Cardiac-targeted and ROS-responsive liposomes containing puerarin for attenuating myocardial ischemia-reperfusion injury

Abstract

Aim: This study aimed to construct an ischemic cardiomyocyte-targeted and ROS-responsive drug release system to reduce myocardial ischemia-reperfusion injury (MI/RI).

Methods: We constructed thioketal (TK) and cardiac homing peptide (CHP) dual-modified liposomes loaded with puerarin (PUE@TK/CHP-L), which were expected to deliver drugs precisely into ischemic cardiomyocytes and release drugs in response to the presence of high intracellular ROS levels. The advantages of PUE@TK/CHP-L were assessed by cellular pharmacodynamics, in vivo fluorescence imaging and animal pharmacodynamics.

Results: PUE@TK/CHP-L significantly inhibited apoptosis and ferroptosis in H/R-injured cardiomyocytes and also actively targeted ischemic myocardium. Based on these advantages, PUE@TK/CHP-L could significantly enhance the drug's ability to attenuate MI/RI.

Conclusion: PUE@TK/CHP-L had potential clinical value in the precise treatment of MI/RI.

Plain language summary

Article highlights.

• Myocardial ischemia-reperfusion injury (MI/RI) is a series of pathological changes caused by revascularisation after myocardial infarction, for which there is no effective treatment.

• Excessive production of reactive oxygen species (ROS) is an important predisposing factor for MI/RI.

• Apoptosis and ferroptosis of cardiomyocytes are important pathological mechanisms of myocardial ischemia-reperfusion injury and both are closely related to excessive intracellular ROS.

• We have successfully synthesized ROS-responsive and cardiac-targeted liposomes loaded with puerarin (PUE@TK/CHP-L).

• It was found that PUE@TK/CHP-L could target ischemic myocardium.

• In vivo and ex vivo results showed that PUE@TK/CHP-L can effectively reduce the level of myocardial oxidative stress and decrease cardiomyocyte apoptosis and ferroptosis.

• These results suggested that PUE@TK/CHP-L could hopefully be a promising drug carrier for mitigating MI/RI.

1. Introduction

Acute myocardial infarction (AMI) is one of the most common and aggressive diseases in clinic [1]. In 1976, coronary thrombolytic therapy was the first successful treatment for AMI, and in the 30 years since then, the mortality rate of AMI has declined from about 20% to about 5% [2]. Nevertheless, there is growing evidence that further myocardial ischemia/reperfusion injury (MI/RI) may result from the implementation of various revascularization strategies aimed at rescuing AMI, moreover, there are no very effective solutions available [3,4].

The occurrence of MI/RI involves the intertwining of multiple mechanisms such as bursts of elevated reactive oxygen species (ROS) levels and excess iron deposition [5]. During the initial phase of reperfusion, the mitochondrial electron transport chain could generate large amounts of ROS [6]. Rapidly increasing ROS levels induce the opening of the mitochondrial permeability transition pore (mPTP), leading to a decrease in mitochondrial membrane potential, which prompts the release of cytochrome c (Cyt c) from the mitochondria into the cytoplasm and the activation of the caspase-3 activity, thereby inducing cardiomyocyte apoptosis [3]. Furthermore, the sustained opening of mPTP triggers the segregation of the oxidative respiratory chain, which also drastically increases intracellular ROS levels [6]. Notably, the production of large amounts of intracellular ROS also induces ferroptosis in cardiomyocytes, further exacerbating ischemic myocardial injury. Ferroptosis is a recently discovered iron-dependent programmed cell death, which is mainly characterized by persistently elevated levels of Fe²⁺, accumulation of lipid peroxides, rupture of mitochondrial membranes and reduction or loss of mitochondrial cristae [7]. Among them, glutathione peroxidase 4 (GPX4), a selenoprotein that specifically catalyzes the conversion of lipid peroxides to lipids and alcohols by glutathione (GSH), plays an important role in regulating cellular ferroptosis [8]. Previous studies have found that lipoxistatin-1 (Lip-1) protects ischemic myocardium by largely reducing intracellular ROS production through upregulation of GPX4 protein levels [9]. This suggests that upregulation of GPX4 protein levels to reduce ROS accumulation from lipid peroxidation is an effective strategy to inhibit ferroptosis. In summary, the dramatic increase of ROS levels in ischemic cardiomyocytes and the down-regulation of intracellular GPX4 expression interact with each other to form a vicious circle, leading to massive programmed cell death in ischemic myocardium. Therefore, pharmacologically reducing excess ROS levels and increasing GPX4 expression in ischemic cardiomyocytes could inhibit cardiomyocyte apoptosis and ferroptosis, thereby addressing the common clinical problem of MI/RI.

Puerarin (PUE) is an isoflavonoid active ingredient isolated from the roots of the Chinese herb Pueraria lobata (Willd.) Ohwi. Numerous studies have shown that PUE has excellent pharmacological activities in anti-oxidant, anti-apoptosis and anti-ferroptosis [10], and PUE injection has now become a commonly used drug in the clinical treatment of cardiovascular diseases [11,12]. Nevertheless, the lack of myocardial targeting of PUE severely limits the expected efficacy of the drug, resulting in the need for high doses of the drug to maintain clinical efficacy, which is prone to excessive accumulation in the body and adverse effects [13]. It is assumed that constructing an ischemic cardiomyocyte-targeted drug delivery system that delivers PUE to the lesion site and releases the drug in response to high levels of intracellular ROS, giving full play to the drug's role in inhibiting apoptosis and ferroptosis in cardiomyocytes, which would significantly improve the drug's efficacy and solve the problem of large-dose administration in the clinic.

Nowadays, the construction of ROS-responsive drug delivery system has attracted great research interest in the field of diseases such as cancer, atherosclerosis and diabetes [14]. Among them, many drug-carrying systems use acetone-based thioketal (TK), which breaks down into thiols and acetone in the presence of high concentrations of ROS, leading to the disintegration of the carrier system [15]. Thus, TK-based delivery systems are expected to respond to high levels of ROS in ischemic cardiomyocytes, prompting disassembly of the delivery system to fully release the drug. However, it is still difficult to achieve the desired efficacy by relying on ROS-responsiveness alone. The drug delivery system should first achieve ischemic cardiomyocyte targeting before it has a chance to encounter high intracellular concentrations of ROS to implement responsive drug release. In addition, the beating heart's pulsatile blood flow constantly impacts the coronary arteries during reperfusion, which also greatly affects the retention of the drug at the lesion site.

The current literature suggests that liposomes have better enhanced permeability and retention (EPR) effects at ischemic myocardial sites [16,17]. Notably, cardiac homing peptide (CHP, amino acid sequence CSTSMLKAC) is highly specific for ischemic myocardium by binding to receptors on the membrane of ischemic myocytes [18]. Vandergriff A and Wang X et al. demonstrated that CHP-modified exosomes specifically target ischemic myocardium and significantly enhance the effect of drug carriers in attenuating MI/RI [19,20]. Based on the EPR effect of liposomes at the ischemic myocardial site and the ischemic cardiomyocyte-targeting effect of CHP, we predicted that CHP-modified liposomes could enhance drug targeting and retention at the ischemic myocardial site, thereby attenuating the shock of pulsatile coronary blood flow during reperfusion.

In this study, liposomes could tend to ischemic myocardium, CHP promotes drug carrier targeting to ischemic cardiomyocytes and TK could response to high levels of intracellular ROS. Therefore, we constructed PUE-loaded liposomes co-modified by TK and CHP, which are expected to deliver PUE efficiently into ischemic cardiomyocytes and release it in response to high intracellular levels of ROS, making full use of the drug to inhibit apoptosis and ferroptosis, thus significantly improving drug efficacy (Figure 1).

Figure 1.

Preparation, targeted delivery of PUE@TK/CHP-L and its mechanism of inhibiting apoptosis and ferroptosis in cardiomyocytes.

CHP-PEG-PE: 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-N-[succinimidyl Succinate ester(polyethylene glycol)2000]-Cardiac homing peptide; Cyt c: Cytochrome c; GSH: Glutathione; GPX4: Glutathione peroxidase 4; PEG-TK-PE: Methoxy (polyethylene glycol)2000-Thioketal-1,2-distearoyl-sn-glycero-3-phosphoethanolamine; PUE: Puerarin; ROS: Reactive oxygen species; SL: Soybean lecithin; mPTP: Mitochondrial permeability transition pore.

2. Materials & methods

2.1. Materials

Soybean lecithin (SL) and Cholesterol (Chol) were purchased from Sigma-Aldrich. Cardiac homing peptide (CHP, CSTSMLKAC) was obtained from GL Biochem (Shanghai) Ltd (Shanghai, China). Methoxy (polyethylene glycol)2000-TK-amine (mPEG-TK-NH₂), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[succinimidyl Succinate ester(polyethylene glycol)2000] (DSPE-PEG-NHS) and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-acide (DSPE-COOH) were synthesized by Xi'an ruixi Biological Technology Co., Ltd (Shanxi, China). Coumarin-6 (C6) was provided by J&K Chemical Ltd (Shanghai, China). 4′,6-diamidino-2-phenylindole (DAPI) was acquired from Beyotime Biotech (Jiangsu, China). H9c2 cells derived from rat myocardium were obtained from the American Type Culture Collection (Manassas, VA).

2.2. Synthesis & characterization of CHP-PEG-PE & PEG-TK-PE

The synthesis process of CHP-PEG-PE was performed according to the previously reported method [21,22]. Briefly, 5 ml of anhydrous N, N-dimethylformamide (DMF) was used to dissolve 100 mg of DSPE-PEG-NHS and 35 mg of CHP, followed by the addition of 50 μl of TEA. The mixture was reacted overnight in a light-proof environment. The obtained products were dialyzed with a dialysis bag (molecular weight cut-off: 2 kDa) to remove the by-products, and then further purified by preparative chromatography using binary gradient elution with 40% to 100% acetonitrile for the oil phase and 0.1% trifluoroacetic acid (TFA) for the aqueous phase. The structure of the purified CHP-PEG-PE polymer was confirmed by ¹H NMR and MALDI-TOF mass spectrometry (MALDI-TOF MS).

PEG-TK-PE was synthesized according to the literature method [23], 100 mg of mPEG_2K-TK-NH₂ was weighed and dissolved in 10 ml of chloroform, then 35 mg of DSPE-COOH, 8 mg of EDC and 3 mg of DMAP were added, and the reaction was carried out at room temperature for 12 h. Organic solvents were removed by distillation under reduced pressure, and the residue was dissolved with the addition of appropriate amount of ethanol and transferred to a dialysis bag, which was dialyzed in pure water for 24 h to remove impurities, and then the dialysate was collected and freeze-dried to obtain the final PEG-TK-PE product. The synthesis of the amphiphilic polymer of PEG-TK-PE was verified by ¹H NMR.

2.3. Preparation & characterization of PUE@TK/CHP-L

Briefly, PUE was encapsulated in ROS-responsive liposomes co-modified by TK and CHP (PUE@TK/CHP-L). The formulation was a mixture of Soybean lecithin (SL), cholesterol (Chol), PEG-TK-PE and CHP-PEG-PE with a molar ratio of 33:33:11:23 (Table 1). The preparation of liposomes was carried out by the commonly used thin-film hydration method. The above mixture and PUE were firstly dissolved completely using chromatographic methanol and attached to the bottom of an eggplant shaped flask after rotary evaporation. It was observed that a uniform film formed at the bottom of the vial. An appropriate amount of purified water or PBS was then added and left at 37°C for 30 min before being resuspended by ultrasound. Finally, PUE@TK/CHP-L with uniform particle size was obtained by first squeezing a 0.22 μM polycarbonate membrane several times and then extruded through a 100 nm pore size polycarbonate membrane. PUE-loaded liposomes without (PUE@L) or with single modification of CHP (PUE@CHP-L) were used as the controls to validate the synergistic effect of CHP and TK. Similarly, C6-labeled liposomes were prepared by a similar method as described above using coumarin-6 (fluorescent probe, C6) instead of PUE.

Table 1.

The lipid composition, size, zeta potential, encapsulation efficiency and drug loading of liposomes.

Formulations		PUE@L	PUE@CHP-L	PUE@TK/CHP-L
Composition (% of molar mole)	SL	50	43	33
	Chol	50	43	33
	CHP-PEG-PE	–	14	11
	PEG-TK-PE	–	–	23
Size (nm)		114.3 ± 0.5	120.8 ± 0.6	122.5 ± 0.8
Zeta Potential (mV)		-45.6 ± 0.4	-37.8 ± 0.8	-29.6 ± 0.6
Encapsulation efficiency		82.2 ± 8.6%	80.2 ± 5.7%	84.2 ± 9.6%
Drug loading		5.64 ± 0.72%	5.70 ± 0.98%	5.80 ± 0.68%

Just one drop of PUE@TK/CHP-L was applied onto a carbon-coated grid. The micromorphology of PUE@TK/CHP-L was then well visualized using a transmission electron microscope (TEM) (JEM-2100F, JEOL, Japan). The drug loading efficiency (DL%) and entrapment efficiency (EE%) of PUE@TK/CHP-L were determined by high performance liquid chromatography (HPLC) system (LC-20, Japan). The chromatographic conditions were acetonitrile: water (18:82) as the mobile phase, and the detection wavelength was 250 nm.

2.4. Fourier transform infrared spectrometer

The PUE embedding status in PUE@TK/CHP-L was characterized using a Fourier transform infrared spectrometer (FT-IR) (thermal Fisher Nicolet iS10). Initially, a baseline spectrum of an empty KBr was conducted. Next, the KBr sheets were gently coated with PUE, blank liposomes (TK/CHP-L) and PUE@TK/CHP-L. Then, the FT-IR spectral characteristics of each sample within the wave number range of 400 to 4000 cm^-1 were recorded using a FT-IR spectrometer.

2.5. Drug release

The in vitro release characteristics of PUE@TK/CHP-L were evaluated using a common dialysis method. Briefly, 5 ml of newly synthesized PUE@TK/CHP-L was placed in a dialysis bag (molecular weight cut-off: 3 kDa). The bags were then immersed in phosphate buffered saline (PBS) at pH 7.4 containing 0.1 mM H₂O₂, with PBS without added H₂O₂ as a control. The mixture was then agitated at a temperature of 37°C. Following this, 50 μl of the solution was removed at 0, 2, 4, 8, 12, 24, 36 and 48 h. The obtained samples were diluted with methanol at a 1:1 v/v ratio and analyzed by HPLC. The free PUE obtained from PUE@TK/CHP-L was detected using a CAPCELL PAK C18 column (250 mm × 4.6 mm, 5.0 μm) and the chromatographic conditions described above.

2.6. Establishment of a hypoxia/reoxygenation model in H9c2 cells

In a cell culture chamber at 37°C with 5% CO₂, H9c2 cells were cultivated in DMEM medium with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S) until the cell density reached 80%. To simulate MI/RI in vivo, hypoxia/reoxygenation (H/R) was performed on H9c2 cells. As previously mentioned, hypoxia/reoxygenation (H/R)-injured H9c2 cells model was constructed [24,25]. H9c2 cells were cultured for 8 h in serum and sugar-free medium under hypoxic conditions in a three-atmosphere incubator with 94% N₂, 1% O₂ and 5% CO₂. After hypoxia, H9c2 cells were reoxygenated by exposing them to a mixture of 95% air and 5% CO₂ for 12 h. Subsequently, these cells were treated with the relevant drugs throughout the initial phase of reoxygenation.

2.6.1. Celluar uptake

H9c2 cells were hypoxia-treated for 8 h and then reoxygenated using the method described above. The H/R-injured H9c2 were co-cultured with free C6 and three C6-loaded liposomes (C6@L, C6@CHP-L, C6@TK/CHP-L, with a C6 equivalent dose of 50 ng C6/ml) at the onset of reoxygenation. After 1 h and 4 h of co-incubation, H9c2 cells were subjected to two rinses with PBS. Subsequently, cells were stained with DAPI for 10 min and stabilized with 4% formaldehyde for 10 min at room temperature. Finally, the cellular uptake was partially assessed using fluorescence microscopy. In the same way, the identical experimental approach was employed to co-cultivate free C6 and three C6-loaded liposomes (C6@L, C6@CHP-L, C6@TK/CHP-L) with normal H9c2 cells. This facilitated comparison with cellular uptake in H/R-injured H9c2 cells.

Additionally, cellular uptake was quantified using flow cytometry. Free C6, C6@L, C6@CHP-L and C6@TK/CHP-L were co-incubated with H/R-injured H9c2 cells using the above methods. After 1 h of drug incubation, H9c2 cells were harvested and washed. The fluorescence intensity of C6 in each group of cells was quantified using flow cytometry (BD LSR II, Biosciences). The same methodology was used for quantification for C6@TK/CHP-L in normal H9c2 cells (excluding H/R treatment).

2.6.2. Detection of ROS

ROS were detected by the 2′,7′-dichlorodihydrofluorescin diacetate (DCFH-DA) as described previously [26]. H9c2 cells were subjected to hypoxia in 96-well plates (1 × 10⁴ cells per well) for 8 h. The cells were subsequently co-incubated with free PUE, PUE@L, PUE@CHP-L, or PUE@TK/CHP-L (with a PUE equivalent dose of 20 μM) during the reoxygenation process. After a 12-h period of cell culture, 500 μl of serum-free medium containing 10 μM DCFH-DA was added to each well and incubated for 20 min, followed by rinsing the cells three-times with serum-free medium to remove residual DCFH-DA. Fluorescence images were generated using an inverted fluorescence microscope to capture intracellular ROS in a timely and direct manner. The fluorescence of DCFH-DA was quantified using Image J software. Increased fluorescence intensity corresponds to the elevated ROS levels.

2.6.3. Detection of mPTP & caspase-3 activity

According to the kit requirements (C2009S, Beyotime, China), Calcein AM loading/CoCl₂ quenching test was used to determine the opening of the mitochondrial permeability transition pore (mPTP) as previously described [27]. Following an 8 h period of hypoxia, H9c2 cells were co-incubated with the above drugs and oxygenation was restored. The cells were then incubated for an additional 12 h. Cells were digested by trypsin and collected by centrifugation, each tube was incubated with 1 ml of Calcein AM (1X) staining solution and 10 μl of CoCl₂ (1X) for 30 min, washed twice using PBS solution and cells were resuspended by the addition of 200 μl of assay buffer, while a negative control group was established. The mPTP opening of each group was analyzed using flow cytometry.

Following the guidelines of the caspase-3 activity test kit (BC3830, Solarbio), the H9c2 cells were subjected to H/R treatment and then incubated by introducing the assay solution. Cell samples were collected, ruptured at low temperature for 15 min and then centrifuged. The resulting absorbance was quantified at a wavelength of 405 nm.

2.6.4. Cell viability

The Cell Counting Kit-8 (CCK-8) method was used to evaluate cell viability. H9c2 cells were subjected to H/R treatment as described above and co-incubated with co-incubated with each group of drugs for 12 h. Then, 10% CCK-8 solution was added to each group of cells and incubated for 1 h, and then the absorbance value of each group of cells was measured at 450 nm to determine the relative cell viability.

2.6.5. Inhibition of apoptosis

The quantification of cell apoptosis was detected by Annexin V-FITC/PI double staining (BD Bioscience). Briefly, H9c2 cells were inoculated in 6-well plates at a density of 2 × 10⁵/well, then they were subjected to H/R treatment and incubated with each group of drugs described above. The supernatant was collected and digested with EDTA-free trypsin for 5 min, and the corresponding supernatant was added to terminate the digestion, and the cells were collected in centrifugation tubes and the cells were centrifuged at 1000 g/min for 5 min and the supernatant was discarded and the cells were washed two to three-times with PBS solution. Add 200 μl of buffer and transfer to 2 ml EP tubes, add 5 μl of Annexin V-FITC to each tube and mix well, incubate for 15 min away from light, and then add 5 μl of PI dye for apoptosis detection by flow cytometry.

2.6.6. Lipid ROS measurement

After hypoxia treatment of H9c2 cells in a 6-well dish (2 × 10⁵ cells/well) for 8 h, free PUE, PUE@L, PUE@CHP-L, or PUE@TK/CHP-L (PUE equivalent dose: 20 μM) were added at the beginning of reoxygenation. Lipid ROS were detected using the BODIPY™ 581/591 C11 fluorescent probe (Thermo Fisher, Catalog No. D3861) as previously described [28], and at the end of the 12 h incubation, cells were incubated for 30 min with 2 μM BODIPY C11 in the cell culture medium. Excess BODIPY C11 was then removed by washing the cells with PBS twice. Labeled cells were trypsinized and resuspended in PBS plus 2% FBS. Oxidation of BODIPY C11 resulted in a shift of the fluorescence emission peak from 590 nm to 510 nm, which was proportional to lipid ROS generation and was analyzed using a flow cytometer.

2.6.7. SOD & MDA level measurement

The H9c2 cells were subjected to H/R and drug treatments as above. The WST-8 Total Superoxide Dismutase Assay Kit (S0101S, Beyotime, China) was utilized to evaluate SOD activity. An enzyme marker was used to measure SOD activity in the test samples according to the protocol of the kit instructions. A malondialdehyde content test kit (S0131S, Beyotime, China) was utilized to measure the levels of MDA. Following the manufacturer's instructions, the working solution was prepared and the cells were treated. Subsequently, the absorbance of each sample was measured at wavelengths of 532 nm and 600 nm. The MDA content was then estimated according to the instructions method.

2.6.8. Quantification of Fe²⁺, Glutathione (GSH)

A hypoxia-reoxygenation model of the H9c2 cells was prepared in the same way and incubated with the above drugs. The intracellular Fe²⁺ content was detected using a FerroOrange fluorescent probe (F374, Dojindo) as described previously [29], and each well was incubated with 500 μl of 1 μM FerroOrange working solution for 30 min and observed under a fluorescence microscope.

GSH levels were detected by Reduced Glutathione Assay Kit (A006-1-1, jiancheng, Nanjing), the experimental operation was referred to the instruction manual, and the absorbance at 405 nm was detected by an enzyme marker.

2.6.9. western blot analysis

H9c2 cells via homogenization in RIPA buffer (R0010, Solarbio, China) containing protease inhibitors. In total, 20 μg protein was separated by 12% sodium dodecyl sulfate polyacrylamide gel (SDS-PAGE) and subsequently transferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore, MA, USA). and blocked with 5% Blotting Grade (P0216-300 g, Beyotime, China) for 2 h at 37°C. Then, the anti-GPX4 primary antibody (BM5231, Boster Biological Technology, Wuhan) was incubated with the strips overnight at 4°C, followed by incubation with the secondary antibody. Finally, the signal was detected using an enhanced chemiluminescence detection kit (AR1171, Boster Biological Technology, Wuhan). The signals were detected using an electrochemiluminescence ECL system. The WB bands were analyzed using ImageJ software.

2.7. Animals

C57BL/6J male mice (20–25 g) were provided by Hunan Slack Scene of Laboratory Animal Co., Ltd (Changsha, China). All animal experiments were approved by the Institutional Animal Care and Use Committee (lACUC), The Second Xiangya Hospital, Central South University, China.

2.8. Establishment & administration of MI/RI model in mice

C57BL/6J male mice weighing between 20 and 25 g were used in the study and divided into six groups of 10 mice each. The mice were kept in a controlled environment with a light-dark cycle of 12 h and a temperature of 25 ± 1°C. The previously published methods in the literature were used to develop the mice model of MI/RI [30]. Mice were anesthetized by intraperitoneal injection of sodium pentobarbital (60 mg/kg). Depth of anesthesia was detected by performing a tug reflex test on at least two toes, and anesthesia was confirmed to be adequate if no reflexes were present [19]. The mice were then mechanically ventilated at a respiratory rate of 110 breaths/min and a tidal volume of 1.8 ml/min. As reported in the classical approach, temporary blockade of the left anterior descending artery resulted in myocardial ischemia, and then the ligature was loosened to restore blood flow, which mimicked the state of myocardial ischemia-reperfusion in clinical patients [31]. The left anterior descending coronary artery of mice was ligated with a surgical knot using a No. 8-0 silk suture to achieve myocardial ischaemia, which was confirmed by the appearance of cyanosis in the region of ischemic myocardium [31]. After 30 min of ischemia, a hyperaemic response with gradual filling of the ischemic myocardial area was observed when the suture was loosened, confirming successful reperfusion, which lasted for 24 h. Approximately 5 min prior to reperfusion, free PUE, PUE@L, PUE@CHP-L and PUE@TK/CHP-L were given to MI/RI mice by tail vein injection at an administration dose of 15 mg/kg PUE.

2.8.1. Ex vivo biodistribution

Free DiR and DiR-labelled liposomes (DiR@L, DiR@CHP-L, DiR@TK/CHP-L, 30 μg DiR/ml) were injected into MI/RI mice via the tail vein. At 2, 4, 6, 8 and 12 h after injection, three mice in each group were taken to observe the fluorescence distribution of DiR in MI/RI mice using an in vivo imaging system (IVIS Spectrum, PerkinElmer). Subsequently, the mice were executed under anaesthesia and the heart, liver, spleen, lung and kidney were harvested to track the distribution of DiR in the isolated organs. These isolated organs were first rinsed with saline to remove the blood, and then gently squeezed with forceps to remove the blood hidden inside, before finally undergoing the in vivo imaging study of the isolated organs. The hearts of MI/RI mice were then divided into three parts and arranged vertically to show the cross-section. In addition, the same method was taken to detect the fluorescence intensity of major organs in some of the MI/RI mice 4 h after administration.

2.8.2. Determination of myocardial infarct size

The mice were subjected to euthanasia 24 h following reperfusion. The obtained hearts were frozen at a temperature of -80°C for a duration of 10 min. They were then cut into six equal slices and placed in a 2% TTC solution for a period of 30 min at a temperature of 37°C in the absence of light. 4% paraformaldehyde was used to fix the hearts. Infarcted myocardium is stained white, while normal myocardial tissue is stained red. Infarct size was calculated using Image J software as previously reported [32,33]. The calculation formula was as follows:

$$\text{Infarct size}(\%)=\left(\frac{\text{Infarct size}}{\text{Viable size}}\right)\times 100\%$$

2.8.3. H&E staining

The hearts were subjected to overnight fixation in a 4% paraformaldehyde solution with a pH of 7.4. Subsequently, they were embedded in paraffin and sectioned in a serial manner, with each section being 5 μm thick. The sections were then stained with Hematoxylin and Eosin (H&E) for periodic histological analysis using a microscope.

2.8.4. Tunel staining

Heart samples were collected 24 h after reperfusion, fixed with 4% paraformaldehyde and embedded in paraffin. DNA fragmentation was detected with the use of TUNEL Staining Kit (Abbkine Scientific, Wuhan, China) following the manufacturer's instructions. DAPI-staining was used for nuclear staining. Nuclear density and TUNEL-positive nuclei were counted by two blinded investigators examining the same fields. Three fields from seven sections were examined for each heart.

2.8.5. Dihydroethidium staining

The fixed myocardial tissues were cut into 5 μm sections, adhered to the slides, excess liquid was aspirated, dried at room temperature for 1 h, 20 μl of 5 μM dihydroethidium (DHE) working solution was added dropwise and incubated at 37°C for 30 min away from light, rinsed with PBS solution three-times for 5 min each time and covered with coverslips. After 3 washes with PBS, the sections were then washed three-times with PBS, sealed and photographed with a fluorescence microscope with an excitation wavelength of 490 nm and an emission wavelength of 520 nm.

2.8.6. GPX4 protein expression

The GPX4 level was detected by the method in reference [34,35]. Mice were executed under anesthesia, their hearts were removed and placed in 4% paraformaldehyde for 4 h. The tissues were dehydrated and processed, embedded in paraffin wax and cut into 5 μm slices, deparaffinized and hydrated at room temperature and rinsed three-times with PBS. Drops of 5% (v/v) normal goat serum sealing solution was added to seal the non-specific sites, incubated at 37% for 2 h, remove the sealing solution and add anti-mouse-Glutathione Peroxidase 4 (1:200) (ab125066, 1:200, Abcam), incubated at 4°C overnight, dropwise added anti-rabbit secondary antibody (1:500 dilution, Proteintech) at room temperature for 10 min, washed three-times with PBS solution, incubate with DAB coloring solution and then terminate the staining under microscope until the positivity is obviously strengthened. The results of immunohistochemistry were obtained using a microscope.

2.8.7. Mitochondrial morphology

The preparation process of mitochondrial sections was performed as previously described [36]. After 24 h of reperfusion, the anterior left ventricular tissue was taken and cut into small pieces of 1–2 mm³ and immediately fixed in electron microscope fixative for 48 h. These sections were then preserved using a solution of 2% glutaraldehyde in 0.1 m PBS, followed by fixation with 1% osmium tetroxide. Ultrathin sections were made by dehydrating them in ethanol and embedding them in Epon resin. These sections were then counterstained with uranyl acetate and lead citrate. Finally, the micromorphology of mitochondria in myocardial tissue was observed using transmission electron microscopy (TEM).

2.8.8. Biosafety evaluation

PUE, PUE@L, PUE@CHP-L and PUE@TK/CHP-L were injected into the MI/RI mice according to the above method, and then the mice were killed by cervical dislocation under anesthesia after 24 h of reperfusion, and the liver, spleen, lung and kidney were taken and fixed in 4% paraformaldehyde overnight, paraffin-embedded and stained with hematoxylin and eosin (H&E). The structure and morphology of the different tissues were observed under the optical microscope.

2.9. Statistical analysis

Experimental data are presented as mean ± standard deviation. Statistical analyses between groups were performed using Analysis of variance (ANOVA) test, with *p < 0.05 indicating significant differences and **p < 0.01 and ***p < 0.001 indicating highly significant differences.

3. Results

3.1. Synthesis & characterization of CHP-PEG-PE & PEG-TK-PE

The quantitative reaction between DSPE-PEG-NHS and CHP was conducted in equimolar ratios, as shown in Supplementary Figure S1A. The MALDI-TOF mass spectrometry revealed that the mean molecular weight of the resultant product CHP-PEG-PE was precisely equivalent to the combined molecular weights of CHP (with an average mass peak of 983 Da) and DSPE-PEG-NHS (with an average mass peak of 2600 Da). In addition, the ¹H NMR spectra of CHP-PEG-PE showed the chemical shift of CHP (δ: 7.5∼8.0 ppm), which further confirmed the successful reaction of CHP with DSPE-PEG-NHS to obtain the target product (Supplementary Figure S1B). The amino group of DSPE-NH₂ was esterified with the carboxyl group of PEG-TK-COOH to produce the amide-bonded product PEG-TK-PE. Supplementary Figure S1C showed the chemical shift of TK (δ: 1.5 ppm) and the characteristic peak of PEG (δ: 3.5 ppm) in the ¹H NMR spectrum of CHP-PEG-PE, confirming the successful synthesis of PEG-TK-PE.

3.2. Preparation & characterization of PUE@TK/CHP-L

In this study, we prepared PUE-loaded liposomes (PUE@L), CHP-modified PUE-loaded liposomes (PUE@CHP-L) and CHP and TK dual-modified PUE-loaded liposomes (PUE@TK/CHP-L). As shown in Table 1, the particle sizes of these three types of PUE-loaded liposomes ranged from 110 nm to 130 nm, with a gradual increasing trend. Among them, the particle size of PUE@TK/CHP-L was the largest, which could be attributed to the fact that the PEG-TK-PE and CHP-PEG-PE polymers on the surface of the liposomes could freely diffuse in aqueous solution, thus increasing the particle size of the liposomes. The literature suggested that nanoparticles with diameters of 20–200 nm were the optimal particle size for passive targeting of the injured left ventricle [37]. Therefore, the particle size of PUE@TK/CHP-L fell with this range and had the potential to penetrate and retain in the ischemic myocardial site via EPR effect. Interestingly, the zeta potential of PUE@TK/CHP-L also showed an increasing trend (Table 1), which may be attributed to the fact that the long-chain polymers of PEG-TK-PE and CHP-PEG-PE were entangled on the surface of the liposomes, thus weakening the negative potential of the liposomes. Figure 2A of Malvern laser particle sizer results also suggested that the particle size distribution and zeta potential of PUE@TK/CHP-L were well distributed, with a PDI of less than 0.2. Transmission electron microscopy (TEM) of Figure 2B also showed that the microscopic morphology of PUE@TK/CHP-L presented as a spherical structure with uniform size. The drug loading and encapsulation rates of PUE@TK/CHP-L as determined by high performance liquid chromatography (HPLC) were 5.80 ± 0.68% and 84.2 ± 9.6%, respectively (Table 1). Figure 2C displayed the encapsulated state of PUE in liposomes. The IR spectral characteristics of the blank TK/CHP-L and PUE@TK/CHP-L samples exhibited a relatively similar proximity, e.g. the infrared spectral peaks in the fingerprint region (1254.61, 1105.94, 952.17) of the two samples are basically the same. Moreover, the distinctive peaks associated with PUE (1515.34, 1397.40, 836.49 and 797.96) essentially disappeared in the PUE@TK/CHP-L. This observation implicitly suggested that the PUE had been fully encapsulated within the liposome core.

Figure 2.

Characterization of PUE@TK/CHP-L. (A) Determination of the particle size and zeta potential of PUE@TK/CHP-L by DLS. (B) TEM image of PUE@TK/CHP-L. (C) Infrared spectra of PUE, blank TK/CHP-L, PUE@TK/CHP-L. (a) PUE, (b) blank TK/CHP-L and (c) PUE@TK/CHP-L. (D) Release profile of PUE@TK/CHP-L in PBS and in the presence of 0.1 mM H₂O₂. Data are shown as mean ± SD (n = 3).

CHP: Cardiac homing peptide; DLS: Dynamic Light Scattering; L: Liposome; PBS: Phosphate buffer saline; PUE: Puerarin; SD: Standard deviation; TEM: Transmission electron microscopy; TK: Thioketal.

During myocardial ischemia-reperfusion injury, ischemic cardiomyocytes showed a trend of persistently elevated ROS levels [6], therefore, we investigated the release characteristics of PUE@TK/CHP-L in response to high ROS levels. The result in Figure 2D showed that PUE@TK/CHP-L gradually released PUE in PBS, reaching approximately 62.3 ± 9.8% after 24 h. However, when 0.1 mM of H₂O₂ was added to PBS, PUE@TK/CHP-L significantly increased the rate of drug release, up to full release within 4 h. These results indicated that PUE@TK/CHP-L could release PUE rapidly in response to high ROS levels. The current study found that ischemic myocardial sites exhibited microenvironments with rapidly increasing ROS levels [6], and the results of the in vitro release study showed that PUE@TK/CHP-L had ROS-responsive properties, so it was hypothesized that PUE@TK/CHP-L could responsively release the drug in ischemic cardiomyocytes.

3.3. Cellular uptake

C6 exhibited green fluorescence and is commonly used to show the distribution of drugs within the cells [38]. Therefore, we prepared C6-loaded liposomes to study cellular uptake process of liposomes in this study. As shown in Figure 3A & Supplementary Figure S2A, the fluorescence intensity of all liposomes in normal cardiomyocytes varied little, but all were significantly stronger than that of free C6. However, in H/R-injured H9c2 cells, the green fluorescence intensity of C6@CHP-L and C6@TK/CHP-L was higher than that of C6@L. The intracellular fluorescence intensity increased with incubation time from 1 to 4 h, suggesting that CHP-modified liposomes promote uptake by H/R-injured H9c2 cells in a time-dependent manner (Supplementary Figure S2B & C). Quantitative uptake results by flow cytometry also confirmed that CHP significantly promoted liposome uptake by H/R-injured H9c2 cells. As shown in Figure 3B & C, cellular uptake of C6@CHP-L and C6@TK/CHP-L by H/R-injured H9c2 cells was increased by 2.6-fold and 2.7-fold, respectively, compared with C6@L within 1 h. In addition, the findings also showed that the mean fluorescence intensity of C6@TK/CHP-L was 3.6-fold higher in H/R-injured H9c2 cells compared with normal H9c2 cells (Figure 3D & E). The result suggested that a larger amount of C6@TK/CHP-L were taken up by H/R-injured H9c2 cells. It may be attributed to the fact that receptors exposed on H/R-injured cardiomyocytes bind tightly to CHP modified on the surface of the liposomes, thereby facilitating cellular uptake of C6@TK/CHP-L [18]. In addition, H/R injury may induce an increase in cardiomyocyte membrane permeability [39], which also contributes to increased uptake of C6@TK/CHP-L by H/R-injured H9c2 cells. Based on the two aspects, it is possible to explain the significantly higher uptake of C6@TK/CHP-L in H/R-injured cardiomyocytes compared with that in normal cardiomyocytes.

Figure 3.

Normal H9c2 cells or H/R-injured H9c2 cells were incubated with free C6, C6@L, C6@CHP-L and C6@TK/CHP-L, and intracellular fluorescence intensity was measured by flow cytometry. (A) Fluorescence Images of free C6, C6@L, C6@CHP-L and C6@TK/CHP-L uptake by normal H9c2 cells and H/R-injured H9c2 cells at 1 h (Scale bar: 100 μm). (B & C) Flow cytometry quantification of cellular uptake by H/R-injured H9c2 cells. (D & E) Flow cytometric quantification of C6@TK/CHP-L uptake by normal H9c2 and H/R-injured H9c2 cells. Statistical analysis were performed by one-way ANOVA. Data are shown as mean ± SD (n = 3). *p < 0.05.

ANOVA: Analysis of variance; C6: Coumarin-6; CHP: Cardiac homing peptide; DAPI: 6-Diamidino-2-phenylindole; H/R: Hypoxia/reoxygenation; L: Liposome; NS: Not significant; SD: Standard deviation; TK: Thioketal.

3.4. PUE@TK/CHP-L inhibited apoptosis in H/R-treated H9c2 cells

During the reperfusion phase, the oxidative phosphorylation process associated with aerobic respiration in the mitochondrial electron transport chain generates a large amount of ROS [6]. We used a DCFH-DA fluorescent probe to observe intracellular ROS levels and found that the intensity of green fluorescence in H/R-treated H9c2 cells was significantly higher than that in normal H9c2 cells (Figure 4A). The PUE@TK/CHP-L group can significantly reduce the level of ROS in H/R-injured H9c2 cells, which is significantly better than other drug-treated groups (Figure 4B). These rapidly rising ROS not only directly impair cardiomyocyte injury, but also trigger the opening of mPTP. This leads to the release of Cyt c into the cytoplasm and activation of caspase-3, which induces cardiomyocyte apoptosis [40]. In this study, the inhibition of mPTP opening in H/R-treated H9c2 cells by PUE@TK/CHP-L was investigated using mitochondrial Calcein-AM assay. As shown in Figure 4C, intra-mitochondrial fluorescence was significantly reduced in the H/R group, indicating that mPTP on the mitochondria of H/R-treated H9c2 cells was in an open state. However, PUE@TK/CHP-L, PUE@CHP-L and PUE@L could enhance the fluorescence intensity of the H9c2 cells. Among them, the highest fluorescence intensity was observed in mitochondria after PUE@TK/CHP-L treatment, indicating that PUE@TK/CHP-L significantly inhibited the opening of mPTP. We further assayed the intracellular caspase-3 activity, and according to Figure 4D, H/R-injured H9c2 cells treated with PUE@TK/CHP-L significantly reduced caspase-3 activity, which was also significantly better than the other groups. Based on the above apoptotic mechanism studies, PUE@TK/CHP-L significantly reduced apoptosis and enhanced cardiomyocyte viability in H/R-injured cardiomyocytes. We evaluated the viability of H9c2 cells using the CCK-8 assay. As depicted in Figure 4E, compared with PUE@L and free PUE, both PUE@TK/CHP-L and PUE@CHP-L significantly increased the viability of H/R-injured H9c2 cells. It was also found that PUE@TK/CHP-L was superior to PUE@CHP-L in protecting H/R-injured cardiomyocytes. The results of apoptosis rate detected by flow cytometry similarly revealed that the PUE@TK/CHP-L group inhibited apoptosis of H/R-injured H9c2 cells significantly more effectively than the other groups (Figure 4F & G). Specifically, the rate of cardiomyocyte apoptosis in the PUE@TK/CHP-L group was 7.91 ± 0.23%, which was significantly lower than that in the PUE@CHP-L group (Figure 3G). This may be related to the fact that CHP promotes liposome uptake by H/R damaged H9c2 cells, and subsequently TK breaks chemical bonds in the presence of high intracellular ROS levels, inducing liposome disassembly to release PUE. Based on the above studies, PUE@TK/CHP-L significantly reduced the intracellular ROS level, inhibited the opening of mPTP and lowered caspase-3 activity, thereby inhibiting cardiomyocyte apoptosis. This may be related to the fact that CHP promotes liposome uptake by H/R injured H9c2 cells, and subsequently TK breaks chemical bonds in the presence of high intracellular ROS levels, inducing liposome disassembly to release PUE.

Figure 4.

Inhibition of apoptosis in H/R-injured H9c2 cells by PUE, PUE@L, PUE@CHP-L and PUE@TK/CHP-L. (A & B) Typical fluorescence images and quantitative fluorescence intensity analysis of DCFH-DA staining in H/R-injured H9c2 cells (Scale bar: 100 μm). (C) Quantitative fluorescence intensity analysis of Calcein AM in H/R-injured H9c2 cells. (D) Detection of caspase-3 activity in H/R-injured H9c2 cells. (E) Cell viability of H/R-injured H9c2 cells in each group. (F) Cardiomyocyte apoptosis in each group was analyzed by Annexin V-FITC/PI staining. (G) Statistical analysis of apoptotic cells in H/R-injured H9c2 cells by flow cytometry. Statistical analysis were performed by one-way ANOVA. Data are shown as mean ± SD (n = 3). *p < 0.05.

ANOVA: Analysis of variance; Calcein AM: Calcein acetoxymethyl ester; DCFH-DA: 2′,7′-Dichlorodihydrofluorescein diacetate; H/R: Hypoxia/reoxygenation; PI: Propidium Iodide; SD: Standard deviation.

3.5. PUE@TK/CHP-L inhibited ferroptosis in H/R-treated H9c2 cells

Ferroptosis is an iron-dependent form of programmed cell death that could be activated by iron accumulation or GPX4 inactivation, ultimately leading to the accumulation of lipid peroxides to lethal levels, resulting in cardiomyocyte death [41]. Thus, we examined intracellular levels of Fe²⁺ and oxidative stress to assess the role of PUE@TK/CHP-L in inhibiting ferroptosis in H/R-treated cardiomyocytes. As shown in Figure 5A, the fluorescence intensity level of Fe²⁺ in H/R-treated H9c2 cells was significantly increased compared with the normal group. In contrast, the degree of excess iron in H/R-treated cardiomyocytes showed a decreasing trend after drug treatment in all groups, with the level of Fe²⁺ in cardiomyocytes of the PUE@TK/CHP-L group being significantly lower than that of the other groups. This indicated that PUE@TK/CHP-L could effectively remove Fe²⁺ from the cells. Excessive accumulation of intracellular iron triggers the Fenton reaction and lipid peroxidation [31]. As shown in Figure 5B & C, intracellular lipid ROS and malondialdehyde (MDA) levels were higher in the H/R group than in the normal group. After drug treatment, intracellular lipid ROS and MDA levels in H/R-injured cardiomyocytes showed a decreasing trend. Among them, lipid ROS and MDA levels were lowest after PUE@TK/CHP-L treatment. Furthermore, the PUE@TK/CHP-L group significantly reduced oxidative stress. Glutathione (GSH) served as a cofactor in the process of converting peroxides into alcohols via the action of GPX4. The deficit of GSH initiated a shortage of cysteine, which subsequently resulted in the deactivation of GPX4 [41]. The results shown in Figure 5D & E indicated that the levels of GSH and SOD in cardiomyocytes were much lower than normal after H/R injury, whereas PUE@TK/CHP-L significantly reversed these changes in H/R-injured H9c2 cells, with a better effect than all other groups. GPX4 uses reduced GSH to attenuate lipid peroxidation and inhibit ferroptosis by converting lipid hydroperoxides to lipids and alcohols [42], therefore, increased GPX4 expression could inhibit ferroptosis in H/R-injured cardiomyocytes. As shown in Figure 5F & G, both PUE@TK/CHP-L and PUE@CHP-L significantly increased GPX4 expression, especially PUE@TK/CHP-L was better than PUE@CHP-L in enhancing GPX4, and the difference was significant. Based on these findings, PUE@TK/CHP-L could scavenge excess iron ions, reduce lipid ROS and MDA levels and increase GSH, SOD and GPX4 levels, thereby inhibiting ferroptosis in H/R-injured cardiomyocytes.

Figure 5.

Inhibition of ferroptosis in H/R-injured H9c2 cells by PUE, PUE@L, PUE@CHP-L and PUE@TK/CHP-L. (A) Fluorescence images and fluorescence intensity analysis of Fe²⁺ staining in H/R-injured H9c2 cells (Scale bar: 100 μm). (B) Flow cytometric quantification of C11-BODIPY fluorescence intensity in H/R-injured H9c2 cells. (C) The MDA levels in H/R-injured H9c2 cells. (D) The GSH content in H/R-injured H9c2 cells. (E) The SOD levels in H/R-injured H9c2 cells. (F) western blot analysis of GPX4 expression levels in H/R-injured H9c2 cells. (G) Quantitative analysis of GPX4 expression levels in H/R-injured H9c2 cells. Statistical analysis were performed by one-way ANOVA. Data are shown as mean ± SD (n = 3). ***p < 0.001; **p < 0.01; *p < 0.05.

ANOVA: Analysis of variance; CHP: Cardiac homing peptide; GSH: Glutathione; H/R: Hypoxia/reoxygenation; L: Liposome; MDA: Malondialdehyde; PUE: Puerarin; SD: Standard deviation; SOD: Super oxide dismutase; TK: Thioketal.

3.6. Myocardial targeting of PUE@TK/CHP-L

According to the methods reported in the literature, we constructed MI/RI mice to study the in vivo targeting behavior of CHP and TK bimodified liposomes [30]. A series of DiR-loaded liposomes were prepared using fluorescent DiR instead of PUE and then injected into MI/RI mice via tail vein. In vivo fluorescence imaging of MI/RI mice was performed at different time points from 2 h to 12 h. The result shown in Supplementary Figure S3 indicated that the free DiR group and DiR@L group had minimal targeting to the myocardium, with the majority in the liver. In contrast, the fluorescence of DiR@CHP-L and DiR@TK/CHP-L in the cardiac region was significantly enhanced with time. In both groups, a clear fluorescent signal in the cardiac region was seen at 4 h after injection administration, which persisted until 12 h of administration. The results of 4 h ex vivo organ fluorescence assay also showed that the cardiac fluorescence intensity of DiR@CHP-L and DiR@TK/CHP-L were significantly stronger than that of other groups (Supplementary Figure S4). It was very meaningful that the area of strong fluorescent signals of DiR@TK/CHP-L aggregated at the heart site was larger than that of DiR@CHP-L after 4 h of drug treatment. The reason may be the dual combined effect of CHP and TK. In the early stage of reperfusion, CHP targeted binding to receptors on the membrane of ischemic cardiomyocytes induced the entry of DiR@TK/CHP-L into cardiomyocytes, and then TK induced liposomal disassembly in response to high intracellular concentration of ROS, which facilitates the intracellular distribution of DiR. In vivo imaging showed that DiR@TK/CHP-L aggregated in the cardiac myocardial region rapidly. It was hypothesized that PUE@TK/CHP-L could rapidly exert its efficacy at the early stage of reperfusion, thus alleviating the critical symptoms of myocardial infarction. In addition, CHP targeted to ischemic cardiomyocytes to enhance the retention of liposomes in the ischemic myocardium, and strong fluorescence intensity was still observed in isolated hearts 12 h after CHP-modified liposomes were administered. The 12 h main tissue organ fluorescence results in Figure 6A & B also showed that DiR@CHP-L and DiR@TK/CHP-L accumulated more in myocardial tissue compared with DiR@L. Each group of hearts was divided into three segments flatly from bottom to top for fluorescence imaging, and the results are shown in Figure 6C & D. DiR@CHP-L and DiR@TK/CHP-L accumulated predominantly in the ischemic region of the ligature, suggesting that CHP promotes liposome targeting to the ischemic myocardium.

Figure 6.

In vivo fluorescence imaging of various organs in MI/RI mice. (A) Ex vivo fluorescence images of major organs obtained from each group were taken 12 h post-injection. (B) Quantification of radiant efficiency of the organs in each group at 12 h after injection. (C) Fluorescence Distribution of DiR and DiR-labeled liposomes in the hearts of MI/RI mice after 12 h of administration. (D) Ex vivo optical imaging of cardiac cross-sections from base to apex in each group after 12 h of administration. Statistical analysis were performed by one-way ANOVA. Data are shown as mean ± SD (n = 3). *p < 0.05.

ANOVA: Analysis of variance; CHP: Cardiac homing peptide; DiR: 1,1′-Dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide; L: Liposome; MI/RI: Myocardial ischemia/reperfusion; ROI: Quantitative region of interest; SD: Standard deviation; TK: Thioketal.

3.7. PUE@TK/CHP-L reduced myocardial infarction size & ischemic myocardial damage

Measurement of myocardial infarct size is the gold standard for assessing the extent of myocardial infarction, which is usually assessed by staining ischemic myocardium with 2% TTC solution [43]. As indicated in Figure 7A & B, The TTC staining study showed that the mean infarct size in I/R group was 32.86 ± 0.81%. However, myocardial infarct area was substantially reduced to 22.48 ± 1.32% and 18.34 ± 0.82% after treatment with PUE@CHP-L and PUE@TK/CHP-L, respectively. Among them, the reduction of myocardial infarction area in the PUE@TK/CHP-L group was significantly better than that in the other drug groups, indicating that PUE@TK/CHP-L could significantly reduce myocardial infarction area.

Figure 7.

Preliminary evaluation of safety and efficacy of PUE@TK/CHP-L in MI/RI mice. (A & B) 2% TTC staining and quantitative analysis of myocardial infarct area in MI/RI mice heart. (C) H&E staining of myocardial tissue in MI/RI mice (Scale bar: 100 μm). Statistical analysis were performed by one-way ANOVA. Data are shown as mean ± SD (n = 3). **p < 0.01; *p < 0.05.

ANOVA: Analysis of variance; CHP: Cardiac homing peptide; H&E: Hematoxylin and eosin; L: Liposome; MI/RI: Myocardial ischemia/reperfusion; PUE: Puerarin; TK: Thioketal; TTC: 2,3,5-Triphenyl tetrazolium chloride; SD: Standard deviation.

MI/RI is often accompanied by some pathological changes of ischemic myocardium, such as myocardial structural disruption and disordered arrangement of myocardial fibers. The results of pathological changes in myocardial injured tissues assessed by H&E staining were shown in Figure 7C. Myocardial fibers with I/R injury exhibited significant misalignment, gap enlargement and edema compared with the normal group. In contrast, I/R-injured myocardium showed varying degrees of improvement and injury reduction after treatment with free PUE and a series of PUE-loaded liposomes. In particular, PUE@TK/CHP-L significantly attenuated the pathological symptoms of ischemic myocardium, such as disturbed myocyte arrangement and disordered arrangement of myocardial fibers. Overall, the improvement of ischemic myocardium in the PUE@TK/CHP-L group was better than that in the other drug groups, which may be related to the combined effect of TK and CHP, which promoted the responsive release of a large amount of drugs at the site of ischemic myocardium.

3.8. Inhibition of apoptosis & ferroptosis effect of PUE@TK/CHP-L

During the period of MI/RI, the burst of intracellular ROS in ischemic myocardium induces a large number of apoptosis and ferroptosis, and these programmed cell death patterns also cause a continuous rise in intracellular ROS levels, creating a vicious cycle that further exacerbates ischemic myocardial injury [44]. Therefore, it is valuable to study the effect of PUE@TK/CHP-L in inhibiting apoptosis and ferroptosis. Ischemic myocardial tissue was stained using a tunel staining assay kit to quantify the rate of cardiomyocyte apoptosis. As shown in Figure 8A & B, a large number of cardiomyocytes in the I/R group showed apoptosis, and the rate of TUNEL-positive cells decreased to varying degrees in all groups, especially in the PUE@TK/CHP-L group with an apoptosis rate of 7.50 ± 0.75%, which was significantly more effective than the other drugs in inhibiting cardiomyocyte apoptosis. DHE staining results also showed that PUE@TK/CHP-L had a better inhibitory effect on ROS levels in ischemic myocardial tissues. As shown in Figure 8C & D, excessive ROS generation was observed in the I/R group after 24 h of reperfusion. Quantitative analysis of DHE fluorescence in myocardial tissue showed 36.60 ± 2.85 in I/R group, whereas the PUE@TK/CHP-L group reduced fluorescence intensity to 12.4 ± 0.8, with the lowest DHE signal intensity in all drug-treated groups. This may be related to the fact that CHP promotes cardiomyocyte uptake, TK immediately releases PUE in an intracellular microenvironment of high ROS levels and sufficient amount of PUE scavenge the excessive intracellular ROS, thereby minimizing ROS levels.

Figure 8.

Inhibition of apoptosis and ferroptosis by PUE@TK/CHP-L in MI/RI mice. (A & B) Typical TUNEL staining of myocardial tissues after 24 h of reperfusion and its statistical analysis (Scale bar: 200 μm). (C & D) DHE staining of myocardial tissues after 24 h of reperfusion and its statistical analysis (Scale bar: 200 μm). (E) Transmission electron microscopy images of mitochondrial morphology of cardiomyocytes in MI/RI mice (Mitochondrial membrane rupture appearing as severe damage is indicated by yellow arrows) (Scale bar 1 μm). (F & G) Typical GPX4 immunofluorescence staining of myocardial tissues after 24 h of reperfusion and its statistical analysis (Scale bar: 200 μm). Statistical analysis were performed by one-way ANOVA. Data are shown as mean ± SD (n = 3). ***p < 0.001; **p < 0.01; *p < 0.05.

ANOVA: Analysis of variance; CHP: Cardiac homing peptide; DHE: Dihydroethidium; GPX4: Glutathione peroxidase 4; L: Liposome; MI/RI: Myocardial ischemia/reperfusion; PUE: Puerarin; SD: Standard deviation; TK: Thioketal; TUNEL: Terminal oxynucleotidyl transferase mediated dUTP biotin nick end labeling.

Unlike apoptosis, ferroptosis led to specific alterations in the structure of cardiomyocytes, including rupture of the cell membrane, mitochondrial atrophy, reduction or even disappearance of mitochondrial cristae and increased mitochondrial membrane density [45]. Therefore, we observed the electron microscopic morphological changes of mitochondria in myocardial tissues to investigate the effect of PUE@TK/CHP-L in inhibiting ferroptosis [44]. Compared with the sham group, I/R group showed signs of mitochondrial atrophy, mitochondrial membrane and reduction of mitochondrial cristae (Figure 8E). However, PUE@TK/CHP-L treatment significantly improved mitochondrial micromorphology, such as a more intact mitochondrial outer membrane and a significant increase in the number of mitochondrial cristae, suggesting that PUE@TK/CHP-L had a better repairing effect on ischemically injured myocardial mitochondria. A previous study demonstrated a significant downregulation of glutathione metabolic pathways, particularly GPX4, which protected cells from the metabolic pathway of ferroptosis in MI/RI mice [46]. Similarly, in the present study, immunohistochemical analysis of the myocardium of MI/RI mice showed a significant decrease in GPX4 expression, as shown in Figure 8F & G. According to the immunohistochemical data, the PUE@TK/CHP-L group significantly increased the activity of GPX4, which was significantly better than the other groups. In conclusion, PUE@TK/CHP-L could reduce the ROS level in myocardial tissue, decrease the degree of mitochondrial damage, and increase the activity of GPX4, thereby inhibiting ferroptosis.

3.9. Biosafety evaluation

Due to the clinical risk of acute intravascular haemolysis of PUE injections [47,48], perhaps with damaging effects on the vascular-rich liver, spleen, kidney and lung, we used H&E staining method to evaluate the biosafety of PUE@TK/CHP-L. As shown in Supplementary Figure S5, no significant pathological changes were observed in the liver, spleen, lung and kidney. Overall, no signs of inflammatory response were observed in the experimental group as compared with the sham group. Therefore, the liposomes prepared in this study were not significantly toxic to experimental animals. The result indicated that PUE@TK/CHP-L did not cause pathological changes to the tissues of the major organs and have good prospects for application.

4. Discussion

Although percutaneous coronary intervention could largely reduce mortality in patients with AMI, post-procedural complications and mortality remain high, mainly because MI/RI is still not well resolved [2]. More and more studies showed that oxidative stress could induce apoptosis and ferroptosis in cardiomyocytes, thus aggravating MI/RI [49]. It has been found that ischemic cardiomyocytes produce large amounts of ROS, which cause oxidative stress when they exceed the scavenging capacity of antioxidant systems such as glutathione, superoxide dismutase and catalase [50]. Excess ROS induces the opening of mPTP, leading to increased mitochondrial permeability and Cyt c release into the cytosol, triggering the caspase cascade reaction and inducing cardiomyocyte apoptosis. In addition, there is excessive iron deposition in the ischemic myocyte, and the excess of Fe²⁺ generates free radicals through the Fenton reaction, which induces lipid peroxidation. Meanwhile, the inactivation of GPX4, the main inhibitor of lipid peroxidation, further triggers the continuous elevation of intracellular ROS levels and induces ferroptosis. Therefore, timely inhibition of bursting ROS levels in ischemic cardiomyocytes and alleviation of oxidative stress is thought to greatly reduce cardiomyocyte apoptosis and ferroptosis, thereby attenuating MI/RI [32,51,52]. Several pharmacological studies have shown that PUE has excellent antioxidant capacity and could alleviate oxidative stress in ischemic myocardium [53,54]. Therefore, PUE was loaded into CHP and TK dual-modified liposomes, which efficiently delivered the drug to ischemic cardiomyocytes and released the drug in response to high ROS levels. The cellular uptake results indicated that CHP contributed to the uptake of CHP-modified liposomes by H/R-treated H9c2 cells. In vitro responsive release studies confirmed that TK induced rapid release of PUE from PUE@TK/CHP-L in 0.1 mM H₂O₂. Based on the combined effects of CHP and TK, PUE@TK/CHP-L can efficiently target cardiomyocytes and release drugs in response to high intracellular ROS levels, fully utilizing the anti-oxidative stress effects of the drug. At the cellular pharmacodynamic level, PUE@TK/CHP-L significantly reduced intracellular ROS levels, inhibited the opening of mPTP, decreased caspase-3 activity and reduced cardiomyocyte apoptosis. In addition, PUE@TK/CHP-L reduced intracellular iron content and MDA levels, increased GSH, SOD and GPX4 activity and contributed to the conversion of cytotoxic lipid peroxides to fatty alcohols, thereby inhibiting the occurrence of cellular ferroptosis. At the animal pharmacodynamic level, PUE@TK/CHP-L significantly reduced ROS levels in ischemic myocardial tissues, decreased the rate of apoptosis in cardiomyocytes, increased GPX4 levels and reduced the degree of mitochondrial damage, thereby largely reducing myocardial infarction size. In this study, the inhibitory effects of PUE@TK/CHP-L on cardiomyocyte apoptosis and ferroptosis were significantly superior to those of PUE@CHP-L and PUE@L-L and PUE@L at both cellular and animal levels. It was inferred that the combined effects of CHP and TK promoted the enrichment of PUE in ischemic cardiomyocytes and ROS-responsive release, thus largely enhancing the efficacy of the drug. In vivo imaging also confirmed that PUE@TK/CHP-L could target the ischemic myocardial site and maintain strong fluorescence intensity in the ischemic myocardium 12 h after injection. This result indicated that PUE@TK/CHP-L could sustain full efficacy at the site of the lesion. In addition, the results of in vivo imaging revealed that liposomes were more accumulated in the liver and spleen, which was due to the fact that conventional liposomes were easily taken up by the hepatic and splenic RES system [55] and thus the toxicity of liposomes to major organs needs to be considered. H&E staining results in the present study showed that no significant damage was observed on the tissue sections of the liver, spleen, lung and kidney in the PUE@TK/CHP-L group, which was basically in line with that of the sham group. This indicates that PUE@TK/CHP-L have no obvious toxicity to major organs and have a good safety profile.

Considering that conventional nanocarriers are more intercepted at the liver site, we will subsequently use platelet membranes fused with liposomes to construct novel biomimetic liposomes. On the one hand, the endogenous platelet membrane could reduce the phagocytosis of liposomes by hepatic and spleen RES system due to its excellent biocompatibility. On the other hand, the platelet membrane has the natural property of tending to the injured ischemic myocardial vessels [30,56]. Based on the two advantages of platelet membrane bionic liposomes, more drugs could be efficiently delivered to the ischemic myocardial lesion site, thus having the potential to significantly improve drug efficacy to a greater extent.

5. Conclusion

In conclusion, we developed PUE-loaded liposomes dual-modified with TK and CHP, which efficiently target drugs into ischemic cardiomyocytes and release drugs at high intracellular concentrations of ROS. Both in vitro cellular and in vivo animal experiments confirmed that PUE@TK/CHP-L had a favorable inhibitory effect on cardiomyocyte apoptosis and ferroptosis. In H/R-injured cells, PUE@TK/CHP-L could inhibit oxidative stress effectively, thereby reducing mitochondria-mediated apoptosis. In addition, PUE@TK/CHP-L scavenged excess Fe²⁺ and prevented the Fenton reaction from generating lipid ROS, thereby reducing iron depletion. It also increased GSH and GPX4 levels and prevented lipid peroxidation. In the MI/RI mouse model, PUE@TK/CHP-L exhibited excellent targeting and biosafety, significantly reducing infarct size, which may be related to the efficient delivery of PUE by PUE@TK/CHP-L to scavenge excessive ROS in myocardial tissues, repair the mitochondria of the damaged myocardium as well as enhance the expression of GPX4. In conclusion, PUE@TK/CHP-L are expected to accurately treat MI/RI and have good potential for application in this area.

Funding Statement

This research work is supported by the National Natural Science Foundation of China (81673614), Hunan Provincial Natural Scientific Foundation (No. 2024JJ8126, 2020JJ4128), Changsha Natural Science Foundation (kq2403088), Health Research Project of Hunan Provincial Health Commission (No. W20243065), Teaching Reform Project for Postgraduate Education at Central South University (No. 2024JGB158) and the Fundamental Research Funds for the Central Universities of Central South University (No. 2024ZZTS0995).

Supplemental material

Supplemental data for this article can be accessed at https://doi.org/10.1080/17435889.2024.2402678

Author contributions

Y Wang and S Li: conceptualization, investigation, writing-original draft. W Li: investigation, writing-original draft. J Wu: writing-original draft, data curation. X Hu: investigation, writing-review & editing. T Tang: investigation, writing-review & editing. X Liu: conceptualization, methodology, validation, formal analysis, investigation, resources, data curation, writing-original draft, writing-review & editing, visualization, supervision, funding acquisition.

Financial disclosure

This research work is supported by the National Natural Science Foundation of China (81673614), Hunan Provincial Natural Scientific Foundation (No. 2024JJ8126, 2020JJ4128), Changsha Natural Science Foundation (kq2403088), Health Research Project of Hunan Provincial Health Commission (No. W20243065), Teaching Reform Project for Postgraduate Education at Central South University (No. 2024JGB158) and the Fundamental Research Funds for the Central Universities of Central South University (No. 2024ZZTS0995). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Ethical conduct of research

All animal experiments were approved by and performed in accordance with the Laboratory Animal Ethics Committee of Xiangya Second Hospital of Central South University (Approval No. 20230892).

Data availability statement

Data will be made available on request.