Membrane fusogenic liposomes facilitate the production of immunostimulatory extracellular vesicles for enhanced cancer therapy

Abstract

Cancer therapy based on extracellular vesicles faces several practical challenges, such as low production, inadequate stability, and inefficiency, making it difficult to achieve ideal therapeutic effects. To address these issues, a novel engineering technology of EVs was developed based on the autonomous “fusion-exocytosis” process. Specifically, membrane fusogenic liposomes (MFLs) with unique physical and chemical properties were engineered by tailoring the components, which could interact with cells to facilitate mutual fusion of lipid membranes and release fusogenic extracellular vesicles (FEVs). As an inducer of immunogenic cell death (ICD) and enhancer of EV production, monensin (Mon) was incorporated into MFLs. The synergistic action of MFLs and Mon not only increased the production of released FEVs but also significantly enhanced their immunogenicity, effectively promoting maturation of dendritic cells (DCs). Consequently, the obtained FEVs effectively inhibited tumor growth by activating anti-cancer immune responses. This engineering method of generating FEVs could substantially increase production and enhance immunogenicity, which would facilitate their clinical translation for cancer therapy.

---

1. Introduction

Extracellular vesicles (EVs) are naturally occurring, spherical vesicles secreted by cells with double-layer membrane structures.^1,2 EVs can hijack the membrane components and cytoplasmic contents of parental cells, making them serve as important vectors for the exchange of information and substance between cells.³ As the epitome of cancer cells, the EVs derived from cancer cells can carry antigen information.^4,5 EVs transfer antigens to dendritic cells (DCs), which then process cancer-related antigens and present them to T cells, thereby activating the anti-cancer immune response to kill cancer cells.^6,7 Therefore, EV-mediated cellular communication plays a vital role in the immune regulation process in cancer therapy.^8,9 Despite their therapeutic potential, the cancer treatment technologies based on EVs are limited by challenges such as low production, poor stability, and inefficient immune activation capacity, making it difficult to achieve ideal therapeutic effects. To address these shortcomings, the current common strategy is to modify EVs through physical, mechanical, chemical or genetic engineering methods. However, these processes would inevitably cause damage to the structure and biological function of EVs.¹⁰ Therefore, it is necessary to develop an engineering method that could increase yield, achieve functional modification and does not affect the natural structures and properties of EVs.

The surface properties of nanomedicines determine how they interact with cells.^5,11–14 As the widely used commercial nanomedicines, conventional liposomes are internalized through the endocytosis pathway.^15,16 In contrast, the cationic liposomes with specific surface positive charges interact with cells via fusion.^17–21 Such liposomes with fusogenic properties are called membrane fusion liposomes (MFLs).^22,23 MFLs interact with cells through membrane fusion, facilitating the integration of liposome components into the cell membrane.^24–29 Then, cells secreted EVs through the multivesicular body (MVB) pathway or the budding pathway. The EVs produced through the “fusion-exocytosis” process are hereinafter referred to as fusogenic EVs (FEVs). They carry both the materials of the parent cells and the functional components of the MFLs. This process does not affect the natural properties and structures of the vesicles. The utilization of MFLs alone can yield FEVs with specific characteristics, but the production efficiency and immunogenicity of these FEVs still fall short in achieving effective therapeutic outcomes.

Monensin (Mon), a lipophilic antibiotic, can affect ion transport carriers on cell membranes and mediates Na⁺/H⁺ exchange.³⁰ It causes a continuous increase in intracellular Na⁺. The elevated intracellular Na⁺ would activate the Na⁺/Ca²⁺ exchanger,³¹ leading to an increase in Ca²⁺ within the cytoplasm.³² Then, the elevated Ca²⁺ stimulated the MVBs, thereby promoting the secretion of EVs. Therefore, Mon could be loaded into the lipid layers of MFLs, subsequently referred to as Mon@MFLs. Due to the fusion of Mon@MFLs with the cell membrane, Mon was anchored to the cell membrane, thus maintaining everlasting efficacy. Mon@MFLs could serve as a potential tool to improve the production of FEVs (Fig. 1A).

Fig. 1 Schematic illustration of the design and antineoplastic application of FEVs. (A) Mon@MFLs boosted the production of FEVs via the “fusion-exocytosis” process. (B) Mon@MFLs induced ICD of cancer cells. (C) Immunostimulatory FEVs initiated the anti-cancer immune response. The figure was drawn by Figdraw (https://www.figdraw.com).

Colorectal cancer frequently progresses to involve the peritoneum, resulting in peritoneal metastasis denominated as peritoneal carcinomatosis from colorectal cancer (PCCC). As the disease develops, cancer cells proliferate uncontrollably within the peritoneal cavity, facilitated by a highly permeable neovascular network, leading to the accumulation of malignant ascites.^33,34 Current treatments mainly include surgery and chemotherapy, but the varied nature of colorectal cancers necessitates the development of novel therapeutic strategies to improve patient outcomes. Cancer cell-derived EVs are rich in biomolecules and antigenic information, making them a promising delivery platform. These EVs have been utilized as cell-free vaccines to potentiate the immune response against tumors and suppress the proliferation of malignant cells. Herein, Mon@MFLs were used to stimulate cancer cells to produce a large number of FEVs. The Ca²⁺ elevation induced by Mon not only promoted the secretion of FEVs, but also induced immunogenic cell death (ICD) in cancer cells (Fig. 1B). ICD was driven by stress, in which damage-related molecular patterns (DAMPs) and cancer-related antigens were released during the process. Therefore, FEVs carried immunostimulatory molecules and effective cancer antigens, which can activate the anti-cancer immune responses and achieve enhanced immunotherapy. The engineering technology based on Mon@MFLs would yield a substantial quantity of immunostimulatory FEVs, thereby offering an effective approach for cancer treatment (Fig. 1C).

2. Materials and methods

Materials and reagents

Mon was purchased from MedChemExpress (China). Vitamin E (VE) and N,N-dimethyl-1,3-propylene diamine were purchased from Aladdin (China). 1,2-Dipalmitoyl-sn-glycero-3-phosphorylcholine (DPPC), cholesterol and 1,2-distearoyl-sn-glycero-3-phospho-ethanolamine-rhodamine B (DSPE-RB) were purchased from Xi’an Ruixi Biological Technology (China). LysoTracker Green and Hoechst 33342 were purchased from Yeasen Bio (China). Antibodies were purchased from Cell Signaling Technology (USA) and Abcam (UK). 1,1′-Dioctadecyl-3,3,3′,3′-tetramethylindodicarbocyanine, 4-chlorobenzenesulfonate salt (DiD) and 3,3′-dioctadecyloxacarbocyanine perchlorate (DiO) were purchased from Beyotime Biotechnology (China). 4′,6-Diamidino-2-phenylindole (DAPI) was purchased from Biosharp (China). Penicillin–streptomycin solution (100×), Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS) and trypsin were obtained from Hyclone (USA). Phosphate-buffered saline (PBS) powder was purchased from Servicebio (China). Centrifuge tubes were obtained from Kirgen Bioscience (Shanghai).

Cell lines and animals

MC38 cells and DC2.4 cells were cultured in DMEM medium containing 10% FBS and 1% penicillin–streptomycin in a humidified atmosphere incubator with 5% CO₂ at 37 °C. All animals were purchased from the Laboratory Animal Resources of Huazhong University of Science and Technology (HUST) and managed under specific pathogen-free (SPF) conditions at the Animal Center of HUST. All animal experiments were conducted in accordance with the principles from the Institutional Animal Care and Use Committee (IACUC) of HUST and approved by the IACUC of HUST (2024 IACUC Number: 4100).

Preparation and characterization of MFLs and Mon@MFLs

The vitamin E cationic lipid (VECL) used in this study was obtained by the same synthetic route as described in previous studies.¹⁴ First, VE reacted with acrylic chloride under the catalysis of triethylamine to synthesize acryl-VE. Then, acryl-VE reacted with N,N-dimethyl-1,3-propanediamine through the Michael addition reaction. Finally, the compound reacted with methyl iodide to synthesize the final cationic lipids. After successful synthesis, column chromatography was employed for purification. The final product was characterized by mass spectra. MFLs were prepared by thin-film hydration. No cationic lipid liposomes (NFLs) were prepared as a control.¹⁸ VECL, DPPC, and cholesterol were dissolved in chloroform at a molar ratio of 2 : 4 : 4. The solvent was removed by the rotary evaporation of the lipid mixture at 55 °C, allowing the lipids to form a uniform thin film on the flask wall.²⁵ Then, the ultrapure water or saline was added to the flask and stirred under ultrasonic conditions to make it completely hydrated, and finally the membrane fusion liposomes were obtained. The monensin-loaded MFLs (Mon@MFLs) were prepared by the same method. Mon was dissolved in methanol and added to the lipid mixture at 1% of the total lipid content to prepare Mon@MFLs. Dynamic light scattering (DLS, zeta PALS, Brookhaven Instruments, USA) was used to determine particle size, polydispersity index (PDI) and zeta potential of MFLs.

Membrane fusion efficiency of MFLs

To prepare DSPE-RB labeled liposomes, they were added and mixed with the lipid mixture, followed by rotary evaporation and hydration. The ratio of DSPE-RB lipid concentration for cell uptake was 20 μg mL⁻¹. MC38 cells were seeded in confocal microscopy dishes. The original media was then removed, and blank media containing NFLs or MFLs was added for co-incubation for 2 h or 4 h. The liposomes-containing medium was removed and then washed with an appropriate amount of PBS solution to remove the excess liposomes.²² Next, the cells were fixed using paraformaldehyde and stained with DAPI for CLSM observation.²⁹

Preparation and characterization of FEVs

MC38 cells were inoculated into 60 mm culture dishes and cultivated in DMEM until the cell density reached 5 × 10⁶ cells per dish. The culture condition was set as 5% CO₂ and the temperature of 37 °C. Then, fresh blank media was replaced, and Mon@MFLs (at the Mon concentration of 8 μM³¹ and the total lipid concentration of 500 μg mL⁻¹) were added to further incubate for 4 h. Mon@MFLs entered the cells through membrane fusion. To ensure the purity of cell-released vesicles, excessive liposomes were removed, and the same volume of fresh media was supplemented for continuous cultivation for 24 h. The media was collected and centrifuged at 1500 rpm for 10 min to remove excess cells and cell debris. Subsequently, centrifugation was performed at 13 000 rpm for 60 min to collect the obtained vesicles, FEVs or EVs were obtained for the subsequent experiments.

To characterize these nanostructures of EVs or FEVs, drop the vesicles onto the copper grid, let them settle for 30 min, then stain them with phosphotungstic acid for 1 min and take images on the transmission electron microscope (TEM). To determine the protein content in the FEVs or EVs, the bicinchoninic acid assay (BCA) protein quantification method was used. First, the BCA working solution was prepared according to the instructions in the kit, and a standard curve was established using BSA.⁹ Then, the samples were appropriately diluted and added to 96-well plates, with three replicates for each sample. The samples were incubated with standard substances and blank controls at 37 °C for 30 min; finally, the absorbance values were measured using a microplate reader, and the protein concentration in the samples was calculated based on the standard curve.¹⁴

SDS-PAGE was utilized to characterize the proteins. EVs and FEVs were lysed on ice by RIPA lysis buffer for 30 min, and the protein concentration was determined using a BCA assay kit. Then, the lysate was mixed with protein loading buffer and boiled for 5 min. The protein loading contents were 20 μg. Proteins were separated by SDS-PAGE (15% gel).

Immunogenic cell death (ICD) induced by Mon@MFLs in cancer cells

To explore the Mon@MFL-induced immunogenic cell death effect of cancer cells in vitro, MC38 cells were cultured to the appropriate density in 12-well plates, and then MC38 cells were incubated with Mon@MFLs (at the Mon concentration of 8 μM) for 4 h. During other time periods, the cells were cultured in blank medium. At the designated time point, we collected the cells for subsequent processing. The levels of ICD-related biomarkers were analyzed. The probe DCFH-DA (Beyotime) was used to detect the intracellular reactive oxygen species (ROS) and the released ATP level was detected by an ATP assay kit (Beyotime). In order to detect calreticulin (CRT) concentrations, MC38 cells were seeded in confocal microscopy dishes. CRT was labeled using primary rabbit anti-CRT antibody (Ab2907, Abcam) and then incubated with Alexa Fluor 647-conjugated goat anti-rabbit IgG antibody. In the final step, cells were stained with DAPI for CLSM observation.

In vitro cellular uptake behavior and lysosome escape capacity of FEVs

To investigate the cellular uptake behavior, FEVs were labeled with DiD, then they were co-incubated in confocal dishes with MC38 cells for 4 h (the protein concentration of FEVs: 50 μg mL⁻¹) and treated in the same manner as before. The cell membranes were labeled with DiO, and DAPI-labeled nuclei were observed using CLSM. LysoTracker Green was used to label lysosomes, and Hoechst 33342 staining was used to observe the colocalization conditions of lysosomes and FEVs.

In vitro immune cell activation capacity of FEVs

The pro-maturation capacity of FEVs was analyzed. Bone marrow-derived dendritic cells (BMDCs) were derived from mouse bone marrow and incubated with PBS, EVs and FEVs. The expression levels of CD80 and CD86 were measured by flow cytometry (FCM).⁹ The maturation of DCs was analyzed. BMDCs were derived from mouse bone marrow and incubated with PBS, EVs and FEVs (the protein concentration of EVs or FEVs: 50 μg mL⁻¹) for 24 h, while lipopolysaccharide (LPS) served as the positive control. The fluorescent probe-labeled antibodies such as anti-CD11c, anti-CD80 and anti-CD86 were used for staining. The fluorescence of each marker was measured by FCM, and the expression level of surface markers was analyzed to evaluate the maturation status of BMDCs.

In vivo anti-cancer effect of FEVs

To test the anti-tumor effect of FEVs, a peritoneal metastasis model was established by intraperitoneally injecting 5 × 10⁵ MC38 cells into the peritoneal cavity of female C57BL/6J mice.³⁴ The changes in body weight and abdominal bulge of the mice were monitored. The tumor mice were randomly divided into three groups and treated with PBS, EVs, and FEVs (dosage: 100 μg protein of EVs or FEVs per 20 g body weight of mouse) at day 3, 7, and 11 after implantation. The experiment was terminated and mice were sacrificed on day 20. The whole metastatic tumor and ascitic fluids were collected. The weight of metastatic nodules and the volume of ascites were measured to compare the therapeutic effects.

3. Results and discussion

Preparation and characterization of MFLs

Vitamin E-based quaternary ammonium cationic lipids (VECLs) were successfully synthesized as described in the previous method (Fig. S1, ESI†) and subsequently utilized to construct liposomes. Compared to the conventional endocytosis pathway, membrane fusion has several notable advantages. It directly releases the encapsulated drugs into the cytosol, bypassing the endosome pathway and reducing drug degradation in lysosomes. The preparation process is shown in Fig. 2A, using the thin-film hydration method. MFLs were prepared with the main component VECL, DPPC, and cholesterol in a molar ratio of 2 : 4 : 4. As displayed in Fig. 2B, liposomes without VECLs were prepared for comparison and referred to as non-fusogenic liposomes (NFLs). Next, we characterized the particle size and potential of MFLs, and the results are shown in Fig. 2C and D. The average diameter of MFLs was 153.9 ± 5.6 nm, which was smaller than that of the NFLs without cationic lipids (185.9 ± 6.7 nm). It may be due to the presence of cationic lipid VECLs enhancing the stability of the liposome structure. The two quaternary ammonium groups of VECLs endowed liposomes with strong positive electrical property. The presence of a positive charge on the surface is a key factor for membrane fusion liposomes. Under physiological conditions, the surface potential of MFLs was approximately +20 mV, which enabled the liposomes to interact with the negatively charged cell membrane and promote the membrane fusion process.

Fig. 2 Preparation and characterization of MFLs or NFLs. (A) Schematic diagram for the preparation method of MFLs or NFLs. (B) Table of the lipid composition of MFLs or NFLs. Basic characterization of MFLs or NFLs, including (C) particle size and (D) zeta potential. (n = 5); data were represented as mean ± SD. (E) The CLSM images (scale bar = 50 μm) and (F) colocalization analysis of DSPE-RB-labeled MFLs or NFLs (red) incubated with MC38 cells (DAPI, blue) for 4 h to verify the membrane fusion ability of MFLs.

To ascertain whether liposomes possessed the property of membrane fusion, we utilized the fluorescent lipid DSPE-RB to label NFLs or MFLs and subsequently co-incubated them with MC38 cells, monitoring the manner in which MFLs penetrated the cells using CLSM. As shown in Fig. 2E and Fig. S2 (ESI†), the cellular uptake efficiency of MFLs was higher than that of NFLs. MFLs were promptly internalized by the cells within 2 h and achieved saturation after 4 h. However, the uptake rate of NFLs was distinctly slow. The cells took up very little within 2 h and only showed a certain degree of improvement after 4 h. After the cellular uptake, the red signal of the NFLs had already entered the interior of cells and distributed in a spot-like manner around the nucleus, indicating that early endosomal phagocytosis had occurred and NFLs had entered the lysosomes. However, the red signal from MFLs was observed to appear very obvious around cells, showing evident colocalization with the cell membrane. This suggested that MFLs had higher internalization efficiency and excellent membrane fusion ability. Gray value analysis further proved the colocalization situation as shown in Fig. 2F. The Pearson correlation coefficient was used for comparing the co-localization between liposomes and nucleus. MFLs were located on the cell membrane, so the Pearson correlation coefficient was negative (−0.19), indicating no correlation with the nucleus. NFL were distributed in the cytoplasm, so the Pearson correlation coefficient was positive (0.31), indicating a correlation with the nucleus. These results proved that MFLs were an efficient delivery vehicle for cargo delivery into cells.

Mon@MFLs preserved the membrane fusion function

Monensin is a lipophilic antibiotic that can affect ion transport carriers on the cell membrane and mediates Na⁺/H⁺ exchange. It causes an increase in cellular Na⁺ and stimulates a continuous pump out of Na⁺/K⁺-ATPase. The increasing intracellular Na⁺ will activate the Na⁺/Ca²⁺ exchanger of cells, resulting in the increase of Ca²⁺ degree in the cytosol. The increase of Ca²⁺ concentration can bring about the enlargement of multivesicular bodies (MVBs). Therefore, Mon, as a potential drug, can be used to improve the production of EVs, providing a new strategy for research and application.

As mentioned above, MFLs possess the ability to fuse with membranes and exhibit rapid internalization effects. Using MFLs as the carrier to deliver Mon not only facilitates rapid cellular uptake of Mon but also ensures its anchoring on the cell membrane, allowing it to fully play an ionophore role. Similarly, we used the thin-film hydration technique to produce membrane fusion liposomes containing monensin (Mon@MFLs), as shown in Fig. 3A. Given the good lipophilicity of Mon, it was combined with lipid components and the Mon content was 1% of the total lipid mass. The characterization of the obtained Mon@MFLs showed that their particle size (163.1 ± 2.9 nm) was similar to that of MFLs (Fig. S3, ESI†).

Fig. 3 Preparation and characterization of Mon@MFLs or FEVs. (A) Schematic diagram for the preparation method of Mon@MFLs. (B) The CLSM images (scale bar = 50 μm) and (C) colocalization analysis of DSPE-RB-labeled Mon@MFLs (red) incubated with MC38 cells (DAPI, blue) at different time periods, 2 h or 4 h. (D) Schematic diagram of the preparation method of EVs or FEVs. Basic characterization of FEVs or EVs such as (E) particle size and (F) zeta potential (n = 5). (G) SDS-PAGE analysis of proteins of different vesicles. (H) Protein content of EVs obtained by different treatments (n = 3). Data are represented as mean ± SD, ***p < 0.001.

The cellular behavior of above liposomes was further investigated. After interacting with cells for 2 or 4 h, CLSM images revealed that the red signal of the labeled liposomes still appeared around the cells. This suggested that despite the presence of Mon, MFLs were capable of effectively fusing with the cell membrane and delivering its cargo intracellularly (Fig. 3B). The results indicated that the addition of Mon did not affect the membrane fusion ability of MFLs themselves. Fig. 3C shows the gray value analysis obtained based on CLSM images. The red signals were mainly distributed near the cell membrane and had little overlap with DAPI-labeled nuclear signals. This further confirmed the highly efficient membrane fusion ability and intracellular delivery effect of Mon@MFLs. Additionally, this finding implied that Mon can remain on the cell membrane, thereby enabling it to exert its intended effects.

The above results indicated that MFLs served as an effective delivery vehicle, maintaining the membrane fusion ability even in the presence of Mon, which supported its potential application in therapeutic strategies.

Mon@MFLs promoted the production of EVs

MC38 cells exhibit a limited capacity for vesicle production and are relatively insensitive to external stimuli. Upon treating MC38 cells with Mon@MFLs, a significant enhancement in vesicle production was observed, and these vesicles were termed fusion extracellular vesicles (FEVs). The procedure for preparation of EVs or FEVs is detailed in Fig. 3D. Initially, Mon@MFLs were co-cultured with MC38 cells for a duration of 4 h, following which FEVs were isolated using gradient centrifugation. We characterized FEVs by DLS and found that their particle size was slightly larger than that of blank EVs derived from cells without Mon@MFL treatment. This may be due to the incorporation of foreign lipid components into the cell membrane (Fig. 3E). Given that the cell membrane's surface inherently bears a negative charge, vesicles generated through conventional methods also exhibit a negative surface charge. The vesicles possessed the zeta potential of around −20 mV. The negatively charged EVs experienced partial neutralization upon the integration of membrane fusion liposomes, resulting in the zeta potential for FEVs of approximately −10 mV (Fig. 3F). The TEM images show that EVs and FEVs have similar morphological structures (Fig. S4, ESI†), both being hollow circular structures with the size of over two hundred nm. This also indicated that Mon@MFLs were successfully fused with the cell membrane. In addition, the protein expression pattern of the vesicles obtained by this method was basically consistent with that of EVs (Fig. 3G), indicating that Mon@MFLs did not cause obvious additional interference. In addition, the cohort comprised PBS, free Mon, MFLs and ultraviolet light (UV, irradiation for 30 min at the beginning). EVs derived from different treatments were collected for comparison. Upon completion of the 4-hour co-cultivation period, vesicles secreted by MC38 cells were collected to determine their protein content, thereby estimating the quantity of vesicles produced. As shown in Fig. 3H, compared with the control groups, the protein content of FEVs was significantly increased, indicating that Mon@MFLs successfully promoted cells to produce more extracellular vesicles. The results also showed that Mon and positively charged liposome MFLs could increase vesicle production to some extent when utilized individually, but the effects were not as effective as those of Mon@MFLs. Compared with the commonly used UV-induced EV production method,³⁵ the vesicles yielded from cells treated with Mon@MFLs increased by more than 3 times. These results proved that the delivery of Mon to the cell membrane by the fusion pathway was an effective method to increase the production of FEVs.

Mon@MFLs induced intracellular Ca²⁺ overload

Upon fusion with the cell membrane, Mon became anchored on the cell membrane from Mon@MFLs, functioning as an ion channel. This resulted in an intracellular Ca²⁺ overload, which in turn stimulated MVBs and facilitated the release of EVs (Fig. 4A). To verify this process, we incubated MC38 cells with Mon@MFLs for 4 h and measured the intracellular Ca²⁺ content at different time points (2 h or 24 h). The control group referred to MC38 cells treated with PBS instead of Mon@MFLs (Fig. 4B). We found that Mon@MFLs indeed promoted Ca²⁺ influx, with the highest intracellular Ca²⁺ content at 2 h, followed by a gradual decrease. However, MFLs showed the opposite trend, with intracellular Ca²⁺ content gradually increasing (Fig. 4C). MFLs could anchor Mon to the cell membrane by mediating membrane fusion, which allowed Mon to function as an ion channel, causing persistent cellular Ca²⁺ internal flow. Consequently, MFLs facilitated the maintenance of functionality of Mon at the cell membrane, which led to intracellular Ca²⁺ overload and elevated EVs’ production.

Fig. 4 Mon@MFLs promote ICD effects. (A) Schematic diagram of the process of ICD induced by Mon@MFLs. (B) Schematic diagram for detecting biomarkers related to ICD. (C) The intracellular Ca²⁺ content, (D) intracellular ROS content and (E) extracellular ATP content within cells after FEV treatment at different time points, 2 h or 24 h (n = 3). (F) The flow cytometry detection (FCM) of the percentage of CRT-positive cells at different time points (n = 3) and (G) CLSM images of MC38 cells incubated with CRT antibody (scale bar = 20 μm). Data are represented as mean ± SD. *p < 0.05 and ***p < 0.001.

Mon@MFLs induced immunogenic cell death (ICD) within cancer cells

Inducing ICD within cancer cells is a highly promising strategy for cancer immunotherapy.³⁶ By enhancing the immunogenicity of tumors, dendritic cells (DCs) are attracted to the tumor site and their function of antigen presentation could be enhanced, along with the ability to activate specific cytotoxic T lymphocytes (CTLs) to attack the tumor.³⁷ Under normal conditions, the concentration of Ca²⁺ in cells is strictly regulated, and the intracellular Ca²⁺ concentration is maintained at a physiological level, which is a balance achieved by Ca²⁺ channels, pumps, and other cellular organelles. The ion carrier action mediated by Mon caused an increase in intracellular Ca²⁺ level, and the Ca²⁺ concentration in the mitochondria also increased. High Ca²⁺ level can affect mitochondrial energy metabolism, leading to mitochondrial damage and a reduction in ATP synthesis efficiency. The intracellular level of reactive oxygen species (ROS) is also continuously elevated. Cells in the high oxidative stress state initiate a series of cascade biological processes. Besides the upregulation of the ROS level, a series of damage-associated molecular patterns (DAMPs), including adenosine triphosphate (ATP), high mobility group box 1 protein (HMGB1), and heat shock proteins, were released during this process. We conducted tests to assess alterations in indicators of ICD impact. As depicted in Fig. 4D, we observed the ROS levels in cells post-stimulation with Mon and noted a continuous elevation in ROS levels over time, indicating that cells remained in the state of oxidative stress. The ATP content at various time intervals was quantified using an ATP assay kit, with the outcomes depicted in Fig. 4E. Following treatment with Mon@MFLs, the ATP content in the extracellular environment exhibited an increase.

Calcium ion overload can also lead to severe endoplasmic reticulum stress and induce the endoplasmic reticulum-related molecular chaperone calreticulin (CRT) to flip out from the endoplasmic reticulum to the cell membrane surface of tumor cells. It can serve as an “eat me” signal to mark cells that are about to be phagocytosed, thereby attracting DCs to presenting antigens. The percentage of CRT-positive cells was detected by flow cytometry (FCM) at different time points. The percentage of CRT-positive cells approached 60% at 24 h, indicating that CRT has been sufficiently exposed to the cell membrane surface (Fig. 4F). The results of CLSM were consistent with results from flow cytometry, and compared with the control group, the red fluorescence signal of CRT increased with time (Fig. 4G).

The cellular uptake behavior of FEVs

Generally speaking, EVs are internalized by cells through the process of endocytosis. To elucidate the cellular uptake mechanism of FEVs, we used CLSM to observe the localization of FEVs within DC2.4 cells. FEVs were labeled with the fluorescent dye DiD, and the cell membrane was labeled with DiO to visualize them. As depicted in Fig. 5A, the CLSM images revealed that the red signal from the FEVs exhibited poor co-localization with the green signal from the cell membrane. The FEVs were mainly located around the nucleus, exhibiting that they were mainly distributed in the cytoplasm. This suggested that FEVs entered the cell primarily through endocytosis, crossing the cell membrane without lingering on its surface and subsequently internalizing into the cellular interior. The mechanism by which FEVs entered cells differed from that of membrane fusion at Mon@MFLs.

Fig. 5 Analysis of the cellular uptake mechanism of FEVs and the ability to promote DC maturation. (A) The CLSM images (scale bar = 10 μm) and (B) colocalization analysis of the cellular uptake of FEVs, in which FEVs were labeled with DiD, and the cell membrane was labeled with DiO. (C) The CLSM images (scale bar = 10 μm) and (D) colocalization analysis of FEVs and lysosomes, in which DiD was used to label FEVs, and Lyso-Tracker Green was used to label lysosomes. (E) Schematic diagram of extracting BMDCs and detecting DC maturation. Figure was drawn by Figdraw. (F) The flow cytometry scatter plots and (G) quantified analysis after different treatments, in which the cells were labeled with antibodies, including anti-CD80 and anti-CD86, to analyze the percentage of mature BMDCs (n = 3). Data are represented as mean ± SD. ***p < 0.001.

In order to more precisely define the localization and dispersion of FEVs inside cells, we utilized LysoTracker Green to label the lysosomes following the vesicles’ incubation with the cells, thereby enabling us to monitor the lysosomal escape. Lysosomes represent the initial significant barrier encountered by numerous drugs and nanocarriers upon their entry into cells. FEVs should be able to efficiently evade the lysosome, otherwise they risk degradation or failure to reach their intended target site, thereby impacting their biological activity and therapeutic efficacy. The CLSM images in Fig. 5C showed that most of the FEVs have dispersed throughout the entire cytoplasm of cells, and only a small amount of fluorescent signal overlapped with the lysosome labeling, indicating that FEVs had good lysosomal escape ability. Gray value analysis also showed signal overlap as shown in Fig. 5B and D. FEVs were able to quickly escape the lysosome upon entering cells, subsequently carrying out their intended biological function. Based on the observations, we proposed that the highly efficient lysosomal escape performance of FEVs may be due to the inclusion of the VECL in their structure or the action of Mon. These components may enhance the escape efficiency of FEVs by altering their physicochemical properties.

FEVs promoted DC maturation

Dendritic cells (DCs), as one of the important members of the immune system, are responsible for recognizing and capturing antigens and presenting them to T cells to initiate an adaptive immune response. In this part, we studied whether FEVs could promote the maturation of DCs and further understand their role in the immune response. This process is shown in Fig. 5E. We extracted BMDCs from mouse bone marrow and treated them with FEVs for 24 h. We subsequently employed flow cytometry (FCM) to detect the expression levels of CD80 and CD86 on the cells. These molecules are critical costimulatory factors for T cell activation and are typically upregulated during DC maturation. By the analysis of FCM on the BMDC suspension, we found that there were significant changes in the expression levels of DC maturation markers CD80 and CD86 in the FEV group compared to the blank EV group, as shown in Fig. 5F and Fig. S5 (ESI†).

Analyzing the FCM results indicated that the proportion of CD11c⁺CD80⁺CD86⁺ cells increased in the FEV group, suggesting that FEVs promoted the maturation of DCs. The maturation effect was higher than that of the blank vesicles (Fig. 5G). The ability of FEVs to promote DC maturation indicated that FEVs may have broad application prospects in the field of immunotherapy. These results may provide a strong scientific basis for the application of FEVs in anti-cancer therapy.

In vivo anti-cancer effects of FEVs

To evaluate the efficiency of FEVs as a therapeutic approach and their potential anti-cancer effects, the PCCC model based on MC38 cells was established to conduct in vivo experiments. First, MC38 cells were i.p. injected into mice to establish the ascite tumor model and the mice were treated on a predetermined schedule (Fig. 6A). On day 3, 7, and 11, FEVs or EVs were administered to the experimental group, while the control groups received PBS of the same volume. During the experiments, we continuously monitored tumor growth and collected tumor and ascite samples from all mice on day 20 for analysis. Tumor weight is commonly used as a measure of tumor size and growth, with a lower tumor weight indicating that a drug or therapeutic intervention is effectively inhibiting tumor growth. As shown in Fig. 6B and C, the average tumor weight of mice in the FEV group was significantly lower than that of the other two groups, indicating that FEVs effectively inhibited tumor growth. Next, we collected the ascites from mice and measured their volume (Fig. 6D). These results showed that the ascite volume in the FEV group was observably reduced, suggesting that tumor vesicle FEVs may have the ability to reduce ascite formation and accumulation. In summary, FEVs showed a good therapeutic effect in the mouse PCCC model which not only effectively inhibited tumor growth but also significantly alleviated ascites symptoms. These results provided strong experimental support for the application of FEVs in colorectal cancer treatment. However, not all the mice showed good anti-tumor efficacy. The reasons for the unsatisfactory anti-tumor efficacy might be attributed to the significant individual differences. The individualized immune background, including genetic background, tumor mutational burden, and baseline immune status (such as the degree of lymphocyte infiltration), significantly affects the efficacy of FEVs. Another possible reason is that the targeting ability and delivery efficiency of FEVs were insufficient. FEVs might be rapidly cleared by the mononuclear phagocytic system such as the liver and spleen. Therefore, the future optimization strategies and directions include: (1) engineering modification: by means of gene editing or chemical conjugation, target molecules can be modified on the surface of FEVs. (2) Combination therapy: FEVs can be combined with immune checkpoint inhibitors (such as anti-PD-1), immune stimulatory molecules (such as IL-12, interferon-gamma stimulator agonists), chemotherapy or radiotherapy to reverse the immunosuppressive microenvironment. (3) Personalized customization: FEVs can be customized based on tumor-specific antigens to enhance immunogenicity and specificity.

Fig. 6 In vivo anti-cancer effect of FEVs. (A) Schematic diagram of FEVs for the treatment of peritoneal carcinomatosis. (B) The tumor weight and (C) tumor images of the PCCC after different treatments (n = 6). (D) The ascite volume in PCCC mice after different treatments (n = 6). Data are represented as mean ± SD.

4. Conclusions

A novel preparation method for cell-derived functional EVs with high yield has been developed to facilitate cancer immunotherapy. First, cationic Mon@MFLs facilitated the delivery of Mon to the surface of the cell membrane. Then, Mon was embedded on the membrane lipid layers, initiating a cascade of biological processes within cells that could alter the function of cancer cells, including inducing ICD effects. Through the synergistic action of MFLs and Mon, the capacity of cancer cells to release EVs was improved. The released functional EVs possessed immunostimulatory properties and have been utilized in cancer immunotherapy, exhibiting favorable anti-cancer effects. The utilization of Mon@MFLs not only enhanced the therapeutic potential of FEVs but also delved into the intricate interplay between cells and nanomedicines.^38,39 The application of Mon@MFLs exhibits promising prospects in EV-based immunotherapy, thereby warranting further investigation into the feasibility across diverse cancer types to make a more profound impact on cancer treatment.^40–42 In the future, the efficacy of FEVs can be further optimized and enhanced through engineering modifications, combination therapy, and personalized customization strategies.

Author contributions

Qi Xie: equal contribution, writing – original draft, visualization, validation, methodology, investigation, data curation, conceptualization. Yuan Gao: equal contribution, writing – original draft, visualization, validation, methodology, investigation. Wei Chen: visualization, validation, methodology, data curation. Haoyu Zhang: validation, methodology. Chuansheng Fu: validation, methodology. Samira Batur: validation, methodology. Yixuan Zhou: validation, methodology. Yang Li: validation, methodology. Jiao Zhang: validation, methodology. Conglian Yang: validation. Li Kong: writing – review & editing, validation, project administration, funding acquisition, conceptualization. Zhiping Zhang: writing – review & editing, project administration, funding acquisition, conceptualization.

Data availability

All data generated or analyzed during this study are included in this article and its ESI.† The data used to support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (32271454, 82173760 and 82373816), the Wuhan Science and Technology Plan (2022023702025187), and the Natural Science Foundation of Hubei Province (2024AFB582). The authors are thankful for the animal care provided by the Laboratory Animal Center of Huazhong University of Science and Technology. We are also thankful for the technical support provided by Teacher Sisi Li of the Analytical and Testing Center of Huazhong University of Science and Technology.

References

1. E. I. Buzas, Nat. Rev. Immunol., 2022, 23, 236–250.
2. C. Marar, B. Starich and D. Wirtz, Nat. Immunol., 2021, 22, 560–570.
3. P. Fu, J. Zhang, H. Li, M. Mak, W. Xu and Z. Tao, Adv. Drug Delivery Rev., 2021, 179, 113910.
4. G. van Niel, G. D’Angelo and G. Raposo, Nat. Rev. Mol. Cell Biol., 2018, 19, 213–228.
5. C. A. Marquez, C.-I. Oh, G. Ahn, W.-R. Shin, Y.-H. Kim and J.-Y. Ahn, J. Nanobiotechnol., 2024, 22, 6.
6. E. I. Buzas, Nat. Cell Biol., 2022, 24, 1322–1325.
7. A. Yokoi and T. Ochiya, Semin. Cancer Biol., 2021, 74, 79–91.
8. S. Du, Y. Guan, A. Xie, Z. Yan, S. Gao, W. Li, L. Rao, X. Chen and T. Chen, J. Nanobiotechnol., 2023, 21, 231.
9. Y. Yu, Y. Tian, Y. Li, X. Qin, X. Li, Q. Hu, C. Fu, B. Niu, C. Yang, L. Kong and Z. Zhang, Nano Today, 2024, 55, 102173.
10. H. Mengyuan, L. Aixue, G. Yongwei, C. Qingqing, C. Huanhuan, L. Xiaoyan and L. Jiyong, J. Nanobiotechnol., 2024, 22, 604.
11. Q. Xie, S. Li, X. Feng, J. Shi, Y. Li, G. Yuan, C. Yang, Y. Shen, L. Kong and Z. Zhang, J. Nanobiotechnol., 2022, 20, 226.
12. M. Sousa de Almeida, E. Susnik, B. Drasler, P. Taladriz-Blanco, A. Petri-Fink and B. Rothen-Rutishauser, Chem. Soc. Rev., 2021, 50, 5397–5434.
13. K. Debnath, S. Pal and N. R. Jana, Acc. Chem. Res., 2021, 54, 2916–2927.
14. L. Shang, Q. Xie, C. Yang, L. Kong and Z. Zhang, ACS Appl. Mater. Interfaces, 2023, 15, 42378–42394.
15. J. Kim, O. A. Santos and J.-H. Park, J. Controlled Release, 2014, 191, 98–104.
16. A. Mukherjee, B. Bisht, S. Dutta and M. K. Paul, Acta Pharmacol. Sin., 2022, 43, 2759–2776.
17. F. Chen, M. Bian, M. Nahmou, D. Myung and J. L. Goldberg, RSC Adv., 2021, 11, 35796–35805.
18. B. Kim, S. Sun, J. A. Varner, S. B. Howell, E. Ruoslahti and M. J. Sailor, Adv. Mater., 2019, 31, 1902952.
19. C. Shi, Q. Zhang, Y. Yao, F. Zeng, C. Du, S. Nijiati, X. Wen, X. Zhang, H. Yang, H. Chen, Z. Guo, X. Zhang, J. Gao, W. Guo, X. Chen and Z. Zhou, Nat. Nanotechnol., 2022, 18, 86–97.
20. H. Kong, K. Yi, C. Zheng, Y.-H. Lao, H. Zhou, H. F. Chan, H. Wang, Y. Tao and M. Li, J. Mater. Chem. B, 2022, 10, 6841–6858.
21. X. Gao, E. Ren and G. Liu, Smart Mater. Med., 2022, 3, 254–256.
22. J. Lee, J. Kim, M. Jeong, H. Lee, U. Goh, H. Kim, B. Kim and J.-H. Park, Nano Lett., 2015, 15, 2938–2944.
23. J. Lee, H. Lee, U. Goh, J. Kim, M. Jeong, J. Lee and J.-H. Park, ACS Appl. Mater. Interfaces, 2016, 8, 6790–6795.
24. J. Lee, U. Goh, H.-J. Lee, J. Kim, M. Jeong and J.-H. Park, Mol. Pharmaceutics, 2016, 14, 423–430.
25. H. Kim, J. Lee, C. Oh and J.-H. Park, Nat. Commun., 2017, 8, 15880.
26. X. Ji, Y. Ma, W. Liu, L. Liu, H. Yang, J. Wu, X. Zong, J. Dai and W. Xue, ACS Nano, 2020, 14, 7462–7474.
27. P. Bao, Z.-T. Zheng, J.-J. Ye and X.-Z. Zhang, Nano Lett., 2022, 22, 2217–2227.
28. N. T. Ho, M. Siggel, K. V. Camacho, R. M. Bhaskara, J. M. Hicks, Y.-C. Yao, Y. Zhang, J. Köfinger, G. Hummer and A. Noy, Proc. Natl. Acad. Sci. U. S. A., 2021, 118, e2016974118.
29. Y.-C. Chen, Y.-T. Li, C.-L. Lee, Y.-T. Kuo, C.-L. Ho, W.-C. Lin, M.-C. Hsu, X. Long, J.-S. Chen, W.-P. Li, C.-H. Su, A. Okamoto and C.-S. Yeh, Nat. Nanotechnol., 2023, 18, 1492–1501.
30. F. Xu, H. Zhong, Y. Chang, D. Li, H. Jin, M. Zhang, H. Wang, C. Jiang, Y. Shen and Y. Huang, Biomaterials, 2018, 158, 56–73.
31. Y. He, K. Wang, Y. Lu, B. Sun, J. Sun and W. Liang, Nano Lett., 2022, 22, 1415–1424.
32. S. Bai, Y. Lan, S. Fu, H. Cheng, Z. Lu and G. Liu, Nano-Micro Lett., 2022, 14, 145.
33. A. E. Shin, F. G. Giancotti and A. K. Rustgi, Trends Pharmacol. Sci., 2023, 44, 222–236.
34. J. Zhang, C. Fu, Q. Luo, X. Qin, S. Batur, Q. Xie, L. Kong, C. Yang and Z. Zhang, J. Controlled Release, 2024, 373, 201–215.
35. S. Ruan, N. Erwin and M. He, J. Extracell. Vesicles, 2022, 11, e12194.
36. G. Kroemer, L. Galluzzi, O. Kepp and L. Zitvogel, Annu. Rev. Immunol., 2013, 31, 51–72.
37. L. Tu, C. Li, Q. Ding, A. Sharma, M. Li, J. Li, J. S. Kim and Y. Sun, J. Am. Chem. Soc., 2024, 146, 8991–9003.
38. M. Li, L. Lu, Q. Xiao, A. A. Maalim, B. Nie, Y. Liu, U. D. Kahlert, K. Shu, T. Lei and M. Zhu, Exploration, 2024, 20240027.
39. J. Jiang, X. Cui, Y. Huang, D. Yan, B. Wang, Z. Yang, M. Chen, J. Wang, Y. Zhang, G. Liu, C. Zhou, S. Cui, J. Ni, F. Yang and D. Cui, Nano Biomed. Eng., 2024, 16, 152–187.
40. S. E. Glass and R. J. Coffey, Gastroenterology, 2022, 163, 1188–1197.
41. S. Rahmati, A. Moeinafshar and N. Rezaei, J. Transl. Med., 2024, 22, 452.
42. L. Xiong, Y. Wei, Q. Jia, J. Chen, T. Chen, J. Yuan, C. Pi, H. Liu, J. Tang, S. Yin, Y. Zuo, X. Zhang, F. Liu, H. Yang and L. Zhao, J. Nanobiotechnol., 2023, 21, 143.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4tb02807d

‡ These authors contributed equally to this work.

---