Macrophage Membrane-Camouflaged Mesoporous Polydopamine Nanoparticles for Targeted Delivery of Apigenin in Sepsis-Induced Acute Lung Injury

Abstract

Introduction

Sepsis-induced acute lung injury (ALI) is a critical illness that poses a serious threat to human health, and effective treatments remain lacking. This study aimed to develop a macrophage membrane-biomimetic nano-delivery system loaded with mesoporous polydopamine (MPDA)-encapsulated apigenin (API) and to investigate its therapeutic potential in ALI.

Methods

In this study, we developed a macrophage membrane-coated biomimetic nano-delivery system, designated MEM@MPDA/API nanoparticles (NPs). We first systematically characterized the physicochemical properties of MEM@MPDA/API, and then evaluated its cellular uptake, reactive oxygen species (ROS) scavenging capacity, and mitochondrial function restoration through in vitro experiments. In vivo, after constructing an ALI mouse model, we assessed its active targeting to lung tissue, anti-inflammatory efficacy, pulmonary barrier repair capacity, and tissue safety following tail vein administration.

Results

In vivo results demonstrated that MEM@MPDA/API NPs effectively scavenged ROS, reduced inflammatory cytokine levels, and restored mitochondrial function. In the ALI mouse model, these NPs efficiently accumulated in the injured lung regions, alleviated pulmonary histopathological damage, and repaired the alveolar epithelial barrier. Integrated single-cell transcriptomics analysis and network pharmacology studies revealed that mitophagy-related pathways were activated during the progression of ALI.

Conclusion

This study proposes a promising biomimetic nano-delivery strategy, providing a safe and efficient potential new approach for the application of biomimetic nanomedicine in the treatment of ALI and oxidative stress-related critical diseases.

Introduction

As a globally recognized respiratory disease, ALI has an acute onset and a prolonged course, with damage to the alveolar epithelial and pulmonary capillary endothelial barriers and gas exchange dysfunction as its core pathological features, and can rapidly progress to acute respiratory distress syndrome (ARDS), posing a serious threat to patients’ lives.A study involving 145 PICU patients across 27 countries reported an incidence of ALI/ARDS of 17% and a mortality rate of 3.2%.¹ Another study covering 459 ICU patients from 50 countries found that 10.4% of ICU admissions were diagnosed with ARDS, among which the incidence of severe ARDS was 23.4%, with a corresponding mortality rate approaching half (46.1%).² Moreover, the annual incidence and mortality of ALI increase with age, placing a heavy burden on global public health.³ Currently, clinical management of ALI remains largely limited to supportive measures such as treatment of the underlying disease, protective lung ventilation, and mesenchymal stem cell therapy, lacking specific therapeutic interventions. Although drugs including glucocorticoids, pulmonary surfactants, and antioxidants have been tried to modulate inflammation and oxidative stress, their clinical utility is constrained by significant systemic side effects, poor targeting, and limited efficacy, making it difficult to effectively reduce patient mortality.^4–6 Therefore, the development of safe and effective novel drugs and therapeutic strategies for ALI/ARDS represents an urgent challenge in the prevention and treatment of respiratory diseases.

In the pathological progression of ALI, alveolar macrophages, vascular endothelial cells, and alveolar epithelial cells play distinct yet interrelated roles. Macrophages, as important immune cells in the body, can initiate inflammatory cascades and are also the primary clearance cells for NPs.⁷ Endothelial cells regulate vascular permeability and neutrophil extravasation, participating in the early formation of pulmonary edema. Alveolar epithelial cells—especially type II epithelial cells—are the ultimate executors of pulmonary barrier function. Studies have reported that during ALI/ARDS, pulmonary epithelial injury often precedes vascular endothelial injury and is a key factor in the development and progression of ALI, with the integrity of their tight junctions directly determining the recovery of gas exchange and alveolar fluid clearance.⁸ Although modulating macrophage and endothelial cell functions is crucial for suppressing inflammatory amplification, the core therapeutic goal of this study is to repair the pulmonary barrier and alleviate inflammatory damage; thus, we focus on pulmonary epithelial cells.

Oxidative stress and excessive inflammatory responses are the core pathological mechanisms underlying the initiation and progression of ALI.^4,9 Excessive production of ROS in the lungs induced by pathogenic factors, triggers lipid peroxidation and mitochondrial dysfunction,¹⁰ while simultaneously activating inflammatory signaling pathways such as NLRP3 inflammasome and NF-κB signaling pathways,^11–13 forming a vicious cycle of “oxidative stress–inflammatory storm”, which further disrupts pulmonary barrier integrity and exacerbates pulmonary edema and inflammatory cell infiltration. Therefore, the key to treating ALI lies in the ability to precisely and actively target inflamed lung sites to scavenge ROS, reduce oxidative stress, and inhibit the activation of inflammatory signaling pathways to alleviate inflammatory responses.

API is a flavonoid compound abundant in vegetables, fruits, and medicinal plants, exhibiting a broad spectrum of biological activities.^14,15 Numerous in vitro and animal model studies have demonstrated the protective effects of API in various inflammation- and oxidative stress-related diseases. It exerts anti-inflammatory and antioxidant effects in neurodegenerative diseases, protecting neurons; and in animal models of inflammation, it promotes macrophage polarization toward the M2 phenotype to exert anti-inflammatory effects.^16,17 API can exert anti-inflammatory responses by inhibiting the expression of NF-κB and transcription factors, and it exhibits antioxidant activity by scavenging free radicals and protecting cells from oxidative stress. Moreover, it is safe, with no toxicity cases observed even at high doses,¹⁸ making API a potential therapeutic agent for ALI and oxidative stress-related diseases. However, the clinical application of API is limited by its low oral bioavailability and poor solubility.¹⁹

The use of emerging nanotechnology to construct drug carriers for ALI treatment has become a promising therapeutic strategy to overcome these limitations,⁸ examples include novel nano-delivery systems based on nanomicelles loaded with the traditional drug berberine²⁰ and biomimetic metal-organic framework nanoplatforms.²¹ MPDA, as a novel multifunctional nanomaterial, integrates the broad-spectrum ROS scavenging ability of polydopamine with the high drug-loading capacity of mesoporous materials, offering favorable biocompatibility and ease of surface modification,²² making it an ideal drug carrier widely applied in various fields.^23–26 Wang et al designed and synthesized a multifunctional injectable nanocomposite hydrogel system (mPDA@Mel-AB NPs/OCS-CMC/BG), providing a new approach for the treatment of osteoarthritis.²⁷ In the study by Zhu et al, composite NPs (D/Ce@mPDA-C/P) using MPDA as the carrier significantly reduced oxidative stress and modulated the inflammatory microenvironment, offering a promising therapeutic strategy for myocardial ischemia-reperfusion injury (MIRI).²⁸

Cell membrane biomimetic modification technology represents an emerging trend in nano-drug delivery systems, capable of overcoming the limitations of conventional nanomedicine. NPs coated with natural cell membranes acquire diverse complex biological functions derived from donor cells. During the pathological progression of ALI, damaged pulmonary epithelium significantly upregulates chemokines (eg, MCP-1, CCL2)²⁹ and adhesion molecules (VCAM-1, ICAM-1),³⁰ while the macrophage membrane is rich in inflammation-targeting receptors (such as CCR2 and CCR5)³¹ and integrins (α4 and β1 integrins), and possesses strong phenotypic plasticity. Through receptor-ligand binding, it enables cell-to-cell contact and exerts immunomodulatory effects,³² thereby endowing nanocarriers with immune evasion capability and active homing effects to inflammatory sites,³³ avoiding rapid clearance by the body. Moreover, the macrophage membrane has a large surface area and excellent mechanical properties, allowing efficient coating of NP cores without compromising the structural integrity of the NPs.³⁴ Based on this, the present study utilized mesoporous MPDA to load API and coated the surface with a macrophage membrane to obtain MEM@MPDA/API NPs. This novel nano-delivery system, synergizing the advantages of MPDA, API, and the macrophage membrane, holds promise for providing new approaches for the clinical treatment of ALI.

In this study, MEM@MPDA/API NPs were successfully prepared. Unlike the conventional approach of directly using API as a monotherapy for ALI, this innovative combination provides a new strategy to overcome the limitations of the traditional drug API, while also offering novel insights for the clinical treatment of ALI. By modifying the MPDA surface with macrophage membrane coating, we enhanced the NPs’ ability to target inflamed lung tissues and reduced the metabolic clearance of API. Simultaneously, the broad-spectrum ROS scavenging capacity of MPDA, synergistically combined with the anti-inflammatory and antioxidant properties of API, provides an effective strategy for alleviating inflammatory damage and oxidative stress in ALI. This approach promotes a paradigm shift in the current treatment of pulmonary inflammation and oxidative stress-related diseases, and facilitates the broader application of nanomaterials in the therapy of pulmonary disorders.

Materials and Methods

Preparation of MEM@MPDA/API

Extraction of Macrophage Membrane

Mouse monocyte-macrophage cells (RAW264.7, purchased from Haixing Biotechnology Co., Ltd., product code: TCM-C766-UD) were washed with phosphate-buffered saline (PBS) and lysed using a cell membrane extraction kit (Beyotime, Shanghai, China) in lysis buffer containing protease inhibitors on ice with an ultrasonicator. The lysate was subjected to differential centrifugation to enrich the plasma membrane fraction: nuclei were removed by centrifugation at 1,000 ×g, mitochondria were removed at 10,000 ×g, and the macrophage membrane fraction was finally collected at 100,000 ×g. The obtained membrane pellet was dialyzed to remove nucleic acids and soluble protein impurities, and the membrane protein concentration was determined using a BCA protein assay kit (Beyotime, Shanghai, China).³⁵

Synthesis of MPDA

MPDA NPs were prepared via a one-pot method: first, 150 mg of dopamine hydrochloride (Aladdin Biochemical Technology Co., Ltd., Shanghai, China) and 100 mg of polyoxyethylene polyoxypropylene ether (F127, Guanghua Technology Co., Ltd., Guangdong, China) were dispersedin 10 mL of deionized water and ethanol, respectively. After complete dissolution, the two solutions were mixed, and 160 μL of 1,3,5-trimethylbenzene (TMB, Aladdin Biochemical Technology Co., Ltd., Shanghai, China) was added. The mixture was thoroughly mixed and subjected to water bath ultrasonication for 6 minutes. Subsequently, under magnetic stirring, 375 μL of ammonia solution was slowly added dropwise, and the mixture was allowed to react at room temperature for 2 hours. After centrifugation at room temperature (9,000 rpm, 10 minutes), the precipitate was washed three times with ethanol and three times with ultrapure water sequentially to obtain MPDA. The product was vacuum freeze-dried for 24 hours and stored at −20°C.³⁵

Synthesis of MPDA/API

The prepared MPDA NPs were weighed and dissolved in deionized water. API (Aladdin Biochemical Technology Co., Ltd., Shanghai, China) was weighed, dissolved in dimethyl sulfoxide (DMSO, Sigma, USA), and then further dissolved in anhydrous ethanol. The mass ratio of MPDA to API was 1:2. The mixture was subjected to magnetic stirring in the dark for 24 hours, followed by centrifugation (9,000 rpm, 10 minutes) and washed three times with deionized water to successfully obtain MPDA/API.³⁵

To prepare the membrane-coated MPDA/API (MEM@MPDA/API), cell membrane fragments and NPs were mixed at a 2:1 ratio in physiological saline and sonicated. The mixture was then repeatedly extruded 30 times through polycarbonate porous membranes with pore sizes of 800 nm and 400 nm, respectively, using a nano-liposome extruder. The resulting macrophage membrane-coated NPs were collected as MEM@MPDA/API.³⁵

Synthesis of Fluorescently Labeled Nanocomposites

Fluorescently labeled nanocomposites were synthesized by incorporating Cy5.5(Shanghai Yuanye Technology Co., Ltd., China) into the core of MEM@MPDA/API . Cy5.5 dye (excitation wavelength = 683 nm, emission wavelength = 703 nm) was incorporated into the nanocomposite structure, and these fluorescent nanocomposites were used for fluorescence-based evaluations in vitro and in vivo.

Characterization of MEM@MPDA/API NPs

The morphology of MEM@MPDA/API was observed using scanning electron microscopy (SEM, Hitachi, 5 kV) and transmission electron microscopy (TEM, Hitachi, 120 kV). Particle size and Zeta potential were determined by a Zetasizer Nano ZS90 particle size analyzer using dynamic light scattering for size and electrophoretic light scattering for Zeta potential. Chemical functional groups on the NP surface were detected using a Fourier transform infrared spectrometer (FT-IR, Nicolet 6700; Thermo Fisher Scientific, USA). The absorption spectrum was analyzed using ultraviolet-visible-near infrared spectroscopy (UV, UV-3600 spectrophotometer; Shimadzu, Japan).Finally, the suspension stability was evaluated by measuring the changes in particle size of the NPs stored in PBS and Dulbecco’s Modified Eagle Medium (DMEM) at 4 °C under dark conditions over a period of 0–14 days.

Drug Loading Capacity (LC) and Encapsulation Efficiency (EE)

To determine the drug loading efficiency of MPDA, during the preparation of MPDA/API, 20 mg of API was dissolved in DMSO and then added to 20 mL of anhydrous ethanol, mixed with 10 mg of MPDA NPs, stirred for 24 hours, and then centrifuged. The precipitate and supernatant were collected. The API content in the supernatant was determined using a UV-Vis spectrophotometer, and the measurement was repeated three times. For the UV-Vis determination of API in the supernatant, a standard curve was first established by preparing API solutions at different concentrations (2, 4, 6, 8, 10 μg/mL), placing them in a UV-Vis detection plate, measuring the absorbance at 336 nm, and plotting the standard curve, as shown in Figure S1. The collected supernatant was diluted 100-fold, and the UV absorbance of API was measured at 336 nm. The free drug content in the supernatant was calculated using the standard curve equation. The loading capacity (LC) and encapsulation efficiency (EE) were calculated using the following formulas:

The drug loading and encapsulation efficiency were calculated based on three repeated measurements.

Verification of Macrophage Membrane Coating by Western Blotting (WB)

The retention of macrophage membrane proteins on the surface of MEM@MPDA/API NPs was confirmed by WB analysis. Key proteins, including F4/80 and CD47, were detected in RAW264.7 macrophages, the extracted macrophage membrane, and MEM@MPDA/API samples. The presence of these characteristic macrophage membrane proteins on the surface of MEM@MPDA/API NPs demonstrated the successful integration of the macrophage membrane onto the MEM@MPDA/API NP platform.

CCK-8 Assay for Pulmonary Epithelial Cell Viability

Mouse lung epithelial cells (MLE-12, purchased from Haixing Biotechnology Co., Ltd., product code: TCM-C745) (5 × 10³ cells/well) were seeded into 96-well plates. After cell attachment, various concentrations of the drug were added to each well and incubated for 24 hours. The cytotoxicity of API and MEM@MPDA/API against lipopolysaccharide (LPS)-induced MLE-12 cell injury was evaluated using the CCK-8 assay, with the control group treated with an equal volume of PBS. Absorbance was measured at 450 nm using a microplate reader (BioTek, USA). Cell viability was expressed as a fold change relative to the control group (set as 1), ie,

Cellular Uptake of MEM@MPDA/API

Cy5.5-labeled NPs were co-incubated with cells in good growth condition for 4 hours, with two groups: the MPDA/Cy5.5 group and the MEM@MPDA/Cy5.5 group. The culture medium was then discarded, and the cells were washed with PBS to remove extracellular NPs. Finally, the nuclei were stained with Hoechst, and fluorescence was observed under a fluorescence microscope (Carl Zeiss AG, Germany).

Quantitative Real-Time PCR (qRT-PCR)

Total RNA was extracted from MLE-12 cells using Trizol reagent according to the manufacturer’s instructions. The concentration and purity of RNA were determined, with qualified samples exhibiting a 260 nm/280 nm absorbance ratio between 1.80 and 2.00. A total of 1 μg of RNA was reverse transcribed into complementary deoxyribonucleic acid (cDNA), followed by polymerase chain reaction (PCR) amplification. Primers were synthesized by Sangon Biotech (Shanghai) Co., Ltd., and the primer sequences are listed in Table 1.

Table 1.

Specific primer of sequences used in PCR

Gene	Forward Primer (5′to 3′)	Reverse Primer (5′to 3′)
IL-1β	TGCCACCTTTTGACAGTGATG	AAGGTCCACGGGAAAGACAC
IL-6	TGATGGATGCTACCAAACTGGA	TGTGACTCCAGCTTATCTCTTGG
TNF-α	ACTGAACTTCGGGGTGATCG	CCACTTGGTGGTTTGTGAGTG
GAPDH	ACTCAGGAGAGTGTTTCCTCG	TTTGCCGTGAGTGGAGTCAT

ROS Staining

Intracellular ROS levels were detected using 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA, Beyotime, Shanghai, China) as a fluorescent probe. Cells from different treatment groups were incubated with DCFH-DA in serum-free DMEM for 30 minutes at 37°C in a 5% CO₂ atmosphere protected from light. After washing three times with PBS, the cells were observed under a fluorescence microscope, where intracellular ROS levels were proportional to green fluorescence intensity. Images were captured, and fluorescence intensity was quantitatively analyzed using ImageJ.

Mitochondrial Membrane Potential (JC-1 Assay)

Mitochondrial membrane potential was assessed using the JC-1 staining method (Beyotime, Shanghai, China). Cells were seeded into 24-well plates at a density of 3 × 10⁴ cells per well and treated according to the experimental conditions. After treatment, the cells were washed with PBS and stained with 10 μg/mL JC-1 dye for 20 minutes at 37°C protected from light. The cells were then washed twice with PBS and observed under a fluorescence microscope.

Transmission Electron Microscopy (TEM) of Cells

MLE-12 cells (seeded at 1×10⁵ cells/well in 6-well plates) were subjected to different treatments. After 24 hours, the cells were trypsinized, centrifuged at 1,000 rpm for 5 minutes, and fixed with 3% glutaraldehyde at 4°C for 5 minutes. The fixed cells were transferred to 1.5-mL microcentrifuge tubes and pelleted by centrifugation at 12,000 rpm for 10 minutes. The cell pellets were post-fixed with 1% osmium tetroxide, dehydrated through a graded acetone series (30%→100%), and infiltrated with Epon812 resin at ratios of 3:1, 1:1, and 1:3. The samples were embedded in Epon812, and ultrathin sections (60–90 nm) were prepared using an ultramicrotome and loaded onto copper grids. The sections were stained with uranyl acetate (10–15 minutes) and lead citrate (1–2 minutes) at room temperature, and imaged at 8,000× magnification under a JEM-1400FLASH transmission electron microscope.

LPS-Induced ALI in Mice

Seven-week-old male C57BL/6 mice were purchased from the Animal Center of Guangxi Medical University and housed in a specific pathogen-free (SPF) environment. This study was approved by the Medical Ethics Committee of the Second Affiliated Hospital of Guangxi Medical University (Approval No. 2024-KY(1062)). During the experiments, we strictly adhered to the 3R principles (Replacement, Reduction, Refinement).

Mice were randomly divided into 5 groups with 12 mice per group. They were first anesthetized by intraperitoneal injection of 1% sodium pentobarbital (supplier: Merck; catalog number: Y0002194; batch number: 20170525; dose: 50 mg/kg), and then the ALI model was established by intratracheal instillation of LPS (5 mg/kg), while the control group received an equal volume of PBS. Four hours later, API, MPDA, and MEM@MPDA/API were administered via tail vein injection at a volume of 200 μL, and the control group received an equal volume of PBS. At 24 hours after administration, the mice in each treatment group were anesthetized and euthanized for subsequent experiments.

In vivo Fluorescence Imaging in Mice

Cy5.5-labeledMPDA/Cy5.5 and MEM@MPDA/Cy5.5 (200 μL, 200 μg/mL) were injected into the tail vein of acute lung injury mice, and fluorescence imaging was performed using an in vivo imaging system (Guangzhou Biolight Biotechnology Co., Ltd., China) at 30 minutes and 2, 4, 6, 8, 12, and 24 hours post-injection. Subsequently, the mice were euthanized, and their major organs (heart, liver, spleen, lungs, and kidneys) were collected for ex vivo imaging analysis.

Lung Histopathology

After washing with PBS, lung tissues were fixed in 4% paraformaldehyde, embedded in paraffin, sectioned, and subjected to hematoxylin and eosin (H&E) staining and immunohistochemical staining according to standard procedures, and then observed under a microscope. Immunohistochemistry was used to detect the expression level of myeloperoxidase (MPO) in lung tissues to assess the degree of neutrophil infiltration. The histopathological scoring criteria for lung tissues are shown in Table 2. For blinding, grouping, experimentation, and outcome assessment were performed separately by three different researchers in the team. Frozen sections for ROS staining were treated with autofluorescence quencher, then stained with ROS probe and DAPI, and observed under a fluorescence microscope.

Table 2.

Lung Injury Scoring Sheet

Score	Histopathological Features
0	Normal alveolar structure
1	Mild interstitial thickening
2	Thickening of the alveolar wall with a small amount of inflammatory cell infiltration
3	Destruction of the alveolar structure with extensive inflammatory cell infiltration
4	Severe pulmonary consolidation or hemorrhage

Collection of Bronchoalveolar Lavage Fluid (BALF)

Mice were euthanized 24 hours after administration, and BALF was collected and centrifuged at 3,000 rpm for 15 minutes. The total number of pelleted cells in the BALF from each mouse was counted using an automated cell counter (Thermo Fisher Scientific, USA), and the supernatants were collected for determination of protein concentration and levels of inflammatory cytokines (IL-6, TNF-α, and IL-1β) according to the instructions of the BCA protein assay kit (Servicebio, Wuhan, China) and ELISA kits (MultiSciences (Lianke) Biotech, Hangzhou, China), respectively.

Biosafety Evaluation

To evaluate the hemolytic potential of MEM@MPDA/API, blood was collected from the hearts of healthy mice and centrifuged at 1,000 rpm for 5 minutes to isolate red blood cells, which were then washed three times with PBS and resuspended in PBS to prepare a 2% red blood cell suspension. Equal volumes of MEM@MPDA/API solutions at various concentrations (15.63–1000 μg/mL) were mixed with the red blood cell suspension, with PBS as the negative control and pure water as the positive control. After thorough mixing, the mixtures were incubated in a 37°C water bath for 1 hour, then centrifuged at 3,000 rpm for 15 minutes and photographed. A 100 μL aliquot of supernatant from each tube was transferred to a 96-well plate, and the absorbance was measured at 542 nm using a microplate reader. The hemolysis rate was calculated using the following formula:

hemolysis rate (%) = (absorbance of sample - absorbance of negative control)/(absorbance of positive control - absorbance of negative control) × 100%.

Blood Biochemical Analysis and Body Weight Monitoring in Mice

Six healthy male mice, aged 7 weeks and weighing 22–25 g, were randomly divided into two groups (n = 3). One group received tail vein injections of MEM@MPDA/API (10 mg/kg) for 14 consecutive days, while the control group received an equal volume of PBS. Body weight was monitored every two days. On day 14, blood samples were collected via the retro-orbital sinus. Complete blood counts (RBC, WBC, PLT, HGB) and serum biochemical markers (ALT, AST, BUN, CREA) were measured using an automatic hematology analyzer and a biochemical analyzer, statistically compared with the control group. After dissection, the heart, liver, spleen, lung, and kidney were collected for H&E staining.

In addition, to further observe toxic reactions after an extended administration period, an additional 30-day treatment group was established in this experiment. The injection method was the same as that for the 14-day group. On day 30, blood samples were collected from the heart for complete blood count and serum biochemical analysis, and the results were statistically compared with those of the control group.

Immunofluorescence Staining

Lung tissues were fixed in 4% paraformaldehyde for 24 hours and then cryoprotected with 30% and 40% sucrose solutions. After dehydration, the tissues were embedded, sectioned, and blocked with 2% bovine serum albumin, 0.2% Triton X-100, and 5% fetal bovine serum for 1 hour. The sections were incubated with primary antibodies overnight at 4°C, washed three times with PBS, incubated with corresponding secondary antibodies for 1 hour, washed again, and then stained with DAPI for nuclear visualization. Images were captured under a fluorescence microscope. The primary antibodies used for immunofluorescence included: Anti-LC3B (ABclonal BiotechnologyCo., Ltd., Wuhan, China), Anti-TOMM20 (Cell Signaling Technology), Anti-ZO-1 Rabbit antibody, and Anti-Occludin Rabbit antibody (Proteintech, Wuhan, China).

WB Assay

Lung tissues were lysed using RIPA lysis buffer containing phosphatase inhibitors, and tissue proteins were extracted according to the manufacturer’s instructions. Equal amounts of total protein were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto PVDF membranes. The membranes were blocked at room temperature for 15 minutes, incubated with primary antibodies overnight at 4°C, and then incubated with secondary antibodies for 1 hour at room temperature protected from light. After washing three times with TBST, the membranes were scanned for detection. The antibodies used for Western blotting included: Anti-GAPDH (Cell Signaling Technology), Anti-ZO-1 Rabbit antibody, Anti-VE-cadherin Rabbit antibody (Proteintech, Wuhan, China), and Anti-Occludin Rabbit antibody (Proteintech, Wuhan, China).

Network Pharmacology Target Prediction and Molecular Docking

Active Ingredient Screening and Target Prediction

API targets were obtained from the SwissADME (https://swissadme.ch/), TCMSP (https://tcmspw.com/tcmsp.php), and CTD (https://ctdbase.org/) databases, consolidated, and deduplicated.

ALI Target Acquisition

ALI-related datasets (GSE2411) were downloaded from the GEO database, and differentially expressed genes (DEGs) were screened using the “Limma” package. Meanwhile, disease target databases including GeneCards (https://www.genecards.org) and OMIM (https://www.omim.org) were searched using the keyword “acute lung injury” to identify relevant targets, which were then combined and deduplicated.

Drug–Disease Target PPI Network Construction and Core Target Screening

An intersection diagram was generated between the identified active ingredient targets of API and ALI-related disease targets. The intersecting genes were then imported into the STRING protein–protein interaction database to construct a PPI network. Core target genes were screened based on node and edge scoring metrics.

Functional Enrichment Analysis of Intersection Targets

Gene Ontology (GO) and Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed on the intersecting genes of disease and drug targets. The top-ranked biological processes (BP), molecular functions (MF), cellular components (CC), and signaling pathways were presented as bar charts to elucidate the biological functions and pathways associated with the potential targets in ALI.

Molecular Docking

The 3D structure data of the core protein genes were obtained from the RCSB PDB database (https://www.rcsb.org/) and processed using PyMOL software to remove small molecules and bound ligands. The 3D structures of key active components were downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov/) and subjected to energy minimization using Chem3D. Semi-flexible docking was performed using AutoDock Vina software, with the active pocket defined and docking parameters (such as box center and size) set accordingly. A binding energy less than −5.0 kcal/mol was used as the selection criterion, where a lower binding energy indicates more stable ligand–receptor binding. The docking conformations were visualized using PyMOL to analyze key interactions, including hydrogen bonds and hydrophobic interactions.

Single-Cell RNA Sequencing Data Analysis of ALI and Control Mice

Data Source and Download

The mouse ALI single-cell RNA sequencing datasets GSE292681 and GSE264032 were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/geo/). The raw count matrices were analyzed using the Seurat package (version 4.3.0) in R software (version 4.2.1) for downstream analysis.

Quality Control and Data Filtering

The raw expression matrix was converted into a Seurat object using the CreateSeuratObject function. The filtering criteria were as follows: (1) cells with UMI counts <200 or >5000 were removed to exclude empty droplets and potential doublets/multiplets; (2) cells with mitochondrial gene content >5% were removed to exclude apoptotic or membrane-damaged cells; (3) cells with fewer than 200 detected genes were also removed.

Data Normalization and Highly Variable Gene Selection

The gene expression matrix was normalized using the NormalizeData function with the LogNormalize method, with a scale factor set to 10,000. The top 3,000 highly variable genes were identified using the FindVariableFeatures function for subsequent principal component analysis. All genes were scaled and centered using the ScaleData function.

Principal Component Analysis (PCA) and Batch Effect Correction

PCA was performed for linear dimensionality reduction using the RunPCA function based on the highly variable genes. The number of principal components was determined by combining the elbow plot and the JackStraw method, and the top 30 principal components (cumulative variance contribution >85%) were selected for downstream analysis. Since two datasets were integrated, batch effect correction was performed using the FindIntegrationAnchors and IntegrateData functions based on canonical correlation analysis (CCA).

Cell Dimensionality Reduction and Clustering

The FindNeighbors function was used to construct a K-nearest neighbor graph based on the top 30 principal components (k.param = 20). Cell clustering was performed using the FindClusters function with the resolution parameter set to 0.5. The UMAP algorithm was used for two-dimensional visualization and dimensionality reduction, with parameters set as follows: n.neighbors = 30, min.dist = 0.3.

Cell Annotation and Functional Enrichment Analysis

Differentially expressed genes (DEGs) in each cell subpopulation were identified using the `FindMarkers` function. Combined with previously reported classical cell lineage-specific marker genes for mouse lung tissue, manual and precise cell type annotation was performed for each cell cluster.

The DEGs obtained from the screening were imported into the clusterProfiler package for Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis, with an adjusted P value < 0.05 set as the significance threshold.

Screening and Visualization of Lysosome-Related DEGs

Lysosome gene sets were retrieved and downloaded from the Gene Ontology (GO) database (https://www.geneontology.org/). The lysosome-related DEGs were extracted by intersecting the lysosome gene sets with the identified DEGs, and the expression distribution of key lysosome genes across various cell types was visualized using the DotPlot and VlnPlot functions.

Statistical Analysis

Experimental data are expressed as mean ± standard deviation (SD). Before applying parametric tests, the normality of data distribution was assessed using the Shapiro–Wilk test, and homogeneity of variances was verified by Bartlett’s test. All datasets met the assumptions of normality and equal variance (P > 0.05 for each test), thus justifying the use of one-way analysis of variance (ANOVA) followed by Dunnett’s post-hoc test for multiple comparisons against the control group. A value of P < 0.05 was considered statistically significant, with significance levels denoted as P < 0.05, *P < 0.01, **P < 0.001, and ***P < 0.0001.

Results

Characterization of MEM@MPDA/API NPs

MPDA NPs were synthesized via a one-pot method under alkaline conditions, with the simultaneous encapsulation of API. Subsequently, macrophage membranes were coated onto the surface of the MPDA/API NPs through ultrasonication and extrusion techniques, yielding biomimetic NPs designated as MEM@MPDA/API NPs (Figure 1A).

Figure 1.

Preparation of MEM@MPDA/API. (A) Synthesis of MEM@MPDA/API NPs. (B) SEM image of MPDA (i), TEM image of MPDA (ii), TEM image of MPDA/API (iii), negative staining electron microscopy image of MEM@MPDA/API (iv). (C) Particle size distribution of MPDA (i), MPDA/API (ii), and MEM@MPDA/API (iii).

As shown in Figure 1B, the morphology of MPDA was characterized by SEM (Figure 1B i) and TEM (Figure 1B ii), revealing spherical particles with a uniform size distribution. For MPDA loaded with API (Figure 1B iii), the porous structure on the surface was markedly reduced or even disappeared. Negative-staining TEM of MEM@MPDA/API NPs (Figure 1B iv) showed a bilayer membrane with strong electron-lucent properties coated on the particle surface. The size distribution results indicated that MPDA had an average diameter of 167.17 ± 3.68 nm (Figure 1C i), MPDA/API 197.17 ± 2.48 nm (Figure 1C ii), and MEM@MPDA/API 256.40 ± 1.21 nm (Figure 1C iii). Comparison revealed that the particle size of MEM@MPDA/API increased significantly after macrophage membrane coating.

To verify the successful synthesis of MEM@MPDA/API, UV-Vis and FT-IR spectroscopic analyses were performed on each component (Figure 2A and B). The UV-Vis spectra showed that the macrophage membrane component exhibited an absorption peak at 204 nm, corresponding to the π→π transition of aromatic amino acid residues in membrane proteins, while the ester functional groups of phospholipids gave a peak around 278 nm. API mainly displayed a broad band with absorption near 330nm, which is characteristic of flavonoids and assigned to the A band of the conjugated flavonoid nucleus. The strong peak of MPDA appeared at approximately 221 nm, attributable to the π–π transition of the benzene ring in polydopamine; the high conjugation degree of MPDA resulted in a rapid decrease in intensity after this strong peak. The absorption peaks of MEM@MPDA/API were essentially a superposition of those of the macrophage membrane, API and MPDA: the peak near 200 nm arose from the aromatic group transitions of the membrane, API and polydopamine, while the peak around 260 nm was broader than that of the macrophage membrane alone, indicating that interactions among the components affected the electronic environment of the functional groups. The UV-Vis results corroborated the occurrence of interactions between the functional groups of the macrophage membrane, API, and MPDA during the composite formation process, indicating sufficient integration among the components.

Figure 2.

Characterization of MEM@MPDA/API. (A) UV-Vis absorption spectra of NPs in each group: (A) MEM, (B) API, (C) MPDA/API, (D) MEM@MPDA/API NPs. (B) FTIR spectra of NPs in each group: (A) MEM, (B) API, (C) MPDA/API, (D) MEM@MPDA/API NPs. (C) Zeta potential measurements of MPDA, MEM, API, and MEM@MPDA/API NPs. (D) Colloidal stability of MEM@MPDA/API NPs in PBS and DMEM solutions. (E) WB analysis of F4/80, CD47, and GAPDH protein expression in RAW264.7 cells, MEM, and MEM@MPDA/API NPs.

Figure 2C presents the Zeta potential results. The Zeta potentials of MPDA, API, macrophage membrane, and MEM@MPDA/API were −23.27 ± 0.46 mV, −31.77 ± 0.61 mV, −12.67 ± 0.21 mV, and −29.67 ± 0.42 mV, respectively. After membrane coating and drug loading, the Zeta potential of the MEM@MPDA/API complex became significantly more negative compared to that of pristine MPDA and closely approached the potential level of free API, strongly confirming the efficient loading of the negatively charged API and the successful coating of the negatively charged macrophage membrane. This enhanced negative surface charge is expected to impart strong electrostatic repulsion to the NPs, which is beneficial for maintaining colloidal stability and reducing nonspecific clearance in vivo. These findings demonstrate that the macrophage membrane has been successfully coated onto the outer surface of MEM@MPDA/API. The macrophage membrane camouflage enables MEM@MPDA/API to mimic macrophage functions, thereby laying a foundation for subsequent experiments. In addition, the drug loading content and encapsulation efficiency of MPDA were determined using a UV-Vis spectrophotometer at 336 nm. From three replicate measurements, the drug loading content was calculated to be 26.37% ± 2.894, and the encapsulation efficiency was 18.33% ± 1.992.

The storage stability of NPs is particularly important for their storage and transportation. After storing the NPs in different solutions for 14 day and measuring the particle size at various time points, it was found that MEM@MPDA/API exhibited superior stability in both PBS and DMEM, with no obvious aggregation or degradation observed (Figure 2D).

To further confirm the successful functionalization of MEM@MPDA/API NPs, WB analysis was performed to detect the signature proteins F4/80 and CD47 on RAW264.7 cells, extracted macrophage membranes, and MEM@MPDA/API NPs (Figure 2E). The results demonstrated that the physical sonication and extrusion-based membrane coating method employed in this study could retain the key functional proteins of macrophages during the construction of MEM@MPDA/API NPs. Meanwhile, GAPDH, a typical cytosolic protein, was not detected in MEM@MPDA/API NPs, indicating that the extracted membrane was free of cytoplasmic contamination.

MEM@MPDA/API NPs Enhance Inflammatory Targeting and Cellular Uptake

Alveolar epithelial cells constitute the core of the lung barrier and play a key role in maintaining tissue homeostasis; therefore, they are widely used as a model for studying pulmonary diseases and cellular responses to various stimuli.³⁶ In this study, an LPS-induced MLE-12 cell injury model was employed to investigate the protective effect of MEM@MPDA/API NPs against oxidative stress in MLE-12 cells. First, the CCK-8 assay was used to evaluate the cell viability of MLE-12 cells after treatment with different concentrations of API and MEM@MPDA/API for 24 hours. Figure 3A and B showed that at the same drug concentrations, MEM@MPDA/API NPs had less effect on cell viability. When the concentrations of API and MEM@MPDA/API were in the range of 0–200 μg/mL, some cell viability values in both groups were above 1, indicating that both drug interventions exerted a proliferative effect on cells. However, API exhibited obvious inhibition at concentrations greater than 200 μg/mL. For concentration uniformity in this study, the concentration of 200 μg/mL, which was the maximum concentration that caused no damage to cells in both groups, was selected for subsequent experiments. Additionally, a hemolysis assay was performed to verify the hemocompatibility, and it was found that the hemolysis rates of MEM@MPDA/API NPs at different concentrations were all below 5%, indicating that this nanoformulation exhibits good hemocompatibility upon contact with red blood cells after intravenous administration, thereby providing an important safety basis for subsequent tail vein injection in mice (Figure 3C).

Figure 3.

Biocompatibility and in vitro and in vivo uptake of MEM@MPDA/API NPs. (A) Cell viability of MLE-12 cells after co-incubation with different concentrations of API. (B) Cell viability of MLE-12 cells after co-incubation with different concentrations of MEM@MPDA/API NPs. (C) Hemolysis rate assessment of MEM@MPDA/API NPs at different concentrations. (D) Representative whole-body imaging of mice at different time points after injection of MPDA/Cy5.5 NPs and MEM@MPDA/Cy5.5 NPs(n=3). (E) Uptake of MPDA/Cy5.5 NPs and MEM@MPDA/Cy5.5 NPs by MLE-12 cells(n=3). (F) Quantitative fluorescence analysis of the uptake of MPDA/Cy5.5 NPs and MEM@MPDA/Cy5.5 NPs by MLE-12 cells(n=3).**P<0.01.

To validate the targeting capability of these NPs, verification was performed in both a mouse model of septic lung injury and a cellular model, using in vivo imaging technology and cellular uptake experiments to investigate the targeting effect after cell membrane modification. The in vivo uptake results in mice (Figure 3D) showed that, compared with the control group, modification of the NP surface with macrophage membrane significantly enhanced their targeting recognition of pulmonary inflammation. In the MEM@MPDA/Cy5.5 group, fluorescence intensity in the lungs gradually increased at 6 hours post‑injection, whereas no such phenomenon was observed in the MPDA/Cy5.5 group. At 24 hours post-injection, the fluorescence intensity in the MEM@MPDA/Cy5.5 group remained higher than that in the MPDA/Cy5.5 group, further confirming the effectiveness of macrophage membrane coating technology in achieving lung-targeted drug delivery. Additionally, relatively stronger fluorescence intensity was observed in the liver compared with other organs, suggesting hepatic metabolism of the nanodrug (Figure S2).

A similar phenomenon was observed in the cellular uptake assay (Figure 3E and F). Compared with NPs without membrane encapsulation, lung epithelial cells exhibited significantly higher uptake of membrane-encapsulated NPs (P < 0.01), further suggesting that inflammation-targeting modifications on the membrane surface enhanced the targeting effect of the NPs. This property provides new strategies and approaches to overcome the difficulties in targeted drug delivery to the lungs, and is expected to offer more effective drug delivery strategies for the treatment of pulmonary inflammatory diseases.

In vitro Anti-Inflammatory and ROS Scavenging Capacity of MEM@MPDA/API NPs

In ALI, the release of inflammatory cytokines is a key driver of the cytokine storm. To investigate the in vitro anti-inflammatory effect of MEM@MPDA/API NPs, the expression levels of inflammatory cytokines were measured after treatment with MEM@MPDA/API NPs. The results showed that, compared with the control group, the expression levels of IL-6 (P < 0.001), TNF-α (P < 0.0001), and IL-1β (P < 0.0001) were significantly upregulated in the model group. After treatment with API, MPDA, and MEM@MPDA/API NPs, the expression levels of inflammatory cytokines in each treatment group were decreased to varying degrees (P < 0.05). Moreover, compared with the monotherapy groups, the MEM@MPDA/API NPs group exhibited a more pronounced downregulation of TNF-α and IL-1β (P < 0.05), demonstrating the excellent anti-inflammatory efficacy of this nanodelivery system in vitro (Figure 4A–C).

Figure 4.

In Vitro Anti-Inflammatory and ROS Scavenging Capacity of MEM@MPDA/API NPs. (A–C) qPCR detection of IL-6, TNF-α, and IL-1β in MLE-12 cells(n=3). (D and E) In vitro ROS scavenging fluorescence assay results and quantitative fluorescence analysis of MEM@MPDA/API NPs(n=3). (F) JC-1 staining for mitochondrial membrane potential after different treatments(n=3). (G) TEM images of MLE-12 cells after different treatments(Orange arrow: Mitochondria, n=3). ns, no significance; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Excessive ROS production is a hallmark of ALI and is a key factor driving the uncontrolled pulmonary inflammation. Therefore, scavenging the overproduced ROS represents an effective strategy to alleviate ALI symptoms. To evaluate the ROS scavenging capacity of the NPs in vitro, the DCFH-DA probe was used for specific ROS staining, and the fluorescence intensity was quantitatively analyzed (Figure 4D and E). The results showed that, compared with the control group, the LPS-treated group exhibited green fluorescence from the probe, indicating excessive ROS production. The green fluorescence signal intensities in all treatment groups were weaker than that in the model group. Notably, the MEM@MPDA/API treatment group showed significantly lower green fluorescence signal than the API monotherapy group, with a statistically significant difference (P < 0.01). These findings clearly confirm that MEM@MPDA/API NPs possess excellent antioxidant capacity.

Mitochondria are essential organelles for maintaining cellular homeostasis and play a critical role in inflammatory responses, with their functional stability governing cell damage and repair. To investigate the effect of MEM@MPDA/API NPs on mitochondrial function, JC-1 staining was employed to detect changes in mitochondrial membrane potential (MMP), thereby inferring alterations in mitochondrial function. The results showed that in LPS-treated cells, the red fluorescence signal was diminished and green fluorescence predominated, whereas in the model group treated with MEM@MPDA/API NPs, the red fluorescence signal was significantly enhanced (Figure 4F). These findings indicate that MEM@MPDA/API NPs effectively repaired LPS-induced mitochondrial damage in MLE-12 cells, restored the MMP, and improved mitochondrial function.

To further investigate the effect of MEM@MPDA/API NPs on mitochondrial morphology, TEM was used to observe LPS-treated cells, which exhibited mitochondrial swelling, deformation, and loss of mitochondrial cristae (indicated by orange arrows). After intervention with MEM@MPDA/API NPs, mitochondrial morphology was restored and the number of damaged mitochondria was reduced (Figure 4G).

In summary, MEM@MPDA/API NPs can alleviate inflammatory damage by scavenging excessive ROS and promoting the recovery of mitochondrial function, demonstrating the potential of this nanomaterial in antioxidant protection and the treatment of oxidative stress-related diseases.

MEM@MPDA/API NPs Alleviate LPS-Induced ALI in Mice

To verify the therapeutic effect of MEM@MPDA/API NPs in vivo, a mouse model of ALI was established by intratracheal instillation of LPS. As expected, HE staining results showed that LPS-stimulated mice exhibited cellular and interstitial edema, alveolar wall thickening (black arrows), and increased inflammatory cell infiltration (red arrows), along with elevated histopathological scores. Compared with the other treatment groups, the MEM@MPDA/API treatment group exhibited the most pronounced improvement, effectively suppressing cellular and interstitial edema, significantly reducing inflammatory cell infiltration, and decreasing the histopathological score (P < 0.0001) (Figure 5A–C).

Figure 5.

MEM@MPDA/API NPs alleviate LPS-induced ALI in mice. (A) Representative H&E staining of lung tissue in different treatment groups (Red arrows: cellular infiltration; Black arrows: alveolar wall thickening). (B) MPO immunohistochemical staining of lung tissue in different treatment groups(n=6). (C) Immunopathological scores of lung tissues in different treatment groups (Pathological scores were evaluated independently by three blinded investigators, and the final score for each sample represents the mean value of their assessments,n=6). (D) Total protein levels in BALF(n=3). (E) BALF cell counts(n=3). (F–H) ELISA measurement of IL-6, TNF-α, and IL-1β levels in BALF from different treatment groups(n=3). (I) In vivo ROS scavenging fluorescence imaging results of MEM@MPDA/API NPs(n=3). ns, no significance; *P<0.05,**P<0.01,***P<0.001,****P<0.0001.

Myeloperoxidase (MPO), as a neutrophil marker, can be used to evaluate inflammatory cell infiltration in lung tissues. Immunohistochemical staining for MPO (Figure 5B) showed that, compared with LPS-stimulated mice, the expression of MPO was reduced after treatment with MEM@MPDA/API NPs. This result suggests that MEM@MPDA/API NPs exert a modulatory effect on pulmonary immunopathology.

The total cell count and total protein in BALF can be used to assess the degree of pulmonary inflammation. Compared with the LPS-treated group, the total cell count and total protein leakage were gradually reduced in the API, MPDA, and MEM@MPDA/API treatment groups (Figure 5D and E). Meanwhile, the concentrations of IL-6, TNF-α, and IL-1β in the BALF supernatant were measured. As shown in Figure 5F–H, compared with the control group, the levels of pro-inflammatory cytokines IL-6, TNF-α, and IL-1β were significantly elevated in the LPS-induced ALI model group (P < 0.0001), and after administration of API, MPDA, and MEM@MPDA/API NPs, the cytokine levels in all treatment groups were gradually reduced. Notably, the MEM@MPDA/API NPs group exhibited the lowest expression levels of inflammatory cytokines. In addition, the regulatory effects of macrophage membrane-coated versus uncoated MPDA/API NPs on the mRNA expression levels of pulmonary inflammatory cytokines were compared in this study. Consistently, the results showed that macrophage membrane-coated MEM@MPDA/API exerted significantly greater inhibition of inflammatory cytokine levels than uncoated MPDA/API (Figure S3).

ROS staining on frozen sections of mouse lung tissues was performed, and the results are shown in Figure 5I. Lung tissues from LPS-stimulated mice exhibited extensive red fluorescence, indicating abundant ROS production in ALI, which was consistent with the in vitro ROS staining results. After treatment with API and MPDA, the red fluorescence signals decreased to varying degrees, demonstrating the ROS scavenging ability of both materials. Notably, the lowest red fluorescence signal was observed after MEM@MPDA/API treatment, indicating its excellent ROS scavenging capacity.

In vivo Safety of MEM@MPDA/API NPs

To further evaluate the potential systemic toxicity of the NPs upon in vivo application, hematological parameters, hepatic and renal function, and histopathological changes in major organs (heart, liver, spleen, lungs, and kidneys) were monitored after 14 consecutive days of intravenous injection via the tail vein. HE staining results showed no obvious pathological damage in the major organs of the treatment group compared with the control group (Figure 6A). Hematological and biochemical measurements (Figure 6B) indicated that all parameters were within the normal range and showed no significant differences from the control group (P > 0.05). Body weight monitoring further demonstrated no obvious abnormalities in body weight changes between the control and treatment groups throughout the administration period (Figure 6C). To further evaluate long-term biocompatibility, the administration period was extended to 30 days, and hematological parameters, serum biochemical indices, and hepatic and renal function were re-assessed in both groups; the results were consistent with those obtained at 14 days, showing no toxic damage to the mice (Figure S4). Collectively, these findings indicate that MEM@MPDA/API NPs at the current dose did not cause any toxic side effects in mice and exhibited favorable long-term biocompatibility, providing an important safety basis for subsequent research and clinical application.

Figure 6.

In vivo safety of MEM@MPDA/API NPs. (A) H&E staining analysis of heart, liver, spleen, lung, and kidney tissues from mice in the NP-treated group and normal mice(n=3). (B) Biochemical parameters of liver and kidney function in NP-treated mice and normal mice(n=3). (C) Body weight change trends of NP-treated mice and normal mice(n=3).

Single-Cell Sequencing Data Combined with Network Pharmacology Reveals the Mechanism of API in Treating Sepsis-Induced Lung Injury

To investigate the potential mechanism of MEM@MPDA/API NPs in treating ALI, this study analyzed single-cell sequencing datasets of mouse ALI from public databases. The results indicated that in the ALI model, the cell types with the highest proportions were various immune cells, such as B cells, neutrophils, T cells, and macrophages, and the proportion of M2 macrophages was higher than that of M1 macrophages (Figure S5A and B). This suggests that upon the onset of ALI, the body initiates an immune response and activates anti-inflammatory programs to maintain homeostasis.

This study found that mitophagy plays an important role in ALI, and lysosomes play an indispensable role in mitophagy.³¹ LRGs extracted from the GO database were intersected with DEGs and displayed by cluster (Figure 7A and B). The results showed that the Fth1 gene was differentially expressed across multiple cell types and ranked among the top in each; this gene primarily mediates iron metabolism and iron storage, and is a key link in antioxidant defense and iron recycling.³² Next were the inflammation-driving genes Il1b, as well as Atp6v0c and Chmp4b, key genes involved in maintaining lysosomal acidification and phagocytosis. The above results indicate that LRGs are altered in ALI.

Figure 7.

Single-cell transcriptomic analysis and molecular docking of API in the prevention and treatment of acute lung injury (ALI). (A) Cell clustering plot from single-cell sequencing of ALI mice. (B) Cell proportion in single-cell sequencing of ALI mice. (C) Schematic diagram of molecular docking between hub genes and API.

To further clarify the critical role of autophagy in ALI, we performed GSEA enrichment analysis on the mitophagy pathway using transcriptomic data from ALI mice in the GEO database, and the results indicated differential regulation of the mitophagy pathway in ALI (P = 0.017) (Figure S5C). Meanwhile, through network pharmacology analysis, we obtained 19 key intersecting genes, with the top seven scored being IL1B, IL6, ICAM1, IL1A, CCL2, and CXCL1 (Figure S5D and E), most of which are associated with inflammatory responses. Enrichment analysis of these 19 key genes revealed significant enrichment in multiple inflammation-related pathways, such as NF-κB, IL-17, P53, and MAPK signaling pathways (Figure S5F), providing a basis for the application of API in the treatment of inflammatory diseases. Finally, to explore the interactions between small active components and protein targets, molecular docking was performed using MOE 2019 software. The docking energy ranged from −4.6252 to −5.2175 kcal/mol; a docking energy value of < −4.25 kcal/mol indicates good binding affinity, while < −7.0 kcal/mol indicates strong binding affinity. The docking results showed that two groups exhibited good binding affinity and five groups showed moderate binding affinity, preliminarily validating the findings of the network pharmacology analysis (Figure 7C).

Based on the above single-cell database and GSEA analyses, we inferred that autophagy plays a critical role in ALI. To verify this finding, we employed lysosome tracking technology to stain inflammatory cells after MEM@MPDA/API treatment (Figure S5G). The results revealed significant colocalization between the NPs and lysosomes (Pearson’s coefficient = 0.667). This finding indicates that upon cellular entry, the NPs establish physical contact with lysosomes, suggesting that this physical contact may be responsible for promoting mitophagy and lysosomal processes.

MEM@MPDA/API NPs Promote Mitophagy in a Mouse Model of Septic Lung Injury

In the above studies, we found that mitophagy plays a critical role in the development and progression of sepsis-induced ALI, and that MEM@MPDA/API, upon cellular entry, establishes physical contact with lysosomes. To further investigate the activation of mitophagy in vivo, immunofluorescence staining was performed to detect the colocalization of LC3B (a autophagy marker) and TOMM20 (a mitochondrial outer membrane marker) in lung tissues. As shown in Figure 8A and B, compared with the LPS-induced ALI model group, the colocalization coefficients of LC3B and TOMM20 immunofluorescence in the lung tissues of ALI mice treated with API, MPDA, and MEM@MPDA/API successively increased, with the MEM@MPDA/API treatment group showing the highest colocalization coefficient (P < 0.0001). These findings indicate that MEM@MPDA/API can significantly promote the selective targeting of damaged mitochondria by autophagosomes, thereby initiating mitophagy.

Figure 8.

MEM@MPDA/API NPs promote mitophagy in a mouse model of septic lung injury. (A) Immunofluorescence staining of LC3B (red) and Tomm20 (green) in lung tissues from different treatment groups(n=6). (B) Quantitative analysis of the colocalization coefficient of LC3B and Tomm20 immunofluorescence in lung tissues from different treatment groups (n=6). (C) TEM images of MLE-12 cells in different treatment groups (Red arrows: mitochondria; Yellow arrows: autolysosomes; n=3). **P<0.01, ****P<0.0001.

Additionally, TEM was used to observe the ultrastructural changes of mitochondria in LPS-stimulated MLE-12 cells. The results showed that, compared with the control group, LPS stimulation caused mitochondrial swelling and deformation, with reduction or disappearance of mitochondrial cristae (red arrows), accompanied by the appearance of a small number of autolysosomes (yellow arrows). After treatment with MEM@MPDA/API, autophagic activity was significantly enhanced, as evidenced by the observation of numerous autolysosomes, indicating that autophagosomes had successfully fused with lysosomes and initiated the degradation process. Meanwhile, the cristae structures were clearly visible, and the mitochondrial morphology was essentially consistent with that of normal mitochondria (Figure 8C).

In summary, based on the LC3B/TOMM20 colocalization results and the TEM observations, we confirmed from both spatial localization and ultrastructural morphology perspectives that MEM@MPDA/API NPs can clear damaged mitochondria and restore mitochondrial morphology by activating the complete mitophagy process—from autophagosome targeting of mitochondria to eventual fusion with lysosomes for degradation—thereby alleviating lung injury.

MEM@MPDA/API NPs Promote Lung Barrier Repair in a Mouse Model of Septic Lung Injury

Mitochondria are not only the core organelles of energy metabolism but also key hubs in regulating inflammatory signaling. Mitochondrial damage can activate the NLRP3 inflammasome and the NF-κB signaling pathway, driving the sustained production of pro-inflammatory cytokines.³⁶ Therefore, the MEM@MPDA/API-induced mitophagy described above—by clearing dysfunctional mitochondria—may not only restore mitochondrial function but also cut off the sustained amplification of inflammatory signals at the source. On this basis, we further investigated the impact of mitophagy activation on lung barrier integrity.

The pulmonary barrier is a multi-layered defense system whose primary function is to protect the lungs from harmful substances while maintaining efficient gas exchange. Previous studies have shown that disruption of the pulmonary epithelial–endothelial barrier can lead to protein-rich pulmonary edema and uncontrolled inflammation, and that restoration of the lung barrier may serve as a therapeutic strategy to eliminate edema and alleviate the pro-inflammatory state. Tight junction proteins between pulmonary cells influence lung barrier function.⁸ The above in vivo experiments confirmed that MEM@MPDA/API NPs intervention could reduce tissue leakage. To explore the underlying mechanism, immunofluorescence and WB analyses were performed to detect the expression of lung barrier-related proteins. As shown in Figure 9A–D, immunofluorescence results revealed that, compared with the control group, the fluorescence signal intensities of the tight junction proteins ZO-1 and Occludin were significantly decreased in LPS-induced ALI (P < 0.0001), while they gradually increased after intervention with API, MPDA, and MEM@MPDA/API. Consistently, the Western blot results (Figure 9E–H) also directly showed that the expression of tight junction proteins ZO-1, Occludin, and VE-cadherin was reduced in ALI mice after LPS stimulation, and these protein expressions were rescued to varying degrees after various therapeutic interventions. In agreement with the immunofluorescence findings, the MEM@MPDA/API treatment group exhibited the most pronounced upregulation of ZO-1, Occludin, and VE-cadherin, with P values of 0.0001, 0.001, and 0.0001, respectively, compared with the LPS model group, indicating statistically significant differences.

Figure 9.

MEM@MPDA/API NPs promote lung barrier repair in a mouse model of septic lung injury. (A) Immunofluorescence staining of ZO-1 protein in lung tissues from different treatment groups. (B) Immunofluorescence staining of Occludin protein in lung tissues from different treatment groups. (C and D) Quantitative analysis of ZO-1 and Occludin immunofluorescence (n=6). (E) Western blot analysis of ZO-1, Occludin, and VE-cadherin protein expression in lung tissues from different treatment groups. (F–H) Relative protein expression levels of ZO-1, Occludin, and VE-cadherin in lung tissues from different treatment groups (n=6). ns, no significance; *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Therefore, through this study, we found that MEM@MPDA/API NPs promoted the recovery of the alveolar epithelial and vascular endothelial barrier, which may be one of the reasons for its ability to reduce pulmonary leakage and alleviate inflammation.

Discussion

ALI is characterized by core pathological features including oxidative stress imbalance, mitochondrial dysfunction, and disruption of the alveolar-capillary barrier, and currently lacks specific therapeutic options. Sepsis is one of the most common triggers of ALI, and its pathological process involves immunosuppression, coagulation dysfunction, and uncontrolled inflammatory responses.³⁷ Although the development of nanomedicine has provided new ideas for ALI treatment, how to achieve the functional synergy of “targeting inflammatory sites,” “scavenging reactive oxygen species and reducing oxidative stress,” and “repairing the lung barrier” within a single platform remains an urgent challenge to be addressed.

The biomimetic nanodelivery platform MEM@MPDA/API NPs constructed in this study integrates three functional advantages (Figure 10): the MPDA core provides high drug loading capacity and broad-spectrum ROS scavenging ability; API exerts anti-inflammatory activity; and the macrophage membrane surface is naturally rich in inflammation-homing receptors such as CCR2 and integrins, endowing the NPs with immune evasion and active targeting capabilities to injured lung tissue. This platform remains stable under normal physiological conditions while achieving efficient targeted delivery and synergistic therapy in the inflammatory microenvironment, a design that distinguishes it from conventional nanocarriers relying solely on the passive EPR effect for delivery.³⁸ The synthesis and characterization of MEM@MPDA/API NPs constitute the core work of this study, and successful synthesis was confirmed by TEM, FT-IR, and UV-Vis-NIR spectroscopy. In vitro experiments demonstrated that macrophage membrane camouflage enhanced the uptake of NPs by lung epithelial cells and their targeted accumulation in damaged lung tissue, attributed to the active recognition of upregulated MCP-1/CCL2 and VCAM-1 in the inflammatory microenvironment by chemokine receptors and integrins on the membrane surface. Functionally, MEM@MPDA/API NPs exhibited excellent ROS scavenging capacity both in vitro and in vivo, and effectively restored mitochondrial membrane potential. Network pharmacology analysis suggested that API can target ROS-mediated inflammatory pathways such as NF-κB and MAPK; consistent with this, we observed significant downregulation of the pro-inflammatory cytokines IL-6, TNF-α, and IL-1β at the mRNA level and in BALF after MEM@MPDA/API intervention.

Figure 10.

Schematic diagram of the mechanism of action of MEM@MPDA/API NPs.

The key mechanistic finding of this study is that MEM@MPDA/API NPs establish a complete causal chain from oxidative stress scavenging to lung barrier repair by triggering mitophagy. Specifically, LPS-induced excessive ROS not only directly damage mitochondrial structure (manifested as mitochondrial swelling, deformation, loss of cristae, and decline in membrane potential), but also prompt damaged mitochondria to continuously release ROS, which act as second messengers to activate the NLRP3 inflammasome and NF-κB pathway, driving the cascade release of pro-inflammatory cytokines, which in turn disrupt tight junction proteins (ZO‑1, Occludin, and VE-cadherin) between alveolar epithelial and vascular endothelial cells——a central molecular event underlying the loss of lung barrier function.

Within this pathological cascade, mitophagy plays a critical role as an “upstream brake.” Our experimental results demonstrated that after cellular entry, MEM@MPDA/API NPs establish physical contact with lysosomes, followed by a significant increase in colocalization of LC3B and TOMM20, indicating that autophagosomes are recruited to damaged mitochondria; the formation of autolysosomes observed under TEM confirmed that autophagosomes had fused with lysosomes and completed the degradation of dysfunctional mitochondria—thus, a complete mitophagic flux was activated. By clearing dysfunctional mitochondria, mitophagy cuts off the sustained amplification of inflammatory signals at the source.

The functional outcomes of mitophagy were manifested at three levels: first, the mitochondrial membrane potential and ultrastructure were restored; second, the pro-inflammatory cytokines IL-6, TNF-α, and IL-1β were significantly downregulated; and third, the expression of tight junction proteins ZO-1, Occludin, and VE-cadherin was rescued at the protein level, with recovery of immunofluorescence signals and reconstruction of alveolar-capillary barrier integrity, ultimately resulting in marked reductions in protein leakage and inflammatory cell infiltration in BALF.

Although the macrophage membrane biomimetic strategy endows the nanoplatform with active targeting and immune evasion capabilities, its clinical translation still faces challenges related to immunogenicity and biological variability. The former mainly stems from the potential host immune rejection elicited by allogeneic membrane proteins (eg, MHC molecules),³² while autologous sources could circumvent this risk but have limited universality; the latter is associated with donor-to-donor variability and the current lack of standardization in membrane extraction and coating procedures, which may lead to batch-to-batch heterogeneity. Future research should focus on establishing standardized production and quality control systems, as well as systematically evaluating long-term immunological safety. In addition, membrane orientation and long-term storage stability of membrane-coated nanoparticles are critical quality attributes affecting functional consistency. In this study, the successful membrane coating was confirmed through combined characterization by TEM, size, zeta potential, and WB; in the future, systematic membrane orientation analysis could be performed using methods such as flow cytometry.

This study has the following limitations. First, we did not use autophagy inhibitors (eg, chloroquine or bafilomycin A1) for quantitative assessment of autophagic flux; however, as mentioned above, the observation of autolysosomes by TEM has already provided morphological evidence of completed autophagic flux. Second, this study only preliminarily evaluated the short-term (14-day and 30-day) biosafety of the NPs, but the observation endpoints used still fall within the subacute category and are insufficient to infer lifelong or chronic safety. Systematic long-term biodistribution (eg, NP residues in various organs at 3–6 months post-injection) and chronic toxicity studies are still necessary to fully validate the safety of this platform for clinical translation. These limitations will be progressively addressed in future studies.

Conclusion

In this study, a macrophage membrane-biomimetic modified mesoporous polydopamine-loaded API nanodelivery platform (MEM@MPDA/API) was successfully constructed. By integrating the native inflammatory-homing receptors of macrophage membranes with the high drug-loading capacity and broad-spectrum ROS-scavenging property of the mesoporous MPDA framework, this platform effectively overcomes the classic bottlenecks of poor solubility and low bioavailability of API. Different from conventional nanocarriers that rely on the passive EPR effect, this biomimetic design endows NPs with the capability of active targeting to inflamed lung tissues. In vivo imaging verified that its lung accumulation was markedly superior to that of the non-membrane-coated control group. This study confirms that MEM@MPDA/API NPs possess outstanding anti-inflammatory and ROS-scavenging capacities. Furthermore, a 30-day in-vivo safety evaluation validated the favorable biocompatibility of this platform within the sub-acute time window.

This work not only provides a theoretical basis for biomimetic nanoplatform-mediated mitophagy modulation to treat oxidative stress-associated inflammatory diseases, but also demonstrates the translational potential of the macrophage membrane biomimetic strategy for targeted pulmonary inflammatory delivery. This synergistic therapeutic strategy integrating “active targeting-antioxidation-barrier repair” opens up new avenues for the design of nanomedicines against acute lung injury and other oxidative stress-driven critical illnesses.

Data Sharing Statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Author Contributions

All authors made a significant contribution to the work reported, whether in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; participated in drafting, revising, or critically reviewing the article; agreed on the journal to which the article will be submitted; reviewed and agreed on all versions of the article before submission, during revision, the final version accepted for publication, and any significant changes introduced at the proofing stage; and agree to take responsibility and be accountable for the contents of the article.

Yingying Zhang: Conceptualization, Methodology, Writing-original draft; Lizhen Huang: Methodology, Writing-original draft, Data curation; Hanghang Han: Methodology, Formal analysis, Investigation; Siyu Zhou: Formal analysis, Investigation; Wuling Liang: Writing-review & editing; Huiming Chen: Writing-review & editing; Xueke Du: Conceptualization, Supervision, Project administration, Funding acquisition, Writing-review & editing.

Disclosure

The authors declare that there are no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.