Extracellular vesicles from TNF-α and TGF-β1-treated ADSCs promote tissue-engineered cartilage regeneration by modulating macrophages via the miR-378a-3p/SIRPα axis

Abstract

Background

Mesenchymal stem cells (MSCs) possess immunomodulatory properties and are concurrently influenced by the local inflammatory microenvironment. Extracellular vesicles (EVs) derived from MSCs (MSCs-EVs) replicate the therapeutic effects of their parent cells while avoiding the limitations of cell therapy. Elucidating the impact of diverse inflammatory factors on the composition and functionality of MSCs-EVs is crucial for their optimal use, though this remains incompletely understood. The aim of this study was to explore the immunomodulatory effects and underlying mechanisms of EVs derived from adipose-derived stem cells (ADSCs) pretreated with TNF-α and TGF-β1 (α-EVs and β-EVs) on macrophages and tissue-engineered cartilage regeneration.

Methods

Isolation and identification of EVs from ADSCs cultured in 3D spheres. The effects on chondrocyte and macrophage proliferation, migration, and polarization were evaluated in vitro. In vivo, chondrocytes-laden porous GelMA hydrogel with EVs were implanted into mice to assess chondrocyte regeneration and macrophage infiltration. Small RNA sequencing revealed distinct EVs-miRNA expression profiles, and the target validation confirmed the molecular mechanism.

Results

In vitro, both α-EVs and β-EVs demonstrated the capacity to modulate macrophage phenotypes. α-EVs more effectively reduced M1 macrophage markers and enhanced M2 polarization. Besides, β-EVs exhibited a stronger inhibitory effect on macrophage proliferation and migration, while also promoting chondrocyte proliferation and extracellular matrix (ECM) formation. In vivo, β-EVs significantly improved ECM deposition and chondrocyte maintenance, while both EVs groups reduced M1 infiltration and increased M2 presence. Small RNA sequencing identified miR-378a-3p upregulation in both α-EVs and β-EVs, targeting Signal-regulatory protein alpha (SIRPα) to modulate the immune status of macrophages.

Conclusions

Both TNF-α and TGF-β1 enhanced the immunomodulatory effects of EVs, with TGF-β1 showing a stronger capacity to promote chondrocyte proliferation and ECM synthesis. The miR-378a-3p/SIRPα axis was identified as a key mechanism underlying the protective effects of both α-EVs and β-EVs. This study provides valuable insights into optimizing EVs-based regenerative strategies to regulate the local inflammatory microenvironment and promote the regeneration of engineered tissues.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-026-04947-0.

Background

Biomaterials and tissue engineering provide promising strategies for tissue repair and regeneration [1]. However, one major challenge impeding the widespread clinical application is the foreign body reaction (FBR), an immune response marked by inflammatory and fibrotic processes that can compromise the efficacy of implanted materials [2]. Macrophages play a pivotal role as the primary effector cells in the FBR. They can migrate to the implantation site and fuse to form foreign body giant cells (FBGCs) for phagocytosis and transform between pro-inflammatory and pro-regenerative phenotypes depending on the local microenvironments [3]. Thus, modulating macrophage activity offers a promising approach to mitigate FBR and foster tissue regeneration.

Mesenchymal stem cells (MSCs) have immunomodulatory effects, which enable them to inhibit inflammation and promote tissue repair [4, 5]. Adipose-derived stem cells (ADSCs), as a kind of MSCs, have garnered significant attention due to their abundant availability and ease of isolation [6]. Extensive research has demonstrated that ADSCs can modulate macrophage activity and enhance tissue regeneration in the inflammatory microenvironment [5, 7]. Our previous study confirmed that ADSCs facilitate cartilage regeneration by modulating the inflammatory response at the implantation site and promoting ECM synthesis in cartilage [5].

Extracellular vesicles (EVs) are key bioactive substances secreted by MSCs [8, 9]. These lipid bilayer vesicles, which are unable to replicate autonomously, carry a variety of bioactive molecules, including proteins, miRNAs, and other molecular signals [10, 11]. Compared to their parental cells, MSCs-EVs replicate the therapeutic benefits while avoiding potential drawbacks such as tumorigenicity, immunogenicity, and ethical concerns. The composition of EVs is closely linked to the physiological state of the secreting cells [12, 13]. Research has shown that EVs derived from inflammation-stimulated MSCs (IAE) are more effective in modulating inflammation than those from normal MSCs. Liu et al. demonstrated that IAE overexpressed miR-27b-3p and regulated macrophage activity by targeting macrophage colony-stimulating factor-1 (CSF1), significantly promoting the differentiation of M2 macrophage. An in vivo rabbit model of temporomandibular joint (TMJ) condylar osteochondral defects revealed that IAE had the most pronounced therapeutic effect in promoting TMJ regeneration [14].

FBR consists two typical phases. In the early phase, elevated levels of pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6), trigger an acute inflammatory response. In the later phase, the increased expression of pro-regenerative growth factors, such as transforming growth factor-β1 (TGF-β1) and platelet-derived growth factor (PDGF), facilitates ECM assembly and remodeling. In this study, we utilized the pro-inflammatory cytokine TNF-α and the pro-regenerative cytokine TGF-β1 to stimulate ADSCs separately. We then compared the similarities and differences in the acquisition of EVs and their respective effects under these conditions., and further explored the underlying mechanisms. This study aims to provide an experimental foundation for the rational and effective use of EVs in mitigating FBR and promoting tissue regeneration.

Methods

Cell isolation and culture

All human samples used in this study were collected and handled in strict accordance with the ethical guidelines and principles approved by the Ethics Committee of the Plastic Surgery Hospital and Institute, Chinese Academy of Medical Sciences, and Peking Union Medical College (Approval No.: 2023(200); Date: November 1, 2023). Written informed consent was obtained from all participants prior to their inclusion in the study.

ADSCs were isolated from adipose tissues samples obtained from four patients with microtia (age range: 5–7 years; male: n = 3, female: n = 1) undergoing second-stage ear reconstruction surgeries, as previously described [15]. The isolation and culture of ADSCs were performed in mesenchymal stem cell (MSC) medium (Sciencell, USA), supplemented with 10% fetal bovine serum (FBS, Sciencell, USA), 1% MSC growth supplement, and 1% penicillin-streptomycin (PS, Sciencell, USA), under standard culture conditions of 37 °C, 95% humidity, and 5% CO2. Phenotypic characterization of ADSCs at passage 4 was conducted by flow cytometry using a Human MSC Analysis kit (BD Biosciences, USA). Primary chondrocytes were isolated from the above four patients with microtia. In accordance with established protocols, the auricular cartilage was aseptically fragmented into 1mm³ pieces and subjected to 30 min of digestion with 0.25% trypsin (Gibco, UK), followed by 10 h of digestion with 0.2% collagenase type Ⅱ (Sigma-Aldrich, USA) under gentle agitation at 37 °C [16]. The mouse monocytic cell line, RAW264.7, was obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Both microtia-derived chondrocytes and RAW264.7 monocytes were cultured and expanded in high-glucose Dulbecco’s Modified Eagle’s Medium (H-DMEM, Gibco, Grand Island, NY), supplemented with 10% FBS and 1% PS. Passage 4 ADSCs and passage 2 chondrocytes were used in subsequent experiments.

Production, isolation and purification of EVs

The EVs collection medium consisted of low-glucose Dulbecco’s Modified Eagle’s Medium (L-DMEM, Gibco, Grand Island, NY), supplemented with 5% EVs-depleted FBS (prepared by ultracentrifugation) and 1% PS. Passage 4 ADSCs were seeded at a density of 1*10⁴ cells per well in ultra-low adhesion SpheroX™ 96Ukit (EFL-SP101, Engineering for Life). On the second day, when the cell spheroids had formed, the cells were washed twice with phosphate-buffered saline (PBS, Solarbio, China). The spheroids were then exposed to media containing 100ng/ml TNF-α (HY-P7058, MedChemExpress), 10ng/ml TGF-β1 (HY-P7118, MedChemExpress), or cytokine-free EVs collection medium, respectively. The supernatant was harvested every 48 h over a 6-day period, with fresh cytokine-containing medium replenished at each medium change, rather than a single pretreatment. All collected supernatants were temporarily stored at − 80 °C prior to EVs isolation. To isolate the EVs, the supernatants were processed according to a previously described protocol [17]. Briefly, the supernatant was first centrifuged at 300 x g for 10 min to remove cells, 2000 x g for 10 min to eliminate dead cells, 10 000 x g for 30 min to clear cell debris, and 100 000 x g for 70 min to collect EVs and remove contaminating protein. The resulting pellet was washed in PBS at 100 000 x g for 70 min to concentrate the EVs and then resuspended in PBS. All centrifugation steps were performed at 4°C. Concentration measurement was performed using a BCA Protein Assay Kit (Beyotime, China). Freshly purified EVs were used for subsequent experiments.

Identification of EVs: morphology, particle size and protein markers

In accordance with the guidelines outlined in the Minimal Information for Studies of Extracellular Vesicles (MISEV 2023) [10], morphological characterization of EVs was performed using a JEM-1200EX Transmission Electron Microscope (TEM) (JEOL, Tokyo, Japan). The particle size distribution of EVs was determined by Nanoparticle Tracking Analysis (NTA) using the Zetaview PMX 110 (Particle Metrix, Meerbusch, Germany). EVs-related protein markers, including CD9 (ab263019, Abcam), HSP70 (ab181606, Abcam), TSG101 (ab125011, Abcam) and calnexin (ab133615, Abcam), were assessed by Western blotting.

Preparation of GelMA hydrogels with porous structures

For cartilage tissue engineering, Gelatin Methacrylate (GelMA) hydrogel (SunP Biotech, China) was used. GelMA was completely dissolved in culture medium to achieve a final concentration of 10%. Lithium phenyl-2,4,6-trimethyl benzoyl phosphinate (LAP) (SunP Biotech, China), a photoinitiator, was fully dissolved in the solution to a final concentration of 0.25%. Poly(ethylene oxide) (PEO) powder with an average molecular weight of 300,000, serving as a pore-forming agent, was added to the mixture to a final concentration of 1%. The porous GelMA hydrogel was fully dissolved in the medium under pasteurization and was subsequently preserved in an incubator shielded from light. Photocrosslinking of the hydrogel formulation was induced by exposure to blue light (wavelength: 405 nm; light source: LED (Uvata Precision Optoelectronics Co., Ltd.); intensity: 20 mW/cm ²; distance: 10 cm; exposure time: 60 s) [1].

Study design of EVs treatment

The following in vitro and in vivo experiments were conducted with four distinct experimental groups: EVs derived from TNF-α-treated ADSCs were designated as the α-EVs group, EVs from TGF-β1-treated ADSCs were labeled as the β-EVs group, and EVs from untreated, normal ADSCs were referred to as the n-EVs group. An equal volume of PBS was used as the control group. In the in vitro experiments, each group was exposed to EVs for 48 h, with media replacement every 48 h to maintain the treatment conditions. Notably, a concentration of 50ug/ml EVs was used for the in vitro experiments, while 0.5ug/ul EVs were administered in the in vivo experiments.

Cellular uptake of EVs

The PKH-26 kit (Sigma, USA, PKH26GL-1KT) was used to label EVs according to the manufacturer`s instructions. Briefly, EVs were diluted with Diluent C, and a 2× PKH-26 solution was prepared in Diluent C. The EVs were then rapidly mixed with the PKH-26 solution by pipetting. The EVs/dye suspension was incubated for 5 min with periodic mixing. The staining was terminated by adding an equal volume of EVs-free FBS, followed by for a 1 min incubation to bind any excess dye. The mixture was subsequently centrifuged at 100 000 x g for 70 min at 4 °C, and the supernatant was carefully removed. The EVs were resuspended in PBS. For labeling of RAW264.7 monocytes and chondrocytes, the cytoplasm was stained with carboxyfluorescein diacetate succinimidyl ester (CFDA SE) (Beyotime, China, C1031) and the nucleus with Hoechst 33,342 (Solarbio, China, C0031), following the manufacturer`s protocol. The labelled EVs were added to the confocal dish and incubated for 24 h. Confocal microscopy was employed to observe the cellular uptake of the labeled EVs.

Cell proliferation assay

Cell proliferation of RAW264.7 monocytes and chondrocytes was assessed using the Cell Counting Kit-8 (CCK-8, Dojindo, Japan) according to the manufacturer`s instructions. Briefly, 2000 cells were seeded per well in 96-well culture plates. Following incubation, 10 ul CCK8 reagent was added to each well and co-incubated with the cells for 2 h. The optical density (OD) at 450 nm on days 0, 1, 3 and 5 using a multifunctional microplate reader to evaluate cell proliferation.

Cell migration assay

For the cell scratch wound assay, 2 × 10⁵ cells per well were seeded into 12-well plates and cultured to confluence. A scratch was made in the monolayer using a pipette tip, followed by washing with serum-free medium to remove detached cells. The cells were then cultured in H-DMEM medium, either supplemented with or without EVs. RAW264.7 monocytes were photographed at 0 h, 12 h, 24 h and 48 h post-scratching. For chondrocytes, images were captured at 0 h, 12 h, 24 h and 36 h post-scratching. The migration area was quantified using the following formula: migration area (%) = (M0-Mn)/M0 × 100, where M0 represents the initial scratch area and Mn represents the remaining scratch area at the measured time point.

Quantitative real-time polymerase chain reaction (RT-qPCR)

Total RNA was extracted from RAW264.7 monocytes and chondrocytes using TRIzol™ Reagent (Invitrogen, USA). Reverse transcription of RNA into cDNA was performed using the PrimeScript™ RT Reagent Kit (Takala, RR0367A). RT-qPCR was carried out in 20ul reaction volumes with SYBR Green I Master (Roche), following the manufacture`s protocol. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal reference for mRNA expression. The primer sequences used in this study were designed based on NCBI and were provided in Additional file 1. Relative mRNA expression levels were calculated using the 2^−Δ(ΔCT) method.

Western blot analysis

Western blotting was performed according to standardized protocols. Briefly, proteins were denatured and separated on 4–12% polyacrylamide gels, followed by electroblotting onto PVDF membranes. The membranes were incubated with primary antibodies, followed by incubation with a horseradish peroxidase (HRP)-conjugated secondary antibody. Protein bands were visualized by incubating the membranes with SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific). The details of the primary and secondary antibodies used in this study were provided in Additional file 2.

Cell immunofluorescence

To assess the effects of different EVs on macrophage polarization, RAW264.7 monocytes were seeded in confocal dishes at a density of 1 × 10⁵ cells per dish. The cells were treated with or without EVs for 48 h. To detect macrophage polarization markers, cells were fixed with 4% paraformaldehyde (PFA, Solarbio, China) for 15 min, permeabilized with 0.2% TritonX-100 (IT9100, Solarbio, China) for 10 min, and blocked with 5% BSA (SW3015, Solarbio, China) for 30 min at room temperature. The cells were then incubated overnight at 4 °C with a mouse anti-CD80 monoclonal primary antibody (66406-1-Ig, 1:200, Proteintech, USA) and a rabbit anti-CD206 polyclonal primary antibody (ab64693, 1:200, Abcam, Cambridge, UK). Following primary antibody incubation, the cells incubated with goat polyclonal secondary antibodies against mouse IgG-H&L (Alexa Fluor^® 488) (ab150113, 1:500, Abcam, Cambridge, UK) and rabbit IgG-H&L (Alexa Fluor ^® 594) (ab150080, 1:500, Abcam, Cambridge, UK) for 1 h at room temperature. The nuclei were counterstained with DAPI (C1005, Beyotime, China). Images were captured using a fluorescence microscope (Leica DM IRB).

Luminex liquid suspension chip detection

Luminex liquid suspension chip detection was performed by Wayen Biotechnologies (Shanghai, China) using the Bio-Plex Pro Mouse Chemokine Panel 31-plex kit, following the manufacturer’s instructions. Briefly, conditional media from macrophages treated with n-EVs, α-EVs, β-EVs or an equivalent volume of PBS were incubated in a microbead-embedded 96-well plate for 30 min. The samples were then incubated with the detection antibody for an additional 30 min. Subsequently, streptavidin-PE was added to each well for 10 min, and the results were read using the Bio-Plex MAGPIX System (Bio-Rad).

In vivo culture of tissue-engineered cartilage

The animal experiments conducted in this study were approved by the Biomedical Ethics Committee of Animal Care and Experiment at the Plastic Surgery Hospital and Institute of the Chinese Academy of Medical Sciences & Peking Union Medical College (Approval No.: 2023animal (99); Date: November 1, 2023). The work has been reported in line with the ARRIVE guidelines 2.0. All procedures adhered to the Guidelines for the Care and Use of Laboratory Animals. A total of 20 female C57BL/6 mice (weighing 18–22 g, aged 6–8 weeks) were sourced from Beijing Wei Tong Li Hua Laboratory Animal Technology Co., Ltd. The mice were housed under a 12-hour light/dark cycle at 22 ± 2 °C and 50 ± 10% humidity, with free access to food and water. All animals underwent a 7-day acclimatization period before surgery. The mice were randomly assigned to four experimental groups: (1) n-EVs, (2) α-EVs, (3) β-EVs, and (4) PBS blank control. Briefly, 1 × 10⁶ chondrocytes at passage 2 were harvested and resuspended in 100 µl of 10% GelMA hydrogel. Next, 50 µg of freshly purified EVs or an equivalent volume of PBS were mixed with the hydrogel. The resulting mixture was then placed into a pre-designed cylindrical mold and photocrosslinked using a 405 nm blue laser for 60 s to form the photocurable composite hydrogel. For the implantation procedure, C57BL/6 mice were anaesthetized and disinfected. A 1.0 cm median longitudinal incision was made, and subcutaneous pockets (1 × 1.5 cm) were created on the back of the mice. The cell-laden GelMA hydrogel was subcutaneously implanted into the pockets, with each mouse receiving four implants. The incisions were then closed with sutures. The subcutaneous implant model was specifically selected to simulate auricular cartilage transplantation in a non-load-bearing subcutaneous environment. Animals were housed individually and allowed unrestricted movement post-surgery to prevent interference with the surgical site and to allow for close postoperative monitoring. During this time, environmental enrichment including nesting material and a hideout was provided. After the wound had fully healed, mice were allowed to interact with each other. The C57BL/6 mice were euthanized via CO₂ inhalation, and samples were harvested for analysis after 2 weeks.

Histological and immunohistochemical analysis

The collected samples were fixed in 4% PFA for 24 h, embedded in paraffin, and section into serial slices with a thickness of 5µm. The sections were stained with hematoxylin and eosin (HE) to examine the histological structure, and with Toluidine blue and Alcian blue to assess collagen matrix deposition in the regenerated tissue. Immunohistochemical staining for CD80 and CD206 was performed to evaluate macrophage infiltration in the capsule tissue surrounding the tissue-engineered cartilage implants in C57BL/6 mice. Briefly, after deparaffin and rehydration, the paraffin slides were treated with complex digestive fluid (AR0022, Boster, China) at 37°C for 20 min to unmask the antigens. Endogenous peroxidase activity was blocked by incubating with 3% H₂O₂ for 10 min, followed by incubation with 5% BSA (SV0004, Boster, China) at room temperature for 30 min. The slides were then incubated overnight at 4°C with the primary antibodies: anti-CD80 (66406-1-Ig, 1:200, Proteintech, USA) and anti-CD206 (ab64693, 1:200, Abcam, Cambridge, UK). The secondary antibodies were applied for 30 min at 37°C, followed by incubation with 3,3’-diaminobenzidine tetrahydrochloride (DAB, Dako) for color development. The nuclei were counterstained with hematoxylin.

Small RNA sequencing and analysis

Total RNA was extracted using mirVana™ miRNA Isolation Kit(Cat #AM1561, Thermo Fisher Scientific, Austin TX, USA)according to the manufacturer`s instructions. RNA integrity was assessed by measuring the RNA Integrity Number (RIN) using an Agilent 4200 TapeStation (Agilent Technologies, Santa Clara, CA, USA). High-qualified total RNA was further purified using the RNAClean XP Kit (Cat #A63987, Beckman Coulter, Inc., Kraemer Boulevard Brea, CA, USA) and the RNase-Free DNase Set (Cat #79254, QIAGEN, GmBH, Germany). After RNA quantification and qualification, 3 µg of total RNA per sample was used for small RNA library preparation. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed on significantly differentially expressed genes [18]. The detailed small RNA sequencing data for this study were provided in Supplementary File.

Luciferase reporter assay and transfection

Transfection was performed using the lipofectamine 3000 reagent (Invitrogen, USA, L3000150) according to the manufacturer’s instructions. The luciferase report vectors, including pmir-report-sirpα wt and pmir-report-sirpα mut, were obtained from SyngenTech (Beijing, China). Briefly, the luciferase report vectors were transfected into HEK293T cells in a presence of miR-378a-3p mimics or a negative control (nc). After 24 h of transfection, the cells were cleaved using passive lysis buffer. Firefly and Renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega, USA).

Statistical analysis

Quantitative data are presented as the mean ± standard deviation (SD) from at least three independent experiments. The two-tailed Student’s t-test was used for comparison of two groups. One-way or two-way analysis of variance followed by Tukey’s post hoc test was carried out for multiple comparisons. Statistical analysis was carried out using GraphPad Prism version 9.5 (GraphPad, San Diego, CA, USA) statistical software. Differences were considered statistically significant when P < 0.05.

Results

2D/3D culture and identification of ADSCs

Passage 4 ADSCs exhibited a characteristic spindle-like morphology (Fig. 1A). Flow cytometry analysis confirmed that ADSCs were strongly positive for MSCs surface markers, including CD73, CD90 and CD105, while being negative for CD34, CD45, CD11b, CD19 and HLADR, consistent with the established criteria for MSCs identification (Fig. 1B) [19]. A live-dead staining assay of ADSCs spheroids demonstrated sustained cell viability on D0, D2, D4 and D6, indicating robust growth within the spheroids (Fig. 1C).

Fig. 1.

Identification of ADSCs and characterization of ADSCs spheres. A Representative image showing the spindle-like morphology of ADSCs cultured in 2D dish. Scale bar: 100 μm. B Flow cytometry analysis of characteristic surface markers of ADSCs revealed that the cells were positive for CD73, CD90, and CD105, and negative for CD34, CD45, CD11b, CD19, and HLA-DR. Data were shown for unstained control cells (orange) and ADSCs (blue). C Representative images showing live and dead ADSCs spheres on D0, D2, D4, D6. Green: live ADSCs. Red: dead ADSCs

Identification of EVs from ADSCs spheres with different stimuli

EVs were isolated from ADSCs treated with TNF-α (α-EVs), TGF-β1 (β-EVs), or without any cytokine treatment (n-EVs). TEM revealed that EVs in all groups displayed the typical bilayer membrane structure (Fig. 2A). NTA showed peak diameters of115.2 nm, 108.8 nm, and 122.6 nm, and median diameters of 111.1 nm, 110.6 nm, and 122.1 nm for n-EVs, α-EVs, and β-EVs, respectively (Fig. 2B). Notably, the median diameter of β-EVs was significantly larger than that of n-EVs and α-EVs (P < 0.05) (Fig. 2C). The peak and median diameters of all EVs groups fell within the established size range for EVs identification. Western blotting confirmed the presence of established EVs markers (CD9, HSP70, and TSG101) in n-EVs, α-EVs, and β-EVs, while the negative marker calnexin was absent in all groups (Fig. 2D). Taken together, these findings verified the successful isolation of EVs from ADSCs spheroids and demonstrated that the addition of inflammatory cytokines did not no significantly alter the morphology or surface marker expression of the EVs.

Fig. 2.

Identification of EVs isolated from ADSCs spheres under different stimuli (n-EVs, α-EVs and β-EVs). A Morphological characterization via transmission electron microscopy. Scale bar: 100 nm. B Particle size distribution determined by nanoparticle tracking analysis, and C the median sizes of three groups of EVs. D Western blot analysis of protein markers of EVs and their corresponding parental ADSCs spheres. Full-length blots were presented in Supplementary Figure. Data were presented as the mean ± SD from three replicates. ∗p < 0.05, ∗∗p < 0.01, ns: no significance

EVs uptake by RAW264.7 monocytes

To evaluate whether EVs interact directly with RAW264.7 monocytes, PKH26-labelled EVs were incubated with the cells and observed via confocal microscopy over a 24 h period. The labelled EVs were predominantly localized within the cytoplasm, suggesting their internalization and potential for direct communication with RAW264.7 monocytes (Fig. 3A).

Fig. 3.

In vitro cellular uptake of EVs and effects of EVs under different stimuli on macrophage proliferation and migration. A Confocal microscopy images showing the uptake of PKH-26-labeled (red) EVs by RAW264.7 cells (cytoplasm, green; nucleus, blue) after 24 h. Scale bar: 25 μm. B Cell proliferation of RAW264.7 cells assessed by the Cell Counting Kit-8 at specific time points after exposure to PBS, n-EVs, α-EVs, and β-EVs. C Representative images of RAW264.7 cell migration following exposure to different EVs. Scale bar: 100 μm, and D quantification of the migrated cell area using ImageJ software. Data were compared with the n-EVs group and presented as the mean ± SD from three replicates. ∗∗p < 0.01, ∗∗∗p < 0.001, ns: no significance

Effects of EVs on the proliferation and migration of RAW264.7 monocytes

The in vitro effects of n-EVs, α-EVs and β-EVs on the proliferation and migration of RAW264.7 monocytes were examined. Cell proliferation assays revealed that, beginning on D3, treatment with β-EVs significantly inhibited proliferation compared to n-EVs (P < 0.001) (Fig. 3B). Migration assays demonstrated that β-EVs markedly suppressed RAW264.7 monocyte migration at 24 h and 48 h (P < 0.01). Additionally, α-EVs exhibited significant inhibitory effects on migration after 48 h of treatment (P < 0.05) compared to the n-EVs group (Fig. 3C and D). Overall, both α-EVs and β-EVs were more effective than n-EVs in inhibiting RAW264.7 monocyte proliferation and migration, with β-EVs exerting the strongest inhibitory effects.

Effects of EVs on the polarization of RAW264.7 monocytes

The impact of EVs treatment on the polarization of RAW264.7 monocytes were assessed by quantifying the expression of polarization-associated macrophage genes using RT-qPCR (Fig. 4A). The results demonstrated that the expression of M2 macrophage marker genes, including found in inflammatory zone 1 (FIZZ1), chitinase-3-like 3 (Ym1), Arginase 1 (Arg1), and interleukin-10 (IL10), was significantly upregulated following 48 h of treatment with α-EVs compared to the control group (P < 0.05). Conversely, α-EVs significantly suppressed the expression of M1 macrophage marker genes, such as Cluster of Differentiation 80 (CD80), inducible nitric oxide synthase (iNOS), IL6 and interleukin-1β (IL1β) (P < 0.05). In the β-EVs group, only iNOS and IL1β exhibited a significant reduction in expression after 48 h of treatment compared to the control (P < 0.05). These findings suggest that both α-EVs and β-EVs can effectively reduce the expression of M1 macrophage marker genes, with α-EVs additionally increasing the expression of M2 macrophage marker genes.

Fig. 4.

In vitro functional studies showing distinct effects of EVs on macrophage proliferation and polarization. A qRT-PCR analysis of gene expression at 48 h post-treatment, assessing markers for M1 macrophage phenotype (CD80, iNOS, IL6, IL1β), M2 macrophage phenotype (FIZZ1, Ym1, Arg-1 and IL-10), and B proliferation marker (PCNA). C Western blot analysis of the protein expression levels of iNOS, COX-2, CD80, CD163, CD206, and Arg-1, and D quantification of relative protein expression using ImageJ software. Data was presented as the mean ± SD from three replicates. Full-length blots were presented in Supplementary Figure. E Immunofluorescence staining showing M1 macrophage marker (CD80, green), M2 macrophage marker (CD206, red), and nuclei (DAPI, blue) after 48 h of exposure to PBS, n-EVs, α-EVs or β-EVs. Scale bar = 100 μm. Compare data from the four groups in pairs and presented as the mean ± SD from three replicates. Only the parts with statistical differences are expressed. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001

Further analysis of proliferation-associated gene expression revealed that all EVs-treated groups significantly downregulated the mRNA levels of proliferating cell nuclear antigen (PCNA) relative to the control group (P < 0.01) (Fig. 4B). Notably, β-EVs exhibited the strongest inhibitory effect, consistent with the findings of the CCK-8 assay.

Western blotting analysis (Fig. 4C) revealed protein bands for M1 phenotype markers (iNOS, Cyclooxygenase-2 (COX-2), CD80) and M2 phenotype markers (CD163, CD206, Arg-1) along with the internal reference protein GAPDH across the four groups. Quantitative analysis of protein expression (Fig. 4D) showed that α-EVs and β-EVs significantly reduced the levels of M1 phenotype-associated proteins, including iNOS, COX-2, and CD80, compared to the control group (P < 0.01). For M2 phenotype-associated proteins, α-EVs significantly increased the expression of CD163 and CD206 (P < 0.01), while β-EVs significantly elevated CD206 levels compared to the control group (P < 0.01).

We then performed cellular immunofluorescence staining. The results indicated a significant decrease in the number of CD80⁺M1 macrophages and a marked increase in CD206⁺M2 macrophages in the α-EVs and β-EVs groups compared to the control and n-EVs groups. This effect was particularly pronounced in the α-EVs group (Fig. 4E).

Effects of EVs on cytokine secretion from RAW264.7 monocytes

To further investigate the impact on inflammatory cytokine secretion, we performed a Luminex liquid suspension chip assay on the culture supernatants of RAW264.7 monocytes (Fig. 5A and B). The heatmap revealed distinct cytokine profiles in EVs-treated macrophages compared to the control group (Fig. 5A). The expression levels of M1 macrophage-related markers, such as C-X-C motif chemokine ligand 16 (CXCL16) and C-C motif chemokine ligand 5 (CCL5), were significantly lower in the EVs-treated groups than in the control group (P < 0.05). In addition, the expression of granulocyte macrophage colony stimulating factor (GMCSF) was significantly reduced in the β-EVs group compared with the control group (P < 0.05). This suggested that β-EVs could effectively inhibit the proliferation, activation and polarization of macrophages via GMCSF, thereby inhibiting local immune reactions [20]. Conversely, the expression of M2 macrophage-related markers, such as C-C motif chemokine ligand 2 (CCL2) and IL10, was significantly elevated in the β-EVs group compared to the control group (P < 0.05) (Fig. 5B).

Fig. 5.

Effects of EVs on cytokine secretion from macrophages. A Cytokine secretion profile heatmap of macrophages treated with PBS, n-EVs, α-EVs or β-EVs for 48 h, analyzed using Luminex liquid suspension chip, and B violin plots illustrating significant changes in M1-related cytokines (GM-CSF, CXCL16 and CCL5) and M2-related cytokines (CCL2 and IL-10) among the four groups of macrophages. Data were compared with the control group and presented as the mean ± SD from three replicates. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001

Uptake of EVs by chondrocytes and the effects on chondrocyte proliferation, migration, and ECM synthesis

To evaluate the effects of EVs on chondrocytes, cells were incubated with PKH26-labelled EVs for 24 h and analyzed using confocal microscopy. The labeled EVs were predominantly localized in the cytoplasm, indicating the direct interaction and internalization of EVs by chondrocytes (Fig. 6A).

Fig. 6.

In vitro cellular uptake of EVs and effects of EVs under different stimuli on the phenotype of microtia chondrocytes. A confocal microscopy images showing the uptake of PKH-26-labeled (red) EVs by microtia chondrocytes (cytoplasm, green; nucleus, blue) at 24 h. Scale bar: 25 μm. B Proliferation of microtia chondrocytes evaluated using the Cell Counting Kit-8 at specific time points after exposure to PBS, n-EVs, α-EVs, or β-EVs. C Representative images of chondrocyte migration analysis upon exposure to different EVs, Scale bar: 200 μm, and D quantification of the migrated cell area using ImageJ software. E qRT-PCR analysis of gene expression at 48 h post-treatment, including markers associated with proliferation (PCNA) and ear cartilage matrix proteins synthesis (COL2A1, SOX9). Data were compared with the control and the n-EVs group, and presented as the mean ± SD from three replicates. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001

Chondrocyte proliferation was measured on D0, D1, D3, and D5 (Fig. 6B). The results revealed that none of the EVs treatments negatively affected chondrocyte survival, while β-EVs significantly enhanced chondrocyte proliferation compared to the control and n-EVs groups. The migratory capacity of chondrocytes was further assessed (Fig. 6C and D). Treatment with β-EVs significantly promoted chondrocyte migration at 12 h, 24 h and 36 h (P < 0.01) compared to the control group.

Gene expression analysis via RT-qPCR demonstrated a marked increase in the mRNA levels of PCNA after 48 h of β-EVs treatment compared to the control group (P < 0.05), corroborating the CCK-8 assay results (Fig. 6E). Furthermore, β-EVs significantly upregulated the expression of chondrocyte-specific genes, including collagen type Ⅱ alpha 1 chain (COL2A1) and SRY-Box Transcription Factor 9 (SOX9) compared to the control group (P < 0.05). These findings highlight the superior effect of β-EVs in promoting chondrocyte proliferation, migration and ECM synthesis.

Effects of EVs on tissue‑engineered cartilage regeneration and macrophage infiltration in vivo

To evaluate the regulatory effects of pre-stimulated EVs on cartilage regeneration, chondrocytes-laden porous GelMA hydrogel with or without EVs constructs were implanted subcutaneously and evaluated after 1 and 2 weeks. Macroscopic evaluation showed that EVs-treated groups formed transparent, white, cartilage-like tissue with a thinner surrounding capsule compared to the control group, with the most pronounced effects observed in the α-EVs and β-EVs groups (Fig. 7A).

Fig. 7.

Regulatory effects of EVs under different stimuli on chondrocytes-laden porous GelMA hydrogels and infiltrating macrophages in mice. A The representative gross image of tissue-engineered implants at 1 week and 2 weeks after implantation. B Histological evaluation of tissue-engineered cartilage by hematoxylin and eosin (HE), Toluidine blue, and Alcian blue staining at week 2 post-treatment. Scale bars: 100 μm. C qRT-PCR analysis of the expression of cartilage proliferation-related gene (PCNA) and cartilage matrix protein synthesis-related genes (COL2A1, SOX9) in tissue-engineered cartilage constructs at 2 weeks post-implantation. D Immunohistochemical staining of CD80 and CD206 at week 2 post-treatment. Black boxes indicated magnified microscopic fields of view. Scale bars: 100 μm. E qRT-PCR analysis of the expression of macrophage M1 phenotypic-related genes (CD80, iNOS, IL6) and M2 phenotypic-related genes (Arg1, Ym1, IL10) in macrophages infiltrating the tissue-engineered cartilage at 2 weeks post-implantation. Compare data from the four groups in pairs and presented as the mean ± SD from three replicates. Only the parts with statistical differences are expressed. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001

Histological analysis of paraffin-embedded tissue-engineered cartilage was performed using HE, Toluidine blue, and Alcian blue staining. The EV-treated groups exhibited more intense cartilage matrix staining, particularly in the β-EVs group, compared to the control group (Fig. 7B). RT-qPCR was performed to quantitatively examine the gene expression levels of PCNA, COL2A1 and SOX9 in the four experimental groups (Fig. 7C). The results revealed that after 2 weeks of in vivo culture, the expression levels of these three genes in the β-EVs group were significantly higher than those in the control group (P < 0.05). Although the n-EVs and α-EVs groups also exhibited higher expression levels compared to the control group, these differences were not statistically significance. In summary, the RT-qPCR data demonstrated that β-EVs had a marked advantage in promoting chondrocyte proliferation and ECM synthesis within the tissue-engineered cartilage in vivo, consistent with the observations from our in vitro studies.

To further elucidate the effects of EVs on local macrophage activity following subcutaneous implantation, macrophage infiltration was assessed after 2 weeks. Immunohistochemical staining revealed a significant reduction in CD80⁺M1 macrophages in the α-EVs and β-EVs groups compared to the control group, along with a marked increase in CD206⁺M2 macrophages (Fig. 7D). These findings align with the in vitro results, demonstrating that α-EVs and β-EVs effectively inhibited M1 polarization while promoting M2 polarization of macrophages. This shift in macrophage phenotype likely contributed to the enhanced formation of tissue-engineered cartilage. We then quantitatively analyzed the macrophage-related gene expression using RT-qPCR after two weeks of implantation (Fig. 7E). The results showed that the inhibitory effect of α-EVs significantly inhibit the expression of M1 macrophage markers, including CD80, iNOS, and IL6, compared to both the control and n-EVs groups (P < 0.05). Similarly, CD80 and IL6 expression levels in the β-EVs group were significantly lower than those in the n-EVs group (P < 0.01). For M2 macrophage markers, both α-EVs and β-EVs markedly upregulated the expression of Ym1 and IL10 compared to the control and n-EVs groups (P < 0.05). Additionally, the α-EVs group demonstrated significantly higher Arg1 expression level than both the control and β-EVs group (P < 0.05). These results indicated that both α-EVs and β-EVs effectively suppressed M1 macrophage polarization while promoting M2 polarization, with α-EVs exerting a more pronounced regulatory effect. Taken together, the in vivo results demonstrated that α-EVs and β-EVs provided superior protective effects in tissue-engineered cartilage implantation compared to n-EVs and the control group. Specifically, α-EVs were more effective in regulating macrophage polarization, while β-EVs played a pivotal role in supporting engineered cartilage regeneration.

Small RNA sequencing and bioinformatics analysis of n-EVs, α-EVs and β-EVs

The miRNA expression profiles of EVs varied significantly depending on exposure to different stimulatory environments [21]. In this study, we performed small RNA sequencing to profile and compare the miRNA expression patterns of α-EVs, β-EVs, and n-EVs, aiming to elucidate the underlying mechanisms of EVs on macrophages (Fig. 8). Heatmap analysis revealed distinct biological variations in the miRNA profiles among the three EVs types (Fig. 8A). Volcano plots highlighted differentially expressed miRNAs (DE-miRNAs) between each pair of groups (Fig. 8B). Notably, nine common DE-miRNAs, including the highly expressed miR-378a-3p and miR-1-3p, were identified in both α-EVs and β-EVs compared to n-EVs. In addition, α-EVs exhibited 13 unique DE-miRNAs, such as miR-370-3p, miR-399-3p, miR-155-5p, miR-140-3p, and miR-365b-5p, all of which were highly expressed. β-EVs, on the other hand, displayed 12 unique DE-miRNAs, including the highly expressed miR-10a-5p (Additional file 3).

Fig. 8.

Small RNA sequencing of EVs. A Cluster heatmap showing differentially expressed miRNAs in n-EVs, α-EVs, and β-EVs. The color gradient from blue (-1 log₂ normalized expression) to red (+1 log₂ normalized expression) represents miRNA expression levels. B Volcano plot illustrating the upregulation and downregulation of miRNAs in different comparison between two types of EVs

KEGG enrichment pathway analysis (Fig. 9A) revealed the upregulation of the PI3K-AKT, mTOR, and Chemokine signaling pathways in both α-EVs and β-EVs, which were absent in n-EVs. Namely, miRNAs associated with these signaling pathways may play pivotal roles in regulating macrophage proliferation, migration, and polarization. Furthermore, GO analysis (Fig. 9B) indicated that different types of EVs modulate various biological processes, including cellular component organization or biogenesis, metabolic process, immune system process, and responses to stimuli.

Fig. 9.

Small RNA sequencing of EVs revealed the potential mechanisms of cell proliferation, migration, and inflammatory responses associated with cartilage regeneration and macrophage infiltration. A Kyoto Encyclopedia of Genes and Genomes enrichment analysis for the upregulated pathways corresponding to the comparisons. B Gene ontology analysis for the upregulated biological processes corresponding to the comparisons

Differential expression of miR-378a-3p in three EVs groups and its targeted interaction with SIRPα

In this study, we focused on the negative regulatory effect of miR-378a-3p on its target genes. Bioinformatics analysis revealed that miR-378a-3p was highly expressed in both α-EVs and β-EVs (Fig. 10A). Using the Target Scan, Miranda and miRTarBase databases, we identified potential target genes of miR-378a-3p based on predicted binding affinity within the 3′-UTR region. Among these, Signal Regulatory Protein α (SIRPα) was successfully predicted by both the Target Scan and Miranda databases. Furthermore, we designed a wild-type 3`UTR sequence of SIRPα containing the predicted binding site of miR-378a-3p and a mutated version of the sequence with disrupted binding (Fig. 10B). SIRPα, predominantly expressed on macrophage surfaces, plays a pivotal role in immune regulation by mediating the recognition of self-and non-self-cells [22]. To validate the regulatory effect of EVs treatment on SIRPα expression in macrophages, we conducted RT-qPCR and western blotting analysis (Fig. 10C and D). The results indicated a significant downregulation of SIRPα at both the mRNA and protein levels in macrophages following 48 h of treatment with α-EVs or β-EVs, compared to the control group (p < 0.05). These findings align with our bioinformatics prediction that miR-378a-3p negatively regulates SIRPα through direct binding to its 3`UTR region. To further confirm this interaction, we performed dual luciferase reporter gene assay. The results demonstrated that co-transfection of the wild-type vector with miR-378a-3p led to a significant reduction in luciferase activity, whereas no inhibitory effect was observed in cells transfected with the mutant vector (Fig. 10E). These findings definitively confirmed that SIRPα is a specific target of miR-378a-3p.

Fig. 10.

The mutual interaction between SIRPα and miR-378a-3p. A Small RNA sequencing analysis showing the expression level of miR-378a-3p in n-EVs, α-EVs, and β-EVs. B Venn diagram of predicted potential target genes of miR-378a-3p, based on Target Scan, Miranda, and miRTarBase databases, and the prediction of the potential binding sites between miR-378a-3p and SIRPα in the 3`UTR, along with the corresponding mutant sequence. C qRT-PCR analysis of SIRPα expression in macrophages treated with EVs for 48 h. D Western blot analysis of SIRPα protein expression in macrophages from the four experimental groups. Full-length blots were presented in Supplementary Figure. E Luciferase reporter assay conducted in 293T cells transfected with luciferase reporter plasmids containing the WT or MUT 3ʹ-UTR of SIRPα, in the presence of miR-378a-3p mimics or negative controls (nc). Luciferase activity was measured using a Dual-Luciferase Reporter Assay System. Data were compared with the control group and presented as the mean ± SD from three replicates. ∗p < 0.05, ns: no significance

Discussion

Adipose tissue offers several advantages as a source for cell-derived EVs: it is more abundant, easier to harvest, and results in less tissue damage compared to other tissue sources. In this study, we employed a 3D spheroid culture system for ADSCs to better mimic the in vivo microenvironment, enhance cell-cell interactions, reduce the required culture space, and increase the concentration of EVs per unit volume.

Previous research has highlighted the immunomodulatory properties of ADSCs-EVs [23, 24]. findings further validated these effects, demonstrating that n-EVs significantly reduced the expression of the macrophage proliferation marker PCNA. Regarding macrophage polarization and cytokine secretion, n-EVs suppressed M1 polarization and promoted M2 polarization. Moreover, EVs derived from pre-stimulated ADSCs exhibited enhanced immunomodulatory effects, including stronger inhibition of macrophage proliferation and migration, regulation of inflammatory cytokine secretion, and an increased M2/M1 macrophage ratio.

In our study, α-EVs empowered by TNF-α exhibited more pronounced immunomodulatory properties than n-EVs. Specifically, α-EVs demonstrated the ability to reduce inflammation and promote the polarization of M1 to M2 macrophages. This was further supported by in vivo experiments which confirmed the immunomodulatory effects of α-EVs on macrophage polarization. Notably, α-EVs did not exert any negative effects on chondrocyte proliferation, migration, or matrix synthesis. We propose that the pro-inflammatory cytokine TNF-α induces ADSCs to adopt an anti-inflammatory phenotype, thereby releasing EVs with enhanced anti-inflammatory properties that can effectively regulate macrophage immune status without impairing chondrocyte regeneration. Previous studies have predominantly focused on the effects of pro-inflammatory cytokines on MSCs. Among these, TNF-α has been one of the most extensively studied cytokines. TNF-α, a representative pro-inflammatory cytokine, plays a pivotal role in the recruitment of additional inflammatory cells and the polarization of pro-inflammatory M1 macrophages during the acute inflammatory phase [25, 26]. Research showed that TNF-α exposure enhanced the therapeutic potential of MSCs-EVs in various inflammation-related diseases, including acute liver failure [27], urethral stricture [28], and periodontitis [29]. For example, Nakao et al. compared the effects of TNF-α, IFN-γ, LPS, and aspirin on the secretion of EVs from gingival tissue-derived MSCs (GMSCs) and found that TNF-α treatment not only increased the number of GMSC-EVs but also significantly induced the polarization of phenotypic macrophage from M1 to M2, which in turn inhibited bone loss in a mouse model of periodontitis [29]. Recent studies revealed that EVs secreted by ADSCs pre-stimulated with IFN-γ or TNF-α combined with IFN-γ, effectively reduced the number of pro-inflammatory M1 macrophages, while simultaneously increasing the number of anti-inflammatory M2 macrophages. This shift in macrophage polarization contributes to the alleviation of inflammation and fibrosis in various disease models, thereby promoting tissue repair and regeneration [14, 30]. Mechanistic studies indicated that inflammatory cytokines upregulated the expression of inflammatory-related miRNAs in MSCs-EVs, such as miR-299-3p, miR-27b-3p, miR-146a, and miR-147b. These miRNAs inhibited the activation of corresponding inflammatory pathways by targeting specific genes, thereby mitigating inflammatory damage and fostering tissue repair [14, 27, 28, 31].

Interestingly, TGF-β1-empowered EVs exhibited stronger inhibitory effects on macrophage proliferation and migration, as well as a reduction in the secretion of macrophage-related chemokines, compared to n-EVs and α-EVs. These effects could potentially mitigate the recruitment of macrophages, thereby inhibiting the progression of FBR. Both in vitro and in vivo experiments demonstrated that β-EVs, in comparison to n-EVs, significantly promoted chondrocyte proliferation and supported the maintenance of engineered cartilage in vivo. TGF-β1, a key pro-regenerative cytokine, is known to induce M2 macrophage polarization and stimulate the secretion of anti-inflammatory cytokines, thereby suppressing inflammation [25, 32, 33]. Additionally, TGF-β1 downregulates matrix metalloproteinases (MMPs), promotes the synthesis of aggrecan and type II collagen in chondrocytes, and facilitates tissue repair and ECM remodeling [25, 32, 33]. TGF-β1also serves as a critical growth factor for inducing MSCs chondrogenesis, promoting cartilage formation, and maintaining cartilage homeostasis [34, 35]. Our findings align with those of a previous study examining the empowering effects of TGF-β1 on BMSCs. Wang et al. demonstrated that TGF-β1-empowered BMSC-EVs promoted chondrocyte proliferation and reduced cartilage degradation in a rat osteoarthritis model. This was achieved by downregulating Sp1, a chondrocyte proliferation inhibitor, via miR-135b, and by facilitating cartilage repair through the targeting of MAPK6, which regulated the polarization of synovial macrophages from the M1 to M2 phenotype [36, 37]. In our study, we not only confirmed that β-EVs could promote the M2 phenotypic polarization of macrophages, but also observed that β-EVs played a more prominent role in inhibiting macrophage proliferation and migration. We hypothesize that TGF-β1 stimulation induces ADSCs to release EVs with enhanced anti-inflammatory and pro-regenerative effects. Anti-inflammatory EVs help to inhibit macrophage proliferation and migration, inhibit the formation of FBGCs around biomaterials, and are conducive to the in vivo integration of engineered tissues. At the same time, pro-regenerative EVs act on chondrocytes, providing a protective effect in the process of cartilage regeneration.

Despite differences in macrophage modulation between α-EVs and β-EVs, both exhibited consistent anti-inflammatory effects and promoted M2 polarization. We focused on 9 upregulated miRNAs, including miR-378a-3p and miR-1-3p, in both α-EVs and β-EVs. Notably, miR-378a-3p was highly expressed in α-EVs, which may explain the stronger immunoregulatory effects observed in our studies. Consequently, we selected miR-378a-3p for target gene prediction and verification, given its role in modulating inflammatory responses. Wang et al. reported a miR-378a-3p upregulation in IL-4-induced M2 macrophages, with growth factor receptor-bound protein 2 (GRB2) identified a one of target genes [38].

We predicted SIRPα as a potential target gene of miR-378a-3p using Target Scan and miRanda databases. Both α-EVs and β-EVs significantly downregulated SIRPα expression at gene and protein levels, and the dual-luciferase reporter gene assay confirmed miR-378a-3p`s direct binding to 3`-UTR of SIRPα.

SIRPα, a member of the immunoglobulin superfamily, interacts with CD47 on macrophages, serving as a “don’t eat me” signal to inhibit macrophage-mediated phagocytosis and promote immune escape in cancer [22, 39]. However, CD47 also acts as an “eat me” signal when oxidative stressed, undergoing conformational changes and binding to thrombospondin-1 (TSP-1). This binding alters the SIRPα-CD47 axis from inhibitory to activating, triggering erythrocyte phagocytosis [40]. Implanting tissue-engineered cartilage triggers biological responses such as inflammation, fibrosis, and healing, leading to local oxidative stress that affects tissue regeneration [22, 41]. Our previous research showed that oxidative damage either in vitro or in subcutaneous transplantation caused conformational changes in CD47 on chondrocyte. This altered CD47 binds to SIRPα via TSP-1, promoting macrophage recognition and phagocytosis, which hampers cartilage regeneration and ECM formation (data not shown). Although these observations support the involvement of the CD47/TSP-1/SIRPα axis in the FBR, direct mechanistic validation was beyond the scope of the present study. Building upon this framework, our current results showed that both α-EVs and β-EVs were enriched in miR-378a-3p and were associated with reduced SIRPα expression in macrophages. These findings suggested that miR-378a-3p–mediated regulation of SIRPα might contribute to attenuated macrophage recognition of implanted constructs, thereby creating a more permissive immune microenvironment for tissue-engineered cartilage formation.

In addition to its role in phagocytosis, SIRPα—also referred to as macrophage fusion receptor (MFR)—has been implicated in macrophage fusion and the formation of FBGCs, a hallmark feature of chronic FBR [42]. Recent studies indicate that SIRPα can interact with macrophage antigen-1 (Mac-1), forming a Mac-1–SIRPα–CD47 signaling axis that facilitates cell–cell adhesion and macrophage fusion [43]. Within this context, our data suggested that downregulation of SIRPα by miR-378a-3p might be associated with reduced macrophage fusion potential. However, macrophage fusion was not directly assessed in this study, and thus the involvement of this pathway remains speculative and warrants further experimental validation.

In summary, this study systematically explored how pre-stimulation of ADSCs with the pro-inflammatory cytokine TNF-α or the pro-regenerative cytokine TGF-β1 influences the immunomodulatory properties of ADSCs-EVs using in vitro assays, a subcutaneous implantation model, and bioinformatics analyses. Both inflammatory conditions enhanced the immunomodulatory capacity of ADSCs-EVs, and miRNA sequencing identified miR-378a-3p as a shared, upregulated miRNA in α-EVs and β-EVs. Functional analyses further indicated that miR-378a-3p was associated with suppression of SIRPα expression. Collectively, these findings provide preliminary insights into the shared mechanisms by which α-EVs and β-EVs modulate local inflammatory microenvironment and support ectopic cartilage formation. Our findings offer novel perspectives on immunotherapeutic targets and the precise regulation of tissue-engineered implants via empowered EVs.

Several limitations of this study warrant further investigation. First, although our data suggest an association between miR-378a-3p–mediated SIRPα regulation and attenuated FBR, direct causal evidence linking the SIRPα/CD47 axis to FBR progression was not established. Second, ADSCs-EVs likely regulate immune responses through multiple coordinated pathways, whereas this study primarily focused on miRNAs shared between α-EVs and β-EVs. miRNAs uniquely enriched in either EVs subtype may exert distinct immunomodulatory effects and merit further investigation, which will provide a theoretical foundation for the targeted and spatiotemporal modulation of FBR using engineered EVs.

Conclusions

In this study, we utilized the pro-inflammatory cytokine TNF-α and the pro-regenerative cytokine TGF-β1 to separately empower ADSCs, from which we isolated α-EVs and β-EVs, respectively. Both α-EVs and β-EVs demonstrated superior abilities in modulating the immune status of macrophages compared to the control and n-EVs groups, although distinct functional profiles were observed.α-EVs exhibited a more pronounced effect in inhibiting M1 macrophage polarization while promoting M2 macrophage polarization, whereas β-EVs more effectively suppressed macrophage proliferation and migration. Importantly, neither α-EVs nor β-EVs adversely affected chondrocyte phenotype, and β-EVs were further associated with improved chondrocyte proliferation, migration, and ECM synthesis. miRNA analysis revealed consistent enrichment of miR-378a-3p in both α-EVs and β-EVs, which was associated with reduced SIRPα expression, suggesting a potential role of this axis in EV-mediated immune modulation. However, direct causal links between this pathway and foreign body response regulation remain to be established. Our findings offer new insights into immunotherapeutic targets and provide experimental evidence supporting the precise application of inflammation-empowered EVs in tissue engineering and regenerative medicine. Further comprehensive mechanistic analyses are required to define the translational relevance of these findings.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

MSCs

Mesenchymal stem cells

EVs

Extracellular vesicles

MSCs-EVs

EVs derived from MSCs

ADSCs

Adipose-derived stem cells

ECM

Extracellular matrix

SIRPα

Signal-regulatory protein alpha

FBR

Foreign body reaction

FBGCs

Foreign body giant cells

IAE

EVs derived from inflammation-stimulated MSCs

CSF1

Colony-stimulating factor-1

TMJ

Temporomandibular joint

TNF-α

Tumor necrosis factor-α

IL6

Interleukin-6

TGF-β1

Transforming growth factor-β1

PDGF

Platelet-derived growth factor

MSC

Mesenchymal stem cell

FBS

Fetal bovine serum

PS

Penicillin-streptomycin

H-DMEM

High-glucose dulbecco’s modified eagle’s medium

L-DMEM

Low-glucose dulbecco’s modified eagle’s medium

PBS

Phosphate-buffered saline

TEM

Transmission electron microscope

NTA

Nanoparticle tracking analysis

GelMA

Gelatin methacrylate

LAP

Lithium phenyl-2,4,6-trimethyl benzoyl phosphinate

PEO

Poly(ethylene oxide

CFDA SE

Carboxyfluorescein diacetate succinimidyl ester

CCK-8

Cell counting Kit-8

OD

Optical density

RT-qPCR

Quantitative real-time polymerase chain reaction

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

HRP

Horseradish peroxidase

PFA

Paraformaldehyde

HE

Hematoxylin and eosin

RIN

RNA Integrity Number

GO

Gene ontology

KEGG

Kyoto encyclopedia of genes and genomes

Nc

Negative control

SD

Standard deviation

FIZZ1

Found in inflammatory zone 1

Ym1

Chitinase-3-like 3

Arg1

Arginase 1

IL10

Interleukin-10

CD80

Cluster of differentiation 80

iNOS

Inducible nitric oxide synthase

IL1β

Interleukin-1β

PCNA

Proliferating cell nuclear antigen

COX2

Cyclooxygenase-2

CXCL16

C-X-C motif chemokine ligand 16

CCL5

C-C motif chemokine ligand 5

GMCSF

Granulocyte macrophage colony stimulating factor

CCL2

C-C motif chemokine ligand 2

COL2A1

Collagen type II alpha 1 chain

SOX9

SRY-Box Transcription Factor 9

DE-miRNAs

Differentially expressed miRNAs

GMSCs

Gingival tissue-derived MSCs

MMPs

Matrix metalloproteinases

TSP-1

Thrombospondin-1

MFR

Macrophage fusion receptor

Mac1

Macrophage antigen 1

Author contributions

Yuchen Wang: Conceptualization, Methodology, Data curation, Writing-original draft. Jianguo Chen and Yanlong Yang: Data curation, Investigation. Siyu Liu, Tian Li, Zhuoqi Chen and Yue Ma: Methodology, Data curation and analysis. Haiyue Jiang: Conceptualization, Supervision, Writing-review & editing. Xia Liu: Conceptualization, Methodology, Supervision, Writing-review & editing.

Funding

This study was supported by National Key R&D Program of China (Grant number: 2024YFA1107800), National Natural Science Foundation of China (Grant Number: 82371796), CAMS Innovation Fund for Medical Sciences (Grant number: 2021-I2M-1-052) .

Data availability

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

Declarations

Ethics approval and consent to participate

All human samples used in this study were collected and handled in strict accordance with the ethical guidelines and principles approved by the Ethics Committee of the Plastic Surgery Hospital and Institute, Chinese Academy of Medical Sciences, and Peking Union Medical College (Title of the approved project: Inflammation-Stimulated ADSCs-Derived Small Extracellular Vesicle Regulates Macrophages to Promote Tissue-Engineered Ear Cartilage Regeneration. The approved number: 2023(200). Date of approval: November 1, 2023). The animal experiments conducted in this study were approved by the Biomedical Ethics Committee of Animal Care and Experiment at the Plastic Surgery Hospital and Institute of the Chinese Academy of Medical Sciences & Peking Union Medical College (Title of the approved project: Inflammation-Stimulated ADSCs-Derived Small Extracellular Vesicle Regulates Macrophages to Promote Tissue-Engineered Ear Cartilage Regeneration. The approved number: 2023animal (99). Date of approval: November 1, 2023).

Consent for publication

Written informed consent was obtained from all participants prior to their inclusion in the study.

Competing interests

The authors declare no competing interests.