Exosome-coated oxygen nanobubble-laden hydrogel augments intracellular delivery of exosomes for enhanced wound healing

Poor wound healing after trauma and surgery affects millions and imposes substantial costs, with complications driven by dysregulated inflammation and persistent hypoxia that delay closure and promote hypertrophic/keloid scarring. Adipose-derived stem cell (ADSC)-exosomes carry bioactive proteins, lipids and nucleic acids and can reduce inflammation, inhibit apoptosis, promote angiogenesis, and enhance fibroblast migration and proliferation, but their intracellular cargo delivery is impaired under hypoxia due to hypoxia-induced endocytic recycling. The study addresses these barriers by creating an exosome-coated, oxygen nanobubble system embedded in a self-healing, tissue-adhesive polyvinyl alcohol (PVA)/gelatin (GA)/borax hydrogel to simultaneously supply oxygen, enhance exosome intracellular delivery under hypoxia, provide hemostasis, and scavenge reactive oxygen species (ROS). The hypothesis is that the multifunctional EBO-Gel will mitigate hypoxia, improve exosome cargo delivery, reduce inflammation, promote angiogenesis, and accelerate high-quality wound healing.

Prior studies have highlighted the promise of exosome-loaded hydrogels for acute and chronic wound healing, showing oxidative stress mitigation, angiogenesis stimulation, and improved fibroblast migration, while exosomes offer safety advantages over stem cells. However, hypoxia activates endocytic recycling that reduces intracellular cargo delivery of exosomes, limiting efficacy in poorly perfused wounds. Glycosylated protein conjugates (e.g., dextran–BSA) provide improved stability and self-assembly and can scavenge free radicals, enabling encapsulation of nanoscale oxygen bubbles to form oxygen nanobubbles by ultrasonic cavitation. Dynamic borate ester crosslinks in PVA/GA/borax hydrogels provide tissue adhesion, shape adaptability and self-healing, and can react with hydrogen peroxide to mitigate inflammation; gelatin contributes hemostasis by facilitating platelet activation. These advances motivate a multifunctional, oxygen-releasing, exosome-delivering, self-healing hydrogel dressing tailored for hypoxic, bleeding wounds.

Ethics: All animal procedures were approved by the University of Illinois IACUC (Protocol#: 23012). Materials included BSA, dextran sulfate (200 kDa), PVA, sodium tetraborate (borax), collagenase, dyes/probes/antibodies for imaging and flow cytometry, and various assay kits. Cell lines: Human dermal fibroblasts (HDF-a) and HUVECs were cultured per supplier instructions under 37°C, 5% CO2. Primary ADSCs were isolated from de-identified human adipose tissue. ADSC isolation and characterization: Tissues were collagenase-digested (0.075% type I, 30 min), filtered, and cultured; cells were phenotyped by flow cytometry (positive: CD90, CD105, CD44; negative: CD106, CD45, CD19). Exosome isolation and characterization: ADSCs (P3–P8) were cultured with exosome-depleted FBS; conditioned media was clarified, ultrafiltered (100 kDa), and ultracentrifuged (120,000 × g, 90 min, 4°C). Exosomes were resuspended in PBS and stored at −80°C. NTA quantified size/concentration; TEM imaged morphology; BCA assessed protein. ONB and EBO preparation: ONB were synthesized by mixing BSA (40 mg) and dextran sulfate (80 mg) in PBS (10 mL) overnight, then ultrasonication in ice (3 s on/off, 50% amplitude, 7 min) while continuously sparging oxygen. The solution was filtered (0.22 µm) and ultrafiltered (100 kDa). EBO were generated by mixing exosomes and ONB at 1:2 and ultrasonication (ice bath, 10 s on/off, 50% amplitude, 5 min, repeated 3×) to reassemble an exosomal membrane around ONB. Characterization: DLS and zeta potential (Litesizer 500), TEM/SEM, UV-Vis (browning A294/A420) and SDS-PAGE to confirm BSA–dextran conjugation, FTIR for functional groups, stability by TEM after storage (2 days at 37°C; 1 month at 4°C). Hydrogel synthesis: PVA and GA were dissolved in water (typical precursor: 10 wt% PVA, 2.5 wt% GA). For nanoparticle formulations, exosomes, ONB, or EBO were dispersed in PVA/GA, then crosslinked 1:1 with 2 wt% borax to form Blank-Gel, Exo-Gel, ONB-Gel, or EBO-Gel. EBO release from gels was measured up to 48 h. Rheology and physical tests: ARES-G2 rheometer with 25-mm parallel plates (1 mm gap, sandpaper) assessed LVE via frequency sweeps (γo=5%), yield behavior via strain sweeps (ω=1 rad/s), injectability via complex viscosity vs stress, and recovery via alternating small-large-small amplitude oscillations. Protorheology demonstrations included tilted vial, syringe extrusion, shape remodeling, macroscopic self-healing (cut/reassemble dyed specimens), and adhesion to diverse substrates, tissues, and skin. Degradation: In vitro degradation in PBS with lysozyme (1000 U/mL) at 37°C, with mass loss measured up to 3 days; in vivo degradation monitoring at wound sites. Oxygen and ROS assays: Dissolved oxygen release monitored up to 10 h and extended profiles up to 40 h; cellular hypoxia assessed by RDPP staining in HDF-a. H2O2 scavenging quantified over 240 min; cellular ROS/SOD by fluorescence imaging and flow cytometry (H2DCFDA). Exosome delivery enhancement: HDF-a were co-cultured with CFSE-labeled exosomes under hypoxia vs normoxia, with Lamp2 immunostaining to quantify colocalization and exosome trafficking; exosome recycling to medium quantified per experimental timeline. Biocompatibility and hemostasis: HDF-a viability by MTT across serial dilutions; hemolysis assay on rat erythrocytes after 4 h incubation; in vitro hemostasis by mixing gels with whole blood or hemocytes and inversion test; in vivo rat liver hemorrhage model to measure blood loss. In vitro wound healing functions: HDF-a proliferation (WST-1), BrdU incorporation, cell cycle analysis; migration assays (scratch under hypoxia with inserts; transwell migration quantification). Angiogenesis: HUVEC tube formation on Matrigel under hypoxia, quantifying branches and total length after 6 h. In vivo full-thickness wound model: Male Sprague-Dawley rats (250–300 g) received 8 mm diameter, 2 mm deep dorsal wounds. Groups: Tegaderm (no gel), Blank-Gel, Exo-Gel, ONB-Gel, EBO-Gel. Gels were freshly crosslinked and applied; dressings refreshed every 2 days. Wound areas photographed and quantified over 14 days. Histology (H&E, Masson’s trichrome) on Days 4 and 14; metrics included scar index (scar area/average dermal thickness), dermis/epidermis thickness, collagen volume fraction. Immunofluorescence on Day 14 for CD31 (angiogenesis), DHE (ROS), macrophage markers (CD86/F4/80 for M1, CD206/F4/80 for M2), and IL-6. Biosafety: H&E of major organs after 14 days. Statistics: Mean ± SD; two-tailed unpaired t-test for two-group comparisons; one-way ANOVA with Tukey’s multiple comparisons; two-way ANOVA with Dunnett’s multiple comparisons; significance at P<0.05.

• ADSC-derived exosomes were uniformly sized (TEM ~100 nm) with NTA mean diameter 125.2 nm and concentration 7.22 × 10^9 particles/mL.
• ONB and EBO formed stable core–shell nanostructures: ONB hydrodynamic diameter 122.50 nm, zeta −40.73 mV; EBO hydrodynamic diameter 192.02 nm, zeta −23.30 mV. TEM/SEM confirmed a bilayer core–shell morphology (≈150–200 nm). EBO retained structure after 2 days at 37°C and 1 month at 4°C.
• Ultrasonication-driven Maillard-type conjugation between BSA and dextran sulfate was verified by increased A294/A420 absorbance, SDS-PAGE band shift, and FTIR signatures, supporting formation of glycosylated protein shells for ONB.
• EBO internalized into HDF-a cytoplasm within 6 h (3D confocal Z-stacks).
• EBO-Gel exhibited rapid gelation with borate ester crosslinks, strong shear thinning (readily extrudable), plateau elastic modulus ~5 kPa, relaxation time ~1.5 s, and large zero-shear viscosity ~2 kPa·s; robust macroscopic and rheological self-healing after cyclic cutting/reassembly and strain recovery.
• Adhesion: EBO-Gel adhered strongly to plastics, glass, rubber, steel, various rat organs, and skin (rat, porcine, mouse, human), maintaining fixation under finger flexion (0–90°).
• Release/degradation: >80% of embedded EBO released within 48 h; in vitro gel mass loss ~80% by 3 days; complete in vivo degradation within ~3 days.
• Oxygenation: EBO-Gel and ONB-Gel elevated dissolved oxygen under hypoxia (sustained above control ≥40 h in extended tests) and reduced RDPP hypoxia indicator fluorescence in HDF-a, indicating cellular hypoxia mitigation.
• Antioxidant activity: Rapid H2O2 consumption within 30 min and continued decline to 240 min; reduced cellular ROS and increased SOD signals; H2DCFDA flow cytometry showed lowest ROS with EBO-Gel versus Blank-, ONB-, and Exo-Gels.
• Enhanced exosome delivery under hypoxia: Hypoxia reduced exosome–Lamp2 colocalization and increased recycling; oxygen supplied by EBO restored Lamp2 colocalization and reduced exosome recycling to the medium, indicating improved intracellular cargo delivery (n=3 biological replicates; one-/two-way ANOVA significance reported).
• Biocompatibility: High HDF-a viability across tested concentrations of Exo-, ONB-, and EBO-Gels (n=5); low hemolysis comparable to negative control (n=3).
• Hemostasis: EBO-Gel rapidly coagulated with whole blood and hemocytes in vitro and significantly reduced blood loss in a rat liver hemorrhage model, comparable to commercial hemostats and superior to non-hemostatic controls.
• Proliferation/migration: Under hypoxia, EBO-Gel increased BrdU incorporation and S/G2/M fractions; enhanced HDF-a scratch closure at 12–24 h and increased transwell migration (n=3).
• Angiogenesis: In HUVEC tube assays under hypoxia, EBO-Gel yielded the highest branch number and total branch length after 6 h (n=3).
• In vivo wound healing (rat full-thickness): EBO-Gel accelerated wound closure especially from day 2–10, avoided infection seen in Tegaderm group, and produced superior healing quality by day 14.
• Histology Day 14: EBO-Gel showed lowest scar index, thicker dermis, epidermal thickness closer to normal, and higher, more mature collagen volume fraction/organization versus Tegaderm and Blank-Gel.
• Immunostaining Day 14: EBO-Gel increased CD31 (angiogenesis), reduced ROS by DHE, shifted macrophages toward M2 phenotype (lower CD86/F4/80, higher CD206/F4/80), and showed lowest IL-6 expression; major organ histology revealed no systemic toxicity.

The multifunctional EBO-Gel addresses key barriers in wound healing by simultaneously mitigating hypoxia, enhancing intracellular exosome cargo delivery, scavenging ROS, and providing robust hemostasis and adhesion. Oxygen nanobubbles in an exosome-coated core–shell architecture supply dissolved oxygen to reverse hypoxia-induced endocytic recycling, thereby improving endolysosomal trafficking and intracellular release of exosomal protein/RNA cargo. The PVA/GA/borax hydrogel imparts dynamic, self-healing adhesion and shape adaptability for irregular, bleeding wounds while borate ester chemistry decomposes hydrogen peroxide, reducing oxidative stress and inflammation. In vitro, EBO-Gel improved fibroblast proliferation and migration and enhanced endothelial tube formation under hypoxia, consistent with exosomal pro-angiogenic cargo and oxygen-dependent matrix remodeling. In vivo, EBO-Gel accelerated wound closure and improved tissue quality, as shown by reduced scar index, normalized epidermal thickness, enhanced collagen maturation, increased microvessel density (CD31), reduced ROS, dampened IL-6, and favorable macrophage polarization (M2). Collectively, these results validate that concurrent oxygenation and targeted exosome delivery in a degradable, hemostatic hydrogel scaffolding can significantly elevate both the speed and quality of wound repair, addressing the central hypothesis and offering translational potential for acute and potentially chronic hypoxic wounds.

This study presents a tissue-adhesive, self-healing PVA/GA/borax hydrogel embedding exosome-coated oxygen nanobubbles (EBO-Gel) that integrates oxygen delivery, enhanced exosome cargo transfer under hypoxia, ROS scavenging, and hemostasis. EBO-Gel demonstrated strong injectability, adhesion, controlled nanoparticle release, rapid degradability, and robust biocompatibility. It improved fibroblast proliferation/migration, endothelial tube formation, and, in vivo, accelerated wound closure with superior healing quality characterized by reduced scarring, enhanced angiogenesis, reduced oxidative stress and inflammation, and favorable immune modulation. These findings highlight EBO-Gel as a promising multifunctional dressing for traumatic and surgical wounds and suggest broader applicability to hypoxic, ischemic, or chronic wounds (e.g., diabetic ulcers). Future work could optimize dosing and release kinetics, explore long-term scar modulation in hypertrophic/keloid models, expand to diabetic and other ischemic wound models, and investigate applications in other hypoxic pathologies.