A nature-derived, strong adhesive hydrogel microsphere for wet tissue repair: Multifunctional and clinical potential

Abstract

In dynamic and fluid-rich environments, achieving stable bioscaffold adhesion to tissue surfaces while facilitating therapeutic delivery always presents significant challenges. Inspired by the nature adhesive mechanisms of skin section of Andrias davidianus (SSAD), we engineered SSAD@GelMA microspheres (SGM) as a novel multifunctional biomimetic platform, combining strong bioadhesion, cell/drug delivery capabilities, and extracellular matrix (ECM) remodeling. Preclinical evaluation in wound models demonstrated SGM exceptional adhesive strength, maintaining stable attachment even under continuous fluid shear stress. Unlike conventional hydrogel microspheres, the unique multifunctionality of SSAD enables SGM to serve dual roles as both an advanced biomimetic platform and an effective tissue-repairing agent for wet wound healing applications. Through efficient delivery of therapeutic agents or cells in wet environments and promotion of ECM remodeling, SGM significantly enhanced tissue repair across various in vivo wet tissue models, including bone tissue, cartilage tissue, soft tissue. These findings introduce a promising class of wet-adhesive microspheres with broad potential for adaptation to diverse fluid-rich tissue regeneration scenarios.

Highlights

• •Novel Hydrogel Microsphere Design: An Andrias davidianus-derived hydrogel microsphere (SGM) that incorporates SSAD.
• •Superior Functional Performance: The SGM exhibits excellent wet adhesion, efficient cell delivery, and high drug-loading capacity.
• •Broad Therapeutic Applicability: SGM demonstrates strong adaptability to in vivo wet dynamic environments, enhancing its clinical potential.

1. Introduction

Tissue defects and injuries are prevalent clinical challenges, often accompanied by inflammatory responses and decomposition processes [[1], [2], [3]]. The regeneration of complex tissue damage, whether caused by trauma or chronic inflammation, requires precise coordination of multiple biological processes, including immune modulation [4,5], cell recruitment [3,6], and extracellular matrix (ECM) remodeling [7,8]. Achieving this organised repair process remains a significant hurdle in regenerative medicine [9,10]. Currently, biomaterial-based strategies, particularly implantation and injectable systems, have emerged as promising approaches for in situ tissue regeneration by serving as platforms for therapeutic agents and stem cells delivery [11,12]. Among these, hydrogel microspheres have gained considerable attention in biomedical applications due to their unique advantages [[13], [14], [15]].

Hydrogel microspheres offer several critical benefits for tissue repair. First, they provide a three-dimensional microenvironment that enhances cell viability and facilitates efficient cellular delivery [[16], [17], [18], [19]]. Second, their tunable physicochemical properties enable the modulation of drug release kinetics, broadening the release profile [20]. These features make hydrogel microspheres a multifunctional tool for developing next-generation platform [21,22].

However, existing hydrogel microspheres systems face substantial limitations in clinical translation, particularly in maintaining interfacial stability under dynamic wet conditions [22]. A major challenge lies in their poor adhesion to hydrated tissue such as cartilage, oral mucosa, and exudative wounds, significantly restricting their therapeutic performance [[23], [24], [25], [26]]. For effective tissue repair, hydrogel microspheres play a dual roles: they must mechanically support the damaged area while ensuring sustained, site-specific delivery of drugs or cells, which demands precise localization and prolonged retention [27].

While localized delivery via microspheres can reduce systemic toxicity and enhance local drug retention [28,29], practical implementation is hindered by physiological barriers. Synovial fluid, saliva, and wound exudates create a flushing effect that displaces microspheres from target sites [[30], [31], [32]], while mechanical stress such as joint motion, mastication forces, and skin deformation, further compromise interfacial integrity [[33], [34], [35]]. Fundamentally, the hydrated molecular layer on wet tissue surfaces impedes direct bonding between the microspheres and the ECM [[36], [37], [38]], rendering conventional systems inadequate for withstanding these multifactorial stresses.

Nature offers innovative solutions to this challenge. Andrias davidianus (the Chinese giant salamander) secretes a unique viscous fluid that forms a protective, adhesive barrier over wounds, enabling efficient healing in aquatic environments [39]. Our prior work demonstrated that SSAD exhibit exceptional wet-adhesion properties [40], attribute to their rich composition of hydrophobic amino acid residues, phenolic hydroxyl groups, and benzene ring-containing compounds [41]. Notably, SSAD-derived adhesive proteins facilitate robust interfacial bonding through synergistic hydrogen bonding, van der Waals interactions, and π-π stacking, making them ideal for wet tissue applications such as hemostasis and wound closure [41]. Building on this, we previously developed SSAD-functionalized bioinks for 3D printing, enabling in situ adhesion to hydrated tissues when combined with gelatin methacryloyl (GelMA) [42]. Beyond adhesion, SSAD also exhibits bioactive functions, including immunomodulation, stem cell recruitment, and pro-migratory effects [[43], [44], [45]], suggesting its potential to enhance hydrogel microspheres with both structural and biological benefits.

Inspired by Andrias davidianus wound healing mechanisms [46], we designed a wet-adhesive hydrogel microspheres (SSAD@GelMA, SGM) for targeted drug/cell delivery and tissue repair. SGM is administered via injection, similar to adjunctive therapies in periodontitis treatment, enabling seamless integration into the existing treatment workflow. As illustrated in Scheme 1, we systematically characterized SGM's wet-adhesion performance, drug-loading capacity, and cellular delivery efficiency. To validate clinical relevance, we established multiple wet tissue defect models, including bone tissue, cartilage tissue, soft tissue. Our results demonstrate that SGM achieves stable adhesion across diverse hydrated tissue environments, offering a versatile platform for regenerative medicine.

Scheme 1.

Schematic illustration of injectable hydrogel microspheres derived from the skin secretions of Andrias davidianus exhibit unique wet-adhesion properties mediated by multiple interaction forces, enabling robust tissue adhesion in dynamic physiological environments. These bioadhesive microspheres serve as versatile platforms for localized delivery of therapeutic cells and pharmaceutical agents. Preclinical studies demonstrated their significant regenerative potential across multiple tissue repair models, including periodontitis-induced bone loss, articular cartilage defects, and chronic diabetic wounds.

2. Results

2.1. Preparation and characterization of SGM

In this study, we developed wet-adhesive microspheres based on SSAD. The fabrication process of SSAD hydrolysate and SGM is illustrated in Fig. 1a. To optimize microsphere composition, we first assessed the rheological properties of GelMA incorporating with varying ratios of SSAD hydrolysate. Viscosity measurements (Fig. S1) revealed a concentration-dependent increase, with 30 % and 50 % SSAD formulations exhibiting comparable profiles.

Fig. 1.

Fabrication and characterization of SGM. a) Schematic illustration of the preparation workflow for SGM. b) Microscopic image of spherical morphology of microspheres. c) SEM micrographs revealing the porous architecture of lyophilized microsphere. d) Quantitative size distribution analysis of microspheres. e) FTIR spectral analysis confirming successful conjugation while preserving characteristic functional groups of GelMA and SSAD. f) Surface charge characterization via zeta potential measurement. g) Proposed adhesion mechanism involving synergistic interfacial interactions (hydrogen bonding, hydrophobic effects, van der Waals forces and cation-π interactions). h) Comparative drug release kinetics showing sustained delivery profiles of microspheres. i) The adhesion of GM and SGM to agarose and subsequent drug release. Fluorescence microscope demonstrating the tissue permeability of drug-loaded microspheres.

An in vitro tissue adhesion assay further demonstrated that SGM containing 30 % or 50 % SSAD displayed robust wet-adhesive properties (Fig. S2). Based on morphological integrity and adhesive performance, the 30 % SSAD formulation (SGM) was selected for subsequent experiments. Microscopic analysis confirmed that both SGM and pure GelMA microspheres (GM) maintained uniform 3D architecture (Fig. 1b). Scanning electron microscopy (SEM) revealed a highly porous internal structures in both SGM and GM (Fig. 1c), facilitating cell adhesion and controlled drug release. For the freeze-dried microspheres, SGM exhibited a slightly larger diameter (218.24 ± 14.26 μm vs. 211.92 ± 18.50 μm for GM, Fig. S3). After water absorption, both GM and SGM showed excellent water absorption and expansion effects. As time increases, the diameters of SGM and GM gradually increased, reaching a plateau after 48 h. At day 5, the swelling rates of SGM and GM are 166.33 ± 5.25 % and 159.46 ± 6.03 % (Fig. S4), respectively, with SGM exhibiting a slightly larger diameter (334.29 ± 27.95 μm vs. 319.71 ± 39.96 μm for GM, Fig. 1d), attribute to SSAD.

Fourier transform infrared (FTIR) confirmed the preservation of key functional groups in SSAD@GelMA post-crosslinking, while peak shifts indicated enhanced crosslinking led to modifications in intermolecular interactions, including hydrogen bonds, contributing to improved material stability (Fig. 1e). Zeta potential measurements further supported successful SSAD integration, with SGM exhibiting a net positive charge (1.98 ± 0.05 mV), intermediate between GelMA (−2.69 ± 0.33 mV) and SSAD (16.16 ± 2.30 mV) (Fig. 1f).

According to our previous research, SSAD contains many amino acids including tyrosine and phenylalanine. And the phenolic hydroxyl group, amino acids, and benzene ring of SSAD were considered to be the functional groups showing the adhesive effect [41]. Thus, SGM adhesion mechanistically is mediated by multiple interfacial including hydrogen bonding, van der Waals interactions, π-π stacking, cation-π effects, and hydrophobic interactions (Fig. 1g) [40,47]. Tris (2-carboxyethyl) phosphine (TCEP)-assisted SSAD hydrolysis exposes additional hydrophobic amino acid residues, enhancing cohesive strength and wet-tissue adhesion through interfacial water displacement [48,49].

To evaluate drug delivery performance, we compared the release kinetics of GM and SGM. Both formulations exhibited an initial burst release (3 days), followed by sustained release over 13 days, ensuring prolonged therapeutic availability for tissue repair (Fig. 1h). Fluorescence imaging confirmed rapid drug penetration into target tissues (Fig. 1i), validating SGM's efficacy as a drug carrier.

2.2. Biocompatibility assessment of the SGM

To establish SGM as a clinically viable platform for cell delivery and tissue regeneration, we systematically evaluated its biocompatibility and cell-supporting functions. Fig. 2a illustrates the successful encapsulation of cells within the microspheres. CCK-8 assays confirmed excellent biocompatibility for both SGM and GM, with no significant cytotoxicity observed (Fig. S5).

Fig. 2.

Functional characterization of cell-laden microspheres. a) Schematic illustration of cell-laden microspheres fabrication. b) Longitudinal viability assessment by Live/dead cell staining (green: live cells; red: dead cells) at day 0, 3 and 7 post-encapsulation. c) Live/Dead cells percentage (%). d) F-actin Cytoskeletal staining (red) and nuclear morphology (blue) with corresponding semi-quantitative analysis. The cytoskeleton was stained in red and the nuclei in blue. e) Mean fluorescence intensity of cytoskeletal staining. f) Time-dependent migratory capacity of encapsulated cells evaluated at day 3, 7 and 14. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Long-term cell viability was validated via Live/Dead cell staining, which revealed >90 % cell survival over 7 days of culture (Fig. 2b and c). We attribute this sustained viability to the microsphere's porous 3D architecture, which facilitates efficient nutrient/waste exchange while maintaining a protective microenvironment. Notably, cytoskeletal staining demonstrated significantly enhanced cell spreading on SGM compared to GM, with a 3.63-fold increase in mean fluorescence intensity (Fig. 2d and e). This aligns with our prior findings that SSAD-derived bioactive peptides (e.g., RGD motifs) actively promote integrin-mediated cell adhesion.

To evaluate the in vivo biological behavior of SGM-mediated cell delivery, we simulated the post-transplantation microenvironment using an in vitro well-plate model. Live cell staining revealed robust migration and sustained proliferation of cells encapsulated in both SGM and GM microspheres (Fig. 2f). This suggests that the cells delivered by SGM can not only maintain cell survival for a long period of time, but also migrate to the tissues and repair the tissue damage after being delivered into the body. The above experimental results confirmed that SGM has good biocompatibility, can be used as a cell delivery bioscaffold, and the introduction of SSAD can promote cell spreading.

2.3. SGM promotes extracellular matrix secretion

Beyond demonstrating good biocompatibility, we investigated SGM's functional modulation of cellular behavior through transcriptomic sequencing (Fig. 3a and b). RNA-seq analysis of SGM-delivered cells revealed three dominant Gene Ontology (GO) clusters: extracellular matrix, synapse and nerves, and stimulus detection.

Fig. 3.

Transcriptomics and functional validation of SGM-mediated ECM remodeling. a) Experimental workflow (Created in part with BioRender.com). b) The GO enriched network. c) GSEA of ECM-related genes. d) Heatmap of differentially expressed ECM-related genes. e) Alcian blue-stained proteoglycans. f) RT-qPCR of chondrogenic ECM genes. g) Mechanistic schematic of SSAD regulate ECM remodeling (Created with BioRender.com). *P < 0.05, **P < 0.01, ***P < 0.001.

Notably, ECM-related terms are the most dominant enriched pathways (Fig. S6a). Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis further implicated cytoskeletal reorganization, MAPK signaling pathway and Protein digestion and absorption (Fig. S6b). Gene Set Enrichment Analysis (GSEA) analysis confirmed that SGM significantly upregulated the expression of genes associated with the ECM and collagen-containing extracellular matrix, including Epyc, Mmp17, Matn2, Col11a2, and Col10a1 (Fig. 3d, Fig. S6c). These findings suggest that SGM may influence cellular biological functions through extracellular matrix remodeling or collagen synthesis, thereby promoting wound healing and enhancing cell adhesion (Fig. 3c and d). To further validate the effects on ECM secretion, we performed Alcian blue staining. The results demonstrated that after chondrocyte co-culturing with SGM and GM for 7 and 14 days, the staining intensity in the SGM group was greater than that in the GM group and the control group, indicating a stronger promoting effect of SGM on ECM secretion (Fig. 3e). To further confirm this promoting effect, we conducted quantitative real-time PCR (qPCR) to analyze the expression levels of key genes involved in ECM synthesis. The qPCR results revealed that the SGM group exhibited higher expression levels of Col II, Acan and Sox9 compared to the GM group (P < 0.05), suggesting that SGM enhanced the secretion of ECM components (Fig. 3f). In summary, SSAD regulates multiple genes associated with the extracellular matrix. Its natural constituents can mimic the components of the extracellular matrix, promote collagen cross-linking and enhance cell migration (Fig. 3g).

2.4. Wet adhesiveness of SGM in vitro

To evaluate SGM's applicability for periodontitis treatment, constant flow irrigation mimicked saliva or gingival crevicular fluid, following SGM and GM application to the exposed root cementum and alveolar bone surface (Fig. 4a). And the clinical periodontitis images from patients at the Affiliated Stomatological Hospital of Chongqing Medical University (Fig. 4b) guided simulated defect creation on the pig's alveolar bone. SGM showed significantly superior retention on periodontal tissue compared to GM, which was predominantly displaced into the collection vessel (Fig. 4c and d; Fig. S7 and Movie S1, Movie S2).

Fig. 4.

Wet adhesiveness of SGM in simulated physiological environments. a) Schematic of hydrodynamic challenge simulating saliva or gingival crevicular fluid over microspheres on alveolar bone (Created with BioRender.com). b) Clinical periodontitis image (human) and reconstructed periodontal defect in the pig's alveolar bone. c) SGM retention on the pig's cartilage under constant-flow irrigation. d) Quantitative analysis of microsphere retention. e) Schematic of hydrodynamic challenge simulating synovial fluid flow over microspheres on cartilage (Created with BioRender.com). f) Clinical osteoarthritis image (human) and reconstructed equivalent osteochondral defect in the pig's knee joint. g) SGM retention on the pig's cartilage under constant-flow irrigation. h) Quantitative analysis of microsphere retention. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Supplementary data related to this article can be found online at https://doi.org/10.1016/j.bioactmat.2026.01.033

The following are the Supplementary data related to this article:

Similarly, to evaluate SGM's applicability for articular cartilage defect repair, SGM and GM were locally injected into defect site and subjected to constant flow irrigation simulating synovial fluid (Fig. 4e). The clinical osteoarthritis images from the First Hospital of Chongqing Medical University informed the reconstruction of equivalent osteochondral defects on isolated pigs' knee joints (Fig. 4f). SGM demonstrated robust adhesion, resisting hydrodynamic displacement, while GM lacking adhesive properties was completely dislodged (Fig. 4g and h; Movie S3, Movie S4). To assess adhesion under dynamic wet conditions, a articular cartilage defect model with SGM were agitated in water to simulate articulation. SGM maintained stable cartilage adhesion, whereas GM exhibited minimal retention (Movie S5, Movie S6).

Supplementary data related to this article can be found online at https://doi.org/10.1016/j.bioactmat.2026.01.033

The following are the Supplementary data related to this article:

To establish a native skin lesion model, clinical diabetic foot images obtained from the First Hospital of Chongqing Medical University were used as reference for reconstruction (Fig. S8a). Subsequently, SGM and GM were applied onto the simulated skin wound, which was then subjected to continuous fluid irrigation to mimic dynamic wound conditions (Fig. S8b). Following these mechanical and fluid challenges, the SGM remained strongly adhered to the tissue surface even after flushing (Fig. S8c and d; Movie S7, Movie S8).

Supplementary data related to this article can be found online at https://doi.org/10.1016/j.bioactmat.2026.01.033

The following are the Supplementary data related to this article:

In addition, we further conducted quantitative tests of the adhesion strength of SGM and GM (Fig. S9). The results showed that the adhesion strength of SGM on the pig's skin reached 18.5 ± 0.7 kPa, significantly higher than the adhesion strength of GM (8.6 ± 1.67 kPa). This indicated that the addition of SSAD significantly enhanced the adhesion of the microspheres.

2.5. In vivo demonstration of SGM mediated functional repair in wet tissue

2.5.1. In vivo efficacy of SGM for periodontal tissue regeneration

Based on superior sustained-release and penetration properties of drug-loaded SGM observed in vitro, an in vivo periodontitis model was established using ligature induction combined with Porphyromonas gingivalis (P.g) injection. This model enabled evaluation of the repair and regenerative potential of drug-loaded SGM under dynamic loading and wet conditions (Fig. 5a). Micro-CT analysis revealed significant reductions in alveolar bone height and the bone volume in the ligature-induced periodontitis model (Fig. 5b and c). Compared to direct minocycline (Mino) injection, both GM-Mino and SGM-Mino treatments effectively reduced alveolar bone loss, with SGM-Mino demonstrating superior efficacy. Notably, the SGM-Mino group exhibited the most significant therapeutic effects, as evidenced by an ABC-CEJ distance nearly restored to normal levels (Fig. 5b and c). Furthermore, the most favorable improvements in bone tissue volume fraction, trabecular thickness, and trabecular spacing were observed in the SGM-Mino group, indicating enhanced recovery of alveolar bone density and mass (Fig. 5b and c).

Fig. 5.

Therapeutic efficacy of SGM in a rat model of periodontitis. a) Experiment procedure (Created in part with BioRender.com). b) Representative images of micro-CT reconstruction. c) Micro-CT quantification of bone parameters: BV/TV, ABC-CEJ distance, Tb.Th and Tb.Sp. Radar plot summarizes overall group performance. d) Histological and immunofluorescence analysis (AB: Alveolar bone; R: root). e) Quantitative of ABC-CEJ distance (H&E) and osteoclast fraction (TRAP⁺ cells). Semi-quantitative of OCN and ALP positive area (%). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Histological and immunofluorescence analyses were subsequently performed to evaluate periodontal bone resorption and new bone formation. Hematoxylin-eosin (H&E) staining demonstrated significant alveolar bone height reduction in the periodontitis group (Fig. 5d and e). All treatment groups (Mino, GM-Mino, SGM-Mino) exhibited bone height restoration, with the SGM-Mino group demonstrating the most significant recovery (Fig. 5d and e). TRAP staining indicated a higher number of osteoclasts in the control and Mino groups compared to SGM-Mino and GM-Mino groups (P < 0.05, Fig. 5d and e). Moreover, SGM-Mino treatment resulted in a near absence of osteoclasts and significantly inhibited osteoclast activity relative to the GM-Mino treatment (P < 0.05, Fig. 5d and e).

Osteogenic activation was assessed via quantitative immunofluorescence analysis of osteocalcin (OCN) and alkaline phosphatase (ALP) expression (Fig. 5d). The expression of OCN and ALP was significantly higher in the GM-Mino group than in the blank and Mino groups (P < 0.001, Fig. 5e). The SGM-Mino group exhibited a pronounced upregulation of both ALP and OCN expression compared to the GM-Mino group (P < 0.001, Fig. 5e). These findings indicate that SGM, leveraging its high drug-loading capacity and robust wet-adhesion properties, attenuates periodontal inflammation and enhances alveolar bone regeneration through the localized delivery of minocycline to inhibit P.g infiltration.

2.5.2. In vivo efficacy of SGM for cartilage tissue regeneration

Building upon SGM's demonstrated efficacy in vitro, we evaluated its regenerative capacity in a challenging in vivo osteochondral defect model under physiological challenges (Fig. 6a). At 6 weeks post-operation, digital photography revealed persistent cartilage defect in the blank group, exhibiting defined margins and surface irregularity (Fig. 6b). The GM-chondrocyte construct (GC) group showed limited neotissue formation and poor integration at the defect-native tissue interface (Fig. 6b). In contrast, defects treated with SGM-chondrocyte constract (SGC) group were nearly completely replaced by cartilaginous tissue with a smooth surface and seamless integration into adjacent native cartilage (Fig. 6b). Quantitative assessment using the International Cartilage Repair Society (ICRS) scoring system confirmed these observations: the SGC group achieved significantly higher scores than other groups, demonstrating superior repair efficacy (Fig. 6c).

Fig. 6.

In vivo therapeutic efficacy of SGM in rat cartilage defect model. a) Animal experiment procedure (Created in part with BioRender.com). b) Digital photography of cartilage defect in the different groups at 6 weeks post-operation. c) ICRS score. d) H&E staining, Safranin O-fast green staining, and Col II immunohistochemistry staining in different groups (A: Articular cartilage; SC: Subchondral bone). e) Mankin scores in different groups. f) Upper: 3D gait pressure distributions (X-axis: the length of the footprint; Y axis: the width of the footprint; Z-axis: the maximum contact intensity of the footprint). Down: Quantitation of peak contact at 6 weeks post-operation. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

H&E and Safranin O-fast green staining revealed extensive cartilage defects with structural disruption in the blank group. The GC group exhibited disorganized neotissue morphology characterized by fibrous tissuse deposition. Conversely, SGC-treated defects displayed a smooth articular surface, well-organized layered structure, and chondrocyte within defined lacunae resembling hyaline cartilage (Fig. 6d). Quantitative assessment via modified Mankin scoring demonstrated significantly superior structural integrity in the SGC group versus blank control at 6 weeks (P < 0.01, Fig. 6e). Immunohistochemical confirmed robust collagen type II (Col II) expression at SGC repair sites, comparable to native cartilage. This indicates SGC's wet-adhesive functionality and pro-secretory microenvironment facilitated rapid hyaline chondrogenesis of delivered stem cells in vivo. Minimal Col II immunoreactivity was observed in blank and GC groups (Fig. 6d).

Functional recovery was assessed via gait analysis at 6 weeks post-treatment, comparing with blank, SGC, GC and normal groups (Fig. 6f). Closer resemblance to normal gait patterns indicated superior functional restoration. At 6 week, the lowest max contact area (0.84 ± 0.42 cm²) appeared in the SGC group, which was smaller than that in the GC (1.17 ± 0.38 cm²) and control group (1.46 ± 0.2 cm², P < 0.05) (Fig. 6f, Fig. S10).

2.5.3. Wound healing efficacy of SGM in an in vivo diabetic wound model

Building on our prior demonstration of microsphere efficacy in repairing cartilage and bone defects under wet, dynamic loading conditions, we extended this evaluation to a diabetic skin wound model to assess broader applicability. Diabetic skin wounds are characterized by substantial exudate and mechanical stress from movement, creating a complex pathological environment that challenges effective drug delivery by compromising retention and residence time. To evaluate microsphere adaptability in complex scenarios, we established a diabetic skin defect model and harvested tissue at 7 and 14 days post-wounding for histological and molecular analyses (Fig. 7a). Wound closure was monitored photographically over time (Fig. 7b). Macroscopically, all treatment groups (DFO, GM-DFO, and SGM-DFO) exhibited significantly accelerated wound closure rates compared to the control group (Fig. 7b). By day 4, SGM-DFO treatment resulted in significantly smaller wound areas compared to both control and DFO groups (Fig. 7c). At day 7 and 14, SGM-DFO treated wounds were smallest among all groups, demonstrating superior efficacy (Fig. 7c). Histological (H&E, Masson's trichrome) and immunofluorescence analyses assessed collagen deposition and angiogenesis. At day 7, both GM-DFO and SGM-DFO groups exhibited significantly shorter wound lengths and enhanced angiogenesis relative to control and DFO groups. Notably, SGM-DFO surpassed GM-DFO in both reduced wound length and angiogenic response, with residual microspheres observable at the wound periphery (Fig. 7d, e and f). By day 14, H&E and Masson staining revealed that SGM-DFO treated wounds displayed the narrowest width, characterized by tightly organized epithelium, dense and continuous collagen deposition, and optimal tissue integrity. Wound widths in GM-DFO and DFO were larger, ranking second and third smallest, respectively (Fig. 7g and h). These results demonstrate that SGM promotes diabetic skin wound repair through robust adhesion to the dynamic, exudative wound enabling sustained DFO release, which ameliorates microcirculatory impairment and enhances tissue regeneration.

Fig. 7.

Therapeutic efficacy of SGM in a diabetic skin wound model. a) Experiment procedure (Created in part with BioRender.com). b) Digital photography of wound healing progression in different groups. c) Quantification of relative wound area over 14 days. (*; compared with the control group. ^#; compared with the DFO group. ^&; compared with the GM-DFO group.). d) H&E staining of different groups on day 7 post-wounding. e) IF staining for endothelial marker CD31 (red) and smooth muscle marker α-SMA (green) in different groups. f) Quantification of wound length and number of blood vessel. g) H&E and Masson staining of different groups on day 14. h) Radar plot comparing group performance based on wound area and microvessel density. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

2.5.4. The biological histocompatibility of SSAD hydrolysate

Histopathological evaluation of H&E-stained tissue sections from the heart, liver, spleen, lung, and kidney revealed no abnormal morphology in microsphere-treated animals, confirming in vivo biocompatibility (Fig. S11). In our study considered that SSAD might enter the gastrointestinal tract via the oral route, thus we evaluated its median lethal dose and toxic effects through oral gavage in mice. The results indicated that the highest concentration of SSAD does not cause mortality in mice (Table S2), and there is no liver or kidney toxicity, with no abnormalities in blood routine tests (Fig. S12). And in our study, the concentration of SSAD was 30 mg/mL. These findings suggested that SSAD has good biocompatibility and does not produce toxic side effects on the body.

3. Discussion

Recent advances in wet-adhesion microsphere design show promise for tissue repair [50,51], yet critical limitations persist in interfacial stability, functional versatility. This study proposed a novel wet adhesion microsphere delivery platform derived from Andrias davidianus that overcomes these constraints through four transformative innovations. First, unlike synthetic polymer-based approaches, SGM leverages Andrias davidianus derived proteins to establish robust wet adhesion via hydrogen bonding, hydrophobic forces, and electrostatic forces at the tissue-microsphere interface. This multi-mechanistic adhension ensure sustained localization of therapeutic payloads. Second, while hydrogel microspheres broadly demonstrate biocompatibility, their efficacy as an integrated delivery platform remains poorly characterized. Here we provide direct experimental evidence that multifunctional SGM as a 3D culture matrix and a controlled-release platform. Third, the SSAD modification strategy, previously unreported in adhesion scaffolds, confers not only enhanced wet tissue adhesion but also promoted extracellular matrix secretion. Finally, clinical relevance is demonstrated across physiologically distinct wet tissue defect model. Collectively, this work establishes a novel therapeutic strategy for wet tissue defect repair by combining natural Andrias davidianus derived bioproteins with a microsphere delivery system [52].

The dynamic, wet microenvironment characteristic of chronic wounds presents a dual challenge to the adhesion stability and functional delivery of biomaterials [52,53]. Here, a bioinspired wet-adhesive hydrogel microsphere was developed to achieve stable adhesion and functional delivery at wet tissue interfaces through the integration of multi-mechanism adhesion coordination and microsphere structural design. The enhanced adhesion performance of SGM was primarily attributed to the incorporation of SSAD. The glycoprotein component within SSAD displaces interfacial water molecules through dynamic hydrogen bond recombination, thereby mitigating the hydration layer interference [49]. This dynamic bonding capability thereby ensures stable adhesion wet environments. Additionally, aromatic amino acids like tryptophan in SSAD anchor to hydrophobic motifs, such as collagen hydroxyproline, via π-π stacking interactions. These combined mechanisms overcome adhesion failures commonly observed in traditional materials, which are often associated with polarity competition or ion shielding effects. In contrast to mussel-inspired materials such as dopamine coatings, which rely on oxidation-sensitive catechol groups [54,55], the natural components of SSAD enable robust adhesion in dynamic wet environments without requiring external modification, while also demonstrating significantly enhanced biocompatibility [56]. Compared to existing clinical adhesives, such as fibrin and cyanoacrylate adhesives, SGM offers significant advantages in adhesive performance, biocompatibility, and therapeutic efficacy. Fibrin adhesives are limited by weak adhesion strength in moist environments [57], while SGM is specifically engineered for such conditions. Additionally, cyanoacrylate adhesives carry cytotoxic risks. SGM, validated through in vitro and in vivo studies, demonstrates excellent tissue compatibility and minimal immune response, marking an improvement over traditional materials. Unlike current adhesives that primarily focus on adhesion, SGM integrates therapeutic functions like drug and cell delivery, thereby enhancing tissue repair in bone, cartilage, and soft tissues. ECM remodeling, recognized as a fundamental mechanism of tissue repair, plays a central role in regenerating cartilage, bone and skin models [58]. This reparative process was optimized and accelerated through SSAD incorporation within SGMs. SSAD has been shown to not only recapitulate native ECM composition [48], but also enhance stem cell homing via regulation of chemokine and adhension molecule expression [59,60], thereby accelerating tissue repair. Notably, SSAD exhibits enhanced efficacy in stimulating cell migration [44], indicating that its multifactorial components, such as growth factors, matrix components, may synergistically enhance migration efficiency. Through upregulation of key genes (Matn2, Epyc and Col11a2), SGM significantly enhanced collagen crosslinking, improved matrix stability, and facilitated tissue-specific ECM assembly. Safranin O and collagen II staining revealed pronounced type II collagen expression within neocartilage of SGC-treated group, with complete neonatal cartilage regeneration. Furthermore, KEGG analysis highlighted the pivotal role of SSAD in regulating the cytoskeleton and MAPK signaling pathways. This enhancement facilitated cellular migration and localization within the ECM, thereby promoting tissue repair. Consistent with our previous findings [60], SSAD was observed to directly activate the MAPK signaling pathway.

As a multifunctional delivery platform, SGM has shown good results in wet tissue repair. For the treatment of periodontitis bone defects, microspheres achieve a stable release of minocycline through their structural design. The porous network within the microspheres providesd physical support for drug loading. In vitro experiments demonstrated that the drug exhibited sustained slow-release characteristics, with a cumulative release rate of approximately 28 % over 30 days and an initial burst release of only about 3 % within the first 24 h. This minimized the risk of toxic side effects associated with rapid initial drug release and ensured a continuous drug concentration. In the repair of articular cartilage, traditional treatments struggle to achieve cartilage regeneration due to the avascular nature and weak regenerative capacity of cartilage tissue. The SGM deliver chondrocyte to directly supplement exogenous cells. Their porous structure mimics the three-dimensional network of the natural cartilage ECM, supporting cell survival and migration. In vitro experiments demonstrated that the viability of encapsulated cells exceeded 97 %, with enhanced cell migration rates. The regenerative efficiency of SGM is attributed to the synergistic action of multiple molecular mechanisms. Transcriptome analysis revealed that the SSAD activated ECM synthesis-related genes such as Col11a2 and Matn2, promoting collagen crosslinking. The complexity of diabetic skin defects stems from the intricate interplay of hyperglycemia, oxidative stress, and vascular disorders. SGM serves as a delivery system for DFO, targeting iron metabolism and angiogenesis while mitigating oxidative stress. Studies have demonstrated that DFO inhibits ferroptosis by chelating free iron and enhances CD31-positive vascular density through activation of the HIF-1α/VEGF pathway. The antioxidant properties of SSAD effectively scavenge 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radicals, thereby reducing collagen damage caused by oxidative stress. Masson staining revealed improved alignment of collagen fibers, while H&E staining confirmed the restoration of epidermal integrity. Furthermore, SSAD contains a variety of endogenous healing-promoting proteins, which synergistically interact with therapeutic agents to accelerate vascularization and enhance wound healing, thereby highlighting its multi-effect repair ability in metabolic disorders.

4. Conclusion

This work pioneers a bioinspired engineering strategy to overcome the critical challenge of therapeutic retention in dynamic, wet tissue environments. By emulating the wet adhesion mediated by Andrias davidianus skin secretions, we engineered SGM microspheres—a multifunctional platform harmonizing with wet adhesion, drug-loading capacity, and bioactivity. The SGM achieved robust interfacial integration across cartilage, bone, and soft skin tissues, resisting mechanical displacement and fluid washout while maintaining structural integrity under physiological stress. Beyond physical retention, SSAD's inherent bioactivity activates cellular migration and matrix remodeling, synergizing with delivered therapeutics to amplify endogenous repair. This work established a paradigm bridging natural bioadhesion with precision regenerative medicine, offering a universally adaptable and clinically relevant solution for complex, fluid-exposed tissue defects.

5. Experimental section

5.1. Synthesis of GelMA

Briefly, added 10 g gelatin (Solarbio, China) to 100 mL phosphate buffered solution (PBS, Dowobio, China) and stirred at 50 °C, 1000 rpm for 1 h until the gelatin was completely dissolved. Added 5 mL MA (Sigma, USA) drop by drop into the above solution with a syringe under the condition of avoiding light and stirred magnetically for 2 h until the reaction was over. The fully reacted liquid was injected into a 12–14 kDa dialysis bag (Solarbio, China), which was then transferred to deionized water at 40 °C for dialysis by stirring magnets at 300 rpm for 5 days. The deionized water was replaced every 12 h to remove the incomplete MA and harmful intermediate products. The liquid in the dialysis bag was then filtered, frozen, freeze-dried, and set aside.

5.2. Preparation of GM and SGM

The construction of the microfluidic device was achieved through the integration of a coaxial electrospinning nozzle, comprising an inner needle of 25 G and an outer needle of 18 G. The assembly was facilitated by the employment of two distinct syringe pumps. The oil phase was constituted by mineral oil (Rhawn, China) containing 10 % (v/v) Span 80 (Rhawn, China) utilized as the surfactant. The water phase was composed of a 10 % (w/v) GelMA solution, augmented with 0.25 % (w/v) lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP, Rhawn, China) as the photosensitiser. The water phase was injected into the inner needle, while the oil phase was injected into the outer needle, with both phases being delivered via syringes. The GelMA droplets were obtained through the cutting of the water phase, achieved by adjusting the speed of the oil phase, followed by crosslinking under 405 nm UV/blue light. This process resulted in the formation of microspheres. The flow rate of the water phase was set at 20 μL/min, and the oil phase was adjusted to a speed of 700 μL/min to facilitate the preparation of microspheres. The removal of the oil phase was achieved through the repeated use of PBS for microsphere washing. To obtain SGM, SSAD hydrolysate was first prepared as previously described [61], and then SSAD hydrolysate was mixed with GelMA at a volume ratio of 10 %, 30 % and 50 %, respectively. The rest of the steps were the same as above.

5.3. Characterizations of the microsphere

The temperature was set at 25 °C, and the viscosity of SSAD and GelMA mixed hydrogels in different proportions was measured by rheometer. Each group of microspheres (microspheres was stained for easy observation) was placed on the surface of rabbit isolated knee cartilage, and then the microspheres were washed with a syringe equipped with PBS at the same rate and time. The remaining amounts of microspheres on the cartilage surface before and after washing was observed to test the adhesion ability of the microspheres. The proportion with the best adhesion performance was selected for subsequent experiments.

The morphology of microspheres was observed under microscope and the diameter of microspheres was measured by Image J software. The GM and SGM were frozen and internal structure were observed by SEM (Hitachi, Japan). FTIR (Waltham, USA) was used to record the infrared spectrum of the sample. The surface charge is assessed by measuring the Zeta potential. The swelling ratio of lyophilized SGM and GM was determined by measuring their diameter after immersion in PBS.

5.4. Evaluation of drug release in vitro

In the drug release experiment, GM and SGM loaded with Mino (Solarbio, China) were immersed in 1 mL of PBS. The samples were incubated in a shaker at 37 °C. At predetermined time intervals, the supernatant was collected, and fresh PBS was added. The concentration of the drug in the solution was determined using a spectrophotometer by measuring the absorbance intensity of the drug at its peak wavelength of 348 nm.

5.5. Penetration of fluorescent microspheres in agarose models

The agarose powder (3 w/v%, BioFroxx, China) was dissolved in PBS by application of heat. The resultant solution was then permitted to cool, thus forming a model with a density similar to that of human skin tissue. Then the microspheres with FITC were placed on the upper surface of the model to trace and simulate the drug diffusion of the drug-containing microspheres, and the penetration depth was observed after 12 h.

5.6. Cell viability on microspheres

To detect cell viability, cartilage stem cells were suspended in a hydrogel precursor solution at a density of 1 × 10⁷ cells/mL and microspheres were prepared as described above. The cell-loaded microspheres were then cultured for 0, 3 and 7 days and the viability of the loaded cells was measured using a Live/Dead Cell Staining Kit (Solarbio, China). The quantity and distribution of living cells (green fluorescence) and dead cells (red fluorescence) in the microspheres were observed under fluorescence microscope (ZEISS, Germany).

5.7. Cell adhesive on microspheres

In order to observe the morphology and cytoskeleton of the cells adhered to the microspheres, FITC-phalloidin (Yeasen, China) was used to stain the cartilage stem cells loading on the microspheres after 24 h of culture. In short, the cells were washed twice with PBS, fixed in 4 % paraformaldehyde (PFA, Solarbio, China) for 15 min, and then washed with PBS again. The cells were permeabilized with 0.2 %Triton X-100 solution for 15 min, followed by incubation with FITC-phalloidin dye (Yeasen, China) in the dark for 30 min, and then stained with 4’, 6-diamidino-2-phenylindole (DAPI, Beyotime, China) for 8 min. The distribution of cells was observed using fluorescence microscopy.

5.8. Cell migration on microspheres

On days 3, 7 and 14, we stained the cells on the cell-loaded microspheres and 6-well plates with the live cell dye from the Live Cell Staining Kit (Solarbio, China), and used a fluorescence microscope (ZEISS, Germany) to observe the range of cells climbing out.

5.9. The isolation and culture of cartilage stem cell

The rat cartilage stem cells were separated from articular cartilage. In summary, the articular cartilage tissues were meticulously minced, and then subjected to a series of enzyme digestions using 0.25 % trypsin (Sigma, USA) and 0.2 % collagenase II (Sigma, USA) for a duration of 30 min and 4 h, respectively. Subsequent to this, the isolated cartilage stem cells were cultivated within an atmosphere of 5 % CO₂ at 37 °C, utilizing Dulbecco's Modified Eagle's Medium/Nutrient Mixture F-12 (DMEM/F12, containing 1 % antibiotics and 10 % fetal bovine serum) as the medium. In order to preserve the phenotype of the cartilage stem cells, it was necessary to utilize cells from just the initial three passages in the in vitro test.

5.10. Alcian blue staining

Cartilage stem cells were directly co-cultured with microspheres. Subsequent to a 7-day and 14-day culturing period, the medium was extracted, washed with PBS, and fixed using 4 % PFA for 15 min. Following a thorough cleansing process with PBS, the samples were subjected to Alcian Blue staining solution (Solarbio, China). Following a 30-min staining period at room temperature, the staining solution was discarded, and the samples were washed twice with PBS. Subsequently, the samples were observed and photographed under a microscope.

5.11. Transcriptome sequencing and data analysis

In a 6-well plate co-cultured with SGM and GM, PBS was added to wash and the floating microspheres were removed. Total RNA was extracted from the tissue using TRIzol® Reagent in accordance with the manufacturer's instructions. The purification of RNA, reverse transcription, library construction and sequencing were conducted at Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd. in accordance with the manufacturer's guidelines. RNA-seq transcriptome library was prepared following Illumina® Stranded mRNA Prep, Ligation (Illumina, America) using 1 μg of total RNA. The raw paired end reads were trimmed and quality controlled by fastp with default parameters. Subsequently, clean reads were separately aligned to the reference genome in the orientation mode using HISAT2 software. The mapped reads of each sample were assembled by StringTi in a reference-based approach. The enrich plot package in R software was used to aggregate similar GO terms and visualize their interactions as an interconnected cluster network to highlight major biological topics. NES (Normalized Enrichment Score) describes the overall richness of activation and inhibition pathways. GSEA analysis was performed using the clusterProfiler package to evaluate the contribution of gene sets to phenotypes.

5.12. RT-qPCR analyses

When cartilage stem cells were co-cultured with microspheres for 7 days and 14 days, RNAeasyTM kit (Beyotime, China) was used to extract total RNA according to the manufacturer's recommended procedure. cDNA was reverse-transcribed from RNA using PrimeScriptTM RT reagent Kit (TakaRa, Japan). TBGreen® pre-mixed ExTaq TM (TakaRa, Japan) was used for quantitative real-time PCR (qPCR) experiments on a Bio-Rad fluorescence quantitative PCR instrument. The primer sequences are provided in Table S1 of the Supporting Information.

5.13. Ex vivo adhesion test

In order to test the adhesion of microspheres to tissues in vitro, we established a periodontitis model, a pig's knee defect model and a skin lesion wound to simulate clinical periodontitis, osteoarthritis and diabetic wound, respectively. Periodontal flap was performed on the gingival margin of the mandibular teeth of pig, microspheres were injected on the exposed root surface, and then the site where the microspheres were located was rinsed with uniform and steady water to simulate the scour effect of saliva/gingival groove fluid on the microspheres, and the area of remaining microspheres at the defect site was observed to evaluate their adhesion to hard tissues.

Similarly, irregular cartilage defects were made on the lateral head of pig femur using a slow speed dental machine and a ball drill. GM and SGM stained with Rhodamine B (Sigma, USA) were injected in situ, respectively, and the parts where the microspheres were located were rinsed with a uniform and stable water flow to simulate the scour effect of the joint capsule's fluid on the microspheres under the resting condition of the joint. The area of microspheres remaining at the defect site was observed to assess their adhesion to the tissue. In addition, we also shook the femur containing the microsphere in the water to simulate the scour effect of the microsphere with greater force under the motion condition of the joint, and observed the adhesion of the microsphere to the tissue.

Furthermore, skin defects were created on the pig's skin using a surgical blade. GM and SGM were applied in situ, respectively, then irrigated with a constant water flow to mimic wound exudation and flushing under physiological conditions. The retention of each material at the defect site was evaluated to assess their adhesion capacity. The flushing off microspheres were documented to quantitatively compare their tissue adhesion performance.

To quantitatively characterize the adhesion property of SGM and GM, a shear adhesion test was performed. Due to the size limitations of SGM and GM microspheres, SGM and GM solutions were used for detection in this experiment. Pig's skin was used as the representative skin tissue. The SGM and GM solutions were adhered to the pig's skin, with a combined area of 10 mm × 10 mm. Subsequently, the pigs' skin were pulled apart at a speed of 2 mm/min using a universal testing machine (MTS Criterion, Model 43, USA).

5.14. Animal experiments

In order to evaluate the tissue repair ability of hydrogel microspheres in different tissue interface wet environment, we constructed 10 week male Sprague-Dawley rats (Beijing Vital River Laboratory Animal Technology Co., Ltd, China) periodontitis bone defect, articular cartilage defect and diabetic skin defect models. In each experiment, the rats were randomly divided into groups using a random number table, with 6 rats in each group. All animal experiments were approved by the Research Ethics Committee of the Affiliated Stomatological Hospital of Chongqing Medical University (CQHS-REC-2023 (LSNo.103)).

5.14.1. Periodontitis model

GelMA-Mino microspheres (GM-Mino) and SSAD@GelMA-Mino (SGM-Mino) microspheres were prepared according to the above method by using 10 % (w/v) Mino-GelMA precursor solution as aqueous phase.

SD rats were randomly divided into Normal, Periodontitis, Mino, GM-Mino and SGM-Mino groups. Except for the normal group, the other groups used the 5-0 silk thread moistened by P.g (ATCC33277) to ligate the maxillary second molar of rats to construct the periodontitis model, and the corresponding intervention measures were given according to the groups. The ligation of the silk thread was checked every other day and local injection of P.g suspension was performed.

Following a 2-week therapeutic regimen, SD rats were euthanized for comprehensive evaluation of osseous regeneration. Dissected maxillae underwent high-resolution micro-CT scanning to quantify bone resorption parameters, with three-dimensional reconstructions performed using Mimics research 19.0 software (Materialise, Belgium). Decalcified tissues were paraffin-embedded, sectioned into 6 μm coronal slices, and subjected to sequential histomorphometric analyses: H&E staining for structural integrity assessment, TRAP staining (Servicebio, China) for osteoclast localization, and immunofluorescence staining targeting osteogenic markers. Primary antibodies against osteocalcin (OCN, Servicebio, China) and alkaline phosphatase (ALP, Servicebio, China) were applied with Alexa Fluor-conjugated secondary antibodies, followed by DAPI nuclear counterstaining. Semi-quantitative analysis was conducted using ImageJ software.

5.14.2. Cartilage defect model

After anesthesia, the cartilage defect with a diameter of 1.5 mm and a depth of 1 mm was created at the lateral femoral head of both legs using a dental slow machine and a ball drill. The rats were randomly divided into Normal group, Blank group, defect group, GC group, SGC group.

In order to conduct the gait analysis, the rats were trained to walk from one side of an illuminated glass plate to the other within a confined corridor for a period of at least 10 s. The collection of footprints was conducted utilizing an innovative internal light footprint refraction technology. In addition to the presentation of the areas of footprints, the relative pressures of the footprints were also demonstrated. It was demonstrated that an increase in the intensity of the light indicated a corresponding increase in the relative pressure of the footprint. The presence of footprints was documented through the utilisation of a high-speed camera. Subsequently, the footprints were analyzed automatically using Catwalk XT 10.0 software (Noldus, Netherlands).

The rats were euthanized after 6 weeks of treatment and were evaluated by ICRS scoring system, histology and immunohistochemistry. Specimens were fixed for 3 days with 4 % formaldehyde, decalcified in 17 % (w/v) EDTA (Servicebio, China) for 10 weeks, embedded in paraffin, and cut into 6-μm-thick sections. The histological evaluations were conducted in accordance with a reported methodology, employing the H&E staining (Solarbio, China) and Safarin o-fast green staining (Solarbio, China) procedures. For immunohistochemistry, rabbit monoclonal anti-collagen II antibody (1:800, Proteintech, China) was used as primary antibodies. Semi-quantitative analysis was conducted using ImageJ software. Furthermore, histological scoring was performed on the repaired tissues in the defect using the Mankin scoring system. The histological sections from the medial and lateral regions of each defect were blindly scored by 3 independent evaluators.

5.14.3. Diabetic skin defect model

GelMA-DFO microspheres (GM-DFO) and SSAD@GelMA -DFO (SGM-DFO) microspheres were prepared with 10 % (w/v) DFO-GelMA precursor solution as the aqueous phase and the remaining steps as described above.

After shaving and disinfecting the back of rats, a circular full-thickness wound with a diameter of 1.0 cm was prepared. Then, the rats were randomly divided into control group, DFO group, GM-DFO group and SGM-DFO group. The control group was not treated, and the other groups were treated with microspheres every 2 days until the wound healed. During the period, the wound area was photographed and recorded, and the wound healing rate was calculated by image processing software ImageJ.

The rats were euthanized on days 7 and 14, and back skin was collected. Fixed with 4 % formaldehyde for 48 h, then dehydrated in graded ethanol (Sorabio, China) and embedded. The sample blocks were cut into 6 μm sections, followed by histological analysis with H&E and Masson trichromatic staining and immunofluorescence double staining with CD31 (Servicebio, China) and α-SMA (Servicebio, China), showing vascularization of the implanted structure. The sections were scanned with an automatic slide scanner (Olympus, Japan) and analyzed with ImageJ.

5.14.4. Biosafety of the SSAD hydrolysate

To assess the potential systemic effects and safety profile of SSAD hydrolysate, 36 KM mice were administered different concentrations of SSAD hydrolysate via intragastric gavage at a volume of 0.8 mL per mouse. The control group received an equivalent volume of normal saline via the same route. Each group consisted of 6 mice, with an equal distribution of males and females. Within 24 h after administration, the general status of the mice was continuously monitored, including survival rate, behavioral changes, and potential toxic manifestations. At the end of the observation period, all surviving mice were anesthetized and collected blood samples for hematological and blood biochemical analyses.

5.15. Statistical analysis

Statistical analysis was performed using SPSS software (SPSS Software, USA). All experimental results were expressed as mean ± standard deviation (SD). The evaluation was performed using a standard one-way ANOVA with Tukey's post hoc analysis (for comparisons of more than two groups). The statistical significance was set as P < 0.05.

CRediT authorship contribution statement

Liwen Zheng: Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Data curation. Jixian Feng: Writing – original draft, Visualization, Validation, Methodology, Investigation. Yaxian Liu: Writing – original draft, Validation, Formal analysis, Data curation. Jianming Zhang: Writing – original draft, Methodology. Yuheng Wang: Visualization, Formal analysis, Writing – review & editing. Xingrui Yan: Software, Methodology, Investigation. Ge Wang: Investigation, Data curation. Lan Li: Investigation, Data curation. Yuan Pang: Writing – review & editing. Hongmei Zhang: Supervision, Resources, Funding acquisition, Conceptualization. Ximu Zhang: Resources, Project administration, Funding acquisition, Conceptualization.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics approval and consent to participate

This study and included experimental procedures were approved by the Research Ethics Committee of the Affiliated Stomatological Hospital of Chongqing Medical University (CQHS-REC-2023 (LSNo.103)). All animal housing and experimental procedures were carried out in strict compliance with the institutional guidelines for the care and use of laboratory animals.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: