as a key biomarker and bioactive factor. Inspired by these findings, this study successfully developed a heme-responsive nanovesicle, DecAS-PC@NM. This carrier innovatively utilizes the heme gradient within the ischemic core to achieve spatially controlled drug release. We constructed a platform with a dual-responsive cascade mechanism by co-assembling A-PC and S-PC with neutrophil membranes. Following intravenous injection, DecAS-PC@NM preferentially accumulates in inflamed areas by leveraging the natural chemotaxis of neutrophils. In the ischemic core, high heme concentrations trigger a cascade response by activating the peroxide bridge reaction of A-PC and the thioether oxidation of S-PC. This significantly enhances the controlled release of decursin. The released free decursin molecules then diffuse and further accumulate in the ischemic core, leading to precise delivery and more significant therapeutic effects.

CRediT authorship contribution statement

Hui Chen: Validation, Project administration, Investigation, Formal analysis, Data curation. Ying Yang: Methodology, Formal analysis. Mirenuer Aikebaier: Data curation, Conceptualization. Haoran Wang: Methodology, Formal analysis. Lan Zheng: Methodology. Yifeng Zhang: Methodology. Qing Zha: Methodology. Shuyao Shan: Methodology. Yanping Wang: Resources. Jiawei Chen: Validation. Fanyi Huang: Data curation. Yiping Li: Validation. Yunkai Tang: Validation. Yawei Du: Writing – review & editing, Writing – original draft, Conceptualization. Wenguo Cui: Writing – review & editing, Conceptualization. Ke Yang: Writing – review & editing, Project administration, Methodology, Data curation, Conceptualization. Yan Liu: Writing – review & editing, Resources, Project administration, Conceptualization.

Ethics approval and consent to participate

The animal study was approved by the Animal Welfare and Ethics Committee of Yunnan University (Ethics Approval Number: YUN20230453).

Declaration of competing interest

The authors declare no competing interests.

Appendix A. Supplementary data

The following is the Supplementary data to this article: