Intravesicular gelation of mesenchymal stromal cell-derived microvesicles for enhanced therapeutic angiogenesis in lower limb ischemia

Abstract

Mesenchymal stromal cell-derived artificial microvesicles (MSC-MVs) hold significant promise as a cell-free alternative to traditional stem cell therapy for the treatment of lower limb ischemia. However, their fragile plasma membrane is highly susceptible to oxidative damage, environmental fluctuations, and long-term storage, often leading to membrane rupture, vesicle disintegration, and leakage of bioactive cargoes. Additionally, MSC-MVs can be contaminated by nuclear genes, limiting their safety and therapeutic applicability. In this study, we developed gelated microvesicles (gel-MVs) derived from enucleated MSCs by incorporating a polyethylene glycol diacrylate (PEGDA) polymer network within the vesicular lumen. This intravesicular gelation process stabilized the structure of MSC-MVs, effectively preventing vesicle degradation and content leakage. In vitro experiments demonstrated that gelation preserved the integrity of bioactive components and maintained their functional activity. In a murine lower limb ischemia model, gel-MVs significantly enhanced angiogenesis, restored blood perfusion, reduced apoptosis, and promoted tissue regeneration in ischemic limbs. This study introduces a novel strategy that integrates artificial polymer networks with natural microvesicles, providing a promising platform for engineering robust and functional MSC-MVs with enhanced therapeutic potential for clinical translation.

---

1. Introduction

Peripheral artery disease (PAD) is defined as chronic vascular dysfunction. The main pathological mechanism is arterial stenosis or occlusion caused by atherosclerosis, resulting in reduced blood flow to the affected limb.^1,2 With the accelerating ageing of the world's population, the incidence of peripheral artery disease is increasing every year. According to statistics, the number of PAD patients worldwide has exceeded 230 million,^3,4 and the incidence rate in people over the age of 60 is 20%.⁵ About 10% of PAD patients go on to develop critical limb ischaemia (CLI), with clinical manifestations of rest pain, ulcers or gangrene. It can even lead to amputation.⁶ Up to date, vascular surgery remains the primary treatment for CLI.^7,8 However, as patients' Rutherford grades advance and their overall health deteriorates, a significant proportion of CLI patients no longer meet the criteria for revascularisation. Moreover, the currently available pharmacological treatments for CLI, such as anticoagulation, antiplatelet and lipid-lowering drugs, only provide symptomatic relief by delaying vascular occlusion. These therapies fail to promote regeneration and functional remodelling of the ischemic tissues, and their side effects further limit their applicability in certain patient populations.^9,10 Therefore, addressing the urgent clinical challenge of improving limb perfusion and alleviating related symptoms in patients with CLI through non-surgical treatment is essential for reducing amputation rates and mortality.¹¹

Extracellular microvesicles produced by mesenchymal stromal cells contain a complex cargo of active components, including proteins, lipids, RNAs (e.g. microRNAs, lncRNAs, and mRNAs), and DNAs. Thus, they serve as a critical class of signal mediators that promote angiogenesis, regulate immune balance, and improve the local microenvironment and other functions largely via horizontal transfer of the cargo to recipient cells.^12–15 These microvesicles, as an alternative to stem cell therapy,^16–19 have been shown to have significant therapeutic effects in myocardial ischaemia-reperfusion injury and lower limb ischaemia.^20–24 However, their application is severely limited by the low yield and poor stability.²⁵ To address these problems, cells are manipulated using top-down engineering methods, such as extrusion and ultrasonic techniques, to create artificial microvesicles.²⁶ This process allows for the production of microvesicles with controllable size, uniform properties, and a relatively high yield while possessing pro-repair properties similar to natural extracellular microvesicles.²⁷ However, during cell fragmentation and recombination, the internal protein fiber network structure disintegrates, leading to a loss of structural support within the microvesicles. This results in increasing fragility of the microvesicles, making them more susceptible to lipid membrane rupture and disintegration, particularly with extended storage or fluctuations in external conditions.^28–30 In addition, the nuclei retained in stem cell-derived microvesicles pose a risk of genetic contamination.

In this study, to increase the stability of MSC-derived artificial microvesicles and eliminate nuclear gene contamination, we developed novel gelated microvesicles derived from enucleated MSCs. We first used cytochalasin incubation to remove genetically risky nuclei from the cells, and then disrupted the cells and reassembled them into microvesicles by extrusion. At the same time, we introduced a polyethylene glycol diacrylate (PEGDA) polymer network in the microvesicles using photoactivated cross-linking chemistry to mimic the support provided by the cytoskeleton to stabilize the microvesicles. Unlike the endogenous cellular skeleton, which is prone to recombination and disintegration when subjected to external stimulation or cell death, the synthetic network can continuously stabilize the microvesicles to prevent the leakage of cytoplasmic contents. We demonstrated that intravesicle gelation effectively preserved the active substance integrity and plasma membrane biological function of MSC-derived microvesicles. These gelated microvesicles (gMVs) were then applied to a lower limb ischemia model, validating their ability to promote angiogenesis and inhibit inflammation (Scheme 1).

Scheme 1 Schematic illustration of the preparation of gMVs and treatment for lower limb ischemia.

2. Materials and methods

2.1. Preparation and characterization of gMVs

Rat adipose mesenchymal stromal cells (rASC) were incubated in a complete medium (α-MEM, Gibco, USA) containing 10% cytochalasin B (Sigma-Aldrich, USA) for 1 h to obtain enucleated cells. Microvesicles (MVs) were obtained by repeated extrusion through a polycarbonate membrane (Whatman, UK) with pore sizes of 3 µm and 1 µm, followed by purification via centrifugation. 0.5 wt% phenyl-2,4,6-trimethylbenzoyl lithium hypophosphate (LAP, Sigma-Aldrich, USA) as a photoinitiator was mixed with FITC (Sigma-Aldrich, USA) labelled PEGDA (Mn = 700 Da, Macklin, China) and dispersed in phosphate buffer saline (PBS), PEGDA content 10–40% (Procell, China). The MVs were suspended in the gel solution and then rapidly frozen in methanol pre-cooled at −80 °C in a refrigerator for 10 min, and then thawed in a water bath at 37 °C. Free PEGDA was removed by washing with PBS 2–3 times. Finally, the internal gel of MVs was cross-linked to obtain gMVs by irradiation with a 365 nm UV lamp for 5 min (Fig. S1).

The concentration of PEGDA within vesicles was determined by iodometry. Briefly, MVs and gMVs (prepared without UV irradiation) were obtained as described previously. Vesicles were disrupted by ultrasound in a water bath for 20 minutes, and the supernatant was collected. The supernatant was then mixed with barium chloride and iodine solution in a ratio of 8 : 2 : 1. After incubation at room temperature for 15 minutes to allow for color development, the optical density (OD) at 535 nm was measured using a multifunctional microplate reader (Tecan, Switzerland) to quantify PEGDA content.

The particle size and zeta potential of MVs and gMVs were analyzed using dynamic light scattering (Malvern, Netherlands), while their morphological features were observed via transmission electron microscopy (TEM, Thermo Fisher, USA). The concentration of specific proteins in gMVs was quantified using an ELISA kit (Elabscience, China). For fluorescent labeling, DiI (Sigma-Aldrich, USA) was used to stain the cell membrane, Hoechst 33342 (Sigma-Aldrich, USA) for the nucleus, and FITC (Sigma-Aldrich, USA) to label the PEGDA gel network. The nucleation efficiency of gMVs was evaluated by flow cytometry (BD, USA), and the co-localization of vesicle membranes with PEGDA gel was visualized via confocal laser scanning microscopy (Zeiss, Germany). To observe this co-localization, DiI-labeled gMVs (without UV irradiation) were incubated in PBS containing 10 µg mL⁻¹ FITC for 2 hours, after which 5 µL of the sample was placed on a microscope slide for confocal imaging.

The morphology and elastic modulus of MVs and gMVs were characterized by atomic force microscopy (AFM, Bruker, USA). For gel core visualization, gMVs were treated with 10% Triton X-100 (Beyotime, China) for 5 minutes, followed by centrifugation to remove the membrane. The resulting gel cores were resuspended in PBS. Fluorescence and morphological changes after membrane removal were observed by confocal microscopy and scanning electron microscopy (SEM, Zeiss, Germany). Dynamic light scattering was also used to assess changes in particle size and zeta potential before and after membrane removal.

2.2. Stability of gMVs

To evaluate stability, MVs and gMVs were dispersed in ultrapure water or 10 wt% sucrose solution and subjected to various storage conditions: refrigeration at 4 °C for 7 days (in ultrapure water), freezing at −80 °C for 30 days, and lyophilization for 30 days (in 10 wt% sucrose solution). Morphological changes were examined via SEM, and particle size alterations were analyzed using dynamic light scattering.

To assess the bioactivity of stored vesicles, MVs and gMVs preserved at −80 °C or freeze-dried for 30 days were co-incubated with human umbilical vein endothelial cells (HUVECs) for 48 hours. Total RNA was extracted using a microRNA extraction kit (Magen, China) according to the manufacturer's protocol. Reverse transcription was performed using a commercial kit (Exongen, China), and quantitative real-time PCR was conducted with the LightCycler 96 system (NEST, China) using All-In-One qPCR Mix (GeneCopoeia, QP001, China). Gene expression was normalized to GAPDH and calculated using the 2^−ΔΔCt method. Primer sequences are provided in Table S1.

2.3. Cell culture

Adipose tissue from the inguinal region of rats was minced and enzymatically digested with 0.075% type I collagenase (Sigma-Aldrich, USA) for 1 hour. The digestion was terminated with a medium containing 20% fetal bovine serum (FBS, Sorfa, USA). The cell suspension was centrifuged, and the resulting pellet was cultured in α-MEM supplemented with 20% FBS, 1% penicillin, and 1% streptomycin (Gibco, USA). HUVECs (iCell, China) were cultured in DMEM supplemented with 10% FBS, 1% penicillin, and 1% streptomycin. All cells were maintained in a humidified incubator at 37 °C with 5% CO₂.

2.4. Cellular uptake

The cellular uptake experiment was conducted in confocal dishes (Biosharp, China). HUVECs (1 × 10⁵ cells) were incubated with 40 µg mL⁻¹ of MVs or gMVs for 24 h. After fixation, the cells were stained with Phalloidin (Bioscience, China) and DAPI (Beyotime, China) to visualize the cytoskeleton and nuclei, respectively. The internalization efficiency of MVs and gMVs was assessed by confocal laser scanning microscopy and flow cytometry. For the co-localization study, HUVECs (1 × 10⁵) were similarly cultured in confocal dishes and incubated with 40 µg mL⁻¹ of MVs or gMVs for 12 h. The medium was then removed, and the cells were washed three times with PBS, followed by incubation in fresh medium (free of vesicles) for an additional 12 or 24 h. Lysosomes were stained using a commercial lysosome-specific probe. The co-localization between gMVs and lysosomes was visualized using confocal microscopy, and Pearson's correlation coefficient was analyzed using the JACoP plug-in in ImageJ. To evaluate the degradation behavior of gMVs, the vesicles were dispersed in complete DMEM (Gibco, USA) and incubated at 37 °C in a humidified atmosphere containing 5% CO₂ for 0–7 days. The morphological changes of gMVs were examined using scanning electron microscopy (SEM).

2.5. Cell proliferation

To evaluate the biocompatibility of MVs and gMVs, HUVECs were seeded into 96-well plates (Costar, USA) at a density of 3 × 10³ cells per well. After cell attachment, various concentrations of MVs or gMVs were added. Cells were incubated at 37 °C in a humidified atmosphere containing 5% CO₂ for 0–7 days. Cell viability was assessed using the CCK-8 kit (Beyotime, China), and the absorbance at 450 nm was measured using a microplate reader.

2.6. Chemotactic migration

Transwell migration assays were conducted using 24-well plates (Costar, USA) with inserts containing an 8 µm pore-size polycarbonate membrane (Corning, USA). HUVECs (2 × 10⁴ cells per well) suspended in 0.2 mL basal medium were seeded into the upper chamber, while 0.6 mL of complete medium containing MVs or gMVs was added to the lower chamber. After 24 h incubation, non-migrated cells on the upper surface of the membrane were gently removed with a cotton swab. The migrated cells on the lower surface were fixed with 4% paraformaldehyde (Biosharp, China) for 30 min, stained with 0.1% crystal violet (Beyotime, China) for 10 min, and observed under a light microscope.

2.7. Tube formation assay

For in vitro angiogenesis assessment, 10 µL of growth factor-reduced Matrigel (Corning, USA) was added to pre-chilled 96-well plates and polymerized at 37 °C. HUVECs pretreated with MVs or gMVs for 48 h were seeded on the Matrigel at 3 × 10⁴ cells per well. After 4–6 h of incubation, tubular structures were observed under a bright-field microscope (Nikon, Japan). The number of branches was quantified using ImageJ software to assess pro-angiogenic activity.

2.8. Real-time polymerase chain reaction (qPCR)

After co-incubation of MVs and gMVs with HUVECs for 48 h, total RNA was extracted using a microRNA extraction kit (Magen, China). Reverse transcription was performed using a commercial kit (Exongen, China). qPCR was conducted using the LightCycler 96 system (Nest, China) and the All-In-One qPCR Mix (GeneCopoeia, QP001, China). Gene expression levels were normalized to GAPDH and analyzed using the 2^−ΔΔCt method. Primer sequences are listed in Table S1.

2.9. Construction of a lower limb ischaemia model

All animal procedures were approved by the Animal Care and Use Committee of South China University of Technology and conducted in accordance with international guidelines. Male BALB/c mice (6–8 weeks old) were obtained from Guangzhou Dean Gene Technology Co., Ltd (Guangzhou, China) and maintained under standard conditions (22 ± 2 °C, 50% humidity, 12 h light/dark cycle). Mice were randomly divided into four groups: Sham, PBS, MVs, and gMVs (n = 6 per group). Anesthesia was induced using 1% sodium pentobarbital (0.1 mL/10 g body weight). A longitudinal incision was made along the left groin to expose the femoral artery. The proximal and distal ends were ligated with 7-0 sutures, and the intervening segment was excised. In the sham group, only the skin was incised without arterial ligation. A 100 µL injection of PBS, MVs, or gMVs was administered intramuscularly into the rectus femoris, adductor, and gastrocnemius muscles. Blood perfusion was monitored using a speckle imaging system on days 0, 7, and 21.

2.10. Blood flow, reperfusion and temperature detection of the ischaemic limb

Blood perfusion in the hindlimbs was measured using a laser speckle contrast imaging system (Perimed, Sweden) on days 0, 7, and 21. Mice were anesthetized and placed on a 37 °C heating pad during imaging to maintain consistent body temperature. Reperfusion was expressed as a percentage relative to the contralateral non-ischemic limb. Additionally, surface temperature changes in ischemic limbs were recorded using an infrared thermal imaging camera (Fluke, USA).

2.11. Semi-quantitative assessment of ischaemic and active injury

Ischaemic injury in the lower limbs was scored on day 21 post-surgery as follows: 0 indicated no difference from the contralateral limb; 1 indicated slight discolouration; 2 indicated moderate discolouration; 3 indicated severe discolouration or tissue necrosis; 4 indicated amputation.³¹ Motor function was assessed as: 0 for toe flexion; 1 indicated plantar flexion; 2 indicated no foot drag but no plantar flexion; 3 meant foot drag.³²

2.12. Histological analysis and immunofluorescence staining

On day 21, mice were euthanized, and gastrocnemius muscles were harvested. For immunofluorescence staining, tissues were embedded in OCT (Sakura Finetek, Japan), frozen, and cryosectioned at 10 µm. Sections were permeabilized with 0.2% Triton X-100, blocked with 5% BSA for 1 h, and incubated with primary antibodies (Abcam, UK) overnight at 4 °C. After washing, sections were incubated with secondary antibodies (Signalway Antibody, USA) for 1 h at room temperature. DAPI was used to stain nuclei. Images were captured by confocal microscopy and analyzed using ImageJ. For histological staining, tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 5 µm. Hematoxylin and eosin (H&E) staining and Masson's trichrome staining were performed to evaluate inflammation and fibrosis. Quantitative analysis was based on five randomly selected fields per section using ImageJ.

2.13. Biosafety assessment

To evaluate the biosafety of MVs and gMVs, healthy mice were randomly divided into four groups and injected with 100 µL PBS, MVs, or gMVs. After 28 days, blood samples were collected for routine hematological analyses. Hematological parameters included white blood cell count (WBC), neutrophil count (Neu#), platelet count (PLT), red blood cell count (RBC), hemoglobin (HGB), and mean corpuscular hemoglobin (MCH). Major organs (heart, liver, spleen, lungs, and kidneys) were harvested, rinsed with PBS, and fixed in 4% paraformaldehyde, embedded in paraffin, sectioned at 4 µm, stained with HE, and subjected to histopathological evaluation.

2.14. Statistical analysis

All experiments were conducted with at least three independent replicates. Data are presented as mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism 8. One-way analysis of variance (ANOVA) followed by Tukey's post hoc test was used to assess differences among groups. A p-value < 0.05 was considered statistically significant.

3. Results and discussion

3.1. Preparation and characterization of gMVs

In this study, to reduce the risks of nuclear gene contamination and tumourigenesis associated with the presence of nuclear material, the nuclei of rat MSCs were removed by incubation with cytochalasin B followed by centrifugation. The cells were then extruded through polycarbonate membranes with different pore sizes to obtain MSC-derived microvesicles (MVs). As shown in Fig. 1A, over 99% of MVs were free of nuclear material, indicating successful enucleation. To achieve intravesicle gelation of microvesicles, the gel monomers must be hydrophilic, and the molecular weight must be low enough to ensure that the gel monomers can penetrate through the membrane into the vesicles. Besides, a controlled chemical cross-linking method with low protein reactivity is required to avoid the destruction of the active components in the vesicles. Finally, to prevent cross-linking of gel monomers outside the vesicles, an effect that can lead to particle aggregation and masking of membrane components, the adsorption of monomers onto the vesicle membrane should be minimized. Based on the above considerations, we chose low molecular weight PEGDA (M = 700) to provide the intravesicle network, which has low biological reactivity and is widely used in the biomedical field. As shown in Fig. 1B, the internalized content of the PEGDA monomer increased with the PEGDA concentration, indicating that the PEGDA monomer entered the microvesicles through the plasma membrane via the freeze–thaw cycle treatment in a concentration-dependent manner. Meanwhile, the water-soluble dye fluorescein isothiocyanate (FITC) could not enter the microvesicles, suggesting the freeze–thaw cycle treatment does not compromise the membrane integrity of the microvesicles (Fig. S2). Following PEGDA cross-linking within the microvesicles, proteomic mass spectrometry revealed that gMVs retained more than seven thousand active proteins (Table S1). Representative pro-angiogenic factors, including VEGF and HIF-1α, were quantified using ELISA, which showed no significant differences in content between MVs and gMVs (Fig. 1C). These findings indicate that the internal gelation process does not compromise the biological regulatory functions of the microvesicles. Besides, Laser confocal microscopy revealed that the plasma membrane structure overlapped with the polymer network, indicating successful internal gelation across various gel monomer concentrations and vesicle sizes (Fig. 1D). Dynamic light scattering analysis demonstrated that the average particle sizes of microvesicles extruded through 0.4 µm, 1 µm, and 3 µm polycarbonate membranes were approximately 312.7 nm, 535.3 nm, and 2176.7 nm, respectively, each slightly smaller than the corresponding pore sizes. The zeta potentials of the vesicles were measured at −21.1 mV, −21.3 mV, and −20.7 mV, respectively (Fig. 1E and F), indicating good colloidal stability across sizes. Considering gel core stability, drug release kinetics, and production efficiency, microvesicles with a diameter of 1 µm and a PEGDA concentration of 20% were selected for subsequent experiments.

Fig. 1 Preparation and characterization of gelated microvesicles: (A) enucleation rate of MVs. (B) Intravesicular PEGDA concentrations in MVs before and after freeze–thaw cycles in gelation buffers with varying PEGDA contents. (C) Quantification of pro-angiogenic proteins (VEGF and HIF) in MVs and gMVs by ELISA. (D) Representative fluorescence images of gMVs prepared using different PEGDA concentrations and vesicle sizes. PEGDA hydrogels were stained with FITC (green), and MSC plasma membranes were labeled with DiI (red). (E) Particle size and zeta potential of gMVs of varying sizes prepared with 20% PEGDA. (F) Particle size and zeta potential of gMVs were prepared using a 1 µm pore size polycarbonate (PC) membrane and different PEGDA concentrations.

To further confirm the formation of the polymer network within the microvesicles, the morphological characteristics of the two nanoparticle types were examined using transmission electron microscopy (TEM). MVs exhibited a hollow vesicular structure, whereas gMVs possessed a compact solid core resulting from internal gelation, displaying a characteristic core–shell architecture with a uniform spherical morphology (Fig. 2A). The morphology and mechanical properties of MVs and gMVs were further analysed using atomic force microscopy (AFM). MVs displayed noticeable surface collapse, which was not observed in gMVs. This structural integrity in gMVs is likely due to the internal polymer network providing mechanical support to the vesicle membrane after dehydration, a feature lacking in MVs. Consistently, Young's modulus of gMVs increased to more than four times that of MVs, indicating enhanced mechanical strength (Fig. 2B and C). To validate the presence of an internal gel core, gMVs were treated with 10% Triton X-100 to remove the plasma membrane. Confocal imaging showed that the DiI signal, which labels the membrane, was completely lost post-treatment, while the FITC-labelled gel core remained intact (Fig. 2D and E). Notably, the average particle size of gMVs decreased from 485.7 nm to 465.5 nm after membrane removal. This reduction, slightly greater than the typical thickness of a phospholipid bilayer, suggests that the gel core did not fully occupy the cytoplasmic volume, supporting the internal localization of the polymer network.

Fig. 2 Verification of gelation within microvesicles: (A) representative transmission electron microscopy (TEM) images of MVs and gMVs. (B) Representative atomic force microscopy (AFM) images showing the morphology of gMVs. (C) Young's modulus of MVs and gMVs. (D) Fluorescence and SEM images demonstrate that gMVs retain their structural integrity in PBS following treatment with 10% Triton X-100. (E) Particle size distribution of MVs and gMVs before and after Triton X-100 treatment, as determined by dynamic light scattering (DLS).

3.2. Stability of gMVs

Next, we set up different storage conditions to examine the effect of internal gelation on the stability of microvesicles. First, MVs and gMVs were dispersed in ultrapure water and stored at 4 °C for 7 days. Scanning electron microscopy and dynamic light scattering analyses revealed that, in contrast to gMVs, MVs gradually lost their spherical morphology, with numerous fragmented membrane remnants observed. The average particle size of MVs decreased significantly from 492.7 nm to 213 nm, indicating structural degradation. Although slight aggregation was noted in gMVs, their shape and size remained largely unchanged (Fig. 3A and B). Next, MVs and gMVs were subjected to freezing and lyophilization. After 30 days, MVs exhibited marked morphological alterations, including collapse and cracking, whereas gMVs retained their structural integrity (Fig. 3C–E). To assess the functional impact of long-term storage, qPCR was performed to measure the expression of growth factors and inflammation-related genes in HUVECs treated with stored MVs and gMVs. Even after 30 days, gMVs preserved the regulatory effects characteristic of parental mesenchymal stromal cells, notably upregulating genes such as VEGF and HIF-1α, while significantly suppressing pro-inflammatory genes ICAM and TNF-α (Fig. 3F). In contrast, MVs showed a marked loss of regulatory function, likely due to leakage of bioactive contents resulting from membrane rupture. These findings suggest that internal gelation, which establishes a supportive polymer network that mimics the cytoskeletal architecture, effectively reinforces the vesicle membrane, shields the contents from environmental stress, and preserves both structural and functional integrity during storage.

Fig. 3 Stability of MVs and gMVs under various storage conditions. (A) and (B) SEM images and particle size of MVs and gMVs after storage in ultrapure water at 4 °C for 7 days. (C)–(E) SEM images and particle size after 30 d of storage at −80 °C or following lyophilization. (F) Quantitative real-time PCR analysis of HUVECs treated with MVs or gMVs (1 µm diameter, 20% PEGDA) after 30 days of storage under the above conditions, showing expression levels of representative growth factors, pro-inflammatory factors, and adhesion molecules.

3.3. Cellular uptake of gMVs

To confirm the uptake of gMVs by HUVECs, gMVs were labeled with the red membrane dye DiI and co-incubated in vitro with HUVECs, whose F-actin cytoskeletons were stained green. Confocal microscopy revealed the presence of gMVs within the cytoplasm of HUVECs, indicating successful internalization (Fig. 3A). Flow cytometry further showed that gMV uptake began as early as 3 h after incubation and increased progressively over time, reaching an uptake efficiency of approximately 95% at 12 h (Fig. 4A and B). Cytotoxicity assays demonstrated that gMVs did not negatively affect the proliferation of HUVECs (Fig. S3). Exogenous microvesicles are often routed to lysosomes following endocytosis and subjected to rapid degradation, which hinders their ability to exert biological functions.³³ To assess this, we performed a co-localization study between gMVs and lysosomes in HUVECs. After 12 h of incubation, the Pearson's correlation coefficient for gMV–lysosome co-localization was 34.7%. Even after extending the incubation to 24 h, the co-localization coefficient only increased to 52.2% (Fig. 4C), indicating that a significant portion of gMVs remained intact and escaped immediate lysosomal degradation. These results indicate that gMVs are efficiently internalized by HUVECs and largely evade lysosomal degradation. Additionally, when incubated under physiological conditions (37 °C) in vitro, gMVs underwent near-complete degradation within 7 days as observed by SEM (Fig. S4), implying that in vivo degradation may occur even more rapidly due to enzymatic activity.

Fig. 4 Cellular uptake of gMVs. (A) Representative fluorescence images showing the internalization of DiI-labeled MVs and gMVs (red) by HUVECs, with F-actin stained green using phalloidin and nuclei stained with DAPI (blue). (B) Quantitative analysis of gMV uptake by HUVECs over time using flow cytometry. (C) Confocal images showing co-localization of internalized gMVs with lysosomes at 12 h and 24 h post-incubation. Pearson's correlation coefficients were used to quantify co-localization.

3.4. Pro-angiogenic and anti-inflammatory potential of gMVs

Mesenchymal stromal cell-derived microvesicles (MSCs-MVs) carry a broad array of proteins and cytokines inherited from their parental MSCs, which are crucial mediators of the MSCs’ biological regulatory functions. For instance, HIF-1α can stimulate the expression of pro-angiogenic and transforming growth factors, thereby promoting endothelial cell proliferation and migration.³⁴ To evaluate the effects of internal gelation on the pro-angiogenic activity of microvesicles, we first assessed the ability of gMVs to regulate the chemotactic migration of HUVECs. The results demonstrated that gMVs significantly enhanced HUVECs migration (Fig. 5A and B). In subsequent tube formation assays, gMV treatment markedly increased the formation of nodes and branches in HUVECs networks (Fig. 5C and D). Moreover, angiogenesis-related genes such as VEGF and HIF-1α were significantly upregulated in HUVECs treated with gMVs, indicating an enhanced angiogenic response (Fig. 5E). Beyond promoting angiogenesis, suppression of inflammation also plays a vital role in the treatment of limb ischemia.³⁵ qPCR analysis revealed that gMVs reduced the expression of pro-inflammatory genes, including IL-6 and TNF-α, in HUVECs (Fig. 5F). Notably, gMVs exhibited comparable efficacy to native MVs at equivalent concentrations in both upregulating angiogenic factors and downregulating inflammatory cytokines, indicating that the internal gelation process preserved the biological functionality of MVs. Collectively, these findings confirm that gMVs enhance the angiogenic phenotype of HUVECs, underscoring their therapeutic potential for promoting angiogenesis in the treatment of CLI.

Fig. 5 Biological functions of gMVs. (A) Transwell migration assay of HUVECs after 24 h of co-culture with MVs or gMVs. (B) Quantification of HUVECs migratory capacity in each group. (C) Tube formation assay of HUVECs after 4 h of co-culture with MVs or gMVs. (D) Quantitative analysis of tube formation in each group. (E) Quantitative real-time PCR analysis of pro-angiogenic gene expression (e.g., VEGF and HIF-1α) in HUVECs. (F) Quantitative real-time PCR analysis of pro-inflammatory gene expression (e.g., IL-6 and TNF-α) in HUVECs.

3.5. Therapeutic effects of gMVs in CLI treatment

Following in vitro validation of gMVs functionality, their therapeutic efficacy was further assessed in vivo using a mouse model of lower limb ischemia. Based on the degree of recovery of the ischemic limbs, mice were categorized into three groups: limb preservation, foot necrosis, and limb loss. Statistical analysis of the proportion of mice in each category revealed that, compared with the sham group, mice in the PBS-treated group exhibited severe foot necrosis and limb loss, with a limb preservation rate of only 20%. In contrast, mice treated with gMVs showed significantly alleviated ischemic symptoms, with a limb preservation rate exceeding 80%, comparable to the MVs-treated group (Fig. 6A). In addition, ischemic severity and limb functionality were evaluated using semi-quantitative scoring systems for both ischemic injury and dynamic limb impairment. Mice in the gMVs-treated group exhibited significantly lower scores compared to the PBS group, indicating substantial improvements in tissue repair and functional recovery (Fig. 6B and C).

Fig. 6 Therapeutic effects of gMVs on lower limb ischemia. (A) Activity impairment scores of mice in each group. (B) Semi-quantitative ischemic injury scores for each treatment group. (C) The proportion of mice exhibiting limb preservation, foot necrosis, or limb loss in each group. (D) Representative laser speckle imaging of blood flow in ischemic limbs before and after treatment, showing perfusion restoration. (E) Quantitative analysis of blood flow recovery, expressed as the ratio of perfusion between ischemic and non-ischemic limbs. (F) Temperature changes in the ischemic limbs across treatment groups.

Blood flow recovery in the ischemic lower limb was evaluated using laser Doppler perfusion imaging prior to surgery and at days 0, 7, and 21 post-surgery. The results demonstrated that gMVs treatment significantly enhanced blood reperfusion in the ischemic limb, with perfusion levels reaching approximately 80% of the pre-surgical baseline by day 21. In contrast, the PBS group exhibited only approximately 40% recovery (Fig. 6D and E). Given that limb temperature correlates closely with blood flow and can serve as an indirect indicator of perfusion recovery,³⁶ we monitored limb temperatures following ischemia. The ischemic limb temperature in the gMVs-treated group gradually increased post-surgery, indicating improved vascular function (Fig. 6F). Consistently, immunofluorescent staining on day 21 showed significantly enhanced microvascular density in gMVs-treated limbs, as evidenced by increased α-SMA and CD31 expression (Fig. 7A–D). These results confirm the potent pro-angiogenic effect of gMVs, supporting their capacity to restore blood perfusion in ischemic tissues.

Fig. 7 Angiogenic effects of gMVs in ischemic limbs. (A) Immunofluorescence staining of α-SMA in the gastrocnemius muscle on day 21 post-surgery. (B) Quantification of capillary density based on α-SMA staining. (C) Immunofluorescence staining of CD31 in the gastrocnemius muscle on day 21. (D) Quantification of microvessel density based on CD31 staining.

To further evaluate the therapeutic efficacy of gMVs, mice were sacrificed on day 21, and gastrocnemius muscles from the ischemic limbs were harvested for histological analysis. Hematoxylin and eosin (H&E) staining showed that the PBS group exhibited severe muscle injury characterized by marked myofibrillar degeneration and inflammatory infiltration compared to the sham group. In contrast, muscle structure was largely preserved in the gMVs group, with significantly reduced degeneration and inflammation (Fig. 8A and B). Masson's trichrome staining demonstrated that gMVs treatment also reduced collagen deposition and fibrosis in ischemic muscle tissue, indicating a protective effect against pathological remodeling (Fig. 8C and D). Furthermore, TUNEL and Ki67 immunostaining analyses showed that gMVs markedly inhibited apoptosis (Fig. 8E and F) and promoted cell proliferation (Fig. 8G and H), further supporting their role in tissue regeneration.

Fig. 8 Evaluation of tissue damage in mice with lower limb ischemia treated with gMVs. (A) H&E staining of gastrocnemius muscle sections on day 7 post-surgery. (B) Quantification of collagen deposition from Masson staining. (C) Masson's trichrome staining of gastrocnemius muscle sections on day 7. (D) Quantification of muscle fiber from Masson staining. (E) TUNEL immunofluorescence staining of gastrocnemius muscle sections on day 7 to assess apoptosis. (F) Quantitative analysis of TUNEL-positive cells. (G) Ki67 immunofluorescence staining of the gastrocnemius muscle on day 7 to assess cell proliferation. (H) Quantification of Ki67-positive cells.

To further compare the in vivo efficacy of MVs and gMVs after long-term storage, both types of microvesicles were stored at 4 °C for 7 days before being re-administered via local injection into mice with lower limb ischaemia. Preoperative and postoperative (days 0 and 7) recovery of lower limb blood flow was assessed using a laser Doppler perfusion imaging system. The results demonstrated markedly enhanced blood reperfusion in the ischaemic limb following gMVs treatment, with perfusion levels more than 1.5-fold higher than those in the MVs group and over twice those in the PBS group (Fig. 9A and D). Concurrently, limb temperature monitoring revealed that gMVs exhibited superior recovery compared to MVs (Fig. 9E). Furthermore, histological staining of muscle tissue at day 7 showed that, relative to the sham group, both the PBS and MVs groups exhibited pronounced inflammatory cell infiltration in the ischaemic limb, along with persistent severe muscle degeneration and myofibril loss, accompanied by extensive fibrosis. These pathological alterations were effectively alleviated following gMVs treatment (Fig. 9B, F and C, G). Collectively, these findings indicate that internal gelation markedly enhances the stability of gMVs, effectively preserving the active components within microvesicles under standard storage conditions and thereby maintaining their prolonged therapeutic efficacy in vivo.

Fig. 9 Evaluation of tissue damage in mice with lower limb ischemia treated with gMVs saved for 7 days. (A) Representative laser speckle imaging of blood flow in ischemic limbs before and after treatment, showing perfusion restoration. (B) H&E staining of gastrocnemius muscle sections. (C) Masson's trichrome staining of gastrocnemius muscle sections. (D) Quantitative analysis of blood flow recovery, expressed as the ratio of perfusion between ischemic and non-ischemic limbs. (E) Temperature changes in the ischemic limbs across treatment groups. (F) Quantification of collagen deposition from Masson staining. (G) Quantification of muscle fiber from Masson staining.

3.6. Biosafety assessment

To evaluate the in vivo biosafety of microvesicles, we assessed potential side effects following injection of MVs and gMVs. Blood routine tests revealed no abnormalities, with all parameters remaining within normal ranges (Fig. S5). Moreover, haematoxylin and eosin (HE) staining of major organs, including the heart, liver, spleen, lungs, and kidneys, showed no significant histological or structural alterations after microvesicle administration (Fig. S6). These results confirm the excellent biocompatibility of the microvesicles and indicate the absence of apparent adverse effects in vivo.

4. Conclusions

In this study, we developed gelated microvesicles derived from mesenchymal stromal cells by introducing a polyethylene glycol diacrylate (PEGDA) polymer network within the vesicular structure. This approach represents an optimised strategy for the fabrication of novel engineered microvesicles. By providing sustained and stable structural support at the plasma membrane interface through the internal polymer network, it effectively overcomes core limitations such as content leakage and microvesicle disintegration, while preserving the integrity of bioactive substances and maintaining plasma membrane functionality. The resulting gelled microvesicles markedly attenuated necrosis in ischaemic lower limb muscle tissue and facilitated functional recovery by promoting collateral vascular regeneration and suppressing inflammation. In summary, this technology shows great promise in addressing the practical challenge of low storage stability in engineered microvesicles, establishing a robust new paradigm for the development of next-generation bioactive materials.

Ethical statement

All animal procedures were approved by the Animal Care and Use Committee of South China University of Technology and conducted in accordance with international guidelines.

Author contributions

Chuang Liu: writing – original draft, investigation, formal analysis, data curation. Wenqi Xie: writing – review & editing. Xian Li: investigation, formal analysis. Zixuan Dong: methodology, investigation. Xiaoling Fu: project administration, funding acquisition, conceptualization.

Conflicts of interest

There are no conflicts to declare.

Data availability

All data generated or analyzed during this study are included in this published article in the main text or supplementary information (SI).

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5tb01411e.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (T2288101, 82272152, 31971266), National Key R&D Program of China (2021YFB3800900).

References

1. B. H. Annex, Therapeutic angiogenesis for critical limb ischaemia, Nat. Rev. Cardiol., 2013, 10, 387–396.
2. W. S. Jones and M. R. Patel, Antithrombotic Therapy in Peripheral Artery Disease: Generating and Translating Evidence Into Practice, J. Am. Coll. Cardiol., 2018, 71, 352–362.
3. A. Farber and R. T. Eberhardt, The Current State of Critical Limb Ischemia: A Systematic Review, JAMA Surg., 2016, 151, 1070–1077.
4. S. S. Virani, A. Alonso, E. J. Benjamin, M. S. Bittencourt, C. W. Callaway, A. P. Carson, A. M. Chamberlain, A. R. Chang, S. Cheng, F. N. Delling, L. Djoussé, M. S. V. Elkind, J. F. Ferguson, M. Fornage, S. S. Khan, B. M. Kissela, K. L. Knutson, T. W. Kwan, D. T. Lackland, T. T. Lewis, J. H. Lichtman, C. T. Longenecker, M. S. Loop, P. L. Lutsey, S. S. Martin, K. Matsushita, A. E. Moran, M. E. Mussolino, A. M. Perak, W. D. Rosamond, G. Roth, U. K. A. Sampson, G. M. Satou, E. B. Schroeder, S. H. Shah, C. M. Shay, N. L. Spartano, A. Stokes, D. L. Tirschwell, L. B. VanWagner and C. W. Tsao, American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee, Heart Disease and Stroke Statistics-2020 Update: A Report From the American Heart Association, Circulation, 2020, 141, e139–e596.
5. M. Søgaard, P. B. Nielsen, F. Skjøth, N. Eldrup and T. B. Larsen, Temporal Changes in Secondary Prevention and Cardiovascular Outcomes After Revascularization for Peripheral Arterial Disease in Denmark: A Nationwide Cohort Study, Circulation, 2021, 143, 907–920.
6. T. S. Polonsky and M. M. McDermott, Lower Extremity Peripheral Artery Disease Without Chronic Limb-Threatening Ischemia: A Review, JAMA, 2021, 325, 2188–2198.
7. V. Amlani, M. Falkenberg and J. Nordanstig, The current status of drug-coated devices in lower extremity peripheral artery disease interventions, Prog. Cardiovasc. Dis., 2021, 65, 23–28.
8. S. E. Altin, Y. S. Castro-Dominguez, K. F. Kennedy, K. C. Orion, A. J. Lanksy, J. D. Abbott and H. D. Aronow, Predictors of Underutilization of Medical Therapy in Patients Undergoing Endovascular Revascularization for Peripheral Artery Disease, JACC Cardiovasc. Interv., 2020, 13, 2911–2918.
9. C. Tu, S. Das, A. B. Baker, J. Zoldan and L. J. Suggs, Nanoscale strategies: treatment for peripheral vascular disease and critical limb ischemia, ACS Nano, 2015, 9, 3436–3452.
10. S. R. Levin, N. Arinze and J. J. Siracuse, Lower extremity critical limb ischemia: A review of clinical features and management, Trends Cardiovasc. Med., 2020, 30, 125–130.
11. Z. Xing, C. Zhao, S. Wu, C. Zhang, H. Liu and Y. Fan, Hydrogel-based therapeutic angiogenesis: An alternative treatment strategy for critical limb ischemia, Biomaterials, 2021, 274, 120872.
12. L. V. Lozano Navarro, X. Chen, L. T. Giratá Viviescas, A. K. Ardila-Roa, M. L. Luna-Gonzalez, C. L. Sossa and M. L. Arango-Rodríguez, Mesenchymal stem cells for critical limb ischemia: their function, mechanism, and therapeutic potential, Stem Cell Res. Ther., 2022, 13, 345.
13. N. Zhao, Z. Yue, J. Cui, Y. Yao, X. Song, B. Cui, X. Qi, Z. Han, Z. C. Han, Z. Guo, Z. X. He and Z. Li, IGF-1C domain-modified hydrogel enhances therapeutic potential of mesenchymal stem cells for hindlimb ischemia, Stem Cell Res. Ther., 2019, 10, 129.
14. J. Q. Yin, J. Zhu and J. A. Ankrum, Manufacturing of primed mesenchymal stromal cells for therapy, Nat. Biomed. Eng., 2019, 3, 90–104.
15. J. Liu, F. Chen, L. Tian, J. Wu, K. Liu, Q. Wan, B. Yuan, X. Zhu, X. Chen and X. Zhang, Therapeutic potential of gelatine methacrylate hydrogels loaded with macrophage-derived exosomes for accelerating angiogenesis and cutaneous wound healing, Collagen Leather, 2024, 6, 13.
16. J. Ma, L. Yong, P. Lei, H. Li, Y. Fang, L. Wang, H. Chen, Q. Zhou, W. Wu, L. Jin, D. Sun and X. Zhang, Advances in microRNA from adipose-derived mesenchymal stem cell-derived exosome: focusing on wound healing, J. Mater. Chem. B, 2022, 10, 9565–9577.
17. Z. Li, Y. Wang, H. Wang, H. Wang, Y. Shang, S. Wang, Q. Han, J. Li, R. C. Zhao, Q. Jiang and B. Ding, Self-Assembled DNA Composite-Engineered Mesenchymal Stem Cells for Improved Skin-Wound Repair, Small, 2024, e2310241.
18. L. Luo, J. Tang, K. Nishi, C. Yan, P. U. Dinh, J. Cores, T. Kudo, J. Zhang, T. S. Li and K. Cheng, Fabrication of Synthetic Mesenchymal Stem Cells for the Treatment of Acute Myocardial Infarction in Mice, Circ. Res., 2017, 120, 1768–1775.
19. N. N. Ng and A. S. Thakor, Locoregional delivery of stem cell-based therapies, Sci. Transl. Med., 2020, 12, eaba4564.
20. S. Sahoo and D. W. Losordo, Exosomes and cardiac repair after myocardial infarction, Circ. Res., 2014, 114, 333–344.
21. H. Zhang, M. Xiang, D. Meng, N. Sun and S. Chen, Inhibition of Myocardial Ischemia/Reperfusion Injury by Exosomes Secreted from Mesenchymal Stem Cells, Stem Cells Int., 2016, 2016, 4328362.
22. B. Zhang, X. Wu, X. Zhang, Y. Sun, Y. Yan, H. Shi, Y. Zhu, L. Wu, Z. Pan, W. Zhu, H. Qian and W. Xu, Human umbilical cord mesenchymal stem cell exosomes enhance angiogenesis through the Wnt4/β-catenin pathway, Stem Cells Transl. Med., 2015, 4, 513–522.
23. W. Du, K. Zhang, S. Zhang, R. Wang, Y. Nie, H. Tao, Z. Han, L. Liang, D. Wang, J. Liu, N. Liu, Z. Han, D. Kong and Q. Zhao, Enhanced proangiogenic potential of mesenchymal stem cell-derived exosomes stimulated by a nitric oxide releasing polymer, Biomaterials, 2017, 133, 70–81.
24. H. Wang, X. Wang, C. Liu and C. Liu, Therapeutic strategies for critical limb ischemia: a focus on endogenous cell behavior modulation, J. Mater. Chem. B, 2025, 13, 12707–12737.
25. A. Grangier, J. Branchu, J. Volatron, M. Piffoux, F. Gazeau, C. Wilhelm and A. K. A. Silva, Technological advances towards extracellular vesicles mass production, Adv. Drug Delivery Rev., 2021, 176, 113843.
26. Y. Du, H. Wang, Y. Yang, J. Zhang, Y. Huang, S. Fan, C. Gu, L. Shangguan and X. Lin, Extracellular Vesicle Mimetics: Preparation from Top-Down Approaches and Biological Functions, Adv. Healthcare Mater., 2022, 11, e2200142.
27. C. Charoenviriyakul, Y. Takahashi, M. Nishikawa and Y. Takakura, Preservation of exosomes at room temperature using lyophilization, Int. J. Pharm., 2018, 553, 1–7.
28. Á. M. Lőrincz, C. I. Timár, K. A. Marosvári, D. S. Veres, L. Otrokocsi, Á. Kittel and E. Ligeti, Effect of storage on physical and functional properties of extracellular vesicles derived from neutrophilic granulocytes, J. Extracell. Vesicles, 2014, 3, 25465.
29. R. Maroto, Y. Zhao, M. Jamaluddin, V. L. Popov, H. Wang, M. Kalubowilage, Y. Zhang, J. Luisi, H. Sun, C. T. Culbertson, S. H. Bossmann, M. Motamedi and A. R. Brasier, Effects of storage temperature on airway exosome integrity for diagnostic and functional analyses, J. Extracell. Vesicles, 2017, 6, 1359478.
30. R. Nieuwland, J. M. Falcon-Perez, C. Soekmadji, E. Boilard, D. Carter and E. I. Buzas, Essentials of extracellular vesicles: posters on basic and clinical aspects of extracellular vesicles, J. Extracell. Vesicles, 2018, 7, 1548234.
31. T. Kinnaird, E. Stabile, M. S. Burnett, M. Shou, C. W. Lee, S. Barr, S. Fuchs and E. S. Epstein, Local delivery of marrow-derived stromal cells augments collateral perfusion through paracrine mechanisms, Circulation, 2004, 109, 1543–1549.
32. R. B. Rutherford, J. D. Baker, C. Ernst, K. W. Johnston, J. M. Porter, S. Ahn and D. N. Jones, Recommended standards for reports dealing with lower extremity ischemia: revised version, J. Vasc. Surg., 1997, 26, 517–538.
33. R. Hartmann, M. Weidenbach, M. Neubauer, A. Fery and W. J. Parak, Stiffness-dependent in vitro uptake and lysosomal acidification of colloidal particles, Angew. Chem., Int. Ed., 2015, 54, 1365–1368.
34. T. Z. Nazari-Shafti, S. Neuber, D. Garcia Duran, Z. Xu, E. Beltsios, M. Seifert, V. Falk and C. Stamm, Human mesenchymal stromal cells and derived extracellular vesicles: Translational strategies to increase their proangiogenic potential for the treatment of cardiovascular disease, Stem Cells Transl. Med., 2020, 9, 1558–1569.
35. G. Marsico, C. Jin, S. A. Abbah, E. M. Brauchle, D. Thomas, A. L. Rebelo, D. Orbanić, S. Chantepie, P. Contessotto, D. Papy-Garcia, C. Rodriguez-Cabello, M. Kilcoyne, K. Schenke-Layland, N. G. Karlsson, K. J. A. McCullagh and A. Pandit, Elastin-like hydrogel stimulates angiogenesis in a severe model of critical limb ischemia (CLI): An insight into the glyco-host response, Biomaterials, 2021, 269, 120641.
36. A. Skomudek, I. Gilowska, R. Jasiński and K. Rożek-Piechura, Analysis of the dynamics of venous blood flow in the context of lower limb temperature distribution and tissue composition in the elderly, Clin. Interv. Aging, 2017, 12, 1371–1378.

---