Abstract Background Extracellular vesicle (EV)-based cell-free therapies have emerged as a powerful alternative to stem cell transplantation in regenerative medicine, owing to their ability to promote tissue repair while avoiding safety concerns associated with live-cell therapies. However, traditional two-dimensional (2D) cell cultures used for EV production are constrained by low exosome (Exo) yields and limited biological activity. Objective This study introduces a novel and scalable three-dimensional (3D) culture platform based on a hyaluronic acid (HA) and L-ornithine methyl ester (Orn) hydrogel to enhance the production and therapeutic efficacy of stem cell-derived exosomes. Methods The HA-Orn hydrogel was fabricated via a simple and mild crosslinking strategy, forming a biomimetic matrix that promotes spontaneous spheroid formation. Exosomes derived from 3D cultures (3D-Exo) were compared with those from 2D cultures (2D-Exo) in terms of yield, molecular composition, and biological functions. Results 3D-Exo exhibited significantly increased yield and superior functional properties, including enhanced stimulation of cell proliferation, migration, angiogenesis, and extracellular matrix remodeling. In vivo, 3D-Exo treatment accelerated wound closure and reduced inflammation in a mouse skin injury model, demonstrating robust therapeutic efficacy and safety. Mechanistic studies revealed distinct miRNA expression profiles and activation of regenerative signaling pathways in 3D-Exo. Conclusion This work presents a cost-effective, scalable, and bioinspired 3D culture system for high-yield and functionally enhanced Exo production. The HA-Orn hydrogel platform offers significant translational potential for advancing cell-free regenerative therapies, particularly in the context of wound healing. Supplementary Information The online version contains supplementary material available at 10.1186/s13287-025-04635-5. Keywords: Regenerative medicine, Extracellular vesicles (EVs), Three-dimensional (3D) cell culture, Hyaluronic acid (HA) hydrogel, Wound healing, Angiogenesis Background Regenerative medicine is revolutionizing approaches to treating tissue damage and organ failure. This interdisciplinary field integrates advanced technologies such as stem cell therapy, tissue engineering, gene editing, and extracellular vesicle-based interventions to efficaciously repair or replace damaged cells, tissues, and organs [[38]1]. Among these, stem cell therapy has demonstrated remarkable regenerative capabilities across various applications by harnessing the self-renewal and multidirectional differentiation potential of stem cells [[39]2]. However, the clinical translation of stem cell therapy is constrained by challenges such as complex procedures, high costs, potential cell heterogeneity, and the risk of tumorigenesis [[40]3]. In contrast, cell-free therapies, particularly those employing stem cell-derived exosomes, have emerged as a safer and more stable alternative. These extracellular vesicles (EVs) are nanosized structures that carry a diverse range of bioactive molecules, including growth factors, cytokines, mRNA, and miRNA, which collectively contribute to tissue repair and regeneration [[41]4–[42]6]. By retaining the therapeutic potential of stem cells while avoiding the risks associated with direct cell transplantation, EV-based therapies offer a promising avenue for advancing regenerative medicine. To maximize the potential of EVs in cell-free therapies, it is crucial to develop efficient, scalable production methods that preserve the biological activity of these vesicles. The conventional two-dimensional (2D) cell culture system is the most widely used method for EV production, wherein EVs are isolated via ultracentrifugation [[43]7].However, this approach yields low quantities of EVs and does not fully replicate the in vivo cellular microenvironment, leading to a loss of some functional properties. In efforts to enhance the yield and functionality of EVs, three-dimensional (3D) cell culture technology has shown significant promise [[44]8, [45]9]. By simulating the natural 3D growth environment of cells, this method more accurately reproduces their physiological state, optimizing cell-cell and cell-matrix interactions and enhancing the quality and biological activity of EVs [[46]10, [47]11]. Studies have demonstrated that EVs derived from 3D cultures outperform their 2D counterparts in promoting tissue repair and modulating immune responses [[48]12–[49]14]. Despite these advantages, the widespread clinical application of 3D culture technology is hindered by high operational costs, lack of standardization, and variability in culture conditions, which can lead to inconsistencies in EV quality [[50]15]. To address these limitations, this study introduces a cost-effective and scalable 3D cell culture system based on a hydrogel matrix constructed from hyaluronic acid (HA) and L-ornithine methyl ester dihydrochloride (Orn). HA, a major component of the extracellular matrix, supports cellular interactions and promotes 3D growth, while Orn, a critical nutrient, sustains cellular growth and physiological function [[51]16]. Using HA and Orn as substrates, a cross-linking strategy was developed to produce HA-Orn hydrogels under mild, aqueous conditions at room temperature (Scheme [52]1A). This approach is characterized by low cost, mild reaction, and industrial scalability. The resulting HA-Orn hydrogels exhibit an ideal microstructure, mechanical properties, biocompatibility and biodegradability, and its low-adhesion property facilitates the spontaneous formation of 3D cellular spheroids during culture and significantly enhancing EV yield (referred to as 3D-Exo). Compared to EVs derived from 2D cultures, 3D-Exo demonstrates superior biological activity in promoting skin cell proliferation, migration, angiogenesis, and extracellular matrix remodeling, highlighting its potential for wound repair applications. Furthermore, in vivo studies confirm that 3D-Exo outperforms 2D-Exo in promoting wound healing through mechanisms such as collagen remodeling, enhanced cell proliferation, and angiogenesis (Scheme [53]1C). This study underscores the unique advantages of the HA-Orn hydrogel-based 3D culture system in reducing the costs associated with 3D culture, increasing EV yields, and enhancing EV biological activity. This innovative approach provides a robust platform for advancing the therapeutic applications of EVs in regenerative medicine. Scheme 1. [54]Scheme 1 [55]Open in a new tab A Chemical structure of HA-Orn and the schematic representation of 3D stem cell exosome production using the HA-Orn hydrogel system; B Mouse model for assessing wound healing mediated by 3D-Exo. C In vivo wound healing effects of 3D-Exo Materials and methods The work has been reported in line with the ARRIVE guidelines 2.0. Materials Sodium hyaluronate and 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride were provided by Shanghai YuanYe Bio-Technology Co., Ltd. L-ornithine methyl ester dihydrochloride and deuterium oxide solvent were supplied by Shanghai Macklin Biochemical Co., Ltd. Hydrochloric acid and anhydrous ethanol were obtained from Sinopharm Chemical Reagent Co., Ltd. PBS was from Cyagen Biosciences Inc. Sodium hydroxide was provided by Shaanxi Panlong YiHai Pharmaceutical Co., Ltd. Potassium permanganate powder was from Shanghai Aladdin Biochemical Technology Co., Ltd. Hyaluronidase was supplied by Beijing Solarbio Science & Technology Co., Ltd. The Calcein-AM/PI staining kit was purchased from Wuhan Elabscience Biotechnology Co., Ltd. Cell culture Human Umbilical Cord hUC-MSCs were acquired from Hunan Shengbao Biotechnology Co., Ltd. hUC-MSCs were cultured in DMEM/F12 medium (Gibco, USA) supplemented with 10% fetal bovine serum (Pronase, China) and 1% penicillin-streptomycin (Solarbio, China) in a humidified incubator at 37 °C with 5% CO₂. Preparation of HA-Orn hydrogel Sodium hyaluronate (0.0108 g) was dissolved in 0.45 mL of sterile PBS and completely solubilized with magnetic stirring. Concurrently, an appropriate amount of L-ornithine methyl ester dihydrochloride was dissolved in 0.45 mL of sterile PBS with stirring. The two solutions were mixed in various ratios and adjusted to pH 6.5 using 0.1 M NaOH or HCl. Different proportions of 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride (DMTMM) were then added, and the mixture was stirred at room temperature for three days to form the HA-Orn hydrogel. The hydrogel was then transferred into a dialysis bag with a molecular weight cutoff of 14,000 Da and dialyzed against PBS (pH 7.4) with stirring to obtain purified hydrogel. The hydrogel matrix was uniformly mixed with a cross-linker solution containing a defined proportion of DMTMM to prepare a solution of desired concentration. After thorough mixing, the solution was immediately placed in a 37 °C water bath and periodically tilted to observe whether the solution flowed or gelled. Gelation was considered complete when the solution ceased to flow upon inverting the sample vial. The gelation state and time were observed and recorded, and each ratio was tested in triplicate. Characterization of HA-Orn hydrogel Hydrogel samples that had reached equilibrium swelling were cut into small pieces, lyophilized, and then dissolved in deuterated water (D₂O) to ensure a clear deuterium lock during NMR analysis. Proton nuclear magnetic resonance (H’-NMR) spectra were acquired using a 400 MHz spectrometer at room temperature. The microstructure of the freeze-dried hydrogels was examined using scanning electron microscopy. Samples were sputter-coated with a thin layer of gold to enhance electrical conductivity and imaged under high vacuum at an accelerating voltage of 10 kV. Images were taken at various magnifications to assess the pore size and distribution within the hydrogel. FTIR spectra of the lyophilized HA-Orn hydrogel samples were acquired over a range of 400 to 4000 cm^− 1 at a resolution of 1 cm^− 1. Cytotoxicity assessment of HA-Orn hydrogels To assess the cytotoxicity of chemically crosslinked HA-Orn hydrogels, both direct and indirect contact methods were employed. For the indirect contact method, the crosslinked hydrogels were sterilized and immersed in 1 mL of complete culture medium for 24 h to obtain hydrogel extracts. After co-culture with MSCs, the medium was supplemented with the Cell Counting Kit-8 reagent, and after a 3-hour incubation, the optical density (OD) at 450 nm was measured to assess cell viability. In the direct contact method, hydrogels were prepared and sterilized via UV irradiation followed by filtration through a 0.22 μm filter. The sterilized hydrogels were laid out at the bottom of a six-well plate and left to gel. Once gelled, DMEM/F12 complete medium was added to replace the medium for 24 h. After the medium replacement, MSCs at 200,000 cells/mL were co-cultured with the hydrogels. The cells adherent to the well plate and the spheroidal cells were observed using Calcein-AM/Propidium Iodide (PI) staining. Calcein-AM stains live cells green as it is converted by esterase activity in viable cells, while PI stains dead cells red by intercalating into the DNA of cells with compromised membrane integrity. Fluorescence microscopy was used to observe and record changes in cell fluorescence to assess the effects of the hydrogel materials on cell viability. Rheological characterization Rheological properties of hydrogel were assessed using a DHR-2 rotational rheometer (TA Instruments, New Castle, DE, USA). Hydrogel samples were equilibrated to swelling in PBS at 37 °C and tested at 25 °C. Oscillatory shear measurements were performed over a frequency range of 0.1 to 100 Hz at a fixed shear stress. The strain amplitude was set within the linear viscoelastic region to ensure accurate measurements of the storage (G’) and loss (G’’) moduli. Swelling ratio determination Prepared hydrogels were first freeze-dried to constant weight and weighed to obtain the dry gel mass (Ma). The hydrogels were then immersed in deionized water and allowed to swell at 37 °C in a temperature-controlled water bath. At designated time intervals, the hydrogels were removed, and surface water was gently blotted off before weighing to obtain the swollen mass (Mt). The equilibrium swelling degree (Swelling Ratio, SR) was calculated once the hydrogels reached swelling equilibrium (Me). Degradation rate determination The hydrogel samples were weighed to record the initial mass (M0). The samples were then incubated in PBS buffer (pH 7.4) containing 100 U/mL hyaluronidase at 37 °C in a water bath to facilitate enzymatic degradation. A control group was set up with hydrogels degrading in PBS without the enzyme. At predetermined time points, the hydrogels were removed, excess surface moisture was blotted off, and their mass (Mt) was measured. The percentage of degraded mass relative to the initial swollen mass was calculated to determine the degradation rate. Exosome isolation Exosomes were extracted from conditioned medium (CCM) obtained from MSCs at passages 3–7. Before collection, the MSCs were cultured under serum-free starvation conditions for 24 h. The CCM was first centrifuged at 1600 × g for 10 min at 4 °C to remove cellular debris. The supernatant was then centrifuged at 3600 × g for 20 min at 4 °C to eliminate large vesicles. Subsequently, the remaining supernatant was subjected to ultracentrifugation at 12,000 × g for 90 min twice to pellet the exosomes. The pellet was carefully resuspended in an appropriate volume of pre-cooled phosphate-buffered saline (PBS). To ensure purity, the exosome suspension was filtered through a 0.22 μm membrane filter prior to use. For long-term storage, exosomes were stored at − 80 °C. Characterization of exosome Transmission electron microscopy (TEM) was performed to characterize the morphology and structure of the exosomes. A 10 µL droplet of exosome suspension was pipetted onto a carbon-coated copper grid and allowed to adsorb for 5–10 min at room temperature. 2% uranyl acetate solution was applied to the grid for 1 min to negatively stain the exosomes. Excess UA was carefully blotted off with filter paper, and the grid was allowed to air-dry at room temperature. The grids were visualized using a TEM system operated at an accelerating voltage of 80–120 kV. DLS measurements were conducted using a Zetasizer Nano ZS (Malvern Instruments) and operating at a detection angle of 173° (backscatter mode). The measurements were performed at 25 °C, with an equilibration time of 2 min prior to data acquisition.A minimum of 3 technical replicates were performed for each sample to ensure consistency, and the results were expressed as the Z-average hydrodynamic diameter and PDI. Protein concentration of exosome was quantified using a BCA protein assay kit. Exosome protein samples were resolved on 10–12% SDS-PAGE gels, transferred to a PVDF membrane, and then blocked in 3% BSA (in TBST) for 1 h to reduce non-specific binding. The membrane was incubated overnight at 4 °C with primary antibodies specific to exosomal markers (Hsp70 and Alix). After washing, the membrane was incubated with HRP-conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using the Bio-Rad ChemiDoc MP Imaging System. Colony formation assay The colony formation assay was used to evaluate the clonogenic potential of HaCaT or HSF cells. Briefly, cells were dissociated into a single-cell suspension using trypsin-EDTA, and 700 cells per well were seeded into 6-well plates containing complete growth medium. The plates were incubated at 37 °C in a humidified atmosphere with 5% CO₂ for 10–14 days to allow colony formation. The culture medium was refreshed every three days, and colony growth was monitored under a phase-contrast microscope. At the end of the incubation period, colonies were fixed with 4% paraformaldehyde for 20 min at room temperature and stained with 0.1% crystal violet solution for 10–20 min. After staining, excess dye was washed off with PBS, and the plates were air-dried. Colonies containing at least 50 cells were counted as valid colonies. The colony formation rate was calculated using the formula: Colony Formation Rate (%) = (Number of Colonies Formed / Number of Cells Seeded) × 100. Representative images of the stained colonies were captured for documentation and further analysis. Scratch wound healing assay The scratch wound-healing assay was performed to assess the migratory ability of cells. A marker was used to draw reference lines on the underside of the wells of a 6-well plate to ensure consistent imaging locations. Cells were seeded at a high density into the wells and cultured until reaching approximately 90% confluence. A sterile pipette tip was used to create a linear scratch across the cell monolayer, and the wells were washed several times with PBS to remove detached cells and debris, ensuring clean scratch edges. Control and experimental treatments were then applied to the respective groups, and cells were incubated under standard culture conditions. Cell migration into the scratch area was monitored at predefined time points, and images of the wound area were captured using a phase-contrast microscope. Hydroxyproline assay Hydroxyproline (HYP) content was measured using a commercially available hydroxyproline assay kit following the manufacturer’s instructions. HSF cells were seeded into 6-well plates and cultured for 48 h under standard conditions. The culture supernatants were collected and centrifuged at 1200 × g for 15 min at 4 °C to remove cellular debris. The cleared supernatants were used as the test samples. Hydroxyproline levels in the samples were determined according to the kit’s protocol, including necessary dilutions, reagent additions, and incubation steps. The hydroxyproline concentration was calculated using a standard curve generated from known concentrations of hydroxyproline. Tube formation assay The tube formation assay was performed to evaluate the angiogenic potential of exosomes. On the day prior to the experiment, Matrigel (Corning) was thawed at 4 °C overnight, and all consumables, including pipette tips, were pre-cooled to ensure proper handling. The entire procedure was conducted on ice to maintain Matrigel integrity. The Matrigel was mixed gently using pre-cooled pipette tips to avoid air bubbles, and 50 µL of the gel was added to each well of a 24-well plate, ensuring an even distribution. The plates were then incubated at 37 °C for 1 h to allow the Matrigel to solidify. HUVEC cells were serum-starved overnight prior to the experiment. The cells were digested, resuspended, and prepared as a single-cell suspension. The cell suspension was mixed thoroughly and seeded onto the Matrigel-coated wells. Cells were treated with 50 µg of 2D-Exo or 50 µg of 3D-Exo in respective experimental groups, and the plates were incubated at 37 °C under standard culture conditions. After 4–12 h, tube formation was observed under an inverted microscope. Images of the tubular networks were captured for analysis. The total tube length was measured and quantified using ImageJ software as an indicator of angiogenic activity. In vivo wound healing evaluation All animal procedures were approved by the Experimental Animal Ethics Committee of Central South University (approval number: CSU-2024-0133) and were conducted in accordance with the National Act on the Use of Experimental Animals (People’s Republic of China). A total of 30 eight-week-old male wild-type C57BL/6 mice were obtained from Hunan Slike Jinda Experimental Animal Co., Ltd., and housed under specific pathogen-free (SPF) conditions at the Laboratory Animal Center, Central South University. To evaluate wound healing, hair was removed from the dorsal region of each mouse, and a circular full-thickness skin wound with a diameter of 1 cm was created under anesthesia using isoflurane administered via inhalation. The mice were randomly divided into three groups: PBS control group, 2D-Exo experimental group, and 3D-Exo experimental group. Exosome preparations were diluted to a final concentration of 50 µg/mL and uniformly mixed with 1% carbomer gel. A volume of 100 µL (corresponding to approximately 5 µg of total exosomal protein) was topically applied to each wound. Following application, the wounds were covered with gauze. Mice were housed individually with free access to food and water in a pathogen-free environment. The wound healing process was monitored on days 0, 3, 7, and 14. Images of the wounds were captured using a digital camera, and wound closure rates were calculated. At day 14, mice were sacrificed by cervical dislocation. Skin samples from the wound area were harvested and fixed in 4% paraformaldehyde for histological and immunohistochemical analyses. Histological evaluation included hematoxylin and eosin (H&E) staining and Masson’s trichrome staining to assess wound morphology and collagen deposition, respectively. For Immunofluorescence assay, at day 14, wound tissue samples were collected and fixed in 4% paraformaldehyde at 4 °C for 24 h. The fixed tissues were then embedded in paraffin and sectioned into 5 μm thick slices. Immunohistochemical staining was performed on these sections to evaluate the expression of proteins involved in wound healing, including collagen I, collagen III, Arg1, Ki67, CD31, VEGF, and α-SMA. Safety evaluation To further assess the biosafety of 3D-Exo, a full-thickness wound model was established in mice, followed by regular administration of the treatment. The mice then were euthanized, and blood samples were collected via retro-orbital bleeding. Serum biochemical analysis was performed to evaluate potential systemic toxicity, including measurements of alanine aminotransferase (ALT), aspartate aminotransferase (AST), urea (UREA), and creatinine (CREA) levels. In addition, major organs (heart, liver, spleen, lung, and kidney) were harvested, fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned for histological examination. The organ sections were stained with hematoxylin and eosin (H&E) and assessed under a light microscope for any pathological changes, including inflammation, fibrosis, or structural abnormalities, to evaluate the localized and systemic effects of the treatment. Statistical analysis Data from three independent experiments were expressed as mean ± standard deviation (mean ± SD). Statistical comparisons among groups were conducted using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test to determine group-wise differences. All analyses were performed using GraphPad Prism (version 10.0). Results and discussion Synthesis and Optimization of HA-Orn Hydrogel The synthesis of the HA-Orn hydrogel, depicted in Fig. [56]1A, utilized DMTMM (4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride), a triazine derivative, as a coupling agent. DMTMM activated the carboxyl groups of HA, thereby facilitating their reaction with the amino groups of L-Orn to form amide bonds. Notably, compared to conventional crosslinkers like divinyl sulfone and 1,4-butanediol diglycidyl ether, DMTMM offers superior reaction efficiency and lower toxicity under aqueous conditions [[57]17, [58]18]. The molar ratio of reactants (HA/DMTMM/L-Orn) was optimized based on the hydrogel’s gelling properties. Initially, the HA-to-L-Orn ratio was fixed at 1:1, while the impact of the DMTMM concentration on gel formation was assessed by varying its dosage. These assessments revealed favorable gel states with HA/DMTMM/L-Orn molar ratios of 1:1:1 and 1:2:1; however, an excessive DMTMM concentration (1:4:1) resulted in overly rigid hydrogels (Fig. [59]1B). To minimize the use of crosslinking agents, the HA-to-DMTMM ratio was set at 1:1, and the influence of varying L-Orn concentrations on gel formation was examined. Insufficient L-Orn (1:1:0.5) led to incomplete reactions, preventing hydrogel formation. Ultimately, we identified optimal hydrogel formation at HA/DMTMM/L-Orn molar ratios of 1:1:1, 1:1:2, and 1:2:1. Fig. 1. [60]Fig. 1 [61]Open in a new tab A Schematic of the chemical synthesis of the HA-Orn hydrogel. B Comparison of hydrogel morphology formed by HA/DMTMM/L-Orn at different molar ratios. C Biocompatibility assessment of HA/DMTMM/L-Orn hydrogels at varying molar ratios with hUC-MSCs. D Effects of HA/DMTMM/L-Orn hydrogels at different molar ratios on the formation of 3D spheroids of hUC-MSCs (scale bar = 100 μm) Subsequently, the biocompatibility of the synthesized hydrogels was evaluated by co-culturing human umbilical cord mesenchymal stem cells (hUC-MSCs) with hydrogel extracts for 48 h, followed by assessing cell viability via MTT assays. Among the three gel formulations with good gel-forming properties, HA/DMTMM/L-Orn (1:1:2) exhibited the highest biocompatibility, with a cell viability exceeding 80% (Fig. [62]1C). To further assess their suitability as cell culture substrates, a 1 mm-thick hydrogel coating was constructed on culture plates, and hUC-MSC suspensions were seeded onto the hydrogel for cultivation. After 24 h, the HA/DMTMM/L-Orn (1:1:2) group showed the most significant cell aggregation (Fig. [63]1D). After 3 days, uniform 3D spheroids of hUC-MSCs began forming exclusively in this group, while irregular elliptical structures were observed in others. By day 7, the differences became more pronounced: the HA/DMTMM/L-Orn (1:1:2) group maintained intact, uniform spheroid structures, whereas other groups exhibited peripheral cell debris, indicative of increased outer-layer cell mortality. Thus, HA/DMTMM/L-Orn (1:1:2) hydrogels provide a low-adhesion surface, promoting compact 3D cell aggregates through intercellular adhesion molecules (e.g., cadherins and integrins) rather than surface adhesion to culture plates. In contrast, other hydrogel formulations demonstrated poor biocompatibility and cytotoxic effects. Considering the collective data on gel morphology, cytocompatibility, and efficacy in supporting 3D cell aggregate formation, the HA/DMTMM/L-Orn hydrogel formulated at a 1:1:2 molar ratio was identified as the optimal candidate. Consequently, this formulation was selected for subsequent investigations. Characterization of HA-Orn hydrogel Scanning electron microscopy (SEM) revealed that the HA-Orn hydrogel possesses a loose, highly interconnected porous network (Fig. [64]2A), with pore sizes predominantly ranging from 50 to 300 μm. This structural feature facilitates effective diffusion of culture media and cellular metabolites. The porosity, determined via the ethanol displacement method, was measured to be 69.37 ± 3.94%, confirming its excellent water absorption capability. The chemical structure of HA-Orn was characterized using infrared (IR) spectroscopy (Fig. [65]2B). Both HA and HA-Orn exhibited stretching vibrations of -OH and -NH groups at 3397.43 cm⁻¹. The appearance of a new peak at 1610 cm⁻¹, corresponding to the stretching vibration of the C = O group in amide bonds, indicates successful amide bond formation between HA and Orn. Furthermore, the carbonyl peak at 1730 cm⁻¹ in HA-Orn confirms the covalent grafting of ornithine to HA. The crosslinking of HA and Orn was further validated by H¹-NMR analysis (Figure [66]S1). While HA showed characteristic sugar ring proton signals between 2.82 and 4.04 ppm, the HA-Orn polymer displayed new resonance peaks at 4.52 ppm (α-H), 1.92 ppm (β-H, 2 H), 2.95 ppm (δ-H, -CH₂-), and 1.70 ppm (γ-H, 2 H), attributed to the grafted ornithine structure. Fig. 2. [67]Fig. 2 [68]Open in a new tab A SEM image of the porous structure of HA-Orn hydrogel. B Infrared spectra of HA and HA-Orn hydrogels. C Storage modulus (G’), loss modulus (G’’), and tan δ of HA-Orn hydrogel during frequency sweep. D Frequency-dependent changes in complex viscosity (η*) of HA-Orn hydrogel. E Swelling behavior of HA-Orn hydrogel in PBS. F Degradation behavior of HA-Orn hydrogel in PBS buffer with or without HAase. G Formation and live/dead staining of 3D hUC-MSC spheroids cultured in HA-Orn hydrogel (Day 3, Day 5, Day 7), with live cells stained green (Calcein AM) and dead cells stained red (propidium iodide). Scale bars = 100 μm The mechanical properties of the HA-Orn hydrogel were evaluated through rheological testing. Dynamic frequency sweep tests indicated that across a frequency range of 0.1–100 rad/s, the storage modulus (G’) was consistently higher than the loss modulus (G’’), signifying the stability of the hydrogel network (Fig. [69]2C). Additionally, the low and stable tan δ values further underscored the hydrogel’s excellent elasticity. The composite viscosity (η*) followed a power-law decrease with increasing frequency (Fig. [70]2D), exhibiting typical shear-thinning behavior advantageous for adaptive deformation under dynamic physiological conditions. Swelling tests demonstrated that the hydrogel rapidly absorbed water within one hour of immersion in PBS and reached equilibrium swelling within 24 h, with an equilibrium swelling ratio of 1450 ± 85% (Fig. [71]2E). This high water retention capacity provides an ideal three-dimensional microenvironment for cell growth. The degradation behavior of the hydrogel was investigated under simulated physiological conditions using hyaluronidase (HAase, 100 U/mL) as the degradation agent. In the absence of HAase, the hydrogel retained over 90% of its mass after 96 h, demonstrating excellent stability (Fig. [72]2F). In the presence of the enzyme, the hydrogel degraded completely within 84 h, highlighting its controllable enzymatic degradation properties. Therefore, the HA-Orn hydrogel exhibited an ideal microstructure, desirable mechanical properties, moderate degradation behavior, and excellent biological characteristics, making it well-suited for 3D cell culture applications. The ability of the HA-Orn hydrogel to support 3D cell spheroid formation was evaluated using hUC-MSCs. After 3 days of culture, 3D spheroid formation was observed (Fig. [73]2G), and the size of these spheroids increased with prolonged culture time. The formation of these spheroids was likely driven by two primary factors: (1) the intrinsic self-assembly properties of stem cells, enabling aggregation in three-dimensional space, and (2) the low adhesion properties of the HA-Orn hydrogel, which reduced direct interactions between the cells and the matrix. This low adhesion environment promoted the formation of stable, multidimensional cellular structures mediated by intercellular adhesion molecules. Additionally, the abundant hydroxyl and carboxyl groups in HA endowed the hydrogel with high hydrophilicity, preventing protein adsorption and limiting direct cell-matrix adhesion [[74]19, [75]20]thereby inhibiting cell colonization. Calcein AM staining revealed strong green fluorescence in most cells within the spheroids, indicating good cell viability. Concurrently, PI staining showed no significant cell death, demonstrating the low cytotoxicity of the HA-Orn hydrogel-based 3D culture system. Extraction and characterization of 3D spherical stem cell exosomes using HA-Orn hydrogel The HA-Orn hydrogel was employed as a matrix for the 3D culture of hUC MSCs to isolate stem cell-derived exosomes (Fig. [76]3A). It is well established that the phenotype and functionality of cells are highly influenced by interactions with neighboring cells, proteins, and the extracellular matrix (ECM) [[77]21–[78]23]. Compared to traditional 2D culture systems, 3D cell culture better mimics the in vivo microenvironment, facilitating richer cell-cell and cell-ECM interactions. To investigate the differences between exosomes produced in 2D and 3D culture systems, we compared the exosomes secreted by hUC MSCs cultured in traditional 2D conditions with those derived from 3D spheroid stem cells cultured in the HA-Orn hydrogel. Under 2D culture conditions, hUC MSCs exhibited typical adherent growth and fibroblast-like morphology (Fig. [79]3B). In contrast, the 3D culture in the HA-Orn hydrogel resulted in uniform spherical cell aggregates. Fig. 3. [80]Fig. 3 [81]Open in a new tab A Workflow for isolating 3D spherical stem cell exosomes from HA-Orn hydrogel culture. B Morphological comparison of hUC MSCs cultured in 2D versus 3D HA-Orn hydrogel systems, Scar bar = 200 μm. C TEM images and (D) DLS particle size distribution of 2D-Exo and 3D-Exo. E Western blot analysis of extracellular vesicle marker proteins Alix and Hsp70. F Quantification of extracellular vesicle protein content using a BCA protein assay, comparing yields from 2D and 3D culture systems Exosomes were then isolated from the culture medium using ultracentrifugation, and their properties were characterized. Transmission electron microscopy (TEM) images and dynamic light scattering (DLS) analysis revealed no significant differences in microstructure or particle size between exosomes derived from 2D (2D-Exo) and 3D (3D-Exo) cultures (Fig. [82]3C, D). The successful isolation of exosomes was confirmed by the detection of the extracellular vesicle marker proteins Alix and Hsp70 using Western blot analysis (Fig. [83]3E). To quantify extracellular vesicle production, a bicinchoninic acid (BCA) protein assay was performed. The results demonstrated that 3D-Exo production was significantly higher than that of 2D-Exo (Fig. [84]3F), indicating that the 3D microenvironment provided by the HA-Orn hydrogel promotes the secretion of extracellular vesicles. Beyond the 2D system, our HA-Orn hydrogel platform also possesses unique advantages over conventional 3D culture systems. Various 3D culture strategies—such as hydrogel encapsulation, scaffold-free spheroid formation, and bioreactor systems—have been employed to enhance EV production, yet each presents limitations. Encapsulation-based methods often suffer from nutrient diffusion barriers and complicated EV recovery, while materials like Matrigel introduce variability and synthetic hydrogels may lack necessary bioactivity. Scaffold-free approaches facilitate cell aggregation and improved intercellular signaling, but often result in heterogeneous, hypoxic spheroids and lack extracellular matrix (ECM) cues. Bioreactor systems enable scalable production but involve mechanical stimuli whose impact on EV properties remains unclear. In contrast, our HA-Orn hydrogel platform offers a defined, bioactive, low-adhesion surface that induces spontaneous spheroid formation without full embedding, promoting physiological 3D structure, enhancing EV functionality, and simplifying recovery. Its cost-effective, scalable synthesis and favorable bioactivity make it a promising platform for efficient, standardized EV production in regenerative medicine. Characterization of the cellular biological properties of 3D-Exo Mesenchymal stem cell-derived exosomes have demonstrated significant potential in cell-free therapies, particularly in wound healing [[85]23–[86]25]. These EVs play critical roles in various stages of wound repair, including promoting angiogenesis, accelerating skin cell migration and proliferation, regulating extracellular matrix remodeling, and inhibiting cell apoptosis [[87]24, [88]26, [89]27]. Therefore, to assess the biological activity of 3D-Exo at the cellular level, wound healing was selected as the application model, with 2D-Exo serving as the control. The pro-proliferative effects of EVs were evaluated using human epidermal keratinocytes (HaCaT) and human skin fibroblasts (HSF). Both 3D-Exo and 2D-Exo significantly enhanced the proliferation of HaCaT and HSF cells in a time- and concentration-dependent manner (Fig. [90]4A-D). Notably, 3D-Exo exhibited a superior pro-proliferative effect compared to 2D-Exo, with the difference becoming more pronounced with increasing concentrations and prolonged incubation times. To further investigate the long-term effects of EVs on cell proliferation, clonogenic assays were performed using crystal violet staining. Following treatment with 3D-Exo, larger and more numerous colonies were observed (Fig. [91]4E). Quantitative analysis revealed that the colony formation rate of the 3D-Exo group was significantly higher than that of the 2D-Exo and control groups (Fig. [92]4F-I), underscoring the enhanced proliferative effects of 3D-Exo on skin cells. Fig. 4. [93]Fig. 4 [94]Open in a new tab A-D Effects of 2D-Exo and 3D-Exo on the viability of HaCaT and HSF cells at various time points and concentrations, measured via MTT assay. E Colony formation of HaCaT and HSF cells treated with 2D-Exo and 3D-Exo. F-I Quantitative analysis of crystal violet-stained colonies of HaCaT and HSF cells. J, K Scratch assay results for HaCaT and HSF cells after treatment with 2D-Exo and 3D-Exo; scale = 100 μm. L, M Quantitative analysis of scratch closure for HaCaT and HSF cells. N Angiogenesis of HUVECs after treatment with 2D-Exo and 3D-Exo; scale = 100 μm. O, P Quantitative comparison of tube formation length after treatment with 2D-Exo and 3D-Exo. Q Hydroxyproline levels in cells treated with 2D-Exo and 3D-Exo. Statistical significance is indicated as * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 The migratory capacity of cells treated with EVs was assessed using scratch assays. The wound closure rate of the 3D-Exo-treated group was markedly faster than that of the 2D-Exo and control groups (Fig. [95]4J, K). In particular, HSF cells treated with 3D-Exo achieved complete scratch closure within 48 h, whereas only minimal migration was observed in the 2D-Exo group at the same time point. Quantitative analysis further confirmed the superior migration-promoting effect of 3D-Exo (Fig. [96]4L, M). To evaluate the therapeutic potential of 3D-Exo in tissue repair, their effects on angiogenesis and collagen synthesis were studied. Using an in vitro human umbilical vein endothelial cell (HUVEC) tube formation model, all EV-treated groups exhibited tube formation after 4 h of culture on Matrigel. After 8 h, the 3D-Exo-treated group displayed significant advantages in tube length and network complexity compared to the 2D-Exo and control groups (Fig. [97]4N). Quantitative analysis of tube length confirmed the superior angiogenic effect of 3D-Exo (Fig. [98]4O, P). The impact of EVs on extracellular matrix remodeling was assessed by measuring hydroxyproline (HYP) levels, a characteristic amino acid in collagen and a key indicator of collagen synthesis [[99]28]. Both 2D-Exo and 3D-Exo significantly increased HYP content compared to the control group (Fig. [100]4Q). Notably, the 3D-Exo-treated group exhibited slightly higher HYP levels, suggesting enhanced collagen deposition. These findings demonstrate that 3D-Exo exhibits superior activity in promoting skin cell proliferation, migration, angiogenesis, and extracellular matrix remodeling compared to 2D-Exo, highlighting its enhanced therapeutic potential for wound repair applications. Investigation of the in vivo wound healing potential of 3D-Exo Following the evaluation of 3D-Exo biological activity at the cellular level, its efficacy in promoting wound healing was further assessed in vivo using an animal model. A circular full-thickness skin wound (1 cm in diameter) was created on the dorsal side of mice, and the wounds were treated locally with PBS, 2D-Exo, or 3D-Exo every three days. No adverse reactions were observed in any of the mice throughout the experiment, with all animals maintaining normal dietary and behavioral patterns. By day 7, the wound healing rate in both the 3D-Exo and 2D-Exo treatment groups was significantly faster than in the PBS group (Fig. [101]5A). By day 14, the 3D-Exo group achieved a wound closure rate of 97%, markedly surpassing the other groups (Fig. [102]5B). These results suggest that 3D-Exo has a stronger capacity to promote wound healing, consistent with findings from the cellular level experiments. Histological analysis was conducted on wound tissues collected after 14 days of treatment. Hematoxylin and eosin (H&E) and Masson staining revealed that the 3D-Exo group exhibited the narrowest scar width, which was significantly smaller than that of the 2D-Exo and PBS groups (Fig. [103]5C, D). Masson staining further indicated that the collagen fibers in the 3D-Exo group were denser and more regularly arranged, reflecting superior tissue reconstruction outcomes. Fig. 5. [104]Fig. 5 [105]Open in a new tab A Wound healing progression in a mouse full-thickness skin defect model under PBS, 2D-Exo, and 3D-Exo treatments, with images recorded from day 0 to day 14. B Wound area changes at different time points. C H&E and Masson-stained histological sections of wound tissues from different treatment groups on day 14, with (D) quantitative comparison of scar widths. E Dual immunofluorescence staining for collagen I and III, Scale bars represent 100 μm. F percentage of collagen deposition, and (G) quantitative analysis of collagen I and III expression. H Ki67 immunofluorescence for cell proliferation and (I) quantitative analysis of proliferation rates. J-K Dual immunofluorescence staining of CD31 and α-SMA for neovascularization and quantitative analysis of vessel density. L VEGF immunofluorescence labeling of VEGF expression and (M) quantitative analysis of VEGF levels. O Ang2 immunofluorescence staining and expression levels across treatment groups. Statistical significance is indicated as * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001 Immunohistochemical analysis was performed to explore the molecular mechanisms underlying the wound healing effects of 3D-Exo. The role of collagen remodeling was investigated by examining the expression of collagen types I and III. Compared with the 2D-Exo, the 3D-Exo group showed a significant increase in collagen III expression, with no significant changes in collagen I expression (Fig. [106]5E-G). Collagen III is synthesized during the early stages of wound healing and facilitates cell migration and granulation tissue formation, while collagen I predominantly appears during later stages to enhance scar strength [[107]29, [108]30]. The selective increase in collagen III expression suggests that 3D-Exo promotes early wound healing stages by fostering soft granulation tissue formation and enhancing elasticity and cell migration. Cell proliferation was assessed using Ki67 immunostaining. The percentage of Ki67-positive cells in the 3D-Exo group was significantly higher than in the 2D-Exo group, demonstrating the superior ability of 3D-Exo to promote cellular proliferation (Fig. [109]5H, I). Angiogenesis was evaluated via dual immunofluorescence staining for CD31 and α-smooth muscle actin (α-SMA). The 3D-Exo group showed significantly higher neovascularization density compared to the 2D-Exo and PBS groups (Fig. [110]5J, K). Additionally, the expression of vascular endothelial growth factor (VEGF) and angiopoietin-2 (Ang2) in the wound area was significantly upregulated in the 3D-Exo group (Fig. [111]5L-O). These findings suggest that the activation of the VEGF/Ang2 axis is a key mechanism through which 3D-Exo promotes angiogenesis. In summary, 3D-Exo exhibits superior wound healing activity compared to 2D-Exo. Its therapeutic efficacy is mediated by mechanisms including collagen remodeling, enhanced cell proliferation, and increased angiogenesis, highlighting its potential for advanced wound repair applications. Mechanistic insights into the regenerative and immunomodulatory potency of 3D-Exo To gain deeper insight into the mechanistic differences underlying the enhanced wound healing effects of 3D-Exo, we conducted a comparative analysis of miRNA profiles between 3D and 2D-Exo. miRNA sequencing identified 28 significantly downregulated and 8 significantly upregulated miRNAs in 3D-Exo relative to 2D-Exo (Fig. [112]6A). Among the upregulated miRNAs, seven remain unannotated in miRBase (e.g., X_29126, 19_12875, 19_12951), suggesting the potential discovery of novel regulatory molecules. Functional enrichment analysis using GO, KEGG, and Reactome databases indicated that these differentially expressed miRNAs are associated with key biological processes and signaling pathways relevant to tissue repair, including Wnt and MAPK pathways. Importantly, several miRNAs known to be upregulated under pathological conditions—such as hsa-miR-21-5p and hsa-miR-423-5p were downregulated in 3D-Exo, suggesting a reduced pathological signal and a culture environment that more closely mimics physiological conditions [[113]31, [114]32]. These results provide mechanistic support for the enhanced therapeutic efficacy of 3D-Exo and highlight the added value of 3D culture in generating functionally superior exosomes for regenerative applications. Fig. 6. [115]Fig. 6 [116]Open in a new tab Comparative miRNA profiling, pathway enrichment analysis, and in vivo immunomodulatory effects of 3D-Exo. A Volcano plot of differentially expressed miRNAs in 3D-Exo versus 2D-Exo. Red and green dots indicate significantly up- and downregulated miRNAs, respectively. B Gene Ontology (GO) enrichment analysis of biological processes associated with differentially expressed miRNAs. C KEGG pathway enrichment analysis of differentially expressed miRNAs. D Reactome pathway enrichment analysis of differentially expressed miRNAs. E Representative immunofluorescence images of wound tissue sections from mice treated with PBS, 2D-Exo, or 3D-Exo. Sections were stained for CD86, IL-6, or TNF-α. Nuclei were counterstained with DAPI. Scale bars = 100 μm To explore the molecular mechanisms underlying the enhanced therapeutic efficacy of 3D-Exo, we performed functional enrichment analyses based on the differentially expressed miRNAs. GO enrichment revealed significant enrichment in biological processes associated with transcriptional regulation, such as “regulation of transcription by RNA polymerase II” and “positive regulation of DNA-templated transcription” (Fig. [117]6B), indicating that 3D-Exo may enhance gene expression activity in recipient cells. Notably, enrichment in terms like “nervous system development” and “actin cytoskeleton organization” suggests a role in promoting neural regeneration and cell motility—two key elements in wound repair. The potential involvement in axonal guidance and neurotrophic signaling underscores the growing recognition that reinnervation at wound sites accelerates tissue healing through neuromodulatory effects on inflammation, angiogenesis, and matrix remodeling. KEGG analysis further identified significant enrichment in several key wound healing-related pathways (Fig. [118]6C). These include the MAPK and EGFR signaling pathways, which are critical for promoting cell proliferation, survival, and migration. Although the pathway term “EGFR tyrosine kinase inhibitor resistance” is typically referenced in oncology, its enrichment here more likely reflects enhanced EGFR activity that supports sustained pro-repair signaling. Enrichment of the AMPK signaling pathway suggests that 3D-Exo may also optimize the metabolic environment of wound sites, while activation of cell adhesion molecule pathways may facilitate cell–cell and cell–matrix interactions essential for tissue regeneration. A particularly striking observation was the high enrichment significance of the membrane trafficking pathway, a core process for exosome biogenesis and secretion. This result supports our hypothesis that the HA-Orn hydrogel-based 3D culture system not only increases vesicle output but may also promote a vesicle secretion profile that more closely mimics in vivo physiological conditions. Reactome pathway analysis reinforced these findings, showing strong enrichment in “signal transduction” and “gene expression (transcription)” pathways (Fig. [119]6D). Additionally, pathways such as “axon guidance” and the “RHO GTPase cycle”—important for cytoskeletal dynamics and cell migration—were significantly upregulated, consistent with enhanced fibroblast motility and neural repair capacity. Collectively, these results suggest that 3D-Exo elicit a multifaceted activation of key regenerative pathways in recipient cells. Through miRNA-mediated modulation of signaling cascades, including Wnt, MAPK, EGFR, AMPK, and neural guidance pathways, 3D-Exo demonstrate an enhanced potential to coordinate complex wound healing processes compared to 2D-derived exosomes. These insights further support the therapeutic advantage of the 3D hydrogel culture platform in producing functionally superior exosomes for regenerative medicine applications. To further evaluate the impact of 3D-Exo on inflammation and immune modulation during wound healing, we performed immunofluorescence staining on wound tissue sections from a murine skin defect model (Fig. [120]6E). The results revealed a significant reduction in the expression of CD86, IL-6, and TNF-α in the 3D-Exo treatment group compared to both the control and 2D-Exo groups. CD86 is a key costimulatory molecule expressed on antigen-presenting cells such as macrophages and dendritic cells, and its upregulation is associated with immune activation. IL-6 and TNF-α are classic pro-inflammatory cytokines, often implicated in sustaining chronic inflammation and impairing tissue regeneration. The observed downregulation of these markers suggests that 3D-Exo exert an immunomodulatory effect by suppressing excessive inflammatory responses and promoting a more balanced immune microenvironment. This anti-inflammatory profile may enhance wound repair by facilitating the resolution phase of inflammation and enabling tissue remodeling. These findings support the notion that the HA-Orn hydrogel-based 3D culture system enhances not only the regenerative but also the immunoregulatory properties of exosomes, contributing to their superior therapeutic performance in vivo. Safety evaluation A comprehensive in vivo safety assessment was conducted for the extracellular vesicles. Following treatment with 2D-Exo and 3D-Exo, blood biochemical indicators—including alanine transaminase (ALT), aspartate transaminase (AST), blood urea nitrogen (BUN), and creatinine (CREA)—did not show significant differences compared to the control group (Fig. [121]7A), indicating the absence of detectable liver or kidney toxicity. To further evaluate potential systemic toxicity, major organs, including the heart, liver, lungs, spleen, and kidneys, were harvested and subjected to histopathological analysis via hematoxylin and eosin (H&E) staining. Microscopic examination revealed that the cellular structures in all tissues remained intact, with no evident signs of necrosis or inflammation. Specifically, kidney tissue exhibited clear glomerular and tubular structures, liver tissue showed neatly arranged hepatocytes, lung tissue maintained intact alveolar structures, spleen tissue displayed distinct lymphoid architecture, and myocardial cells in the heart were arranged orderly. These findings confirm that 3D-Exo derived from HA-Orn hydrogel culture exhibits excellent biological safety, making it a viable candidate for local application in wound healing. Fig. 7. [122]Fig. 7 [123]Open in a new tab A Blood biochemical indices (ALT, AST, BUN, CREA) in mice after treatment with PBS, 2D-Exo, or 3D-Exo. B H&E-stained histological sections of major organs (heart, liver, lungs, spleen, kidneys) after different treatments Conclusion This study demonstrates the development and application of a novel HA-Orn hydrogel system for 3D cell culture and Exo production. By providing a biomimetic environment that mimics the natural extracellular matrix, the hydrogel promotes the formation of 3D cell spheroids and significantly enhances exosome yield and functionality. The resulting 3D-Exo exhibited superior biological activity in vitro, facilitating cell proliferation, migration, and angiogenesis. In vivo, 3D-Exo outperformed 2D-Exo in promoting wound healing, primarily through mechanisms such as collagen remodeling and activation of the VEGF/Ang2 pathway. The HA-Orn hydrogel system offers several advantages, including low cost, scalability, and biocompatibility, making it a practical and efficient solution for the large-scale production of exosomes. Its ability to enhance exosome functionality further positions it as a valuable tool in regenerative medicine, particularly for applications in wound repair and tissue regeneration. This study provides a foundation for future research into the clinical translation of 3D-Exo-based therapies, addressing key challenges in scalability, standardization, and therapeutic efficacy. Supplementary Information Below is the link to the electronic supplementary material. [124]Supplementary Material 1.^ (167.1KB, docx) [125]Supplementary Material 2.^ (15.2MB, xls) [126]Supplementary Material 3.^ (18.8KB, xls) [127]Supplementary Material 4.^ (5.8MB, xls) [128]Supplementary Material 5.^ (846.2KB, xls) [129]Supplementary Material 6.^ (2.1MB, xls) Acknowledgements