Amplifying bacterial cuproptosis by NIR driven biomimetic metal organic framework for nanocatalytic enhanced bacterial pneumonia immunotherapy

Abstract

The clinical therapy of infectious diseases such as bacterial pneumonia poses a major concern for the medical community. Traditional antibiotic therapies can be cytotoxic, and easy to develop bacterial resistance. Herein, we proposed a novel biomimetic nanoplatform called CM consisted of Cu based metal organic framework, and followed by being wrapped by macrophage membrane (MM). In vitro and in vivo results both confirmed that CM inherited the properties of original MM, specifically targeting the infectious site, achieved the near infrared (NIR) mediated Cu ions accumulation inside bacteria, and produced endogenous reactive oxygen species and inflammation related factors, ultimately resulting in the efficient bacteria elimination. In addition, transcriptomics and metabolomics revealed that CM + NIR induced intracellular Cu overload, thereby promoting bacterial cuproptosis. Significantly, CM + NIR roused immunoregulation to eliminate bacteria through inducing macrophage M1 polarization, and adjusting the number and proportion of immune cells. Altogether, CM + NIR promoted bacterial cuproptosis together with immunoactivation for all-stage bacterial elimination, resulting in precise bacterial pneumonia therapy, and also providing a generalized approach for targeted therapy of infectious diseases.

Graphical abstract

A biomimetic nanosized Cu based MOF (CM) was developed combining with photothermal therapy and macrophage re-rousing to achieve high efficiency and safety bacterial pneumonia targeting therapy via the synergistic enhancement of NIR amplified multiple enzymatic activities, bacterial cuproptosis, and macrophage chemotaxis and engulfment as well as immunoregulation activation.

Highlights

• •It developed a biomimetic nanosized MOF for bacterial pneumonia targeting therapy.
• •It effectively targeted the infectious microenvironment, and induced macrophage M1 polarization to kill bacteria.
• •Transcriptome and metabolome analysis validated the NIR amplified cuproptosis mediated antibacterial mechanism.
• •It enhanced the bacterial phagocytic ability of macrophages by CM + NIR.
• •It achieved the enhanced bacterial pneumonia immunotherapy through immunoactivation and bacterial cuproptosis.

1. Introduction

According to the World Health Organization (WHO), pneumonia accounts for 75% of all deaths caused by acute infections of respiratory system, and has become the number one cause of death in children under five years of age, as well as a major contributor to hospitalization and death in people over 65 years of age [1]. Bacterial pneumonia is one of the most common infectious diseases with high mortality, posing a serious threat to the health of children and the elderly. Antibiotic therapy is still the predominant therapeutic strategy of infectious diseases. However, the abuse of antibiotics, and the emergence of multi-drug resistant bacteria are easy to cause recurrent infections and persistent inflammation, ultimately leading to more serious pathological condition in the body such as sepsis and respiratory failure syndrome [2]. Thus, developing an innovative therapeutic strategy for efficient and safe treatment of bacterial pneumonia is an urgent clinical issue that needs to be addressed.

With the development of advanced nanotechnology, nanozymes with multiple mimicking catalytic activities of superoxide dismutase (SOD), catalase (CAT) and peroxidase (POD) have been applied in various biomedical fields [3,4]. Especially, due to their high efficacy of catalytic activities as well as unique chemical toxicity, nanozymes including nanosized metals [5] or metal oxides [6], metal organic frameworks (MOFs) [7], covalent organic frameworks (COFs) [8] and other nanocomposite [9] were designed to achieve antibacterial therapy. To further improve the efficacy of anti-infection therapy, the emerging adjuvant therapeutic techniques such as ultraviolet (UV), near infrared (NIR) [10], ultrasonic (US) [11], and magnetic field [12] stimuli have recently shown great potentials in eliminating bacteria while avoiding antibiotic resistance. The generation of exogenous heat, electric current or reactive oxygen species (ROS) contributes to destroying bacterial membranes, and causing oxidative stress in bacteria. However, continuous high levels of heat, electric current or ROS are essential to kill the bacteria, which also cause severe damage to normal tissue during therapy. Meanwhile, in tissue microenvironment, once the exogenous activation stops, the residual bacteria escaping from immunoregulation may proliferate and colonize in the previously infected sites, causing secondary infection, which is also the reason why the infectious diseases are persistent and easy to recur [13]. Therefore, an ideal antibacterial strategy is to adopt relatively mild methods to achieve the synergistic inhibition of bacterial proliferation, and also continuously exert their effects [14].

Copper (Cu) is one of the essential trace elements for the human body. Previous studies had confirmed that a certain amount of Cu was helpful to strengthening antioxidant and anti-inflammatory effects, promoting the synthesis of extracellular matrix proteins, and accelerating wound healing [15]. On another hand, Cu element has a strong bactericidal effect by the directly interaction with bacterial outer membrane, causing bacterial rupture and hindering its metabolism [16]. Significantly, due to the valence transition of Cu⁺/Cu²⁺ and existence of oxygen vacancy defects, Cu based nanomaterials can generate excessive ROS, and undergo oxidation reactions with the important components inside bacteria to activate oxidative stress related signaling pathways, thereby leading to the death of bacteria [17]. Furthermore, the imbalance of Cu homeostasis can lead to the abnormal accumulation of intracellular Cu ions concentration. The excessive Cu ions can bind to thiocyanate proteins in the tricarboxylic acid cycle (TCA), promoting abnormal oligomerization of thiocyanate proteins [18]. And Cu ions can also reduce the levels of Fe-S cluster proteins [19]. Thus, the above jointly induce proteotoxic stress response, inhibit mitochondrial metabolic function, and consequently promote bacterial cuproptosis [20], which could be enhanced by the external stimuli like NIR irradiation [21] and US stimulation [22] as well. Thus, the strategy of cuproptosis amplification may be an effective method for treating infectious diseases, which has attracted a lot of research attentions.

Despite all this, the poor targeting always affects therapeutic efficacy, and also causes toxic effects on normal cells. A large number of targeted nanosystems have been developed [23]. Among them, cell membrane biomimetic technology is committed to using cell membranes from red blood cells [24], tumor cells [25] or bacteria [26] as carriers to improve biocompatibility, and achieve long term blood circulation, and targeted delivery in vivo. For example, once the body is stimulated by infection or trauma, the release of inflammatory factors guides immune cells including macrophages to infiltrate the disease area. A lot of macrophages could be produced in the inflammatory site. In the meantime, macrophage could recognize diverse foreign bacteria via pattern recognition receptors (PRRs), particularly toll-like receptors (TLRs) during pulmonary infection [27]. Significantly, when macrophages are activated by diversified bacteria, the corresponding expression of membrane receptors has changed, which contributes to the precise bacteria recognition capabilities of macrophages [28]. Thus, nanosystems encapsulating with macrophage membrane (MM) could be enriched in the inflammatory sites due to cell homing effect, thereby achieving specific targeted therapy.

Inspired by the above, we developed a biomimetic nanozyme (CM) through encapsulating Cu based MOF (CB) with MM in combination with NIR irradiation for bacterial pneumonia therapy. The specific receptors on CM facilitated it recognizing both bacteria and macrophage, and accumulating in the infected sites rather than normal tissue, which reduced the cytotoxicity to normal cells. Under NIR irradiation, CM retained the multiple catalytic functions of CB, generated a huge amount of ROS, and induced intracellular Cu overload, resulting in bacteria death. Transcriptomics and metabolomics revealed that the bactericidal capacity of CM + NIR was mainly due to the fact that the intracellular accumulation of Cu could inhibit bacterial TCA cycle, disrupt bacterial membrane function, and limit bacterial substance metabolism, biosynthetic processes and energy metabolism, ultimately causing bacterial cuproptosis. Simultaneously, CM + NIR could effectively trigger the macrophage M1 directional polarization, and adjust the number and proportion of immune cells to achieve immunoactivation, leading to kill invading bacteria through enhanced chemotaxis and phagocytosis. The proposed synergistic therapeutic strategy of NIR triggered cuproptosis amplification and immunoregulation activation efficiently achieved the treatment of bacterial pneumonia, and inhibited its recurrence (Fig. 1). Finally, it provided a new reference for the existing treatment modalities for bacterial pneumonia, and was expected to address the pressing concern of antibiotic resistance.

Fig. 1.

The schematic illustration of the preparation of a biomimetic nanosized Cu based MOF (CM), and its in vivo bacterial pneumonia therapy. By intravenous (IV) injection, the strategy of CM combining with NIR irradiation could achieve the high efficiency and safety bacterial pneumonia immunotherapy via the synergistic mechanisms of NIR amplified multiple enzymatic activities, bacterial cuproptosis, macrophage chemotaxis and engulfment as well as immunoregulation activation.

2. Experimental section

2.1. Material and chemicals

Cupric chloride dehydrate (CuCl₂·2H₂O, ≥99.9%), 2-amino-terephthalic acid (BDC, ≥98%) and N, N-dimethylformamide (DMF, >99.9%) were commercially procured at Macklin Inc. (China). Ethanol (≥99%), phosphate buffered saline (PBS) and dimethyl sulfoxide (DMSO, ≥99.9%) were purchased from Solarbio (China). Hydrogen peroxide (H₂O₂, 30%) were supplied from Junobio (China). 4’, 6-diamidino-2-phenylindole (DAPI, ≥99.9%), RIPA lysis buffer and Triton were purchased from Beyotime (China). All reagents were directly used without any treatments.

2.2. Preparation and physicochemical characterization

CM was prepared by encapsulating CB with MM. Initially, 0.173 mmol CuCl₂ was dissolved in 8 mL DMF with ultrasonic dispersion. And 0.519 mmol BDC was also dispersed in 8 mL DMF. Two solutions were mixed and stirred with magnetic fields for 30 min. The mixture reacted at 140°C for 5 h before cooling down. Later, the mixture was centrifuged at 12000 rpm for 15 min, and re-dispersed in deionized (DI) water for 3 times. Finally, the precipitate was isolated by vacuum drying overnight to obtain CB. On the other hand, the cultured mouse monocyte macrophage leukemia cell line (RAW264.7, American Type Culture Collection (ATCC)) was washed with PBS for 3 times, and re-suspended with PBS. And the MM was sonicated using an ultrasonic cell disruptor. After centrifuge at 4°C and 3000 rpm for 10 min, the supernatant was collected to remove unbroken cells and nuclei. Then, after centrifuge at 15000 rpm for another 30 min to remove the supernatant, the precipitate was re-suspended in PBS. After repeating the above procedures for 3 times, the final precipitate was suspended in PBS to obtain MM suspension. The preparation of CM was implemented as following: 200 mg CB was dispersed in PBS with the concentration of 4 mg/mL, and mixed with 200 mL MM suspension (2 × 10⁷ cells/mL). The mixture was sonicated in an ultrasonic shaker for 15 min, and extruded through polycarbonate porous membrane (200 nm, Merck Millipore, Germany) to obtain a crude product. After centrifuge and re-dispense with DI water for 3 times, the final precipitate was collected and isolated by freeze drying to obtain CM. The detailed recipe for CB and CM preparation was listed in (Table S1).

To investigate the chemical structure and crystallization structure of CB, ultraviolet-visible spectroscopy (UV-vis, PerkinElmer Lambda 950) and X-ray diffraction (XRD, MiniFlex 900, Japan) were applied respectively. And the morphology and element distribution, element composition, and Cu element contents of CB were characterized by transmission electron microscopy (TEM) coupled with energy dispersive X-ray spectroscopy (EDS) (Hitachi, Japan), X-ray photoelectron spectroscopy (XPS, ESCALAB 250XI+, USA) and inductively coupled mass spectrometry (ICP-MS, Thermo, USA) respectively. Specifically, the fitting criteria for distinguishing Cu⁺ and Cu²⁺ were mainly based on the strong peaks in the range of 930∼945 and 960∼965 eV unique to Cu²⁺, and the differences in binding energy between the main peaks of Cu2p_3/2 (Cu⁺: 932.5 eV and Cu²⁺: 933.8 eV). And the detailed parameters were obtained by using Shiriely background and Gaussian Lorentz function for peak fitting. In addition, the specific surface area of CB was analyzed by Brunauer, Emmett and Teller gas adsorption method (BET, Micromeritics ASAP, USA).

To confirm the successful fabrication of CM, the fluorescent co-localization was applied. In brief, the extracted MM was labeled with DIO488 (C1038, Beyotime, China), and CB was labeled with thiol poly(ethylene glycol) cyanine 5 (HS-PEG-Cy5, Lumiprobe, China) (Cy5-CB) before fabricating CM. After incubating with RAW264.6 overnight at 4°C, the treated cells were washed with PBS for 3 times, and stained with DAPI for 10 min before observation by confocal laser scanning microscopy (LSM 980, ZEISS, Germany). Besides, the proteins of CB, MM and CM were extracted by RIPA lysis buffer, and analyzed by protein gel electrophoresis (80 V, 2.5 h). Significantly, the CM was prepared for 3 times in a repeated manner, and their proteins were also extracted, and analyzed by protein gel electrophoresis to verify the repeatability of CM preparation. To further characterize the MM encapsulation of CM, Fourier transform infrared spectrometer (FTIR, Shimadzu, Japan), Raman spectrum (Raman, Thermo Scientific, US), thermal gravimetric analysis (TGA, STD650, TA, USA), and zeta sizer (Nano ZS90, Malvern, UK) were applied to characterize the corresponding molecular structure, thermal stability and zeta potential respectively. In specific, the stability of CM was investigated via monitoring the changes of its size and zeta potential in PBS or Dulbecco's modified eagle medium (DMEM, Solarbio, China) versus time by zeta sizer. In the meantime, the Cu ions release rate of CM was also investigated. In details, CM was immersed in different solutions (pH = 7.4, pH = 5.0 and 5 mM H₂O₂) with shaking. At predetermined time points (0, 4, 8, 12, 24, 48 and 72 h), the Cu contents in the solutions were tested by Cu ions detection kit (Solarbio, Beijing, China) following the manufacture’ s introduction.

2.3. In vitro dispersibility and photothermal effects investigation

The dispersion and stability of CB and CM was implemented by dispersing them in PBS, DMEM, Luria-Bertani medium (LB medium, Biosharp, China), fetal bovine serum (FBS, Solarbio, China), or 5 mM H₂O₂ for a certain of time. And the images were collected at predetermined time points (0, 0.5, 1, 2, 4, 8, 24 and 48 h) by digital camera. In addition, the in vitro photothermal effects were implemented by placing 100 μg/mL CB or CM under NIR irradiation (808 nm, 1 W/cm²), CM with different concentrations of 50, 100 and 200 μg/mL under NIR irradiation (1 W/cm²), and 100 μg/mL CM under NIR irradiation of different power intensities (0.5, 1 or 1.5 W/cm²) for 10 min. Specifically, for photothermal stability testing, 100 μg/mL CM was irritated with 4 repeated “on” and “off” cycles under NIR light (1 W/cm²). The corresponding images and temperatures were collected and recorded by thermal imaging camera (FLIR, USA). And the photothermal transition ratio (η) was calculated as: η = hS(T_max-T_surr)-Q_dis/P(1-10^-A808), where h was the heat transfer coefficient, S was the contact area of laser irradiation, T_max was the highest temperature during NIR irradiation, T_surr was the ambient temperature, Q_dis was the heat released of DI water under NIR irradiation, P was the power intensity, and A808 was the absorbance of material at 808 nm [29].

2.4. ROS generation ability testing

The ROS generation ability was initially tested by electron spin resonance (ESR, Bruker A300, Germany). 5-tert-butoxycarbonyl 5-methyl-1-pyrroline-N-oxide (BMPO, 100 mM), xanthione (10 mM) and xanthione oxidase (XOD, 1 U/mL), and 2, 2, 6, 6-tetramethylpiperidine (TEMPONE, 100 mM) were applied as the working solutions for ·OH, ·O₂⁻ and ¹O₂ generation ability testing respectively. After mixing with 100 μg/mL CM for 10 min, the corresponding signal was recorded by ESR. Particularly, for CM + NIR, NIR irradiation (808 nm, 1 W/cm²) was implemented for 10 min after mixing the working solution with CM. To further test the ROS generation capacities, ROS testing kits were applied by following the protocols. Briefly, 100 μg/mL CM was mixed with the corresponding ROS testing kits (Solarbio, China) for ·OH and ·O₂⁻ generation testing respectively. After incubation at 37°C for 30 min, the mixture was centrifuged, and the supernatant was observed at 536 and 530 nm. Specifically, for NIR alone and CM + NIR, NIR irradiation (1 W/cm²) was implemented for 10 min after mixing.

Besides, the catalase (CAT), peroxidase (POD), glutathione peroxidase (GSH-Px) activity kits (Solarbio, China) were applied to investigate the multiple enzymatic activity by following the instruction of manufactures. In details, for CAT activity testing, CM of different concentrations (50, 100 and 200 μg/mL) were dispersed in the corresponding working solutions. After incubation at 37°C for 30 min with magnetic stirring, the supernatant of solutions was observed at 240 nm by microplate reader (Thermo Scientific, USA). Similarly, the optical density (OD) of corresponding supernatant was recorded at 470 and 412 nm to reflect POD and GSH-Px activity respectively. In specific, for NIR alone and CM + NIR, NIR irradiation was implemented for 10 min (808 nm, 1 W/cm²) during mixing before observation. In addition, to obtain the optimum incubation time, the time dependent CAT and POD activities was also tested. After mixing CM with the corresponding working solutions, the mixture was incubated for predetermined time points at 37°C with magnetic stirring before observation by microplate reader. Significantly, to investigate the enzymatic activity under different conditions, 20 μg/mL CM was mixing with the working solutions of pH = 7.4, 6.2 or 5.0 respectively. After incubation at 37°C for 30 min, the mixture was observed by microplate reader. Meanwhile, 100 μg/mL CM was mixing with the corresponding working solutions, and incubated at 37°C or 54°C for 30 min before observation. Finally, the O₂ generation amount was also tested. In brief, 0.8 M H₂O₂ was prepared in DI water, followed by the addition of 100 μg/mL CM or CM + NIR. The dissolved oxygen levels of the solutions were ultimately detected by dissolved oxygen meter (Thermo Scientific, US) after incubation for 10 min. For NIR alone and CM + NIR, NIR irradiation (1 W/cm²) was implemented for 10 min after mixing.

2.5. Antibacterial ability testing

To investigate antibacterial ability, Gram (+) bacteria: Streptococcus pneumoniae (Spn) and Gram (−) bacteria: Escherichia coli (E. coli) were chose as the research target. The bacteria were inoculated into LB medium with the concentration of 10⁶ CFU/mL, and mixed with 100 μg/mL CB or CM overnight. Herein, for antibacterial testing, no H₂O₂ was added. In specific, the bacteria were also incubated with LB medium of different pH (pH = 7.4, 6.2 and 5.0) with or without 100 μg/mL CM to investigate the antibacterial ability. And for NIR alone and CM + NIR, NIR irradiation was implemented with the power intensity of 1 W/cm² for 3 times, once every hour, 10 min per time. Afterwards, 100 μL mixture was added to an agar plate, and incubated at 37°C for another 12 h. And the number of bacteria was calculated to quantify the antibacterial ratio.

Next, the treated bacteria were washed against with PBS, and stained with SYTO-9 and PI live and dead bacteria stain kit (Thermo Fisher, USA) for 20 min in the darkness. After washing with PBS for 3 times, the bacteria were fixed with PFA, and observed by confocal scanning microscopy. The corresponding images were quantified by Image J. And the bacteria suspension was also quantified analyzed by flow cytometry (Beckman Coulte, US). Besides, the treated bacteria were fixed with 4% polyformaldehyde (PFA, Servicebio, China) after washing with PBS for 3 times, and sequentially dehydrated by ethanol with different concentration gradients (30%, 50%, 70%, 90% and 100%) for 10 min. After gold coating, the bacteria sample was observed by scanning electron microscopy (SEM, JSM-T300, Japan).

The intracellular ROS levels of bacteria were investigated by 2′, 7′-dichlorodihydrofluorescein diacetate (DCFH-DA, maokangbio, China) staining. In brief, the treated bacteria were incubated with fresh LB medium containing DCFH-DA for 30 min. After rinsing with PBS for 3 times, the bacteria were fixed with PFA, and observed by confocal scanning microscopy. Significantly, the total RNA of treated bacteria from control group and CM + NIR was extracted by TRIzol kits (Beyotime, China) for transcriptome and metabolome sequencing analysis. And the corresponding analysis including correlation assessment, differential expression genes (DEGs), genes ontology (GO), kyoto encyclopedia of genes and genomes (KEGG), and differential metabolites analysis was also implemented.

To confirm the bacterial cuproptosis mechanism, the activities of respiratory chain complex I (NADH dehydrogenase) and complex II (reduced flavin dinucleotide) were measured by using a mitochondrial respiratory chain complex activity assay kit (Solarbio, Beijing, China) according to the manufacturer's instructions. And the intra-bacterial Cu ion concentration was tested by Cu ions detection kit following the protocol. In addition, the metabolite of lipid peroxidation was also analyzed. The MDA detection kit (Solarbio, Beijing, China) was applied to determine the malondialdehyde (MDA) content, an indicator of lipid peroxidation, of treated bacteria.

Significantly, to further confirm the cuproptosis mechanism of bacteria, western blotting (WB) was implemented to detect the pathway related proteins’ expression levels of treated bacteria. Briefly, the treated bacteria were washed against with PBS, and lysed with lysis buffer containing protease inhibitors and phosphatase inhibitors (Beyotime, China). After centrifuge at 12000 rpm for 15 min, the supernatant was collected to test the protein concentrations by BCA protein assay kit (Beyotime, China). And the protein samples were separated by protein gel electrophoresis, and transferred to polyvinylidene difluoride (PVDF, Millipore, USA) membrane. After blocking for 2 h, and rinsing with tris buffered saline with Tween-20 (TBST, Servicebio, China) for 3 times, the membrane was incubated with primary antibody (anti-FDX1 and actin, Proteintech, China) overnight at 4°C. After washing against with TBST for 3 times, the membrane was incubated with the secondary antibody (goat anti-rabbit, Beyotime, China) for another 2 h. Finally, the strip was immersed in 1 mL of UltraSignal ECL WB detection reagent (Beyotime, China), and observed by automatic chemiluminescence image analysis system (ODYSSEY, USA).

2.6. Biological functions investigation in cellular levels

RAW264.7 and human normal lung epithelial cells (BEAS-2B) were commercially obtained from American type culture collection (ATCC, USA). For cell culture, DMEM supplemented with 10% FBS, 100 U/mL penicillin (Biosharp, China), and 100 μg/mL streptomycin (Biosharp, China) was applied. And cells were incubated at 37°C in a humid atmosphere containing 5% CO₂. The medium was changed every 2 days, and passaged when reaching 90% confluence. And the third passage was applied for further experiment. The cell viability was investigated by cell counting kit-8 (CCK-8, Biyuntian, China) assay. In details, cells were seeded in 96 well plate with the density of 10⁴. After incubating with CB or CM with the concentrations ranging from 0 to 500 μg/mL for 24 h, the cells were washed with PBS, and added with 100 μL CCK-8 solution. After 1.5 h, the supernatant was observed at 450 nm by microplate reader.

Besides, RAW264.7 was also incubated in 12 well plate with the density of 5 × 10⁴. After induced by 20 ng/mL IL-4 for 24 h [30], the cells were incubated with 100 μg/mL CB or CM for another 24 h. After treatments, the cells were stained with Calcein AM/propidium iodide (PI) kit (Biyuntian, China) following the protocols, and observed by fluorescent microscopy. And the cellular uptake was implemented by incubating RAW264.7 with 100 μg/mL Cy5-CB or Cy5-CM for predetermined time points (0, 3 and 6 h). After washing against with PBS for 3 times, the cells were stained with DAPI for observation by confocal scanning microscopy. Specifically, for preparing Cy5-CB or Cy5-CM, 100 mg CB or CM was dispersed in PBS followed by mixing with 10 mg Cy5. The reaction was implemented overnight before washing with PBS. The final Cy5-CB or Cy5-CM was obtained after vacuum drying. To verify the specific targeting of CM, RAW264.7 was initially treated with 1 μM lipopolysaccharide (LPS, Solarbio, China) for 30 min, and then inhibited with or without Toll-like receptor 4 (TLR4, AmyJet Scientific, China) for 10 μM before incubating with 100 μg/mL Cy5-CM for 6 h. The treated cells were finally observed by confocal scanning microscopy after DAPI staining.

RAW264.7 was seeded in 6 well plate with the density of 2 × 10⁵ per well. And the cells were induced with 20 ng/mL IL-4 for 24 h, and the medium was replaced with fresh medium containing 100 μg/mL CB or CM for another 24 h. Specifically, NIR irradiation (808 nm, 1 W/cm²) was implemented at 0, 1 and 2 h respectively for 10 min after incubating with CM. At the beginning, the supernatant was collected and centrifuged at 3000 rpm for 20 min to remove cell debris. The relative factors (TNF-α, IL-6 and IL-10) expression levels were quantified analyzed by enzyme-linked immunosorbent assay (ELISA, MEIMIAN, China) following the introduction of manufacture. Next, the cells were transferred to centrifuge tubes, and wash against with PBS. And then the cells were incubated with APC anti-mouse CD86 antibody and PE anti-mouse CD206 antibody (Biolegend, USA) at 4°C for 0.5 h. After washing against with PBS for 3 times, the cells were analyzed by flow cytometry. In addition, the ROS levels of treated cells were investigated by ROS testing kit. After staining with DCFH-DA (10 μM) overnight, the cells were washed with PBS for 3 times before being observed by fluorescent microscopy. Furthermore, the treated cells were fixed with PFA at 4°C for 0.5 h, and treated with 0.15% Triton X-100 for 15 min before blocking with the blocking solution (Beyotime, China) for 15 min. After washing with PBS for 3 times, the cells were incubated with primary antibodies (anti-CD86, anti-CD206, anti-iNOS, anti-Arg-1, anti-HSP70 or anti-CD31, 1: 500 dilutions, Proteintech, China) respectively at 4°C overnight. And the cells were incubated with FITC-anti-rabbit IgG (H&L) (Boster, China) in the darkness followed by staining with DAPI. And then the cells were observed by confocal scanning microscopy after washing with PBS for 3 times. Finally, the inflammatory genes (IL-6, iNOS and TNF-α) expression levels were analyzed by reverse transcription-quantitative real-time PCR (RT-qPCR). In brief, the total RNA of treated cells were extracted by RNA extraction kit (Magen, China), and RT-qPCR was implemented by using LightCycler® System (Roche, Switzerland). And the corresponding genes levels were analyzed by 2^−ΔΔCt method, and compared with glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The detailed primer sequences were listed in Table S2.

Significantly, the migration ability of RAW264.7 was also evaluated. The cells (2.5 × 10⁵ per well) were inoculated into an 8 μm transwell chamber, and the medium was replaced by fresh medium containing CB or CM (100 μg/mL). For CM + NIR, NIR irradiation (808 nm, 1 W/cm²) was implemented for 3 times, once every 10 min. And then the cells were incubated for 24 h, and stained with crystal violet (Beyotime, China) before observation. Besides, the phagocytic ability of RAW264.7 was also measured. Spn or E. coli were initially labeled with FITC labeled actin (Beyotime, China), and mixed with treated cell suspension into 12 well plate. The mixture was incubated at 37°C for 3 h, and washed with PBS for 3 times before fixing with PFA. At last, the cells were stained with actin-tracker Red-555 (Beyotime, China) and DAPI before observation by confocal scanning microscopy.

2.7. In vivo therapy evaluation

In vivo experiment was approved by the ethics committee of animal experiments of the first affiliated hospital of Guangxi Medical University (No. 2025-E0936). SD rats (180∼220 g) were operated by the local guide for the care and use of laboratory animals of Guangxi Medical University. The bacterial pneumonia animal models were built as follows. In brief, the mixture of Spn and E. coli (0.5 mL, 10⁸ CFU/mL) was inoculated into the lung of SD rats by intratracheal (IT) instillation. After 48 h, the animal models were completely established for further experiment.

To evaluate lung targeting ability and in vivo biodistribution, small animal imaging systems (IVIS, Pekin Elmer, US) were applied after intravenous (IV) injection of CB or CM. Briefly, after anesthesia, the normal rats or bacterial pneumonia rats were IV injected with 0.8 mL sample solutions (2 μg/mL Cy5, 100 μg/mL Cy5-CB or 100 μg/mL Cy5-CM) respectively. At predetermined time points (0, 0.5, 1, 2, 4, 8, 12 and 24 h), the major organs including heart, liver, spleen, lung and kidney were collected and imaged by IVIS. Additionally, the in vivo photothermal property was also evaluated as follows. Briefly, 0.8 mL CM (100 μg/mL) was IV injected into SD rats. After 30 min, the lung site of rats was irradiated by NIR light (1 W/cm²) for 10 min, and the corresponding images and temperatures were recorded at predetermined time points by thermal camera.

Furthermore, hemolysis testing was applied to assess the blood compatibility of CB and CM. Fresh blood was collected from rats, and centrifuged at 3000 rpm for 10 min to obtain red blood cells (RBC). And 300 μL RBC solution was mixed with 1 mL PBS containing CM with the concentration of 0, 5, 10, 20, 50, 100 and 200 μg/mL, where DI water was used as positive control. After incubating for 4 h, the mixtures were centrifuged at 11000 rpm for 10 min, and observed at 550 nm by UV-vis spectroscopy. And in vivo biosafety was also evaluated by IV injection of CM every 2 days. The body weight of rats was monitored every day. After 14 days, the rats were sacrificed, and their blood was collected for blood routine testing (blood analyzer, BC-6800Plus, Mindray, China), biochemistry testing (biochemical analyzer, 7600, HITACHI, China), and coagulation function testing (coagulation detection analyzer, ACL TOP750, Werfen, China). Additionally, the major organs including heart, liver, spleen, lung and kidney of rats were collected, sectioned and stained by hematoxylin and eosin (H&E) before observation.

After establishing the animal models for 2 days, the rats were IV injected with CB or CM (0.8 mL, 100 μg/mL) at day 0, 2 and 4. In specific, for CM + NIR, NIR irradiation (808 nm, 1 W/cm²) was implemented at 1, 2 and 3 h for 10 min after IV injection. At day 7, rats were sacrificed, and the corresponding blood and major organs were collected for further evaluation. At the beginning, the major organs were collected and imaged by digital camera. And the skin at the lung site was photographed at 0, 0.5, 1, 2, 4 and 8 h after day 7. And the skin at 8 h was fixed, sectioned and stained by H&E before observation. Next, the fresh blood was analyzed by blood routine testing, biochemistry testing and coagulation function testing respectively. And in vivo antibacterial ability was also evaluated by centrifuging the lung homogenate to collect the supernatant for bacterial spreading testing. Next, the inflammatory factors (IL-6, TNF-α and IL-10) expression levels of the lung homogenate was tested by ELISA following the protocols. Besides, the myeloperoxidase assay kit (MPO) and MDA levels of lung homogenate were investigated by MPO and MDA detection kits (Solarbio, China) by following the instruction of manufacture. And then the major organs were collected, sectioned, and stained by H&E for observation. Meanwhile, the Cu contents in the major organs of treated rats were also analyzed by ICP-MS after therapy.

Specifically, for lung tissue, it was sectioned, fixed by 4% PFA for 24 h, and stained with DCFH-DA and DAPI to evaluate the ROS levels. Meanwhile, for immunohistochemical and immunofluorescent staining, the slides were initially incubated with primary antibody (rabbit polyclonal anti-CD86, anti-CD206, anti-HSP70 or anti-VEGF, Zenbio, China) at 4°C overnight before washing against with PBS for 3 times. And the slides were bound with biotinylated secondary antibodies (Servicebio, China). After sealing, the slides were observed by Olympus microscope, and the corresponding AOD was analyzed by Image J. For immunofluorescent staining, the slides were observed by confocal laser scanning microscopy, and the corresponding MFI was also analyzed by Image J. Most importantly, to analyze T cells immunoactivation, the fresh blood, lung and spleen were homogenized with PBS, and washed with red blood cell lysis buffer (BioLegend, USA) to remove RBC. After staining with FITC labeled anti-CD4 (BioLegend, USA), PE labeled anti-CD8 (BioLegend, USA), and APC labeled anti-CD3 (BioLegend, USA) respectively, the cell suspension was analyzed by flow cytometry. Simultaneously, the lung tissue was stained with primary antibody (anti-CD4⁺ and anti-CD8⁺ (Proteintech, China)) following the above procedures for immunofluorescent observation.

2.8. Statistical analysis

All experiments were implemented at least triplicates. All data were analyzed by GraphPad Prism software, and expressed as mean ± standard deviation. Additionally, one-way ANOVA followed by least significant difference (LSD) analysis was conducted for all required data.

3. Results and discussion

3.1. Preparation and physicochemical characterization

To address the clinical challenge of high morbidity and mortality in bacterial pneumonia, we originally developed a novel MM encapsulated Cu based MOF, defined as CM with the ability to targeted induce bacterial cuproptosis in synergy with immunoactivation, leading to efficiently bacterial elimination, and lung tissue repair. As shown in Fig. 2a, CB was fabricated by hydrothermal reaction of Cu²⁺ and BDC at 140°C. The stable framework structure of CM was formed by the coordination reaction between carboxylic acid groups of BDC and Cu ions (Fig. S1). To provide sufficient Cu ions for cuproptosis, CB was chose as the core material, which also acted as a nanozyme with multiple catalytic activities. Besides, it also possessed the excellent photothermal conversion behavior as well as controlled release ability of Cu²⁺. To ensure its specific targeting, CB was coated with MM through a typical cell membrane biomimetic technology as previously reported [31], finally resulting in the formation of CM.

Fig. 2.

The preparation and physicochemical characterization of CB and CM. a) The synthesis procedure of CM. b) XRD results of CuCl₂, BDC and CB. c) TEM-mapping images of CB, and the corresponding element composition (HAADF, C, N, O and Cu images). (Scale bar: 50 nm) d) XPS results of CB (full spectrum, and C, N, O and Cu detailed spectrum). e) BET results of CB. f) The immunofluorescent co-localization images of MM shell (DiO, green), Cy5-CB core (Cy5, red) and nuclei (DAPI, blue) by confocal microscopy. (Scale bar: 5 μm) g) The protein gel electrophoresis image of MM, CB and CM. h) TGA results of CuCl₂, BDC, CB and CM. i) Zeta potential of CuCl₂, BDC, CB, MM and CM. j) The Cu release rate of CM under different conditions (pH = 7.4, pH = 5.0 and 5 mM H₂O₂). k) Photothermal images of CB and CM with the same concentration of 100 μg/mL under NIR irradiation (1.0 W/cm²) versus time. l) Temperature changes of PBS and CB and CM with the same concentration of 100 μg/mL under NIR irradiation (1.0 W/cm²) (i), different concentrations (50, 100 and 200 μg/mL) of CM under NIR irradiation (1.0 W/cm²) (ii), and 100 μg/mL CM under different power intensity of NIR irradiation (0.5, 1 and 1.5 W/cm²) (iii) versus time, and photothermal stability of 100 μg/mL CM under NIR irradiation (1.0 W/cm²) for 4 “on” and “off” cycles (iv). m) The ROS generation ability of CM and CM + NIR by ESR: ·OH (i), ·O₂⁻ (ii) and ¹O₂ (iii). n) The multiple enzymatic activities of CM and CM + NIR with the concentration of 100 μg/mL (i, iii and v), and CM with different concentrations (ii, iv and vi) by POD, CAT and GSH-Px testing kits. o) The multiple enzymatic catalytic mechanism of CM.

As displayed in Fig. S2, the color was blue and khaki for CuCl₂ and BDC respectively. After forming CB, it became brownish black. From UV-vis spectrum, no apparent peaks existed in CuCl₂, while the apparent peak was observed at 342 nm for BDC, attributed by the conjugated double bonds in benzene. However, the corresponding UV absorbance peak disappeared after forming CB (Fig. S3). There were no transmittance curves observed due to the successful formation of CB, which was insoluble in DI water. Besides, XRD was applied to characterize CB. As illustrated in Fig. 2b, the apparent new formed peaks were observed at 11.9° (002), 16.8° (001) and 24.8° (003) for CB respectively compared to those of CuCl₂ and BDC, which was consistent with the previous results [32]. The obvious new crystallization structure was observed for CB, indicating the formation of MOF structure. Besides, the morphology and element composition of CB were also characterized by TEM and XPS. It was spherical with the diameter of around 50.3 ± 3.2 nm for CB. From mapping results, C, N, O and Cu elements were obviously observed for CB (Fig. 2c). Similarly, from XPS results, it was also observed that C, N, O and Cu elements existed in CB (Fig. 2d), also confirming the successful preparation of CB. By peaks fitting, there were 3 types of C1s, 1 type of N1s, 2 types of O1s, and multivalent Cu2p existed in CB. Significantly, after normalization analysis, it displayed the corresponding peaks of Cu⁺ and Cu²⁺ in CB (Fig. S4 and Table S3). By calculation, the Cu⁺/Cu²⁺ ratio was 0.61 (Table S4), indicating the obvious redox ability [33]. From ICP-MS results, the Cu element content was 25.64 ± 0.32% for CB (Table S4). Meanwhile, the porosity of CB was characterized by BET. As indicated in Fig. 2e, it displayed the obvious adsorption and desorption isotherms for CB, indicating the formation of amorphous structures of MOF with the typical H3-type hysteresis [34]. After statistical analysis, the pore size, pore volume and specific surface area was 8.06 nm, 0.10 cm³/g and 49.49 m²/g for CB (Table S5).

After encapsulating CB with MM to form CM, CM and CB were also characterized together. The obvious overlap of green fluorescence from MM and red fluorescence from CB was displayed in Fig. 2f, which gave a proof of the successful formation of CM. And from the result of protein gel electrophoresis, there were no protein bands for CB alone. And it displayed the similar bands for CM and MM (Figs. S5 and 2g), also proving the successful MM encapsulation of CM. Significantly, after preparing CM for repeated 3 times, it displayed almost the similar protein bands (Fig. S6), indicating the excellent reproducibility by this preparation method. In the meantime, their corresponding molecular structure was characterized. As illustrated in Fig. S7, there were a broad curve around 3310∼3700 cm⁻¹ in CuCl₂, and a number of chaotic curves in the range of 500∼1800 cm⁻¹ and obvious sharp curves at 3240 and 3620 cm⁻¹ in BDC. And the chaotic curves still retained in the range from 500 to 1800 cm⁻¹, and obvious broad curves were also observed at 3310 and 3700 cm⁻¹ in CB, indicating that CB was prepared by the reaction between CuCl₂ and BDC. After MM coating, almost no changes happened to CM, consistent with those of CB. Additionally, CB and CM were also characterized by Raman spectroscopy. The obvious D and G bands were observed at 1340 and 1570 cm⁻¹ for CB, and no significant changes happened to CM after MM coating (Fig. S8).

Furthermore, the thermal stability of CB and CM was investigated. As displayed in Fig. 2h, all chemicals maintained a certain of thermal stability at the beginning. When reaching above 300°C, it started to lose weight for all chemicals. When the temperature reached 700°C, the weight remaining ratio was 0.48%, 9.36%, 36.21% and 26.12% for CuCl₂, BDC, CB and CM respectively. After calculation, the encapsulation ratio of MM was 10.09% for CM (Table S6). And the stability of CB and CM was also characterized by zeta sizer. From the results of zeta potential, it was 3.62 ± 0.57 mV and −12.43 ± 0.31 mV for CuCl₂ and BDC. After forming CB, the zeta potential became −9.54 ± 0.95 mV. And the zeta potential was −17.77 ± 0.74 mV for MM. After encapsulating with MM, it became −23.63 ± 0.60 mV for CM (Fig. 2i and Table S6). The higher absolute value of zeta potential always contributed to the good stability and dispersion. Significantly, as displayed in Fig. S9, the size and zeta potential of CM maintained almost the same in PBS and DMEM within 24 h. It indicated that CM maintained the good stability in physiologic conditions, helpful to the practical application of bacterial therapy. And the stability of CB and CM was also evaluated by monitoring their dispersion in different solutions versus time. Herein, PBS was applied to reflect normal physiological condition, DMEM and FBS was used to simulate cell culture conditions, LB corresponded to bacterial culture condition, and 5 mM H₂O₂ represented ROS conditions [35]. As imaged in Fig. S10, at the beginning, CB and CM were well dispersed in different solutions. At 1 h, CB started to deposit in PBS and H₂O₂, while it still maintained stable for CM. And CM started to sink with a little of CM on the bottom at 4 h, while CB was almost totally deposited on the bottom. Significantly, CB and CM kept relatively stable in DMEM, LB and FBS, while only a little of CB was deposited on the bottom of DMEM and FBS at 24 h. It presented the improved stability and dispersion of CB after MM encapsulation. In addition, the Cu ions release rate was also displayed in Fig. 2j. Compared to that of CM in normal condition (pH = 7.4) with relatively slow release rate, the Cu ions release rate was significantly enhanced for CM in acidic condition (pH = 5.0) and high ROS condition (5 mM H₂O₂) with the release rate of 67.58 ± 1.72% and 33.60 ± 3.38% at 72 h respectively.

At last, the photothermal effects of CB and CM were also evaluated. From Fig. 2k, compared to PBS with almost the same temperatures during irradiation, the corresponding temperatures increased versus time for CB and CM. After statistical analysis, the temperature was 31.5°C, 61.1°C and 54.8°C for PBS, CB and CM with the concentration of 100 μg/mL under NIR irradiation (1 W/cm²) for 10 min respectively (i of Fig. 2l). If increasing the concentration from 50, 100 to 200 μg/mL, the corresponding temperatures jumped from 45.0°C, 54.8°C to 61.2°C under NIR irradiation (1 W/cm²) for 10 min respectively (ii of Fig. 2l). Meanwhile, if increasing the irradiation intensity from 0.5 W/cm² to 1.5 W/cm², the corresponding temperature increased from 40.3°C to 65.3°C for 100 μg/mL CM (iii of Fig. 2l). Herein, it displayed the excellent photothermal effects for CB. After MM encapsulation, it slightly decreased the photothemal conversion of CM. And the photothermal effects were also enhanced if increasing the concentration of CB, and intensity of NIR irradiation. Especially, after 4 cycles of heating and cooling procedures, it still maintained the stable photothermal effects for 100 μg/mL CM under NIR irradiation (1 W/cm²) (iv of Fig. 2l). Thus, it possessed the good photothermal stability for CM. By calculation, the η was 56.5% and 44.5% for CB and CM. After MM coating, η was slightly declined [36].

The ROS generation ability was qualified assessed by ESR. As shown in Fig. 2m, compared to control group and CM with almost no intensity of ·OH observed, the obvious intensity was observed for CM with NIR irradiation, indicating the generation of ·OH. And it also presented the similar condition for ¹O₂ and ·O₂⁻ generation ability. Meanwhile, by ROS testing kits, the ·OH and ·O₂⁻ generation ratio was relatively low in NIR alone (1.44 ± 0.94% and 0.73 ± 0.28%) and CM (2.06 ± 0.94% and 0.57 ± 0.24%), significantly jumped to 45.06 ± 1.63% and 43.94 ± 0.64% in CM + NIR (Fig. S11 and Table S7). Similarly, the multiple enzymatic activities were also quantified investigated. As shown in Fig. 2n–Tables S8 and S9, for POD testing kit, the POD activity was 14.38 ± 0.04% for CM with the concentration of 100 μg/mL, and increased to 18.83 ± 0.07% for 100 μg/mL CM + NIR. If increasing the concentration of CM from 50, 100 to 200 μg/mL, the POD activity sequentially became from 10.27 ± 0.08%, 14.38 ± 0.04% to 17.21 ± 0.03%. Similarly, for CAT and GSH-Px generation ratio, it was 55.47 ± 0.99% and 6.08 ± 0.49% for CM, while it became 71.02 ± 0.59% and 19.72 ± 1.06% for CM + NIR. Increasing the concentration of CM from 50 to 200 μg/mL, the corresponding CAT activity and GSH-Px activity jumped from 40.38 ± 1.48% and 1.48 ± 0.76% to 64.64 ± 1.22% and 8.55 ± 1.32% respectively. As a kind of nanoenzymes, CB had already confirmed its multiple enzymatic activities like POD, CAT and GSH-Px due to its unique multivalent MOF [37]. After MM encapsulation, it did not affect the catalytic capacities of CB. For NIR and CM alone, it did not generate ROS from ESR results and ROS testing kits. However, under NIR irradiation, it generated a lot of ROS for CM + NIR, kept up with the previous results [38]. Specifically, for 20 μg/mL CM, the CAT activity was 3.83 ± 0.33%, 22.23 ± 0.17% and 71.49 ± 0.14% under pH = 7.4, 6.2 and 5.0. In the meantime, the POD activity was 4.55 ± 0.05% for 20 μg/mL CM in pH = 7.4, obviously enhanced for CM in pH = 6.2 (10.22 ± 0.07%) and 5.0 (21.44 ± 0.42%) (Fig. S12 and Table S10). Compared to PBS, the release of Cu ions was enhanced, contributing to the improved CAT and POD activities [39]. In addition, the effects of temperature on enzymatic activities were also considered. As displayed in Fig. S13 and Table S11, the CAT and POD activities were 55.05 ± 0.19% and 13.78 ± 0.10% in CM, which were ascended to 72.72 ± 0.16% and 18.51 ± 0.86% in CM + NIR. However, their activities were 57.18 ± 0.11% and 14.80 ± 0.15% in CM incubated at 54°C (CM+54°C), lower than those of CM + NIR. Generally, H₂O₂ is relatively stable at room temperature, and its decomposition was significantly accelerated above 55°C [40]. The enzymatic activities of CM+54°C was lower than those of CM + NIR, indicating that the increased temperature by NIR irradiation was not the main driven force for the enhanced enzymatic activities in CM + NIR. Significantly, the ROS generation was time dependent. To accurately detect the CAT and POD activities, the effect of incubation time was also considered. As shown in Fig. S14, the CAT and POD activities were ascended versus incubation time. When the incubation time was 30 min, the activities tended to be stable. Thus, it also confirmed that the incubation time of 30 min was the optimum condition, consistent with the protocols of manufacture. Moreover, by the dissolved oxygen meter, the O₂ concentration was 0.77 ± 0.01 μg/mL in NIR, significantly jumped to 15.74 ± 0.04 μg/mL and 24.28 ± 0.46 μg/mL in CM and CM + NIR. Similarly, the O₂ concentration was 8.70 ± 0.04 μg/mL in 50 μg/mL CM, increased to 32.02 ± 0.58 μg/mL in CM with the concentration of 200 μg/mL, similar to the results of CAT activity (Fig. S15 and Table S12).

From the above, it confirmed the successful preparation of CM and CB. After CB formation, its color became brownish black with no transmittance curves, and the obvious crystallization structure observed. And CB was in spherical shape with the elemental composition of C, N, O and Cu, and 25.64 ± 0.32% Cu elements existed. After MM encapsulation, it displayed the obvious overlap of green (MM) and red (CB) fluorescence, and still retained the similar proteins expression with MM for CM. Besides, no significant changes in molecular structure happened between CB and CM. And the MM encapsulation ratio of CM was 10.09% by TGA. Simultaneously, the Cu ions release rate had improved by acidic and high ROS conditions. Next, it also confirmed that MM encapsulation was helpful to improve the dispersion and stability of CM. Furthermore, it displayed the excellent photothermal effects and stability for both CB and CM. At last, CM possessed a certain of multiple enzymatic activities including CAT, POD and GSH-Px (Fig. 2o), which were significantly enhanced by NIR irradiation and acidic conditions. NIR irradiation was helpful to speeding up the movement of Cu ions, and increasing the free energy of MOF, contributing to the enhanced enzymatic activities [41]. MM encapsulation did not prominently affect the enzymatic activities of CB. Significantly, it confirmed that an amount of ROS generation happened for CM + NIR, while CM itself and NIR alone did not generate ROS.

3.2. In vitro bacterial cuproptosis therapy

Cuproptosis is a newly emerging form of programmed cell death caused by the excessive intracellular Cu in recent years. Combining with NIR irradiation, it generated the excessive ROS for CM, together with the enrichment of bacterial endotoxin, affecting the mitochondrial respiration and TCA cycle, ultimately leading to NIR enhanced bacterial cuproptosis (Fig. 3a). Herein, we chose two kinds of commonly used bacteria: Spn and E. coli as the models for research. As shown in Fig. 3b, compared to control group, CB and CM could slightly inhibit the growth of both bacteria by spread plate method, which was significantly enhanced by CM + NIR. After statistical calculation, the average amount of bacterial colony was 764.67 ± 18.93 and 676.67 ± 10.69 CFU/mL for Spn and E. coli in control group, sequentially decreased to 338.67 ± 7.09 and 245.67 ± 17.47 CFU/mL, 273.67 ± 14.19 and 239.33 ± 11.37 CFU/mL, and 21.33 ± 3.21 and 7.33 ± 0.58 CFU/mL in CB, CM and CM + NIR (Fig. 3c). Meanwhile, the live/dead staining results of Spn was displayed in Fig. 3d. It was obviously observed that CB could kill a few of Spn with a lot of live Spn (green fluorescence), and a few of dead Spn (red fluorescence) existed, similar to that of CM. Significantly, CM + NIR could efficiently kill a lot of Spn. By statistical analysis, the live bacteria ratio was 99.85 ± 0.13% for control group, decreased to 89.73 ± 1.61 and 94.19 ± 0.09% for CB and CM (Fig. 3e). Similarly, it displayed the same tendency for that of E. coli (Fig. 3f). Among them, CM + NIR most efficiently kill E. coli with the live bacteria ratio of 0.40 ± 0.16%, followed by CB (67.47 ± 1.37%) and CM (74.35 ± 1.97%) (Fig. 3g). Furthermore, flow cytometry was applied to characterize the live/dead ratio of bacteria. As illustrated in Fig. 3h and S16, few of dead bacteria was observed in control group (10.75 ± 0.22% for Spn, and 4.15 ± 0.26% for E. coli), while CB and CM possessed a certain of antibacterial ability with increased dead bacteria ratio (43.85 ± 1.56 and 43.40 ± 1.25% for Spn, and 57.42 ± 0.78 and 57.41 ± 2.44% for E. coli). Remarkably, CM + NIR possessed the most efficient capacity of killing bacteria with the dead bacteria ratio of 68.23 ± 1.29% for Spn and 79.57 ± 0.62% for E. coli respectively. In particular, compared to control group, no obvious differences were observed for the live/dead bacterial staining images of NIR alone. And the similar live bacteria ratio of E. coli and Spn was observed between control group and NIR alone (Fig. S17). It indicated that NIR alone had no effects on antibacterial ability.

Fig. 3.

a) The schematic illustration of cuproptosis antibacterial mechanism of CM + NIR. b) The bacterial spreading experiment of treated bacteria, and the corresponding quantified results (c): Spn and E. coli. d) The live/dead staining images of treated Spn, (Scale bar: 50 μm) and the corresponding quantified results (e). f) The live/dead staining images of treated E. coli, (Scale bar: 50 μm) and the corresponding quantified results (g). h) Live/dead staining of treated bacteria by flow cytometry. i) The intracellular ROS levels of treated bacteria by immunofluorescent staining. (Scale bar: 50 μm) j) SEM images of treated bacteria. (Scale bar: 2 μm) k) The volcano plot of DEGs: 246 upregulated and 279 downregulated genes between control group and CM + NIR. l) The KEGG enrichment analysis of DEGs. m) The GO enrichment analysis of DEGs. n) The heatmap of oxidative phosphorylation pathway related differential metabolites between control group and CM + NIR: genes with relatively high (red) and low (blue) expression levels. o) The heatmap of TCA cycle pathway related differential metabolites between control group and CM + NIR: genes with relatively high (red) and low (blue) expression levels. p) The mitochondrial respiratory chain activity of treated E. coli: complex I (i) and complex II (ii). q) The intracellular Cu concentration of treated E. coli. r) The intracellular MDA content of treated E. coli. s) The FDX1 expression levels of treated E. coli (i), and the relative FDX1/actin expression levels (ii). The corresponding groups were: bacteria without treatments (control group), and bacteria treated with 100 μg/mL CB (CB), 100 μg/mL CM (CM), and 100 μg/mL CM combining with NIR irradiation (1.0 W/cm²) (CM + NIR). (“∗” symbol compared with control group, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001).

From the above results, it confirmed the effective antibacterial ability of CM + NIR, better than that of CM and CB. During this experiment, no H₂O₂ was added as the source of ROS. And the intra-bacterial ROS levels were generated from CM under NIR irradiation [42]. However, for CM and CB, slight antibacterial ability happened due to their unique cytotoxicity. To confirm the above, the antibacterial ability was also investigated under different pH. For Spn, it was observed that almost no red fluorescence (dead bacteria), and only green fluorescence (live bacteria) happened to them under control group, pH = 7.4 and 6.2. However, under pH = 5.0, slight red fluorescence was observed, indicating a small amount of dead bacteria. Remarkably, compared to control group, the number of dead bacteria was obviously ascended in CM under different pH, especially in pH = 5.0 with relatively high red fluorescence (Fig. S18a). Similarly, it presented the similar tendency in E. coli under different pH with or without CM incubation (Fig. S18b). After statistical analysis, the live bacteria ratio of Spn and E. coli was 99.58 ± 0.09 and 99.43 ± 0.10%, 99.03 ± 0.09 and 99.41 ± 0.10%, 98.23 ± 0.10 and 98.41 ± 0.12%, and 84.64 ± 0.37 and 79.84 ± 1.41% in control group, pH = 7.4, pH = 6.2 and pH = 5.0. After incubating with CM, it became 71.83 ± 0.39 and 78.26 ± 0.58%, 64.93 ± 0.73 and 67.89 ± 0.96%, and 0.65 ± 0.11 and 1.69 ± 0.18% in CM + pH = 7.4, CM + pH = 6.2 and CM + pH = 5.0 (Fig. S18c). For most bacteria including E. coli, the optimal culture condition is around weak alkaline. However, the bacteria can survive under slightly acidic conditions for a certain period of time. When the pH is less than 4.5, the bacterial growth is inhibited. Thus, under pH = 5, a few of bacteria were dead. And after incubating with CM, the live bacteria ratio was significantly declined, especially in CM + pH = 5.0 with the lowest live bacteria ratio. The antibacterial ability of CM was mainly attributed by the cytotoxicity of Cu ions. Under acidic condition, the release of Cu ions was significantly enhanced, contributed to the improved antibacterial ability. Additionally, to investigate the anti-infectious mechanism of CM + NIR, the intra-bacterial ROS levels of treated bacteria were initially analyzed. For Spn, the slight green fluorescence, corresponding to low intra-bacterial ROS levels existed in control group (0.11 ± 0.01), while the obvious enhanced ROS levels happened to treated bacteria. And it displayed the strongest green fluorescence for CM + NIR (10.02 ± 0.10), followed by CM (3.75 ± 0.06) and CB (1.57 ± 0.12). Equally, compared to other groups, CM + NIR possessed the highest intra-bacterial ROS levels in E. coli (12.05 ± 0.43) (Fig. 3i and S19). At last, the treated bacteria were also imaged by SEM. From Fig. 3j, it displayed the smooth surface of Spn and E. coli in control group, while the destroyed surface happened in CM, indicating the death of both bacteria. Remarkably, more obvious bacteria surface was destroyed in CM + NIR.

Besides, to further investigate the antibacterial mechanism, the transcriptome sequencing on treated E. coli (control group and CM + NIR) was performed. From Fig. 3k and S20, the significant differences in genes expression were observed with a total of 525 DEGs identified. By constructing a volcano map, it was observed that 279 genes were upregulated, and 246 genes were downregulated. In the meantime, by implementing KEGG enrichment analysis, the 30 most affected pathways were identified. As shown in Fig. 3l, most of these pathways were related to substance and energy metabolism, including fatty acid degradation, oxidative phosphorylation, TCA cycle, and ATP binding cassette (ABC) transporter pathway. Besides, the GO enrichment analysis was also conducted to determine the importance of DEGs in biological processes (BP), cellular components (CC), and molecular functions (MF). From Fig. 3m, it identified the top 30 pathways with greatest impacts for antibacterial mechanism by enrichment analysis, mainly enriched in bacterial membrane related behaviors such as membrane and transporter activity. Herein, it confirmed that CM + NIR efficiently disrupted the normal physiological activity of bacterial membranes. The cell walls and membrane of bacteria served as the protective barriers against external factors, allowing these microorganisms to maintain survival. The disruption of membrane lead to the ultimate death of bacteria [43]. Furthermore, from heat map analysis, it displayed the genes expression differences between control group and CM + NIR during TCA cycle and oxidative phosphorylation process. Compared to control group, there were also significant differences in oxidative phosphorylation related genes such as PPA, FRDD, SDHC, CYOA, CYOB, CYOC, CYOD and CYOE, which were obviously upregulated, while APPB and APPC were obviously downregulated in CM + NIR (Fig. 3n). Meanwhile, CM + NIR resulted in the significant differences in TCA cycle related genes including the upregulation of ACNA, FRDD, SDHC and MDH, and downregulation of MQO (Fig. 3o). The results also confirmed that the excessive Cu significantly regulated the expression levels of TCA cycle and oxidative phosphorylation related genes in bacteria treated by CM + NIR, similar to the cuproptosis mechanism of eukaryotic cells [44].

Significantly, to deeply investigate the antibacterial mechanism, the metabolome sequencing was also applied. The top 8 differential metabolites ranked by fold change (FC), including 6 upregulated (red) and 2 downregulated (green) metabolites, were shown in Fig. S21, displaying the main metabolic alterations between control group and CM + NIR. As illustrated in Fig. S22, it displayed 7 upregulated metabolites (red), 31 insignificant metabolites (grey), and 3 downregulated metabolites (green) between control group and CM + NIR (Fig. S22a). And it was also observed that TCA cycle, fatty acid degradation, and glycolysis/gluconeogenesis pathways were significantly enriched in KEGG pathway (Fig. S22b). Thus, the activities of respiratory chain complexes I and II of treated bacteria were tested. Compared to control group (100.00 ± 1.48% and 100.00 ± 3.96%), the activities of both respiratory chain complexes decreased in CB (63.69 ± 3.39% and 71.94 ± 2.13%) and CM (64.47 ± 2.82% and 69.66 ± 4.55%), which were significantly reduced in CM + NIR (33.82 ± 3.32% and 43.39 ± 5.19%) (Fig. 3p), consistent with the above enrichment analysis. The decrease of both respiratory chain complexes activities directly disrupted the respiratory chain functions of bacteria. It had confirmed that NIR enhanced cuproptosis was an important antibacterial mechanism, where Cu element played a significant role. Therefore, the Cu contents inside bacteria were detected. As shown in Fig. 3q, compared to control group with the Cu content of 0.79 ± 0.02 mg/L, it significantly increased for other groups with the order of CM + NIR (2.68 ± 0.01 mg/L) > CM (1.51 ± 0.10 mg/L) > CB (1.29 ± 0.12 mg/L). Meanwhile, it was also well known that Cu overload always induced the accumulation of lipid peroxides [45]. From MDA testing kit, compared to control group (1.07 ± 0.12 mM), the MDA contents were significantly enhanced in other groups, especially for CM + NIR (3.95 ± 0.31 mM) (Fig. 3r). Significantly, by WB, the cuproptosis specific protein (FDX1) expression of bacteria was in highest levels in control group (1.00 ± 0.01), slightly decreased to 0.72 ± 0.01 and 0.71 ± 0.01 in CB and CM, and highly declined in CM + NIR (0.40 ± 0.02) (Fig. S23 and 3s), consistent with the results previously reported [46].

Summarily, CM + NIR could most efficiently disrupt TCA cycle and respiratory chain functions of bacteria, and induce high levels of MDA, suggesting that it could induce a process similar to Cu poisoning in bacteria (cuproptosis). The amplified cuproptosis was achieved by the strategy of CM + NIR. On one hand, MM encapsulation contributed to the improved biocompatibility, and enhanced bacterial targeting for precise therapy. On the other hand, NIR amplified cuproptosis ensured the bacterial cuproptosis under control, ultimately achieving personalized anti-infectious therapy.

3.3. In vitro anti-infectious immunotherapy

Macrophages are immune cells with the main function of devouring and digesting pathogens and other harmful substances, playing the important roles in both non-specific and specific immunity (Fig. 4a). At the beginning, the cell viability was investigated by CCK-8 assay. As illustrated in Fig. S24, the cell viability of RAW264.7 started to decrease when the concentration of CB and CM was above 100 μg/mL. However, the cell viability of CM was higher than that of CB. At 100 μg/mL, the cell viability was 76.52 ± 7.19% for CB, while it was 91.50 ± 7.03% for CM. Similarly, the cell viability of BEAS-2B was significantly declined with the increase of CB concentration. For CM, the cell viability was above 94.87 ± 5.02% when the concentration was 100 μg/mL, with the better cell biocompatibility than that of CB only with the cell viability of 39.66 ± 4.79%. Thus, 100 μg/mL was considered as the concentration for further experiments. Besides, the live/dead staining of treated RAW264.7 was implemented by fluorescent microscopy. As shown in Fig. S25, compared to control group, a few of dead cells (red) were observed for CB and CM. After statistical analysis, the live/dead ratio was 99.77 ± 0.15, 99.27 ± 0.08 and 99.70 ± 0.01% for control group, CB and CM respectively. Both of CB and CM possessed the good biocompatibility with this dose concentration. And the migration ability of RAW264.7 was evaluated by trans-well experiment (Fig. S26). Compared to control group, the obvious migration of RAW264.7 was observed for CB, significantly enhanced for CM and CM + NIR. After calculation, the migration ratio was 34.00 ± 4.00% for CB, increased to 127.33 ± 14.29 and 245.67 ± 22.90% for CM and CM + NIR respectively under the same condition. NIR irradiation was helpful to improving the migration ability of RAW264.7, strengthening the macrophage targeting bacteria [47]. Significantly, to make full use of their functions, it expected that the nanocatalysts could be uptake by the cells. As shown in Fig. 4b, a little of red fluorescence, corresponding to the cellular uptake of CB, was observed for Cy5-CB at 3 h, appropriately increased at 6 h. However, the obvious improved fluorescence was observed for Cy5-CM at 3 h compared to that of Cy5-CB. Significantly, a huge amount of CM were uptake by RAW264.7 at 6 h. By statistical analysis, the MFI of Cy5-CB was 0.23 ± 0.10 and 1.88 ± 0.06, while it was 1.24 ± 0.14 and 5.00 ± 0.18 for Cy5-CM at 3 and 6 h (Fig. 4c), giving the proof that CB after MM encapsulation contributed to the improved targeting ability of macrophage. To confirm that the targeting ability of CM was due to the receptor recognition of MM, TLR4, a receptor inhibitor was applied [48]. As displayed in Fig. S27, for LPS treated RAW264.7 (LPS), the obvious cellular uptake was observed for Cy5-CM with a lot of red fluorescence. However, almost no red fluorescence was displayed in LPS treated RAW264.7 followed by TLR4 incubation (LPS + TLR4) for Cy5-CM. By statistical analysis, the MFI of Cy5-CM was 8.12 ± 0.51 and 0.00 ± 0.00 in LPS and LPS + TLR4. By LPS stimulation, the obvious inflammation related receptors were enhanced in RAW264.7, helpful to the improvement of CM targeting. After TLR4 treatment, most of receptors were inhibited in LPS induced RAW264.7, leading to the poor targeting of CM [49].

Fig. 4.

a) The schematic illustration of immunoregulation antibacterial mechanism of CM + NIR. b) The cellular uptake of Cy5-CB and Cy5-CM at predetermined time points (0, 3 and 6 h) by confocal microscopy, (Scale bar: 50 μm) and the corresponding quantified results (c). d) The inflammatory factors expression levels in the supernatant of treated cells: IL-6 (i), TNF-α (ii) and IL-10 (iii) by ELISA. e) The inflammation related genes expression levels of treated cells: IL-6 (i), TNF-α (ii) and iNOS (iii) by RT-qPCR. f) The immunofluorescent staining images of CD86 expression levels of treated cells by confocal microscopy. (Scale bar: 50 μm) g) The immunofluorescent staining images of CD206 expression levels of treated cells by confocal microscopy. (Scale bar: 50 μm) h) The immunofluorescent quantified results of CD86 and CD206 expression levels of treated cells. i) The CD86 and CD206 expression levels of treated cells by flow cytometry, and the corresponding M1/M2 polarization ratio (j). k) The immunofluorescent quantified results of iNOS (i) and Arg-1 (ii) expression levels of treated cells. l) The Spn engulfment capacity of treated cells by confocal microscopy, (Scale bar: 50 μm) and the corresponding amplified images. (Scale bar: 20 μm) m) The E. coli engulfment capacity of treated cells by confocal microscopy, (Scale bar: 50 μm) and the corresponding amplified images. (Scale bar: 20 μm) The corresponding groups were: IL-4 induced RAW264.7 treated with PBS (control group), 100 μg/mL CB (CB), 100 μg/mL CM (CM), and 100 μg/mL CM combining with NIR irradiation (1.0 W/cm²) (CM + NIR). (“∗” symbol compared with control group, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001).

The bactericidal mechanism of macrophages involves the production of a huge amount of ROS, secretion of inflammatory factors, and regulation of relevant microenvironment to kill bacteria [50]. To confirm the ROS production, the intracellular ROS levels of treated RAW264.7 were investigated by fluorescent microscopy (Fig. S28). Compared to control group with almost no green fluorescence (1.43 ± 0.20), the green signal was enhanced for CB (4.64 ± 0.30), CM (13.15 ± 1.53) and CM + NIR (20.70 ± 1.94). Especially for CM + NIR, the obvious enhanced green signal was observed, consistent with the results of ESR and ROS testing kits. Generally, almost no ROS generation happened to CB and CM. Under NIR irradiation, a huge amount of ROS was generated for CM + NIR, helpful to killing bacteria. In specific, for NIR alone (1.61 ± 0.07), the green fluorescence was almost the same as that of control group (1.64 ± 0.07) (Fig. S29), indicating that NIR alone could not generate intracellular ROS. Meanwhile, ELISA was applied to characterize the inflammatory factor expression levels of treated cells. As illustrated in Fig. 4d, compared to control group with low inflammatory factors expression, and high antioxidant factor expression, the inflammatory factors expression levels (IL-6 and TNF-α) of treated cells were significantly ascended with the order of CB < CM < CM + NIR, while the antioxidant factor (IL-10) decreased with the order of CM + NIR < CM < CB. Among them, the highest IL-6 (85.01 ± 1.25 pg/mL) and TNF-α (67.89 ± 2.04 pg/mL), and lowest IL-10 (46.76 ± 1.67 pg/mL) expression levels existed in CM + NIR with significant difference of ∗∗∗∗ compared to control group. Next, RT-qPCR was applied to characterize the inflammatory genes expression levels of treated cells. As illustrated in Fig. 4e, the IL-6 gene expression levels were 1.00 ± 0.10 for control group, increased to 1.75 ± 0.15, 2.05 ± 0.13 and 2.35 ± 0.09 for CB, CM and CM + NIR respectively. And the same tendency was observed for TNF-α and iNOS genes expression levels, where CM + NIR most efficiently promoted the corresponding genes expression levels (5.65 ± 0.62 and 5.37 ± 0.29) compared to CM (4.04 ± 0.09 and 3.23 ± 1.42) and CB (3.46 ± 0.18 and 2.29 ± 0.18).

The macrophage polarization levels were also characterized by immunofluorescent staining. As shown in Fig. 4f and g, compared to control group with little green fluorescence (CD86 expression levels), and a lot of red fluorescence (CD206 expression levels), the obvious upregulation of CD86 and downregulation of CD206 expression levels happened for treated macrophages, indicating macrophage M1 directional polarization. Among them, CM + NIR possessed the highest CD86 and lowest CD206 expression levels. After statistical analysis, the MFI of CD86 expression was 11.86 ± 0.51, 14.83 ± 0.93, 21.09 ± 1.34 and 27.14 ± 0.72 in control group, CB, CM and CM + NIR (Fig. 4h), while the MFI of CD206 expression was 9.53 ± 0.31, 7.36 ± 0.77, 6.56 ± 0.45 and 4.26 ± 0.19 in control group, CB, CM and CM + NIR respectively. Besides, flow cytometry was applied to characterize the macrophage directional polarization levels. Consistent with the immunofluorescent results, the obvious M1 polarization existed in CM + NIR with the enhanced CD86 and decreased CD206 expression levels (Fig. 4i). After calculation, the M1/M2 polarization ratio was 31.00 ± 2.64% for control group, increased to 69.59 ± 2.47%, 141.20 ± 14.26% and 284.31 ± 56.71% for CB, CM and CM + NIR respectively (Fig. 4j). In particular, it displayed almost the similar macrophage polarization levels between control group (1.54 ± 0.16 and 2.47 ± 0.10) and NIR alone (1.62 ± 0.08 and 2.58 ± 0.22), with few CD86 expression levels and a lot of CD206 expression levels observed (Fig. S30). It also confirmed that NIR alone could not induce macrophage M1 polarization. Furthermore, the inflammation related factors expression was also characterized by immunofluorescent staining. As displayed in Fig. S31, the iNOS expression was in the lowest levels for control group with little green fluorescence observed, which was successively enhanced in CB, CM and CM + NIR, corresponding to the gradually increased iNOS expression levels. By statistical calculation, the MFI of iNOS expression was 8.67 ± 0.41, 15.73 ± 0.33, 18.40 ± 1.04 and 29.07 ± 1.65 in control group, CB, CM and CM + NIR respectively (i of Fig. 4k). Conversely, it was in the highest levels of Arg-1 expression for control group, separately decreased in CB, CM and CM + NIR (Fig. S32). By calculation, the MFI of Arg-1 expression was 13.69 ± 1.56 in control group, decreased to 8.29 ± 1.30, 3.92 ± 0.52 and 2.72 ± 0.31 in CB, CM and CM + NIR respectively (ii of Fig. 4k). Meanwhile, for NIR alone (2.60 ± 0.05), it could not affect the expression levels of iNOS compared to control group (2.51 ± 0.08), with almost the similar green fluorescence (Fig. S33).

On the other hand, macrophage can engulf and decompose bacteria through its powerful phagocytic activity, thereby achieving the goal of eliminating pathogens. Significantly, M1 type macrophage possesses stronger phagocytic activity [51]. To confirm this, the co-immunofluorescent staining of macrophage and bacteria was implemented. As displayed in Fig. 4l, almost no Spn were observed inside macrophages in control group, while a few of bacteria appeared inside macrophages of CB and CM. Significantly, a huge amount of bacteria were shown inside macrophages of CM + NIR. After co-localization analysis, almost no overlap of green and red signal existed in control group, indicating that no bacteria were engulfed by macrophages. However, the significant overlap of green and red signal happened to CM + NIR, indicating that a huge amount of bacteria were engulfed by macrophages. Similarly, it obviously displayed that a lot of E. coli were engulfed by macrophages in CM + NIR with the apparent overlap of green and red signal observed (Fig. 4m). After statistical analysis, compared to control group with the MFI of 0.72 ± 0.06 and 2.60 ± 0.45 for the engulfment of Spn and E. coli respectively, the MFI of other groups obviously increased with the order of CM + NIR (26.25 ± 1.54 and 32.03 ± 2.96) > CM (12.00 ± 1.26 and 14.83 ± 0.87) > CB (3.20 ± 0.26 and 3.95 ± 0.52) (Fig. S34). CM + NIR most efficiently improved the bacterial engulfment of macrophages. In the meantime, for infectious disease therapy, it also expected that the infected tissue could be repaired after therapy. From Fig. S35a, almost no HSP70 expression was shown in control group, CB and CM, while the obvious green signal of HSP70 expression was observed for CM + NIR. After statistical analysis, the MFI of HSP70 expression was 10.67 ± 0.59, 14.61 ± 1.19, 16.82 ± 0.67 and 25.95 ± 0.75 in control group, CB, CM and CM + NIR respectively (Fig. S35b). Meanwhile, the CD31 expression levels of treated cells were also investigated. As shown in Fig. S36, the relative low green signal (CD31 expression) existed in control group, enhanced for CB, CM and CM + NIR respectively. After statistical analysis, the MFI of control group was 11.64 ± 1.53, while it increased to 23.52 ± 1.28, 28.29 ± 0.99 and 35.92 ± 1.61 for CB, CM and CM + NIR respectively. Specifically, NIR alone (2.00 ± 0.13) also could not affect the expression levels of CD31, with almost the same expression levels of control group (1.85 ± 0.21) (Fig. S37).

From the above, it mainly focused on the anti-infectious immunotherapy by macrophages in cellular levels. And it had proved the dual function of macrophages for immunotherapy. For CM with MM encapsulation, its cell biocompatibility had been significantly improved, helpful to its application in biomedical fields. Besides, it confirmed that CM was helpful to the proliferation and migration of treated RAW264.7. Under NIR irradiation, the migration ability was significantly enhanced, contributing to the improved cell growth and proliferation [52]. Meanwhile, after MM encapsulation, CM possessed the obvious stronger capacity of cellular uptake than that of CB, thereby achieving the effective intracellular biological functions. Next, the enhanced intracellular ROS levels, inflammatory factors expression levels, and macrophage M1 polarization mediated by CM + NIR could all be conducive to efficiently killing bacteria [53]. Furthermore, it also confirmed that M1 type macrophage induced by CM + NIR possessed the most effective engulfment and decomposition functions of bacteria. Finally, the strategy of CM + NIR also contributed to the high expression levels of HSP70, accelerating the tissue repair, and CD31, promoting angiogenesis. However, for NIR alone, it had no effects on inducing inflammation and macrophages M1 polarization, as well as promoting tissue repair. All of the results had demonstrated that CM combining with NIR irradiation could effectively eliminate bacteria, and promote tissue repair, ultimately proving the multimode synergistic enhanced therapy of bacterial pneumonia at the cellular levels.

3.4. In vivo anti-infectious therapy

All in vivo animal experiments were implemented under the approval, and SD rats were selected and housed under standard specific pathogen-free (SPF) grades indoor environment. The time schedule of all animal experiments was listed in Fig. 5a. After IT instillation of the mixture of Spn and E. coli for 24 h, the bacterial pneumonia animal models were built and saved for further experiments. 2 days later, the rats were treated by IV injection every two days. NIR irradiation was implemented for 3 times, 10 min per time after each IV injection. Finally, the rats were sacrificed after 7 days. The IT injection was applied for establishing the animal models to ensure the bacteria were mainly enriched in the lung, and minimize the damages to other organs [54]. To investigate in vivo biodistribution after IV injection, normal rats and bacterial pneumonia rats were considered. For normal rats, almost no fluorescence was observed for all organs except for the liver with a little of fluorescence observed, and kidney with obvious fluorescence existed at 4 h by Cy5 injection. Meanwhile, the obvious fluorescence was shown in the liver and kidney, and it maintained there for 12 h by Cy5-CB injection. However, no fluorescence was observed for other organs like heart and lung. Significantly, for Cy5-CM injection, it displayed the obvious fluorescence in the liver, lung and kidney, while almost no fluorescence happened in the heart and spleen. After 8 h, the fluorescence in the lung disappeared. And the fluorescence of liver and kidney could maintain till 12 h (Fig. 5b). After statistical analysis, it obviously displayed that the MFI of Cy5-CM decreased versus time in the lung, while it maintained the relative low levels in Cy5-CB and Cy5 alone (Fig. 5c). After MM encapsulation, it had improved the blood circulation of CM, and its physicochemical properties also changed, helpful to targeting and enriching in the lung [55]. Significantly, for pneumonia animal models, almost no fluorescence happened to the lung by Cy5 or Cy5-CB injection, and only with a little of fluorescence in the kidney and liver till 8 h. However, the obvious fluorescence was observed in the lung for Cy5-CM injection. And it retained the higher intensity at 8 h, and gradually disappeared after 12 h (Fig. 5d). After calculation, the MFI of lung was in the high levels (10.6 × 10⁸) at 1 h, maintained a relatively high levels till 8 h, and gradually degraded versus time (Fig. 5e). Generally, by IV injection, nanomaterials with the diameter above 20 nm could be easily accumulated in the liver and kidney through blood circulation. When it was below 20 nm, the nanomaterials were easily cleared by the kidney [56]. For CB with the size around 50 nm and its unique physicochemical properties, it was rarely or not enriched in the lung of normal mice, but tend to be enriched in the liver and kidney in the initial stage. After MM encapsulation, the changes in physicochemical properties, improved blood circulation, and cell homing effect of CM jointly contributed to targeting and be enriched in the lung, especially for the lung of pneumonia rats [57]. However, the main reason for specific targeting in the lung of pneumonia rats was mainly attributed by MM coating on CM [58]. And the disappearance of fluorescence in the lung also proved that CM possessed good biodegradability. All of these provided the feasibility of achieving the precise and safe therapy of bacterial pneumonia by IV injection of CM.

Fig. 5.

In vivo bacterial pneumonia therapy evaluation. a) The time schedule of in vivo animal experiment. b) The fluorescent images of major organs (heart (H), liver (Li), spleen (S), lung (Lu) and kidney (K)) isolated from normal rats after IV injection of Cy5, Cy5-CB or Cy5-CM by IVIS, and the corresponding quantified results of lung (c). d) The fluorescent images of major organs (heart (H), liver (Li), spleen (S), lung (Lu) and kidney (K)) isolated from bacterial pneumonia rats after IV injection of Cy5, Cy5-CB or Cy5-CM by IVIS, and the corresponding quantified results of lung (e). f) In vivo photothermal images of treated rats under NIR irradiation (1.0 W/cm²), and the corresponding quantified results (g). h) The bacterial spreading results in the lung tissue of treated rats, and the corresponding quantified results (i). j) The MPO (i) and MDA (ii) contents in the lung tissue of treated rats. k) The macroscopic observation in the lung tissue of treated rats. l) The H&E staining images in the lung tissue of treated rats. (Scale bar: 200 μm) m) The HSP70 expression in the lung tissue of treated rats, (Scale bar: 200 μm) and the corresponding quantified results (n). o) The VEGF expression in the lung tissue of treated rats, (Scale bar: 200 μm) and the corresponding quantified results (p). The corresponding groups were: rats without treatments (sham group), and bacterial pneumonia rats with IV injection of saline (control group), CM (CM), and CM combining with NIR irradiation (1.0 W/cm²) (CM + NIR). (“∗” symbol compared with control group, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001).

At the same time, photothermal therapy (PTT) can dilate local blood vessels, improve blood circulation, promote the dissipation of inflammation, and facilitate tissue repair. NIR irradiation has the characteristic of deep local penetration (0.5∼1 cm) without damaging tissues, and is applied in PTT [59]. To confirm the in vivo feasibility of PTT, the temperature in the lung was monitored by thermal camera. As imaged in Fig. 5f, no obvious temperature changes were observed in the lung of sham group, while the temperature gradually increased versus irradiation time in the lung of CM. By quantified analysis, the temperature became 50.9°C for CM by 10 min’ irradiation, while it maintained around 38.6°C in sham group during irradiation (Fig. 5g). The previous studies confirmed that the temperature of local PTT should be around 50°C, and the duration should not exceed 20 min, which could not cause damages to normal tissue [60].

To further confirm the feasibility of pneumonia therapy, in vivo biosafety was assessed. From the results of hemolysis testing (Fig. S38), compared to DI water with the hemolysis ratio above 100%, it was below 1.5% for CM during the concentrations ranging from 0 to 200 μg/mL, far lower than the normal hemolysis ratio (5%) [61]. It confirmed the favorable blood biocompatibility of CM. In addition, the body weight of treated rats was also monitored. As shown in Fig. S39, the body weight of CM + NIR maintained the similar trend with sham group, gradually increased versus time. And the corresponding blood routine, blood biochemical, and coagulation indicators retained the close ranges without significant differences between sham group and CM + NIR (Fig. S40 and Table S13). As shown in Fig. S41, there were also no obvious changes in H&E staining images between the major organs of sham group and CM + NIR. From the above, it confirmed that the strategy of CM + NIR could ultimately achieve the pneumonia therapy with high biosafety.

For in vivo therapy, after 7 day’ therapy, the skin in the lung of treated rats was observed. As shown in Fig. S42, no significant differences happened for all groups, with smooth skin observed. And from H&E staining images, there were also no apparent blisters and inflammation existed (Fig. S43). The above indicated that the strategy of CM + NIR had no damages to the skin. For NIR irradiation (1 W/cm²) with the wavelength of 808 nm, it had the ability to penetrate tissues within a few centimeters, and did not cause damages to healthy tissues [62]. And it was only 50.9°C by the strategy of CM + NIR, below the irreversible damage temperature (52.0°C) of normal tissue, and impossible to cause any damages to normal tissue. In addition, the blood indicators of treated rats were also evaluated. As shown in Fig. S44 and Table S14, no significant differences were observed for all blood indicators except for PLT, ALT and AST. For ALT and AST, they increased in control group compared to sham group, and were descended in CM and CM + NIR. Conversely, for PLT, it decreased in control group, and recovered to normal state in CM and CM + NIR. From the macroscopic observation images, the other organs including heart, liver, spleen and kidney maintained the same states for all groups (Fig. S45), also indicating the successful modeling of bacterial pneumonia without affecting other organs [63]. Additionally, by ICP-MS, the Cu contents in the major organs were also analyzed. As displayed in Fig. S46, no obvious differences of Cu contents were observed in the blood, heart, liver, spleen, lung and kidney of sham group and CM + NIR, indicating that the Cu elements were completely metabolized from the body.

The in vivo therapy evaluation mainly focused on the lung tissue. At the beginning, the lung homogenate was applied for bacterial re-cultivation. As shown in Fig. 5h, compared to sham group with a few of bacteria observed, a huge amount of bacteria existed in control group. CM could kill a certain of bacteria, while large amount of bacteria were cleared by CM + NIR. After statistical analysis, the average amount of bacterial colony was 9.67 ± 3.06 CFU/mL in the lung of sham group, while it became (8.32 ± 0.31) × 10⁶ CFU/mL, (3.47 ± 0.21) × 10⁴ CFU/mL and 72.67 ± 22.01 CFU/mL in that of control group, CM and CM + NIR respectively (Fig. 5i). It proved that CM + NIR most efficiently eliminated the bacteria of lung. Meanwhile, the MPO and MDA contents of lung homogenate were tested by the corresponding testing kits. As illustrated in Fig. 5j, the MPO and MDA was 0.75 ± 0.00 U/g and 1.00 ± 0.02 nmol/mg in the lung of sham group, ascended to 3.22 ± 0.05 U/g and 2.07 ± 0.01 nmol/mg in the lung of control group, and changed to 1.25 ± 0.02 U/g and 1.54 ± 0.01 nmol/mg in the lung of CM. Specifically, it became 0.82 ± 0.10 U/g and 0.99 ± 0.03 nmol/mg in the lung of CM + NIR, close to those of sham group. And from the macroscopic observation of the lung, the smooth and tender lung was observed in sham group, while the obvious congestion and erosion areas happened in the lung of control group. However, slight congestion was observed in the lung of CM, and it almost returned to the normal smooth and tender state in the lung of CM + NIR (Fig. 5k). Besides, the H&E staining images of lung were shown in Fig. 5l. Compared to the lung of sham group with fine mesh like structure and thin lung wall, the obvious dense structure and thick lung wall happened to that of control group. However, the structure and lung wall had slightly been improved in the lung of CM. Specifically for the lung of CM + NIR, it displayed the obvious fine structure and thin lung wall. Significantly, from H&E staining images of other organs, no obvious differences were observed for the other organs of all groups (Fig. S47). Furthermore, the lung tissue repair was evaluated by the immunofluorescent staining of HSP70 and VEGF expression levels in the lung. As imaged in Fig. 5m, compared to the lung tissue of sham group, control group and CM with little or a little green fluorescence observed, corresponding to the relatively low HSP70 expression levels, a lot of green fluorescence was observed in the lung tissue of CM + NIR. After statistical analysis, the MFI of HSP70 expression was 5.77 ± 0.28 for the lung of CM + NIR, followed by the lung of CM (0.89 ± 0.04), control group (0.21 ± 0.07) and sham group (0.03 ± 0.01) respectively (Fig. 5n). Similarly, the obvious enhanced green signal was observed in the lung of CM + NIR, corresponding to high VEGF expression levels compared to the lungs of other groups (Fig. 5o). By statistical analysis, for the lung of CM + NIR, the VEGF expression was in the high levels with the MFI of 11.87 ± 0.88. However, it was in relatively low levels for CM (4.02 ± 0.13), control group (0.53 ± 0.02) and sham group (0.54 ± 0.08) respectively (Fig. 5p). From the above, it had confirmed that CM + NIR most efficiently achieved the therapy of bacterial pneumonia via eliminating the bacteria together with promoting the expression levels of HSP70 and VEGF.

Most importantly, immunotherapy also played a significant role in in vivo bacterial pneumonia therapy. The role of macrophages not only kill pathogens directly, but also participate in promoting inflammatory responses, secreting various inflammatory cytokines, and helping to recruit more immune cells to infectious sites, enhancing the overall immune capacity of the body [39,64]. As shown in Fig. 6a, it was clearly displayed that the bacterial pneumonia immunotherapy was achieved by the strategy of CM + NIR via increasing intracellular ROS levels, upregulating the inflammatory factors expression levels, inducing macrophage M1 directional polarization, and activating immunoregulation. The ROS levels of lung section were investigated by immunofluorescent staining. From Fig. S48, it was observed that the ROS levels were relative low in the lung of sham group with little green fluorescence observed. However, the ROS levels increased in the lung of control group, obviously improved for the lung of CM and CM + NIR. After statistical analysis, the MFI of ROS levels was 0.33 ± 0.01 in the lung of sham group, went up to 14.51 ± 1.01, 26.61 ± 2.03 and 66.02 ± 1.59 in the lung of control group, CM and CM + NIR. Besides, as illustrated in Fig. 6b, compared to sham group (515.30 ± 67.41 and 93.30 ± 25.86 pg/mL), the IL-6 and TNF-α expression levels of lung homogenate were significantly ascended for all groups, especially in the lung of CM + NIR (1181.75 ± 15.16 and 508.94 ± 30.17 pg/mL). However, the IL-10 expression was in the high levels for the lung of sham group (539.06 ± 4.60 pg/mL), subsequently decreased to 498.28 ± 3.69, 443.62 ± 11.07 and 373.28 ± 27.13 pg/mL for the lungs of control group, CM and CM + NIR respectively. Meanwhile, the inflammatory factors expression levels were also investigated by immunohistochemical staining. For TNF-α expression, it was in the low levels in the lung of sham group, slightly increased in the lung of control group, and significantly ascended in the lungs of CM and CM + NIR respectively. It was similar to the iNOS expression of the lung with the order of sham group < control group < CM < CM + NIR (Fig. 6c). By statistical analysis, the average optical density (AOD) of TNF-α and iNOS expression was 0.002 ± 0.000 and 0.006 ± 0.001 in the lung of sham group, increased to 0.004 ± 0.000 and 0.014 ± 0.001 in the lung of control group, 0.008 ± 0.001 and 0.025 ± 0.002 in the lung of CM, and 0.011 ± 0.001 and 0.043 ± 0.003 in the lung of CM + NIR respectively (Fig. 6d). In addition, the macrophage polarization levels of the lungs were also analyzed. As imaged in Fig. 6e, compared to other groups, the obvious red fluorescence corresponding to CD86 expression, was ascended in the lung of CM + NIR, while the CD206 expression was significantly descended in the lung of CM + NIR with the obvious decreased green fluorescence. By statistical analysis, the MFI of CD86 expression was 5.38 ± 0.18, 4.09 ± 0.12, 5.63 ± 0.14 and 12.20 ± 0.67 in the lungs of sham group, control group, CM and CM + NIR, while it was 13.11 ± 0.54, 8.41 ± 0.22, 5.57 ± 0.40 and 1.16 ± 0.05 for the CD206 expression levels in the lungs of sham group, control group, CM and CM + NIR respectively (Fig. 6f). After calculation, the CD86/CD206 (M1/M2) ratio in the lung of treated rats was in the order of CM + NIR (1052.99 ± 27.19) > CM (101.30 ± 4.96) > control group (48.60 ± 0.21) > sham group (41.02 ± 0.48). The high ROS levels, high expression levels of inflammatory factors together with M1 directional polarization of macrophage mediated by CM + NIR in the lung were helpful to killing bacteria and devouring bacteria, achieving the immunotherapy of bacterial pneumonia.

Fig. 6.

In vivo immunotherapy evaluation. a) The bacterial pneumonia immunotherapy mechanism by induced macrophage. b) The inflammatory factors (IL-6 (i), TNF-α (ii), and IL-10 (iii)) expression levels in the lung homogenate of treated rats by ELISA. c) The immunohistochemical staining images in the lung tissue of treated rats, (Scale bar: 200 μm) and the corresponding quantified results (d): TNF-α (i) and iNOS (ii). e) The macrophage polarization (CD86 and CD206) co-immunofluorescent staining images in the lung tissue of treated rats, (Scale bar: 200 μm) and the corresponding quantified results (f): CD86 (i), CD206 (ii) and M1/M2 polarization ratio (iii). g) The number of T cells in the blood of treated rats by flow cytometry, and the corresponding CD4⁺/CD8⁺ ratio (h). i) The number of T cells in the lung tissue of treated rats by flow cytometry, and the corresponding CD4⁺/CD8⁺ ratio (j). k) The CD4⁺ and CD8⁺ T cells co-immunofluorescent staining images in the lung tissue of treated rats, (Scale bar: 200 μm) and the corresponding quantified results (l): CD4⁺ (i) and CD8⁺ (ii) T cells. The corresponding groups were: rats without treatments (sham group), and bacterial pneumonia rats with IV injection of saline (control group), CM (CM), and CM combining with NIR irradiation (1.0 W/cm²) (CM + NIR). (“∗” symbol compared with control group, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001).

On top of that, the number of CD4⁺ and CD8⁺ T cells in the blood serum, lung and spleen of treated rats was also evaluated. From Fig. S49 and 6g, for the blood serum of treated rats, it was obviously observed that the number of CD4⁺ T cells increased in CM (60.03 ± 0.23%) and CM + NIR (67.57 ± 0.25%) compared to sham group (52.67 ± 0.38%) and control group (52.30 ± 1.04%). However, the number of CD8⁺ T cells was reduced with the order of control group (39.50 ± 0.60%) > sham group (39.17 ± 0.85%) > CM (35.40 ± 0.10%) > CM + NIR (24.83 ± 0.12%). After calculation, the CD4⁺/CD8⁺ ratio was 1.35 ± 0.04%, 1.32 ± 0.04%, 1.70 ± 0.01% and 2.72 ± 0.02% in the blood serum of sham group, control group, CM and CM + NIR respectively (Fig. 6h). Significantly, it also displayed the similar tendency for the number of CD4⁺ and CD8⁺ T cells, and the CD4⁺/CD8⁺ ratio in the lung (Fig. S50, 6i and 6j) and spleen (Fig. S51 and S52) of treated rats. CM + NIR could efficiently increase the number of CD4⁺ T cells, decrease the number of CD8⁺ T cells as well as promote the CD4⁺/CD8⁺ ratio. Besides, the number of CD4⁺ and CD8⁺ T cells in the lung was also characterized by immunofluorescent staining. As imaged by Fig. 6k, compared to other groups, the green fluorescence obviously increased in the lung of CM + NIR corresponding to the increased number of CD4⁺ T cells, while the number of CD8⁺ T cells was significantly reduced, revealed by the decreased red fluorescence in the lung. After statistical analysis, the MFI of CD4⁺ T cells was 0.23 ± 0.04% in the lung of sham group, increased to 0.75 ± 0.09% in the lung of control group, 2.82 ± 0.11% in the lung of CM and 3.89 ± 0.05% in the lung of CM + NIR respectively. And no obvious differences were observed for the MFI of CD8⁺ T cells in the lungs of sham group (0.79 ± 0.15%) and control group (0.83 ± 0.10%), decreased in the lungs of CM (0.70 ± 0.05%) and CM + NIR (0.21 ± 0.02%) (Fig. 6l). Significantly, the CD4⁺/CD8⁺ ratio was 29.09 ± 0.60% in the lung of sham group, sequentially increased to 91.30 ± 21.21%, 402.16 ± 28.37% and 1850.26 ± 167.61% in the lungs of control group, CM and CM + NIR respectively (Fig. S53). Generally, the number of CD4⁺ T cells increased, indicating the immune enhancement, commonly seen in various infectious bacterial diseases. And the CD4⁺/CD8⁺ ratio was above 2.5, corresponding to the overactive cellular immune function, prone to autoimmune reactions [65].

From the above, it confirmed the feasibility of the strategy of CM + NIR for bacterial pneumonia targeted therapy. And it also confirmed the excellent biosafety of this therapeutic strategy including the blood biocompatibility, blood indicators, body weight, and general observation and pathological features of major organs and skin. Significantly, the multimode antibacterial mechanisms of CM + NIR were achieved by NIR amplified ROS generation, Cu overload, and immunoactivation, further producing a huge amount of ROS and inflammatory factors, inducing macrophage M1 polarization, strengthening the engulfment and decomposition function of macrophage, and activating the number of T cells. On the other hand, the Cu overload and PTT had confirmed the ability of accelerating the tissue repair, and promoting angiogenesis [66]. The elimination of bacteria was also helpful to accelerating tissue repair [67]. Similarly, the activation of T cells number was also conducive to tissue repair [68]. All of these confirmed that the strategy of CM + NIR could mostly efficiently arouse autoimmune response, and ultimately achieve immunotherapy of bacterial pneumonia, contributing to the repair and treatment of infected tissues.

4. Conclusion

In summary, it designed a novel biomimetic nanozyme (CM) combining with NIR irradiation for enhanced bacterial pneumonia immunotherapy. The controllable Cu ion release, excellent photothermal effect, and NIR driven ROS generation of CB had been demonstrated. Significantly, it exhibited the excellent therapeutic performance in combating with bacterial pneumonia for CM + NIR in vitro and in vivo through specific targeting infected lung site, excellent antibacterial capabilities, and accelerated lung tissue repair. Particularly, it also confirmed that the intracellular Cu overload by CM + NIR efficiently promoted bacterial cuproptosis via transcriptional regulation and metabolic intervention, and activated immunoregulation via the increased chemotaxis and phagocytosis of immune cells, synergistically achieving the enhanced all-stage bacteria clearance, and providing a novel anti-infectious therapeutic strategy in clinics. However, for future clinical translation, the comprehensive toxicity evaluation and validation in infectious patients after this therapeutic strategy are warranted.

Ethics approval and consent to participate

In vivo experiment was approved by the ethics committee of animal experiments of the first affiliated hospital of Guangxi Medical University (No. 2025-E0936). SD rats (180∼220 g) were operated by the local guide for the care and use of laboratory animals of Guangxi Medical University.

Funding

This study was financially supported by the Joint Project on Regional High-Incidence Diseases Research of Guangxi Natural Science Foundation (Grant No. 2024GXNSFAA010089 and 2025GXNSFBA069275), the National Natural Science Foundation of China (Grant No. 82360066), the Youth Science Foundation of Guangxi Medical University (Grant No. GXMUYSF202334), and the Major Talent Project of Guangxi Autonomous Region.

CRediT authorship contribution statement

Weiqian Jin: Data curation, Writing – original draft. Guoxing Ling: Writing – original draft. Jiaxiao Li: Data curation, Writing – original draft. Shigao Ye: Writing – original draft. Yan Liu: Methodology, Writing – original draft. Ruikai Zhu: Formal analysis. Jing Qian: Software. Yongfeng Lan: Methodology. Wenquan Lv: Data curation. Jing Liu: Data curation. Xiongwei Cai: Methodology. Yongyuan Jian: Formal analysis. Jingwei Jiang: Data curation. Zuyuan Huang: Formal analysis. Yuan Cao: Methodology. Lin Liao: Data curation. Ming Gao: Writing – review & editing. Cheng Luo: Writing – original draft, Writing – review & editing. Baoshi Zheng: Writing – original draft, Writing – review & editing.

Declaration of competing interest

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

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

All data required to assess the conclusions of the paper are included in the paper and/or in supplemental materials. Other data relevant to this article can be obtained from the authors.