Multifunctional injectable chitosan-based adhesive enhanced by limited oxidation for assisting bone grafting

Abstract

In clinical, critical bone defects are often treated with molded bone implants, yet they are prone to loosening, shape mismatch with the defect, and infection and inflammation. Herein, we present a limited oxidation strategy of catechol-chitosan (CC) to fabricate an injectable, adhesive, multifunctional nanocomposite hydrogel (OCCL) to assist bone implants in bone defects. Specifically, the oxidative cross-linking of CC within layered nanoclay (LAP) is moderately restricted to avoid excessive oxidation, thereby enhancing its adhesion to the tissue interface. By moderate oxidation of CC and the physical crosslinking boosted by LAP, the OCCL hydrogels show high adhesive strength, underwater adhesion, high antioxidant activity, good biodegradability and excellent antibacterial activity. Taking advantage of these features, the multifunctional OCCL can effectively promote cell growth, cell adhesion, angiogenesis and osteogenesis to facilitate bone regeneration, as demonstrated by in vitro and in vivo analyses. This work presents a promising strategy of limited oxidation to develop injectable and multifunctional adhesives, providing a safe and easy-to-process solution for reliable bone grafts.

Graphical abstract

1. Introduction

Critical-sized bone defect caused by trauma, surgery, and osteoporosis remains a significant challenge in clinical treatment [1,2]. Currently, implants such as autogenous bone, allograft bone, or new artificial bone materials are often used clinically to treat critical-sized bone defects [3]. Although they have excellent osteoinductive and osteoconductive abilities, these non-deformable implants cannot fill against complex shapes during use and fixate in the defect area, leading to loosening or empty areas, which can cause inflammation and affect bone repair [4]. Moreover, bacterial infections and reactive oxygen species (ROS) are also responsible for bone repair failure [5,6]. Besides, accelerating blood vessel growth and promoting new bone formation are essential [7]. However, most clinical bone adhesives, such as bone cement [8] and cyanoacrylate superglue [9], have only specific single or dual functions, which cannot meet the requirements of clinical multifunctionality. Therefore, there is a pressing need for an injectable, biodegradable, adhesive, antimicrobial, and antioxidative material that can work synergistically with bone grafts to promote effective bone regeneration.

Catechol (CA) chemistry has gained considerable attention, due to its potential for developing multifunctional materials with favorable adhesiveness, antimicrobial, and antioxidant properties [10,11]. In particular, CA-functional hydrogels could provide can be used as extracellular matrix-like environment for cell growth and proliferation [12]. To achieve injectable CA-functional hydrogels, a variety of crosslinking methods have been developed, such as ion crosslinking [13,14], hydrogen bonding [15], oxidative crosslinking [16], photo-crosslinking [17], dynamic covalent bonding [18], free radical copolymerization [19] and so on. However, the hydrogel fabricated by physical crosslink (ion crosslinking and hydrogen bonding) and dynamic covalent bonding show gel behavior during injection, resulting in poor permeability in small pores, which limits the permeability of photo-crosslinking hydrogels [20]. While the injectable CA-functional hydrogel prepared by oxidative crosslinking may satisfy the demand of surgical operation. In particular, catechol-chitosan (CC) is rich in amino groups, which enable Schiff base reactions during oxidative cross-linking [16,21], leading to high cross-linking density and mechanical strength of the injectable hydrogel.

Conducting chemical reactions under confined conditions can alter material synthesis and introduce new material properties [22,23]. For instance, using ice templates to induce crosslinking of chitosan nanofibers under confined space can generate hydrogels with exceptional mechanical performance [23]. To the best of our knowledge, the existing oxidatively crosslinked CA-functional hydrogels are fabricated by adding oxidants, such as sodium periodate, while their oxidation is uncontrolled, leading to the overoxidation and the decreased adhesiveness of the hydrogel [24,25]. Laponite (LAP), a 2D nanosilicate disc, simultaneously offers nanoconfined spaces for various reactions and promotes bone regeneration via bioactive ions release [26,27]. Therefore, we hypothesize that the nanoconfinement effect of LAP could restrict the oxidative crosslinking of catechol-functionalized chitosan hydrogels, thereby potentially enhancing their adhesion, improving antioxidant capacity, and promoting osteogenic activity through ion release.

Herein, we employed LAP to achieve tuned CC oxidation via nanoconfinement effects, yielding an injectable, adhesive, multifunctional CC/LAP (OCCL) hydrogel for bone regeneration (Scheme 1). These hydrogels displayed excellent injectability, biodegradability, antioxidant activity and antibacterial activities. With LAP incorporation, CC was oxidized between LAP layers and protected from excessive oxidation, resulting in higher ROS scavenging activity in vitro. Simultaneously, LAP generated hydrogen bonds and electrostatic interactions with CC. Due to the limited oxidation of CA and the enhanced physical crosslinking by LAP, the OCCL hydrogels exhibited higher adhesive strength to bone, and tightly bonded even underwater, without requiring any subsequent crosslinking. Besides, benefiting from the combination of LAP and CC, the OCCL hydrogels can effectively promote cell growth, cell adhesion, angiogenesis and osteogenesis, resulting in bone regeneration in vivo. Such performances together revealed their great potential for bone regeneration, especially in assisting bone grafts.

Scheme 1.

Diagram for the fabrication and application of the injectable, adhesive, multifunctional OCCL nanocomposite hydrogel.

2. Materials and methods

Materials. Chitosan (CS, Deacetylation degree: 86.58 %, Mw ≈ 20 kDa) was purchased from Shandong Haidebai Biotechnology Co (Shandong, China). Laponite XLG-XR (LAP) was obtained from BYK Additives & Instruments (Germany). Hydrocaffeic acid (HCA) was purchased from Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide (EDC), sodium hydroxide, sodium periodate, 1,1-diphenyl-2-trinitrophenylhydrazine (DPPH), phosphate buffer solution (PBS), anhydrous ferric trichloride, potassium ferricyanide and salicylic acid were purchased from Macklin Biochemical Technology Co., Ltd (Shanghai, China). Trichloroacetic acid, ferrous sulfate, vitamin C (VC) were obtained from Damao Chemical Reagent Factory (Tianjin, China). Hydrochloric acid and hydrogen peroxide (30 %) were purchased from Guangzhou Chemical Reagent Factory (Guangzhou, China). Deionized water (18.2 MΩ cm) was obtained from a Milli-Q purification system (Millipore, USA) and used throughout all experiments. All reagents were of analytical grade and used as received without further purification.

Synthesis and characterization of CC. At room temperature, 1.55 g of chitosan was dispersed in 150 mL of 1 % HCl solution and stirred until complete dissolution. After the pH was adjusted to 3.0, HCA was added into the solution, and then 150 mL of 50 v/v% ethanol/water solution containing EDC was added slowly (10 mL/min). When the EDC was fully added, the pH of the mixture was adjusted to 5.0 and the reaction was carried out for 12 h. Then, the mixture dialyzed using a 10000 Da dialysis bag (MYM Biological Technology Company, USA) at pH = 4.0 for 3 d, and dialyzed in deionized water for 2 h. After dialysis and freeze dring (SCIENTZ-12N, SCIENTZ, China), the catechol chitosan with different degrees of substitution (DS) were obtained and labeled as CC₁, CC₂, and CC₃, while the corresponding reactants dosage were shown in Table S1. The chemical structure of CC was determined using FT-IR spectroscopy (TENSOR27, Bruker, Germany) and ¹H NMR spectrometry (Bruker Avance III HD, 600 MHz, D₂O, Germany). The DS_CA of CC were measured by UV spectroscopy. Briefly, the absorbance value at 280 nm of 1 mg/mL CC solution was measured using a UV spectrophotometer (UV-1800, Shimadzu, Japan). To generate a standard curve, different concentrations (1–800 μM) of HCA were also measured.

Preparation and characterization of OCCL nanocomposite hydrogels. First, various amounts of LAP (2, 4 and 10 mg/mL) dispersion were prepared in 9 mg/mL NaIO₄ solution. 30 mg of CC was added to 0.5 mL of deionized water and stirred until completely dissolved. Then, 0.5 mL of LAP/NaIO₄ solution was added and rapidly mixed under vortex shaking to obtain oxidized CC/LAP nanocomposite hydrogels. They were named as OCC_xL_y hydrogel, where x and y are the serial numbers of CC with different DS and concentration of LAP dispersion (1, 2, and 5 mg/mL), respectively. All components used for hydrogel preparation were sterilized prior to use. Specifically, CC powders were sterilized by UV irradiation for 1 h, while LAP and deionized water were autoclaved. NaIO₄ solution was prepared using sterile water and filtered through a 0.22 μm sterile syringe filter. All hydrogel preparation steps were carried out under aseptic conditions.

The morphology of the hydrogels was observed by scanning electron microscope (SEM, Merlin, Zeiss, Germany). The zeta potentials of hydrogel precursor with different compositions were characterized by Nanoparticles Analyzer (Zetasizer Pro, Malvern, UK). The UV–vis transmittance spectra of the diluted nanocomposite were obtained by automatic microplate reader (Thermo, Multiskan, USA) to elucidate the oxidation of CA during gelation. The tested concentrations of CC and NaIO₄ were kept at 1 mg/mL and 0.15 mg/mL, respectively, while the LAP concentrations of OCC₂L₁, OCC₂L₂ and OCC₂L₅ were 0.033 mg/mL, 0.067 mg/mL and 0.167 mg/mL, respectively. The gelation time of the hydrogels was tested through vial tilting method. 1 mL of hydrogel precursors was added to a 2 mL glass vial, and then incubated at 37 °C. The vial inverted without visible flow was found to be the sol-gel transition time. Swelling ratio of the hydrogels was performed in normal saline at 37 °C. At the predetermined time point (t = 0, 0.5, 1, 2, 4, 8, 12, 24 h), the wet weight (W_t) of the hydrogels was measured after carefully removing excess liquid with filter paper. The swelling ratio of hydrogels was calculated by Equation (1):

$$\text{Swelling}\text{ratio}=\frac{{\mathrm{W}}_{\mathrm{t}}}{{\mathrm{W}}_0}$$ (1)

To determine the degradation rate of the hydrogels, the hydrogel samples were incubated in PBS containing lysozyme (50 μg/mL) for 7, 14, 21, and 28 d, with the buffer refreshed every three days to maintain enzyme activity. The pH of the buffer was monitored during the 3 days and remained stable at approximately 7.4. The weight (Wt) of the hydrogels was measured after multiple rinses in DI water and freeze-drying. The degradation ratio of hydrogels was calculated by Equation (2):

$$\text{Degradation}\text{ratio}=\frac{{\mathrm{W}}_{\mathrm{t}}}{{\mathrm{W}}_0}$$ (2)

Adhesive properties of OCC_xL_y nanocomposite hydrogels. Adhesive strength of the hydrogel was investigated by a universal testing machine (INSTRON VS81452323RE, USA) in stretch mode. The tensile adhesive strength was measured on bovine cortical bone specimens which were rectangular cuboids of a size of 20 mm × 20 mm × 12 mm cut from the bovine femora. The bone adhesive surface (12 mm × 20 mm) covered by a thin layer of hydrogels was pressed and further adhered for 1 h prior to testing.

Antioxidant properties of CC and OCC_xL_y nanocomposite hydrogels. DPPH assay for free radical scavenging activity: 50 μL of CC solution (0.125, 0.25, 0.5, 1.0, 1.5, and 2.0 mg/mL) or hydrogel was added to 96-well plates, followed by 200 μL of DPPH methanol solution (0.4 mM). After 30-min incubation in the dark at room temperature, the absorbance at 517 nm was determined by using an automatic microplate reader. Vitamin C (VC) was chosen as the positive control, and the relative free radical scavenging activity (ROS scavenging activity, %) was calculated by Equation (3) as follows:

$$\text{ROS}\text{scavenging}\text{activity}(\%)=\left(1-\frac{A_1-A_2}{A_W}\right)\times 100\%$$ (3)

Where A_w, A₁ and A₂ are the absorbance of water (negative control), CC (sample group) and blank control (no DPPH), respectively.

Salicylic acid method for the determination of hydroxyl radical scavenging activity: 100 μL of 9 mM FeSO₄ solution, 100 μL 9 mM of salicylic acid-ethanol solution, and 100 μL of 8.8 mM H₂O₂ were homogeneously mixed, and then 100 μL of CC solution (0.25, 0.5, 1, 2.0 mg/mL) or hydrogel was added. After incubation at 37 °C for 15 min, the absorbance of the reaction mixture at 510 nm was measured. VC was used as the positive control. The hydroxyl radical scavenging ability (·OH scavenging ability, %) was calculated according to Equation (4) as follows:

$$·\text{OH}\text{scavenging}\text{ability}(\%)=\left(1‐\frac{A_1‐A_2}{A_0}\right)\times 100\%$$ (4)

Where A₁ is the absorbance of the sample group, A₂ is the absorbance of the water substituted H₂O₂, and A₀ is the absorbance of the water substituted sample solution.

Prussian blue assay for the determination of reducibility: 50 μL of CC solution (0.25, 0.5, 1, 2.0 mg/mL), 50 μL of PBS (0.2 M, pH = 6.6) and 50 μL of 1 wt% K₃Fe(CN)₆ solution were mixed and then incubated at 50 °C for 20 min. After cooling to room temperature, 50 μL of 10 wt% trichloroacetic acid and 50 μL of 0.1 wt% FeCl₃ solution were added to the solution, and incubated at room temperature for 10 min in dark conditions. The absorbance was measured at 700 nm. Using VC as a positive control, the reducibility of CC was calculated according to Equation (5) as follows:

$$\text{Reducibility}={\mathrm{A}}_1-{\mathrm{A}}_2$$ (5)

Where A₁ and A₂ are the absorbance of CC (sample group) and blank control (no Fe³⁺ ions), respectively.

Antibacterial activity of OCC_xL_y nanocomposite hydrogels. The antibacterial activity of the hydrogel against E. coli. and S. aureus. was evaluated. 1 mL of the bacterial suspension (1 × 10⁴ CFU/mL) was added to the 24-well plates, and the hydrogel (80 mg) was soaked in the bacterial suspension for 1 h. The bacterial suspension after treatment was coated on the LB agar plate and incubated at 37 °C for 24 h to measure the viable colony units of E. coli and S. aureus formed. The antibacterial mechanism of the hydrogel was evaluated by morphology, protein leakage, cell membrane permeability, and intracellular malondialdehyde oxidation. The bacteria were first washed twice with saline and then configured into a bacterial dispersion (1 × 10⁷ CFU/mL). This dispersion was incubated with the hydrogel for 10 h. Due to the high concentration of bacteria, the time of interaction of the hydrogel with the bacteria was prolonged. After the incubation, the hydrogel was removed and the bacterial dispersion was centrifuged at 10000 rpm for 5 min. The supernatant and bacterial precipitate were collected separately. The bacterial precipitate was immobilized with glutaraldehyde and then gradient dehydrated with ethanol. Finally, the treated bacteria dispersion was dropped on the silicon wafer. After spraying the sample with gold, SEM was performed to observe the bacterial morphology. BCA Protein Assay Kit (Solarbio, Beijing) was employed to measure the protein concentration of the supernatant. The intracellular malondialdehyde oxidation of the supernatant was determined using a malondialdehyde (MDA) kit (Jiancheng Biological Engineering Institute, Nanjing). Meanwhile, the cell membrane permeability of E. coli was evaluated by the β-galactosidase Assay Kit (Beyotime Biotechnology, Shanghai).

In vitro biocompatibility. The isolation of rat-bone marrow mesenchymal stem cells (r-BMSCs) followed by previous research [28]. The femur and tibia of 1-week-old Sprague-Dawley (SD) rats were aseptically dissected using sterile scissors, followed by removal of the surrounding muscle tissue adjacent to the bones. Subsequently, the midsection of both femur and tibia was excised, the r-BMSCs were extracted from the bone marrow cavities of the tibia and femur using medium, followed by centrifugation prior to inoculation into the culture dish. The MC3T3-E1 cells were acquired from ProCellLife Sciences & Technology Co., Ltd (China). Cell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay kit (CK04, Dojindo, Japan). Briefly, r-BMSCs were seeded into 96-well plates at a density of 10,000 cells mL⁻¹, and then treated with different hydrogel extracts (3.3 mg hydrogel in 1 mL medium). After incubation for 1, 3, and 5 d, 10 μL CCK-8 solution was added to each well and incubated for 2 h. Absorbance values at 450 nm were determined using a microplate reader (VARIOSKAN FLASH, Thermo Fisher Scientific, USA). The OCC₂L₅ hydrogel was evenly distributed in a 6-well plate, followed by seeding r-BMSCs onto the hydrogel surface for a 24 h culture period. Subsequently, the cells were fixed and permeabilized. TRITC phalloidin (100 μM, CA1610, Solarbio, China) and Hoechst 33342 (diluted at a ratio of 1:1000, C1022, Beyotime Biotechnology, China) were used to stain the r-BMSCs for 15 min. Finally, a high-content imaging system (Opera Phenix Plus, PerkinElmer, USA) was employed for detection.

Anti-oxidative stress test. The mitochondrial membrane potential of r-BMSCs was assessed using a mitochondrial membrane potential detection kit (JC-1 assay) (C2006, Beyotime Biotechnology, China). The level of ROS in r-BMSCs was assessed using the ROS Assay Kit (S0033S, Beyotime Biotechnology, China).

In vitro osteogenic ability. The r-BMSCs and MC3T3-E1 cells were cultured in osteoinductive media (RAXMX-90021 and MUXMT-90021, Cyagen Biosciences, China) and co-incubated with the hydrogels of 75 mg/mL. After co-culturing for 7 d and 14 d, the samples were cleaned and fixed for 15 min. On 7 d, the alkaline phosphatase (ALP) activity assays of the cells were measured by ALP activity kit (G1480, Solarbio, China). On 14 d, the differentiation process of osteoblasts was determined by Alizarin Red S (ARS) staining kit for osteogenesis. All these staining images above were captured by an optical microscope (TS2, Nikon, Japan). The expression of osteogenic-related genes in rBMSCs and MC3T3-E1 was analyzed by quantitative real-time polymerase chain reaction (qRT-PCR). The total RNA from each group was extracted using an RNA extraction kit (RN001, Yishan Biotechnology, China), and subsequently reverse transcribed into cDNA following the protocols of the RNA reverse transcription kit (RT001, Yishan Biotechnology, China). The primer nucleotide sequences used in this experiment are presented in Table S3.

For western blot analysis, the r-BMSCs were lysed by protein lysate solution (RIPA: PMSF = 100:1). The extracted proteins were subjected to electrophoresis on a 10 % sodium dodecyl sulfate-polyacrylamide (SDS) gel. The target proteins were subsequently transferred onto a polyvinylidene difluoride (PVDF) membrane (IPVH00010, Millipore, USA), and the bands were blocked with 5 % Bull Serum Albumin (BSA) for 1 h. Following that, the bands were incubated overnight at 4 °C in the respective primary antibody solution, the primer antibody used in this experiment are presented in Table S4. After washing with TBST, the bands were then immersed in the corresponding secondary antibodies (ZB-2301 and ZB-5305, ZSGB-BIO, China) and incubated for 1 h at room temperature. The protein expression of each band was visualized using Tancon 5200 automatic chemiluminescence image analysis system (Tancon, China).

For immunofluorescence staining, the r-BMSCs were washed and fixed, followed by 0.5 % Triton-X100 for 15 min and blocking with 5 % BSA for 1 h at room temperature. Subsequently, the corresponding rabbit primary antibody (Table S4) was incubated with r-BMSCs at a dilution ratio of 1:100 at 4 °C for 12 h. The secondary antibody (K0034G-Cy5, Solarbio, China and bs-0296G-Cy5, Bioss, China) was then diluted to 1:100 and incubated with r-BMSCs for 1 h at room temperature. The r-BMSCs were stained with TRITC phalloidin (100 μM, CA1610, Solarbio, China) and Hoechst 33342 (1:1000 dilution ratio, C1022, Beyotime Biotechnology, China) for 15 min, and finally detected using high-content screening (Opera Phenix Plus, PerkinElmer, USA).

In vitro angiogenesis characterization. Tube formation assay was employed to assess the angiogenic potential of hydrogels. Human Umbilical Vein Endothelial Cells (HUVECs) were starved for 6 h in serum-free and Penicillin-Streptomycin free medium, 96-well plates were coated with Matrigel (356234, CORNING, USA), and HUVECs digestion was performed using 0.25 % trypsin (25200–072, Gibco, USA). Then, a total of 30000 cells were seeded into each well. After 3 h of treatment with the hydrogel extracts, the neovascularization was observed. The angiogenesis correlation analysis was conducted utilizing the Angio Tool. The expression of angiogenesis related genes was evaluated by qRT-PCR, which was described as above. The primer nucleotide sequences used in this experiment are presented in Table S3.

Rat calvarial defect model. The rat calvarial defect model was established using SD rats aged 8–10 w. All animal experiments were conducted according to the protocols approved by the Institutional Animal Care and Use Committee of the Beijing Institute of Radiation Medicine (IACUC-DWZX-2021-713). Following successful anesthesia, the rat's head underwent skin disinfection prior to a layer-by-layer incision of the calvarial bone. The rat calvarial bone was bilaterally trephined using a dental drill to create two circular defects with a diameter of 7 mm each, and the corresponding groups were treated with OCC₂ and OCC₂L₅ hydrogels, the control group did not receive any treatment (n = 3). All rat skulls were collected for micro-CT analysis (PerkinElmer, USA). All samples were scanned using the following settings: 360° total rotation, 70 kV and 114 μA source settings, and 4 min exposure time.

Histological analysis. The rats were euthanized at 6 and 12 w. The obtained skulls were immersed in 4 % paraformaldehyde solution for 48 h, and then immersed in a decalcifying solution which was replaced every 2 d until the skull tissue reached a softened state. Subsequently, the softened skull tissue was embedded in paraffin and sectioned. After deparaffinization, hematoxylin and eosin (H&E) staining were performed following the provided instructions (G1076, Servicebio, China). The Masson staining was performed in accordance with the provided instructions (G1006, Servicebio, China). The Safranin O-Fast Green (SO-FG) staining was carried out in accordance with the provided instructions (G1053, Servicebio, China). Additionally, the toxicological effects of the hydrogel on major organs (heart, liver, spleen, lung, and kidney) were also evaluated by H&E staining.

RNA Sequencing. The samples were extracted and sequenced following BGI (Shenzhen, China) standard operating procedures, while data analysis was conducted on the Dr. TOM II platform provided by BGI.

Statistical analysis. The data were analyzed using GraphPad Prism 8.0 (GraphPad Software Inc., CA, USA). All data were shown as mean ± standard deviation. Comparison between more than two groups was analyzed by one-way analysis of variance followed by Tukey's post hoc test. A P value < 0.05 indicated a statistically significant difference.

3. Results and discussion

3.1. Design and characterization of OCCL hydrogels

CA-functional hydrogels based on oxidative crosslinking have attracted significant attention in biomedical materials due to their injectability and multifunctional properties. Generally, oxidants promote the oxidation of catechol groups to o-quinones, which subsequently undergo oxidative polymerization to form crosslinked networks between molecular chains. However, this oxidation process is typically difficult to control, resulting in a substantial decrease in the content of catechol groups or slightly oxidized o-quinone groups, leading to poor adhesion and antioxidant properties of the hydrogels. To address this issue, we designed a nano-reinforced crosslinked network with limited oxidation of CA groups to develop an injectable multifunctional hydrogel adhesive for supporting bone grafts and promoting bone regeneration.

In this study, we developed a series of injectable multifunctional hydrogel adhesives using LAP as both an oxidation process regulator and nano-reinforced agent, CA-functional chitosan as the network backbone, and NaIO₄ as the oxidant (Scheme 1). The regulation of the network is as follows: in the absence of LAP, excessive oxidative polymerization occurred within the gel; when a small amount of LAP was added, its alkaline environment enhanced the oxidation of catechol groups; with an appropriate amount of LAP, the nanosheets restricted excessive oxidative polymerization of catechol groups trough nanoconfinement effects while simultaneously strengthening the network as nano-fillers; when excessive LAP was added, its negative charge rapidly interacted electrostatically with the positively charged catechol-modified chitosan, forming a heterogeneous system unsuitable for direct application (Fig. 1a).

Fig. 1.

a) Diagram for the crosslinked network of OCCL nanocomposite hydrogel with different LAP addition. b) SEM morphology of different hydrogels. c) The UV–vis spectra during the oxidation process of the CC₂ solution. The UV–vis spectra of CC₂ solution with different LAP contents after oxidation for d) 30 s and e) 15 h. f) FT-IR spectra of OCCL nanocomposite hydrogel with different LAP contents. g) Zeta potential of LAP dispersion, CC₂ solution and CC₂L₅ mixed solution. h) Swelling behavior of different hydrogels. h) Schematic illustration of the possible crosslinked mechanism of OCC₂L₅ hydrogel.

First, CC was synthesized via amidation of chitosan (CS) with hydrocaffeic acid (HCA) under acidic condition using EDC as a catalyst (Fig. S1a). The CC with different degrees of catechol substitution (DS_CA) were successfully synthesized by varying the amount of EDC and HCA added (Table S1 and S2 and Fig. S1b–d) and labeled as CC₁, CC₂, and CC₃, respectively. Their antioxidant properties were measured by typical DPPH method, salicylic acid method and Prussian blue assay (Fig. S2). The results showed that the antioxidant properties of CC were sequentially enhanced according to the increased DS_CA (CC₁ < CC₂ < CC₃), which helped eliminate reactive oxygen species in the bone regeneration microenvironment [29]. Considering the better antioxidant properties and less HCA usage, CC₂ was chosen for the preparation of composite hydrogels.

The OCCL hydrogels were prepared by simply mixing the CC₂ solution with different amounts of LAP and certain NaIO₄. Depending on the final LAP concentrations, the hydrogels were defined as OCC₂L_y, where y is the concentration of LAP dispersion. As shown in Fig. 1e, both the LAP/NaIO₄ mixed solution and CC solution are fluid, while the OCCL nanocomposite hydrogel is formed at a particular time after mixing. Meanwhile, before gelatin, the OCCL nanocomposite hydrogel can be injected and filled with complex shapes due to its good fluidity, and then molded in situ, which can be applied to different defective regions (Fig. S3a). The gelation time of OCCL nanocomposite hydrogels fell within the range suitable for surgical procedures and gradually decreased with an increase in LAP contents, from 742 s to 653 s, 557 s, and 449 s, respectively (Fig. S3b). The above results indicates that the addition of LAP promotes the formation of a hydrogel network. Notably, when the LAP content was excessive, such as 10 mg/mL, obvious flocculation occurred in the mixture of CC and LAP, forming a heterogeneous system (Fig. S4). Therefore, LAP concentrations exceeding 5 mg/mL were not used in the subsequent preparation of OCCL hydrogels.

According to scanning electron microscopy (SEM) analysis, the OCC₂ hydrogel oxidized by NaIO₄ exhibited macroporous structures with pore sizes over 100 μm (Fig. 1b). When the LAP content was raised to 2 mg/mL, LAP promoted the oxidation of CC and formed smaller pores in the OCC₂L₂ hydrogel. At higher LAP content (5 mg/mL), the more lamellar structure of LAP may restrict further chemical cross-linking of CC after oxidation, leading to the formation of macropores in OCC₂L₅ hydrogel. Further, the oxidation processes occurring in CC/LAP solutions mixed with sodium periodate under dilution conditions were investigated using an automatic microplate reader (Fig. 1c–e). After adding NaIO₄, the original intensity of CC₂ at 280 nm decreased rapidly and the absorption peaks attributed to semiquinone (320 nm) and quinone (400 nm) appeared. It is proved that the strong oxidative property of NaIO₄ could rapidly oxidise CC within 30 s, and part of the CA was transformed into o-semiquinone and o-benzoquinone structures. Meanwhile, the absorbance values at 230 nm–350 nm range gradually increased over time. After reacting for 15 h, the peak of o-benzoquinone disappeared and an overall increase in absorbance values from 230 nm to 600 nm occurred, possibly due to the Michael addition and Schiff base between o-benzoquinone and amino [30]. With the addition of varying amounts of LAP, more quinones were produced at 30 s, while the overall absorbance value at 15 h was concurrently higher (Fig. 1d and e), which indicates that LAP facilitated the generation of o-benzoquinone under NaIO₄ oxidation. This phenomenon may be attributed to LAP creating a localized alkaline microenvironment through the release of ions (Mg²⁺, Li⁺) during its dispersion [31]. This alkaline environment triggers the in situ oxidation of CC, while LAP's layered structure provides a nano-confinement effect for CC, achieving controlled oxidation and preventing excessive oxidative polymerization.

In the FT-IR spectra (Fig. 1f), compared to CC₂, the intensity of the peak at 1623 cm⁻¹ for OCC₂ hydrogel increased, suggesting the formation of Schiff base (C=N) [32]. In parallel, the decrease in the intensity of the peak at 1289 cm⁻¹, 820 cm⁻¹ and 780 cm⁻¹ for phenol, and a new peak appears at 775 cm⁻¹, further proving that part of the CA group was oxidized. With the increasing addition of LAP, the peak intensity of the phenolic moiety at 1289 cm⁻¹ in the hydrogel gradually increased, suggesting that LAP limited the excessive oxidation of CA group, which is consistent with the SEM analysis. Additionally, all OCCL hydrogels displayed bands corresponding to Si-O stretching and Si-O bending at 999 cm⁻¹ and 443 cm⁻¹, respectively, which had an obvious shift compared with bare LAP (1008 cm⁻¹ and 460 cm⁻¹). Such observations indicated the formation of hydrogen bonding interactions between the CC chains and LAP [33]. Of note, zeta potential measurements (Fig. 1g) showed that the positively charged CC₂ chain (+68.8 mV) has an electrostatic interaction with the negatively charged LAP (−29.8 mV). These results suggest that the OCCL network of nanocomposite hydrogel is composed of chemical cross-linking (Schiff base and Michael addition reaction) and physical cross-linking (hydrogen bonding, electrostatic interaction).

The swelling behavior of the hydrogels under physiological environment was examined (Fig. 1h). It was found that all OCCL hydrogels showed a slight decrease in mass when immersed in physiological saline for 24 h. This could be attributed to further oxidative polymerization of the hydrogel networks, resulting in increased cohesion and subsequent mild overall shrinkage. Of note, the OCC₂L₂ hydrogel showed the largest shrinkage and the OCC₂L₅ hydrogel showed the smallest shrinkage. It is indicated that OCC₂L₂ hydrogel underwent a stronger oxidative polymerization internally promoted by low LAP, while OCC₂L₅ hydrogel was limited by the lamellar barrier of more LAPs, reducing the excessive oxidative polymerization, which is consistent with the SEM analysis.

Together, the above data suggest that adequate LAP can promote the in situ oxidation of CC while simultaneously mitigating excessive oxidation. This dual effect leads to retention of CA group and reduced formation of chemical crosslinks. Additionally, the introduction of LAP facilitates more physical crosslinking.

3.2. Adhesive properties, degradation and antioxidant activity of OCCL hydrogels

The adhesive property of OCCL hydrogels with different LAP contents was measured by performing in vitro adhesion tests with bovine cortical bone (Fig. 2a). As the LAP addition increased, the adhesive strength of hydrogels significantly increased, which was from 18.25 kPa to 39.69 kPa, 44.15 kPa and 49.76 kPa. This result was attributed to the tuned oxidation of CC and the increased physical crosslinking by LAP, which is consistent with the above analysis. Among them, the OCC₂L₅ hydrogel has the best adhesion strength. Besides bovine bone, OCC₂L₅ hydrogel demonstrated adhesive properties on various hydrophilic and hydrophobic surfaces, including polyterephthalic plastics (PET), silicone, rubber, polypropylene (PP), glass, corundum ceramics, Teflon (PTFE), and steel (Fig. 2b). It is speculated that hydrogels can create interfacial adhesion between bone and various implant materials, thereby reducing the implants loosening. Of note, the OCC₂L₅ hydrogel also has high adhesion to skin, which can tightly bond two pieces of pig skin and withstand pulling and tugging forces exerted by 120 g weights (Fig. 2c). In addition, the hydrogel can be bonded to plastic and metal, and pulled up to 500 g. Since the microenvironment of the bone defect is wet, it is essential to observe the underwater adhesion of the hydrogel. In Fig. 2d, whether the hydrogel adheres first to the wetted bovine bone or directly to the bovine bone underwater, the bovine bone block achieves a tight bond underwater. These results demonstrated the adhesion efficacy of OCCL hydrogels, especially OCC₂L₅ hydrogel, in mediating strong implant integration even under wet conditions, thus exhibiting great potential for mitigating implant loosening.

Fig. 2.

a) Tensile adhesive strength of different hydrogels between bovine bones. (∗∗∗p < 0.001, ∗p < 0.05). b) Photographs of the adhesion of OCC₂L₅ hydrogel on various material surfaces. c) Hydrogel can support weight after adhering to different materials. d) Demonstration of underwater adhesion through different types. e) degradation behavior of hydrogels. f) ROS scavenging capacity and g) ·OH scavenging ability of hydrogels.

The degradation properties of hydrogels are critical for the growth of tissue into the implants. Given that chitosan undergoes enzymatic hydrolysis by lysozyme in vivo, the degradation behavior of the chitosan-based hydrogels was assessed in vitro using a lysozyme-containing buffer [34]. As shown in Fig. 2e, all hydrogels exhibited degradable behaviour in the lysozyme solution. At 28 d, the degradation rates of OCC₂, OCC₂L₁, OCC₂L₂ and OCC₂L₅ hydrogel groups were 16.4 %, 19.5 %, 25.3 % and 31.4 %, respectively. These results indicate that OCCL hydrogels have biodegradable potential. Meanwhile, the degradation products of the hydrogel, including chitosan-derived oligosaccharides and ions (Si⁴⁺, Mg²⁺, Li⁺) released from LAP, are generally considered beneficial for supporting cell proliferation [16,35].

Additionally, since the OCCL hydrogels were formed by the oxidation of NaIO₄, their final antioxidant activities require to be verified (Fig. 2f and g). It was observed that the ROS scavenging rates of the OCC₂-based hydrogels were all greater than 77 %. While the ·OH radical scavenging rates of the hydrogels increased with the increase of LAP content, possibly due to the fact that the introduction of LAP achieves a limited oxidation of catechol, resulting in an increase in the ·OH radical scavenging capacity of the hydrogels. These results render the retention of certain antioxidant activities in the OCC₂-based hydrogels after gelatinization.

3.3. Antibacterial activity of OCCL hydrogels

Bacterial infection is a main cause of bone repair failure, which requires surgical debridement and antibiotic treatment, resulting in multiple surgeries and adverse antibiotic reactions [36]. Therefore, implants inserted into defect sites must ensure minimal infection risks so as to provide a stable healing microenvironment and expedite bone regeneration. Herein, we chose Staphylococcus aureus (S. aureus, Gram-positive) and Escherichia coli (E. coli, Gram-negative) to evaluate the antimicrobial activity of the hydrogels. In Fig. 3a, compared to the control group, the number of colonies for all OCC₂-based hydrogels decreased significantly. The results were further verified by quantitative analysis, and there was no significant difference between the OCC₂-based hydrogel groups (Fig. 3b). Notably, the survival rates of S. aureus and E. coli after OCC₂L₅ hydrogel treatment were 0.5 % and 0.3 %, respectively. These results suggest that all OCC₂-based hydrogels have excellent antibacterial activity, presumably attributable to the important role of OCC₂.

Fig. 3.

a) Representative images of colony-forming units against S. aureus and E. coli after treating with different hydrogels and following 24 h incubation on LB agar plates at 37 °C. b) Bacterial viabilities of E. coli and S. aureus. SEM images of c) S. aureus and d) E. coli induced by different hydrogels. e) The protein leakage and f) MDA oxidation from S. aureus and E. coli after different treatments. g) Possible antibacterial mechanism of the OCCL hydrogels against bacteria. ∗∗∗p < 0.001.

Further, the possible sterilization mechanism of the hydrogel was discussed. The bacteria in OCC₂-based hydrogel groups displayed obvious shrinkage, deformation and fracture by SEM observation, while in the control group showed clear edges and complete smooth morphology (Fig. 3c and d). From the bicinchoninic acid (BCA) protein assay, the OCC₂L₅ hydrogel exhibited higher protein leakage than OCC₂ hydrogel, which may be attributed to the limited oxidation of catechol groups (Fig. 3e). Moreover, these groups showed peroxidation of bacterial membrane lipids, resulting in high MDA content (Fig. 3f). Meanwhile, the cell membrane permeability of E. coli increased in OCC₂ hydrogel and OCC₂L₅ hydrogel groups (Fig. S5). Collectively, upon contact-active with the OCC₂-based hydrogels, the bacteria suffered membrane damage and lipid peroxidation, which led to the leakage of intracellular contents such as proteins and the death of the bacteria. (Fig. 3g). This could be attributed to several possible mechanisms: i) The electrostatic interaction between the protonated amino group in CC and the bacteria destroys the cell membrane. ii) The functional groups of CA after oxidation (benzoquinones, etc.) and the remaining CA have intrinsic antibacterial activity, which seems to be related to their ability to induce a permeability change of membranes and disturb the physiological metabolism [[37], [38], [39], [40]]. Such antibacterial mechanisms endow the hydrogel with the ability to prevent postoperative or implant-associated bone infections, thereby providing a favorable microenvironment for bone regeneration.

3.4. In vitro antioxidant and osteogenic ability of OCCL hydrogels

Next, we sought to explore the impact of OCC₂ hydrogel and OCC₂L₅ hydrogel on the proliferation of r-BMSCs. In Fig. 4a, OCC₂ and OCC₂L₅ hydrogel groups showed high cell viability for 1 d, and significantly promoted the r-BMSCs proliferation after 3 and 5 d. Notably, the addition of LAP demonstrated superior cell proliferation. The results further reveal that r-BMSCs exhibit excellent spreading ability on the surface of OCC₂L₅ hydrogel while maintaining their cellular morphology intact, thereby indicating the remarkable biocompatibility of OCC₂L₅ hydrogel and facilitating cell adhesion (Fig. S6), which may be benefited by the introduction of LAP [41]. Then, the antioxidant activity of hydrogels was further evaluated in vitro. The intracellular ROS was visualized using DCFH-DA probe (a marker for ROS detection) (Fig. S7). The signal of ROS was significantly upregulated in H₂O₂-treated r-BMSCs compared to the control group. After treating with OCC₂ hydrogel and OCC₂L₅ hydrogel, there was a gradual decrease in ROS levels, with the lowest level observed in the OCC₂L₅ hydrogel group. JC-1 staining was utilized to assess the impact of H₂O₂ treatment on mitochondrial function in r-BMSCs. As shown in Fig. 4b,c, H₂O₂ can decrease the mitochondrial membrane potential, while both OCC₂ hydrogel and OCC₂L₅ hydrogel can mitigate the loss of mitochondrial membrane potential. Meanwhile, OCC₂L₅ hydrogel exhibited most favorable role in maintaining mitochondrial function. The results suggested that the OCC₂ hydrogel exhibited effective scavenging of ROS and enhanced mitochondrial function to avoid oxidative stress, thereby demonstrating favorable antioxidant properties, which were similar to Fig. 2h–g. Furthermore, the limited oxidation of CC by LAP can effectively augment the elimination of intracellular ROS. This antioxidant capacity is particularly valuable for mitigating oxidative stress–induced osteoblast dysfunction and inflammation in the bone repair microenvironment, suggesting the potential of this hydrogel in improving bone regeneration.

Fig. 4.

In vitro biocompatibility, antioxidant properties, and osteogenic potential of the OCC₂ and OCC₂L₅ hydrogel. a) Cell viability of OCC₂ and OCC₂L₅ hydrogel for 1, 3, and 5 d. b, c) Representative images and quantification of the JC-1 staining. d) ALP staining of r-BMSCs in 7 d, representative images of ARS staining of r-BMSCs in 14 d. e,f) The mRNA expression levels of ALP, BMP-2, COL-1, OCN, OSX and RUNX2 in 3 and 7d. g) Representative immunofluorescence images of RUNX2, COL-1 and BMP-2 in each group. h) Western blot analysis of COL-1, RUNX2, BMP-2 and OCN protein expression levels in each group. ∗∗∗p < 0.001, ∗∗p < 0.01, ∗p < 0.05, n.s.: not significant (P > 0.05).

To comprehensively assess the biocompatibility and osteogenic potential of the hydrogels, both rBMSCs (representing early osteogenic differentiation) and MC3T3-E1 pre-osteoblasts (representing later stages of osteogenic maturation) were used in this study. As previously reported, LAP can effectively induce osteogenic differentiation, primarily by enhancing ALP activity, upregulating Runt-related transcription factor 2 (RUNX2) expression, and promoting the deposition of osteoblast-related proteins like Bone Morphogenetic Protein 2 (BMP-2), Collagen-I (COL-1), and Osteopontin (OPN), thus laying the foundation for subsequent mineralization [17]. The osteogenic differentiation-promoting ability of OCC₂ and OCC₂L₅ hydrogel was assessed through ALP staining, ARS staining, and qRT-PCR analysis in r-BMSCs. The expression of ALP serves as a crucial indicator for early assessment of osteogenesis, and its high activity level acts as an early marker promoting the differentiation and maturation of osteoblasts [42]. The results reveal that after 7 d of co-culture with r-BMSCs, the OCC₂ hydrogel exhibited positive ALP staining and the OCC₂L₅ hydrogel displayed more intense ALP staining (Fig. 4d). ARS staining was utilized to assess the osteogenic differentiation-promoting ability of OCC₂ and OCC₂L₅ hydrogel on r-BMSCs by highlighting calcium nodules in red. Compared to OCC₂ hydrogel group, OCC₂L₅ hydrogel group exhibited darker and more intense ARS staining, indicating a stronger mineralization capacity with higher levels of calcium nodule production and aggregation (Fig. 4d). These findings indicate that the hydrogel supports both early and late stages of osteogenic differentiation, providing a more complete evaluation of its osteogenic potential.

The expression levels of osteogenesis-related genes, including ALP, BMP-2, COL-1, OCN, OSX and RUNX2, were assessed via qRT-PCR. Our findings indicate that treatment with OCC₂ and OCC₂L₅ hydrogel for 3 and 7 d resulted in a significant upregulation of these genes compared to the control groups. The superiority of OCC₂L₅ hydrogel in promoting osteogenic gene expression was evident over that of OCC₂ hydrogel (Fig. 4e and f), which is consistent with the above analysis. Subsequently, the high-content imaging system revealed that OCC₂L₅ hydrogel demonstrated significant advantages over OCC₂ hydrogel in enhancing the expression levels of RUNX-2, COL-1, and BMP-2 (Fig. 4g). According to the western blot results, the OCC₂L₅ hydrogel exhibited a significant enhancement in osteogenic indicators (COL-1, RUNX-2, BMP-2 and OPN) compared with OCC₂ hydrogel, demonstrating the osteoinductive capacity of LAP (Fig. 4h), as previously reported [43]. Moreover, the osteogenic effects of OCC₂ hydrogel and OCC₂L₅ hydrogel in MC3T3-E1 cells were also validated using ALP staining, ARS staining and qRT-PCR analysis, which is in accordance with the results of r-BMSCs (Fig. S8).

Together, these results indicate that the OCC₂-based hydrogel possesses a certain antioxidant and osteogenic ability, while the addition of LAP results in an obviously enhanced effect of the nanocomposite hydrogel, proving the synergistic effect of LAP and OCC₂-based hydrogel in antioxidant and osteogenic differentiation.

3.5. In vitro angiogenic ability of OCCL hydrogels

Angiogenesis is an important biological process to promote bone regeneration [44,45]. The formation of new blood vessels is an important process to provide oxygen and nutrients to the local defect site to promote bone regeneration [46]. The previous studies have demonstrated that various LAP-based synthetic materials exert substantial effects in promoting angiogenesis [35,47]. To investigate the impact of OCC₂ and OCC₂L₅ hydrogel groups on angiogenesis, we used extracts from these groups to co-culture with HUVECs. Our study found that only part of angiogenesis and mostly incomplete blood vessels could be observed in the control and OCC₂ hydrogel group. After the treatment of OCC₂L₅ hydrogel, a more complete tubular structure and a clearer vascular structure could be observed (Fig. 5a). Subsequently, the parameters related to angiogenesis, including vessel percentage area, total number of junctions and average vessel length, were quantified (Fig. 5b–d), which confirmed the significant advantage in promoting angiogenesis of OCC₂L₅ hydrogel.

Fig. 5.

OCCL hydrogels promote angiogenesis in vitro. a) Representative images of HUVECs cultured with different hydrogels. b) Vesseles percentage area (%) in different groups. c) Total number of junctions in different treatments. d) Average vessel length in each group. Relative angiogenic gene expression (VEGF, TGF-β, HIF-1α, bFGF, CD31, and ANG1) after e) 3 d and f) 7 d assayed by qRT-PCR in different groups. g) Schematic illustration of angiogenesis relative gene induced by OCCL hydrogels. ∗∗∗p < 0.001, ∗∗p < 0.01, ∗p < 0.05, n.s.: not significant (P > 0.05).

Further, the expression levels of angiogenesis-related genes were assessed by qPCR in HUVECs. After 3 d, there was no significant increase observed in the expression levels of Angiopoietin 1 (ANG1), CD31, basic fibroblast growth factor (bFGF), and vascular endothelial growth factor (VEGF), the OCC₂ hydrogel exclusively enhanced the expression of Transforming Growth Factor-beta (TGF-β) and hypoxia inducible factor-1α (HIF-1α). In contrast, the OCC₂L₅ hydrogel significantly enhanced the expression levels of VEGF, TGF-β, HIF-1α, bFGF, CD31 and ANG1 (Fig. 5e). After 7 d treatment, the expression levels of the above genes were significantly elevated in the OCC₂L₅ hydrogel group compared to both the control group and OCC₂ hydrogel group (Fig. 5f and g). These findings suggest that OCC₂L₅ hydrogel exhibits a pronounced potential to promote angiogenesis in HUVECs.

3.6. Potential mechanisms underlying the promotion of osteogenesis and angiogenesis by OCCL hydrogels

To further elucidate the underlying mechanism by which OCC₂L₅ hydrogel promote osteogenesis in r-BMSCs, we utilized RNA sequencing to compare differential gene expression between the OCC₂ hydrogel group versus control group and OCC₂L₅ hydrogel group versus control group, and 433 genes were shared between OCC₂ hydrogel group versus control group and OCC₂L₅ hydrogel group versus control group (Fig. 6a). The volcano plot clearly illustrates the significant differential gene expression in OCC₂ hydrogel group versus control group and OCC₂L₅ hydrogel group versus control group. OCC₂ hydrogel group and OCC₂L₅ hydrogel group exhibited 1090 and 586 differentially expressed genes compared to the control group. The red dots indicate genes that are up-regulated, while the green dots represent genes that are down-regulated (Fig. 6b and c). In light of the preceding findings, our focus will be on elucidating the potential mechanism by which OCC₂L₅ hydrogel facilitates osteogenic differentiation of r-BMSCs. In the heat map, compared with the control group, the expression of BMP6, RUNX1, HIF-1α, VEGF and other key genes related to promoting osteogenesis and angiogenesis was significantly up-regulated in OCC₂L₅ hydrogel group (Fig. 6d). Subsequently, Gene Set Enrichment Analysis (GSEA) reveals that treatment with OCC₂L₅ hydrogel significantly enhances osteogenesis and neovascularization by activating the MAPK and Wnt signaling pathways as well as the HIF-1α and VEGF signaling pathways (Fig. 6e). Gene Ontology (GO) enrichment analysis was performed and molecular functional analysis revealed that the differentially expressed genes of OCC₂L₅ hydrogel enriched in extracellular matrix, growth factors and Wnt protein binding. In the analysis of biological processes, compared with control group, the differentially expressed genes of OCC₂L₅ hydrogel group were mainly enriched in angiogenesis, ossification, cell migration, adhesion and other biological processes related to osteogenesis (Fig. 6f and g). The presence of Mg²⁺, Li⁺, and Si (OH)₄^- in LAP enables them to exert control over a diverse range of cellular processes when degraded [26]. Specifically, Mg²⁺ plays a crucial role in promoting cell adhesion by enhancing interactions with integrin biomolecules. Li ⁺ stimulates osteogenesis through the activation of Wnt signaling pathways. Lastly, the released Si and Si (OH)⁴⁺ are indispensable for metabolic processes, angiogenesis during bone regeneration, and the calcification of bone tissue. Based on these results, we further examined whether the OCCL hydrogels activate key osteogenic signaling pathways. Western blot analyses revealed that treatment with the OCC₂L₅ hydrogel markedly increased the expression levels of p-p38, p-ERK1/2, and β-catenin in r-BMSCs compared with both the control and OCC₂ groups. These findings indicate that the OCC₂L₅ hydrogel robustly activates the MAPK and Wnt pathways, thereby contributing to its enhanced osteogenic effects (Fig. S9). Thus, the incorporation of LAP into the OCC₂L₅ hydrogel significantly enhanced cellular adhesion, promoted osteogenesis, and facilitated neovascularization.

Fig. 6.

RNA sequencing results of r-BMSCs co-cultured with OCCL hydrogels. a) Venn diagram of differentially expressed genes in OCC₂L₅ hydrogel group versus control group and OCC₂ hydrogel group versus control group. Volcano plot of the differentially expressed genes in b) OCC₂ hydrogel group versus Control group and c) OCC₂L₅ hydrogel group versus Control group. d) Cluster analysis of selected differentially expressed genes of OCC₂L₅ hydrogel relative to control group. e) GSEA-enrichment plots of representative gene sets from OCC₂L₅ hydrogel group and control group. f, g) GO analysis of RNA-seq analysis between the OCC₂L₅ hydrogel and control group.

3.7. In vivo bone regenerative capacity of OCCL hydrogels

Our previous in vitro studies have unequivocally demonstrated the substantial benefits of OCC₂L₅ hydrogel in facilitating osteogenesis and angiogenesis. Subsequently, a rat cranial bone defects model was employed to further assess the pivotal role of OCC₂ hydrogel and OCC₂L₅ hydrogel in promoting osteogenesis. After 6 w of OCC₂ hydrogel treatment, only scattered new bone formation was observed in the defect area, whereas significant and pronounced new bone growth originating from the periphery of the defect area was evident after 6 w of OCC₂L₅ hydrogel treatment, accompanied by a higher quantity of newly formed bones (Fig. 7a). After 12 w, it was observed that the OCC₂ hydrogel group exhibited the formation of small sheets of new bone from the edge of the bone defect, while in the OCC₂L₅ hydrogel group, a significant amount of new bone grew into the defect area, effectively covering almost the entire defect area. The BV/TV values of the OCC₂L₅ hydrogel group exhibited a significantly higher value compared to that of the OCC₂ hydrogel group at both 6 w and 12 w, while the BV/TV values in both groups were significantly higher compared to those in the control group, suggesting their excellent new bone formation ability (Fig. 7b and c).

Fig. 7.

OCC₂L₅ hydrogel for rat calvarial defect. a) Micro-CT images of the defect repair sites. Statistical analysis of BV/TV in b) 6 w and c) 12 w. d) H&E staining, Masson staining and SO-FG staining images of skull defect after treatment with OCC₂ hydrogel and OCC₂L₅ hydrogel in 12 w. ∗∗∗∗p < 0.0001, ∗∗∗p < 0.001, ∗∗p < 0.01, ∗p < 0.05. e) Comprehensive comparison between the OCC₂L₅ hydrogel of this work and previously reported injectable adhesive hydrogels or commercial adhesive with respect to adhesive strength, osteogenic capability, biodegradability, antioxidant ability and antibacterial property [16,[48], [49], [50], [51], [52]].

The bone defect area was subsequently subjected to histopathological evaluation in order to assess the efficacy of OCC₂L₅ hydrogel in promoting bone defect repair (Fig. 7d, Fig. S10). The results from H&E, Masson and SO-FG staining at 6 w reveal that the control group exhibited solely physiological self-repair, characterized by a substantial presence of fibrotic and granulation tissue within the bone defect area. The OCC₂L₅ hydrogel and OCC₂ hydrogel groups exhibited a limited amount of new bone formation (Fig. S10). The H&E, Masson and SO-FG staining in 12 w demonstrate that the OCC₂L₅ hydrogel group exhibited substantial new bone formation, effectively filling the bone defects, whereas only minimal new bone generation was observed in the OCC₂ hydrogel group (Fig. 7d). Therefore, the histopathological results once again corroborated the remarkable bone repair potential of OCC₂L₅ hydrogel. Moreover, no abnormal inflammatory infiltration, tissue necrosis, or fibrotic encapsulation was observed around the implantation sites at either 6 or 12 w, indicating that the hydrogel underwent a controllable in vivo degradation process without eliciting adverse tissue responses. Additionally, both OCC₂L₅ hydrogel and OCC₂ hydrogel had superior biosafety, which was demonstrated through histopathological examination of major organs in rats (Fig. S11). This observation is consistent with previous reports on the long-term biodegradability and biological safety of catechol-modified chitosan and Laponite-based materials [16,35]. Taken together with its ability to support angiogenesis and osteogenesis, these properties indicate that the hydrogel is well aligned with the biological requirements of bone-healing tissues and holds strong potential for clinical translation by enhancing bone graft integration in critical defects while also promoting the repair of small or non-load-bearing defects such as those encountered in craniofacial or dental reconstruction.

To more clearly highlight the advantages of the proposed hydrogel, representative hydrogel-based bone adhesive systems reported in previous studies were compared in terms of adhesion, degradation, antibacterial, anti-inflammatory, and osteogenic properties (Fig. 7e). The results indicated that although the adhesive strength of the hydrogel developed in this study was moderate, it exhibited a superior overall balance of multifunctional performance. Combined with its good injectability and shape adaptability, this hydrogel can serve as a multifunctional auxiliary material, effectively enhancing the fixation of non-deformable bone grafts, filling irregular bone defects, and providing a favorable microenvironment for bone regeneration. In addition, the hydrogel components are compatible with clinically accepted sterilization approaches (e.g., γ-irradiation or aseptic manufacturing), supporting its potential for future translational applications [53].

4. Conclusion

In summary, harnessing the nanoconfinement effect of LAP, we developed an injectable, adhesive, multifunctional OCCL hydrogel, which has great potential in assisting bone implant therapy. Specifically, LAP with lamellar structure mitigated NaIO₄-induced excessive oxidation of CC, while enhancing physical crosslinking within hydrogel, enabling injection to conformally fill defects and subsequent in situ gelation with adequate operational time. The remaining CA and the formation of more physical crosslinking (hydrogen bonding and electrostatic interaction) by LAP together reinforced the adhesive strength of OCCL hydrogel. This hydrogel showed excellent wet adhesion, which can mediate strong implant integration under wet conditions. The controlled oxidation also confers antioxidant capabilities to eliminate ROS and relieve oxidative stresses. Moreover, this hydrogel can also kill bacteria through surface contact-active inhibition and prevent infection, while showing good biodegradability to ensure tissue ingrowth. Beyond providing a pro-regenerative microenvironment, the hydrogel markedly stimulated neovascularization and osteogenesis by upregulating expression of osteoblastic and pro-angiogenic markers, thereby profoundly enhancing osseous regeneration within rat cranial bone defects. These multifaceted benefits highlight this chitosan hydrogel's strong potential for comprehensively promoting bone repair, particularly in assisting bone grafts, while maintaining competitive multifunctionality compared with most reported injectable bone adhesives. This work pioneers an approach to strategically regulating the oxidation of catechol-containing polymers, providing new insights for the design of multifunctional bioadhesives toward clinical applications in minimally invasive bone repair and reconstruction.

CRediT authorship contribution statement

Shanshan Li: Writing – review & editing, Writing – original draft, Validation, Funding acquisition, Formal analysis, Data curation. Haichao Yu: Writing – review & editing, Writing – original draft, Validation, Formal analysis, Data curation. Ruirui Guan: Visualization, Validation, Formal analysis, Data curation. Xiaoyun Li: Writing – review & editing, Validation, Resources, Funding acquisition. Junzhao Duan: Visualization, Data curation. Hua Wang: Writing – review & editing, Supervision, Resources, Project administration. Xuesong Zhang: Writing – review & editing, Resources, Project administration. Xiaoying Wang: Writing – review & editing, Resources, Project administration, Funding acquisition, Conceptualization.

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 is the Supplementary data to this article:

Data availability

Data will be made available on request.