Bacterial strains, cells, and animals

E. coli DH5α is preserved in the bacteria library of our laboratory and cultured in Luria-Bertani (LB) Broth (BKMAN, Changsha, China). Murine mammary carcinoma 4T1, B16 melanoma, and human osteosarcoma U2OS cells were acquired from Procell (Wuhan, China). RAW264.7 and L929 cells were sourced from Xiangya Hospital of Central South University (Changsha, China). Female BALB/C mice (4–6 weeks old, 15–18 g) were supplied by Hunan SJA Laboratory Animal Company and housed in the Department of Laboratory Animals at Central South University. Primary bone marrow-derived macrophages (BMDM) were harvested from female BALB/C mice (4–6 weeks) and cultured in DMEM/F12 (Procell, Wuhan, China) containing 10% FBS and 25 ng/mL murine macrophage colony-stimulating factor (M-CSF) (R&D Systems, USA). All animal experiments were conducted following the ARRIVE guidelines and were approved by the Ethics Committee of Xiangya Hospital at Central South University. (202,110,140).

4T1 membrane and OMVs preparation

The 4T1 cell membrane was obtained as previously described [37, 38]. Briefly, 4T1 cells were harvested when the cell density reached 80–90% on cell dishes. The cells were washed twice with PBS by centrifugation at 1,500 g for 5 min. Subsequently, cells were resuspended in double-distilled water and the cell membrane was disrupted using an ultrasound processor (Sonic and Materials Inc., USA) operating at 20 kHz and 130 W, with 5 s on and 5 s off for 5 min, while kept on ice.

Following the disruption, the cell membrane was extruded through a 200 nm polycarbonate membrane in a liposome extruder (Avestin, Germany) for 20 cycles to standardize the membrane particle size. The suspension was then centrifuged at 15,000 g for 30 min to obtain 4T1 cell membrane vesicles. The protein content of the cell membrane was measured using a BCA kit, and the vesicles were stored at -80℃.

E. coli strains are commonly used for producing OMVs in tumor immune treatment [28, 34, 39]. In this study, E. coli DH5α was cultured for OMV production [30, 32]. Following established protocols [37], E. coli DH5α was cultured in LB Broth in a shaking incubator at 37℃. The bacterial suspension was collected when the medium’s OD600nm reached 1.5 and then centrifuged at 5,000 rpm for 5 min at 4℃. The supernatant was extruded through a 200 nm filter twice to remove residual bacteria.

To obtain OMVs, the supernatant was added to an Amicon Ultra15 centrifugal filter tube (10 kDa; Millipore, USA) and centrifuged at 1,500 g for 20 min. The concentrated liquid was transferred to a new tube, and 200 µL of Exoquick TC (System Biosciences, Bay Area, California, USA) was added. The mixture was incubated at 4℃ for 12 h and then centrifuged at 1,500 g at 4℃ for 30 min. The precipitate obtained was OMVs. OMVs were resuspended in PBS, and the protein amount was measured using a BCA kit. The OMVs were then stored at -80℃.

Construction of the 4T1-OMVs hybrid membrane

The hybrid membrane was constructed by infusing 4T1 cell membrane and OMVs at a 1:1 weight ratio. The membrane weight was considered twice as much as the protein weight measured above [32]. The membrane suspension was mixed and placed into an ultrasonic bath at 37 ℃ for 20 min, using an ultrasonic cleaner (Granbo Sonic, Shenzhen, China). The suspension was then extruded through a 200 nm polycarbonate membrane 11 times to physically infuse the membrane. Finally, the mixture was centrifuged at 12,000 rpm for 30 min to obtain the hybrid membrane of 4T1 cells and OMVs [28].

To identify the successful construction, 4T1 membrane, OMVs, and hybrid membrane were observed using a transmission electron microscope (TEM, Hitachi H-7600). Particle sizes and zeta potentials were determined using a dynamic light scattering (DLS) analyzer (Malvern Nano ZS, UK). Western blot analysis was employed to verify the protein components of HM, utilizing anti-VCAM-1 (Abclonal, A19131, Rabbit, 110 KD) and anti-OMPC (abbexa, abx243143, Rabbit, 60 KD). 4T1 membrane and OMVs were labeled with DiO and DiI (Beyotime, Shanghai, China), respectively, before being mixed. Following construction, the labeled HM was cultured with 4T1 cells for 4 h and visualized using confocal laser scanning microscopy (CLSM, Carl Zeiss, LSM 510 META).

Construction of IR780@PLGA@HM nanoparticles

IR780-loaded PLGA nanoparticles (referred to as IR780@PLGA) were constructed using a single emulsion evaporation protocol, protected from light [38, 40, 41]. PLGA, IR780 iodide, and polyvinyl alcohol (PVA) were purchased from Sigma Aldrich (St. Louis, MO, USA). To initiate the process, 100 mg of PLGA and 1 mg of IR780 iodide were dissolved in 3 mL of dichloromethane. After complete dissolution, 10 mL of pre-cooled PVA solution (4% w/v), previously chilled at 4 °C, was added to the solution. An ultrasonic processor (Sonics, VCX150, USA) was employed to emulsify the mixture for 4 min with a 5-second on-and-off cycle. Subsequently, the emulsified solution was introduced into 30 mL of double-deionized water and stirred at room temperature for 3 h to allow for evaporation. The resulting solution was collected, and centrifuged at 10,000 rpm for 7 min, and after washing the centrifuged precipitate twice with double deionized water, the IR780@PLGA nanoparticle sediment was obtained.

The 4T1 cell membrane and HM were both utilized in the construction of IR780@PLGA@4T1 and IR780@PLGA@HM nanoparticles, respectively. A 500 µL solution of IR780@PLGA (1 mg/mL) was mixed with a 500 µL membrane solution (1 mg/mL) and sonicated for 10 min to coat the membrane onto IR780@PLGA. The solution was then centrifuged at 8,000 rpm for 5 min to remove any uncoated membrane. The deposit was washed twice with double-deionized water to obtain the final membrane-coated nanoparticles. The nanoparticles were redispersed in 1 mL PBS for further detection and characterization.

Characterization of IR780@PLGA@HM nanoparticles

To identify the coating, IR780@PLGA, and IR780@PLGA@HM were observed with TEM and analyzed by DLS. Western blotting was also employed to verify the specific components of HM on the particles. Additionally, HM was labeled with DiO, IR780@PLGA nanoparticles were labeled with DiI, and IR780@PLGA@HM were visualized using CLSM. The presence of IR780 within IR780@PLGA and IR780@PLGA@HM nanoparticles was assessed using a UV-Vis-NIR spectrophotometer (Cary 5000, USA).

IR780 loading and encapsulation efficiencies were calculated according to the following equation:

$$\text\left(\text\right)=\frac}\\}\end}}\times 100\%$$

$$\text\left(\text\right)=\frac}\\}\end}}\times 100\%$$

To assess the stability of IR780@PLGA@HM nanoparticles, placed them in PBS buffer at 4℃ for one week, with daily measurements of their particle size and Polydispersity Index (PDI). A fluorescent probe, Singlet Oxygen Sensor Green (SOSG), was employed to quantify reactive oxygen species (ROS) production. In a quartz cuvette, 100 µL of IR780@PLGA@HM nanoparticle solution at varying concentrations (0.5, 1, 1.5, 2 mg/mL) and 1 µL of SOSG (1 mM) were mixed. The ROS generation was induced by irradiating the mixture using a low-frequency US transducer (WED-100, WELLD Medical Electronics, China) at 2 W/cm2, 1 MHz, and 50% duty cycle for 30 s. Subsequently, the fluorescence spectra of SOSG were recorded on a fluorescence spectrometer, with an excitation wavelength of 504 nm. To investigate the time-dependent ROS production, 1.5 mg/mL of IR780@PLGA@HM nanoparticles were subjected to ultrasound treatment for 40, 60, and 120 s.

Evaluation of biosafety of IR780@PLGA@HM nanoparticles in vitro and in vivo

RAW264.7 and L929 cells were employed to evaluate the toxicity of IR780@PLGA@HM in vitro. Briefly, cells were seeded in 96-well plates at a density of 5,000 cells per well. After 24 h, the cells were treated with different concentrations of IR780@PLGA@HM (0, 500, 1,000, 1,500, 2,000 µg/mL). Following 24 and 48 h of incubation, cell viability was assessed using the Cell Counting Kit-8 (CCK8) Kit (New Cell & Molecular Biotech Co., Ltd., Suzhou, China). To evaluate the sonodynamic cytotoxicity of nanoparticles, 24 h after IR780@PLGA@HM treatment, cell culture was removed and cells were washed with PBS 3 times to remove the untaken nanoparticles. Then cells were treated with ultrasound (1 W/cm2, 1 MHz, 10 s on and 10 s off for 2 min). After that, cell viability was evaluated using the CCK8 Kit.

Furthermore, a hemolysis experiment was conducted to evaluate the biosafety of nanoparticles. Fresh RBCs were collected from a BALB/C mouse by centrifuging blood and washing the cells with PBS three times. The RBCs were then resuspended in PBS, and 0.25 mL of a 0.2% RBC solution (v/v) was added to 0.75 mL of nanoparticle solution with different concentrations (final concentrations: 0, 500, 1,000, 1,500, 2,000 µg/mL). After incubating for 0.5, 1, 2, and 3 h, the solution was centrifuged, and the absorbance of supernatants at 540 nm was measured. RBCs treated with deionized water served as a positive control.

To evaluate biosafety in vivo, BALB/C mice were intravenously injected with 200 µL of PBS or 10 mg/mL of IR780@PLGA@HM twice a week (n = 3 per group). 2 weeks later, the mice were euthanized to collect blood for analysis. Detected a complete blood count to reflect the quantity of various types of blood cells. Aspartate aminotransferase (AST), Alanine aminotransferase (ALT), Alkaline phosphatase (ALP), and Total protein (TP) were assessed to reflect hepatic function, while Blood Urea Nitrogen (BUN) and Uric acid (UA) were measured as an indicator of kidney function. Heart, lung, liver, spleen, and kidney were collected for HE staining to observe the toxicity of nanoparticles on organs.

Evaluation of targeting ability of IR780@PLGA@HM nanoparticles in vitro

CLSM observation was employed to assess the targeting ability of nanoparticles, while flow cytometry was utilized to quantify cell uptake. The homotypic targeting ability of nanoparticles was evaluated in the following groups: IR780@PLGA, IR780@PLGA@4T1, and IR780@PLGA@HM (n = 3 per group). The concentration of nanoparticles was 1000 µg/mL.

For CLSM observation, 4T1, B-16, and U2OS cells were seeded in confocal dishes at a density of 50,000 cells per dish and cultured for 24 h. IR780@PLGA nanoparticles labeled with DiI and the cell membrane with DiO during construction were added to the culture medium of 4T1 cells at a concentration of 1000 ug/mL and incubated for 1, 2, and 4 h. Subsequently, free nanoparticles were washed out with PBS, cells were stained with DAPI, and observation was conducted with CLSM. B-16 and U2OS cells were incubated with IR780@PLGA@HM (1000 ug/mL) for 2 h and detected using CLSM.

To discuss the targeting ability of IR780@PLGA@HM towards RAW264.7 cells and 4T1 cells, treated cells with IR780@PLGA@HM (1000 ug/mL) for 4 h and detected them using CLSM. Nanoparticles were labeled with DiI and cells were stained with DAPI.

For flow cytometry detection, 4T1, B-16, and U2OS cells were seeded in 6-well plates at a density of 150,000 cells per well. After 24 h, 4T1 cells were treated with different nanoparticles for 1, 2, and 4 h. B-16 and U2OS cells were incubated with IR780@PLGA@HM (1000 ug/mL) for 2 h. Subsequently, cells were harvested and their fluorescence at 780 nm was measured using a flow cytometer.

Evaluation of macrophage polarization in vitro

RAW264.7 cells were utilized to assess the impact of IR780@PLGA@HM on macrophage polarization. The cells were plated in 12-well plates, and when the density reached 60%, they were treated in various groups (n = 3 per group): control medium, OMV, IR780@PLGA, IR780@PLGA@4T1, and IR780@PLGA@HM. The concentration of nanoparticles was 1000 µg/mL. After 24 hours, the cells were washed with PBS twice and collected for analysis. Flow cytometry was employed to quantify the M1 polarization proportion. Primary BMDM cells were harvested and also treated with different nanoparticles as described above. After 24 hours, detected the M1 polarization proportion of BMDM using flow cytometry. RAW264.7 cells were stained with PE anti-mouse CD86 Antibody (Biolegend, Beijing, China) and APC anti-mouse F4/80 Antibody (Biolegend, Beijing, China) before being analyzed in a flow cytometer. RNA extraction was performed using Trizol (Yeasen, China). Following the measurement of RNA concentration, the expression levels of CD86, CD206, IL-6, TNF-α, and IFN-γ in different groups were assessed using the RT-qPCR SYBR Green Kit (Yeasen, China). The primer sequences used are listed below (5’ to 3’):

CD86 F: TCTCCACGGAAACAGCATCT,

CD86 R: CTTACGGAAGCACCCACGAT.

CD206 F: CCTATGAAAATTGGGCTTACGG,

CD206 R: CTGACAAATCCAGTTGTTGAGG.

IL-6 F: ATCCAGTTGCCTTCTTGGGACTGA,

IL-6 R: TTGGATGGTCTTGGTCCTTAGCCA.

TNF-Α F: AGCCGATGGGTTGTACCTTG,

TNF-α R: ATAGCAAATCGGCTGACGGT.

IFN-γ F: ATGAACGCTACACACTGCATC,

IFN-γ R: CCATCCTTTTGCCAGTTCCTC.

The culture medium from the cells was collected to assess the expression levels of various cytokines, including IL-6, TNF-α, and IFN-γ, using Cytometric Bead Array (CBA).

In vitro anti-tumor efficacy

4T1 cells were seeded in a 12-well plate at a density of 50,000 cells per well. After 24 h, cells were treated with different nanoparticles and subjected to ultrasound. Based on the difference in nanoparticles and ultrasound (US) conditions, the tumor cells were divided into 8 groups: US (-) with control (PBS), IR780@PLGA, IR780@PLGA@4T1, and IR780@PLGA@HM; US (+) with control (PBS), IR780@PLGA, IR780@PLGA@4T1, and IR780@PLGA@HM (n = 3 per group). The concentration of nanoparticles was 1000 µg/mL. After 4 h of treatment, the US (+) group underwent low-frequency ultrasound (1 W/cm2, 1 MHz, 10 s on and 10 s off) for 2 minutes, while the US (-) group did not receive ultrasound treatment.

For cell viability assessment, the culture medium was removed, cells were washed twice with PBS and then stained with the Calcein-AM/PI Double S7 stain kit (Yeasen, Shanghai, China) at 37 °C for 15 min. Live cells were stained green, and dead cells were stained red. The Leica fluorescence microscope (Leica Microsystems) was used to observe the survival status of tumor cells. For ROS detection, the Reactive Oxygen Species Assay Kit (Yeasen, Shanghai, China) was used. After removing the medium, cells were washed twice with PBS, and then stained with a working solution of DCFH-DA at a concentration of 10 mM in serum-free medium at 37 °C for 30 min. Subsequently, the Leica fluorescence microscope was employed to observe ROS production within cells, which appeared green under excitation at 488 nm.

Establishment of an animal model of breast Cancer bone metastasis

After a one-week acclimatization to the new environment, female BALB/C mice aged 4–6 weeks were used to establish the animal model. Following anesthesia, the skin surrounding the tibial plateau was removed. Using a 1 mL syringe needle, a hole was drilled into the tibial plateau, directed towards the distal end of the tibia. Subsequently, a 20 µL suspension containing 4T1 cells (with a cell count of 1 × 105) was injected into the tibial plateau using an insulin needle. Postoperatively, the wound status was monitored daily. One week after the surgery, ultrasound was employed to assess the tibia, evaluating the continuity of the bone cortex and the surrounding blood flow to confirm the success of the modeling. Breast cancer bone metastasis occurred approximately one week after the surgery [36].

In vivo distribution of IR780@PLGA@HM nanoparticles

To assess the targeting efficacy of IR780@PLGA@HM nanoparticles towards 4T1 tumors, BALB/C mice bearing 4T1 bone metastasis received intravenous injections of 200 µL of 10 mg/mL solutions containing different nanoparticles (n = 3 per group): IR780@PLGA, IR780@PLGA@4T1, and IR780@PLGA@HM. In vivo fluorescence was monitored at various time intervals (4 h, 12 h, 24 h, 48 h, and 72 h post-administration) using a Lumina IVIS Spectrum imaging system (PerkinElmer, USA). After 72 h, the limbs and major organs were harvested and subjected to ex vivo fluorescence imaging to assess nanoparticle distribution. The IVIS system was utilized to quantify fluorescence intensity.

In vivo anti-tumor efficacy

Tumor-bearing mice were randomly assigned to eight treatment groups: PBS, US, IR780@PLGA, US + IR780@PLGA, IR780@PLGA@4T1, US + IR780@PLGA@4T1, IR780@PLGA@HM, US + IR780@PLGA@HM (n = 10 per group). Following the successful modeling, nanoparticles were administrated twice a week, intravenously at a dose of 200 µL with a concentration of 10 mg/mL. Ultrasound was applied one day after nanoparticles treatment, with parameters as 1 W/cm2, 1 MHz, with a 10-second on and 10-second off cycle for 2 minutes. Mice survival, leg circumference, and body weight were monitored and recorded every three days.

After a two-week treatment period, mouse legs, main organs, and blood samples were collected to evaluate treatment outcomes. Micro-computed tomography (micro-CT, Viva CT-80) scans of the legs were performed, and SkyScan CT analysis software (SCANCO Medical AG, Zurich, Switzerland) was utilized to analyze percent bone volume (bone volume/tissue volume, BV/TV), bone surface/bone volume ratio (BS/BV), Trabecular number (Tb. N), Trabecular separation (Tb. Sp) and Trabecular thickness (Tb. Th).

Morphological changes in the mouse tibia were assessed through Hematoxylin and eosin (HE) staining. Ki67, TNF-α, and IFN-γ expression were detected by immunohistochemistry to reflect tumor growth and inflammatory regulation. Macrophage polarization was detected by immunofluorescence staining, with CD86 and CD206. Blood samples underwent CBA to detect anti-tumor cytokines expression. Spleen tissues were collected for flow cytometry analysis to investigate dendritic cell (DC) maturation.

Statistical analysis

All results were shown as the mean values ± standard deviations after repeating at least three times. One-way ANOVA and Student’s t-test were adapted to analyze the data by using GraphPad Prism 8.0.1 software. *p < 0.05 and #p < 0.05 were considered statistically significant.