English

• Oral ulcers, regarded as one of the most frequent inflammatory diseases, occur on the oral mucosa and cause injury. When the barrier function of the oral mucosa is disrupted, the area progresses from small red dotted patches with stinging to severe inflammatory ulcers covered with a yellowish-white pus film [1,2]. Oral ulcers cause severe burning pain, increased risk of infection, and impaired oral function, which can seriously affect the quality of life. There is no specific treatment option for oral ulcers, and existing treatments aim at relieving pain and accelerating ulcer healing, such as oral medications, sprayed powders, and topical gels or films [3]. However, oral medications have a first-pass effect that reduces drug bioavailability, while sprayed powders are easily dropped and ingested [4]. A gel or film is used as a pseudomembrane to act as a barrier. It is one of the most effective treatments for promoting ulcer healing while blocking saliva and luminal flora from eroding the ulcer wound, reducing pain and isolating the affected area [5]. Hydrogels with natural antimicrobial properties have been widely used as wound dressings [6]. Guo et al. obtained an injectable hydrogel by cross-linking quaternary chitosan with tannic acid, which showed significant wound repairing ability [7]. Cheng et al. prepared a cationic hydrogel with antimicrobial properties by using trans-1,4-cyclohexanediamine and 1,3-dibromo-2-propanol, which provides a useful treatment for difficult-to-heal diabetic wounds [8].

  However, the humid and highly dynamic oral environment poses a challenge for hydrogel applications. The movement of the facial muscles, the friction of the tongue and the flushing of saliva require hydrogels with strong mechanical properties and adhesion that do not easily swell [9]. Furthermore, the initial modulus must not be excessively high to avoid discomfort and secondary harm to the wound from mechanical mismatch. Therefore, a hydrogel that is reinforced in situ and can be injected precisely into the wound surface is favored, which is able to have a relatively low modulus to adapt to the injured tissues when applied to the wound and spontaneously reinforces itself to support the injured tissues when it becomes cemented [10,11].

  Methacrylated gelatin (GelMA) is similar to the natural extracellular matrix and enables cells to be adhered to and proliferate in GelMA-based scaffolds [12,13]. This means that GelMA is biocompatible and can be degraded by proteases in vivo. GelMA hydrogels with tunable mechanical properties and controlled structure are formed by rapid photo-crosslinking of carbon double bonds in the presence of a water-soluble initiator [14]. It has very mild polymerization conditions, which can occur in aqueous or biological fluids at room temperature, and is one of the best solutions for in situ enhancement of injectable gel oral wound dressings. However, an ideal oral wound dressing should have multiple characteristics simultaneously, such as strong adhesion at the moist interface, high mechanical strength and low swelling [15,16]. In order to further expand the utility of GelMA gels, their modification has been explored extensively [17–20].

  Ionic liquids, composed of organic ions paired with inorganic counterions, represent pure molten salts and are considered cutting-edge materials [21,22]. Their properties can be modulated by adjusting the structure of each of these paired anions and cations [23]. Phenylboronic acid ionic liquid (PIL) monomer is a novel monomer obtained by combining imidazole ions with phenylboronic acid [24,25]. The positively charged imidazole groups have an electrostatic effect on the negatively charged bacterial cell membranes, which disrupts membrane permeability and dynamics [8]. This results in inherent antibacterial properties. The phenylboronic acid possessed by PIL reacts with cis-o-diol to form a dynamic boronic ester bond. This bond typically exhibits a range of typical dynamic features, including shear thinning, self-healing and rapid stress relaxation, and has been shown to have excellent scavenging activity against excess reactive oxygen species (ROS) generated by cellular oxidative stress [26].

  Herein, we designed an injectable gel for repairing oral mucosa (GIL2) with rapid gelation and enhanced mechanical strength in situ (Scheme 1). Photocrosslinking of double bonds and esterification reactions were employed to achieve rapid in situ gelation, and dynamic phenylboronic acid ester bonds and imidazole cation were introduced into the bioactive GelMA to endow the gel with intrinsic antimicrobial properties along with anti-inflammatory properties of ROS scavenging. Through dynamic covalent bonding (specifically boronic ester bonds) and physical interactions (including hydrogen bonds and electrostatic interactions), multiple crosslinked networks are integrated, effectively minimizing hydrogel swelling while simultaneously improving adherence to surfaces of biological tissues. The chemical entanglement is regulated by adjusting the amount of PIL added, and the crosslink density is varied to improve the mechanical properties. This hydrogel offers a balance between high performance and multifunctionality, making it ideal for use in wet, active, microbiologically complex oral environments. In sum, the developed GIL2 hydrogel characterized by low swelling, strong adhesion, rapid gelation, and potent bactericidal and anti-inflammatory attributes offers fresh insights for treating oral ulcers.

  Scheme 1

  

  Scheme 1.  Development and utilization of GIL2 hydrogel in treating bacterial-infected oral ulcers.

  DownLoad: Full-Size Img PowerPoint

  To enhance the adaptability of mucosal dressings in the oral environment, a uniquely designed photo-crosslinked hydrogel is presented in this study. In the synthesis of PIL, an alkylation reaction between 4-(bromomethyl)phenylboronic acid (PBA) and 1-vinylimidazole (VL) is employed. Concurrently, GelMA is produced through a direct reaction involving gelatin and methacrylic anhydride (MA) in phosphate-buffered saline (PBS) [27]. The produced GelMA and PIL were confirmed by ¹H NMR (Figs. 1A, B and Fig. S1 in Supporting information) and Fourier transform infrared (FTIR) (Fig. 1C and Fig. S2 in Supporting information), and their peaks from the chemical structures and functional groups were consistent with their theoretical values. The ¹H NMR spectrum of PIL displayed prominent resonance peaks associated with the carbon-carbon double bond of VL at 7.30, 5.99, and 5.42 ppm (Fig. 1A and Fig. S1) [28]. Additionally, VL's specific peaks were observed at 7.94, 8.23, and 9.60 ppm. PBA's characteristic peaks were noted at 8.12 ppm (—OH) and 5.47 ppm (—CH₂), with additional peaks related to the benzene ring at 7.41 and 7.82 ppm [29–31]. In the ¹H NMR spectrum of GelMA, we observed distinct resonance peaks at 5.43 and 5.67 ppm, which represent carbon-carbon double bonds from methacrylic anhydride (Fig. 1B) [32]. This indicates that methacrylic acid ester has successfully combined with gelatin. Additionally, the appearance of a methyl signal at 1.67 ppm further confirms the successful synthesis of GelMA [33].

  Figure 1

  

  Figure 1.  Preparation and characterization of GIL precursor and hydrogel dressing. ¹H NMR spectra of PIL (A) and GelMA (B). (C) Infrared spectrogram of G-gel, GIL1, GIL2 and GIL3 hydrogels featuring varying PIL concentrations. (D) XPS full spectrum of G-gel, GIL1, GIL2 and GIL3 hydrogels, each with different levels of PIL. (E) B 1s and (F) N 1s XPS signal of GIL2. (G) Synthesis reaction of GIL. (H) Photocrosslinking of GIL. (I) Rheology characterizations of GIL. (J) EDS elemental mapping of GIL2 and SEM images of GIL hydrogels.

  DownLoad: Full-Size Img PowerPoint

  To obtain mucosal dressings with high viscosity, low swelling and excellent mechanical properties, three main structural designs have been used to build hydrogel multifunctional interpenetrating networks: (ⅰ) formation of dynamic boronic acid ester bonds via the interaction between phenylboronic acid groups and free hydroxyl groups in gelatin chains (mainly from tyrosine or serine); (ⅱ) hydrogen bonding and electrostatic interactions of imidazolium cations with the other constituents; and (ⅲ) polymerization of vinyl groups in the PIL with GelMA under ultraviolet (UV) irradiation to induce the strong binding of functional monomers in the hydrogel network. Fig. 1G provides a visual representation of the network formed between GelMA and PIL. The network includes a variety of interactions, such as the covalent and hydrogen bonds described above. They facilitate the enhancement of the mechanical properties of the gel, making it well-suited for wound protection in the oral environment [34,35]. We synthesized a variety of samples with different ratios of gel precursors, and these dressings were named G-gel, GIL1, GIL2, and GIL3 according to the PIL content.

  The chemical composition of GIL hydrogel was determined using FTIR spectroscopy, as shown in Fig. 1C. Each sample displayed distinct bands of the gelatin structure at 1630, 1526, and 1230 cm⁻¹. These bands represent C═O stretching (amide Ⅰ), N—H bending (amide Ⅱ), and C—N stretching coupled with N—H bending (amide Ⅲ) [14]. GIL revealed significant peaks characteristic of PBA at 1330 cm⁻¹ (B—O), 1410 cm⁻¹ (C—B), and 1610 cm⁻¹ (aromatic ring vibrations) [24,36]. Moreover, stretching vibrations of the imidazole ring were noticeable at 3060 cm⁻¹ (C—H), 1501 cm⁻¹ (C—N), and 1650 cm⁻¹ (C═C and C═N) [24,37]. Although all characteristic peaks shifted significantly, all the characteristic peaks of PIL and GelMA appeared in GIL. Employing X-ray photoelectron spectroscopy (XPS), we discovered that the B1s molar percentage rose to 2.34% in GIL1, verifying PIL's integration into the G-gel hydrogel network (Fig. 1D). Based on the high resolution XPS spectra of GIL2, it was observed that new signals of B—O—C, B—OH, C═N and B—C appeared (Figs. 1E and F) [38]. These signals indicate that there are dynamic interactions between the phenylboronic acid groups within PIL and the hydroxyl groups on GelMA.

  Based on photoinitiated cross-linking of carbon-carbon double bonds, the GelMA-PIL mixture could be rapidly converted into a gel within 3 s under UV light (30 mW/cm²) (Figs. 1H and I). Rheological examinations revealed that at a PIL concentration of 1.5 wt% (GIL3), the hydrogel's gelling time reduced to 1.55 s, while its storage modulus (G′) escalated quickly, surpassing the loss modulus (G″). Throughout these tests, the hydrogels consistently displayed a G′ greater than G″, signifying their predominantly elastic nature. Not only that, the G′ value of G-gels increased from 10.409 kPa to 115.477 kPa with increasing PIL content (Fig. 1I). This is possibly due to the interaction between G-gel and PIL molecules, along with the limited mobility of polymer chains within the hydrogel network. The GIL hydrogels were constructed using a photo-initiated method and then freeze-dried. Scanning electron microscope (SEM) was used to analyze the surface texture and micro-porosity of the hydrogels. The GIL series hydrogels all exhibited a porous three-dimensional lattice structure (Fig. 1J). This structure provides an optimal setting for cell proliferation and the exchange of nutrients [39]. Energy-dispersive X-ray spectroscopy (EDS) images revealed boron within the synthesized hydrogels, further validating the effective fabrication of GIL2 (Fig. 1J).

  In view of the frequent movements of the oral mucosa in daily life, an ideal adhesive hydrogel should exhibit appropriate tensile properties. Owing to the brittleness of GelMA hydrogels, they struggle to endure tensile testing, but the addition of PIL moderately enhances the hydrogel's strain and tensile strength (Fig. S3 in Supporting information). As shown in Fig. 2, the tensile strength of GIL3 reached 56.83 kPa, attributed to the high-density crosslinking of PIL with GelMA (Fig. 2F). According to the energy dissipation mechanism, stress disrupts the sacrificial bonds (hydrogen bonds in hydrogels) while preserving the chains, thereby enabling the polymer network to withstand fracture and extend [40,41]. We refer to the standard lap shear strength test (ASTM F2255) for experimental evaluation of the adhesive properties of GIL (Fig. 2A). The discovery was made that upon adding PIL, adhesive shear strength escalated from 23.93 kPa to 63.38 kPa (Figs. 2D and E). Increased addition of PIL contributes to hydrogel adhesion owing to its ability to form hydrogen bonds with amino acid residues (primary amines and carboxylic acids) in mucous membranes [42]. GIL3 exhibits a notably higher adhesive strength compared to commercial adhesives such as fibrin glue, which has an approximate strength of 4 kPa. As shown, GIL has excellent adhesion to animal tissues and inorganic materials (Fig. 2B). Moreover, stress-strain curves depicting the compressive characteristics of the hydrogels were evaluated. With PIL incorporation, both maximum stress and compressive modulus experienced an increase (Fig. S4 in Supporting information).

  Figure 2

  

  Figure 2.  Performance of adhesion and mechanical characteristics of GIL hydrogels. (A) Conventional testing approach for evaluating adhesive strength properties in lap-shear through tension loading (ASTM F2255). (B) Adhesion of GIL2 to various biological tissues and inorganic materials. (C) Swelling conditions of G-gel, GIL1, GIL2 and GIL3 hydrogels in PBS buffer. (D) Force-displacement curves and (E) shear strength of G-gel, GIL1, GIL2 and GIL3 hydrogels. (F) Tensile strength and (G) Water uptake ratio of G-gel, GIL1, GIL2 and GIL3 hydrogels. Data are presented as the mean ± standard deviation (n = 3).

  DownLoad: Full-Size Img PowerPoint

  The low-swelling properties are essential to ensure a tight adhesion to the damaged oral mucosa for an extended period. Moderate swelling aids hydrogels in absorbing wound exudate, thus diminishing fluid accumulation at the interface and securing a firm adherence of the patch to the tissue surface [43–46]. We evaluated the swelling characteristics of the hydrogel dressings (Fig. 2C). As shown by the water uptake curves in Fig. 2G, the GIL series of hydrogels showed similar water absorption behavior. A more pronounced swelling phase occurred during the first 250 min, followed by a gradual equilibrium in 500 min. The water uptake ratio of G-gel reached 182%, while that of GIL3 was only 107% (Fig. 2G). Compared to G-gel, GIL retained its original shape and volume well after 24 h. The transparent GelMA hydrogels became translucent GIL hydrogels due to the addition of PIL. The reduction in transparency, swelling, and pore size of GelMA hydrogels was credited to the substantial cross-linking impact brought about by PIL. The results confirmed that GIL hydrogels could swell in an aqueous environment and maintain good mechanical properties while maintaining an optimal humidity environment for oral ulcer healing.

  As the concentration of added PIL increases, the hydrogel's mechanical and adhesive characteristics enhance. However, determining the most suitable level of addition requires screening for cytotoxicity (Fig. 3A). Firstly, the cell viability and proliferation of human gingival fibroblasts (HGFs) were evaluated using the cell counting kit-8 (CCK-8) assay to assess the GIL hydrogel. As shown in the graph, even at higher doses of PIL (600 µg/mL), the cell viability of HGFs remained above 97% when co-cultured with G-gel, GIL1, and GIL2 hydrogels for 1 day, 3 days, and 5 days. This condition indicates that hydrogels have low cytotoxicity, and the OD values demonstrated significant proliferation of HGFs in all hydrogel groups (Fig. 3B). The results were also corroborated by the live/dead cell staining experiment (Fig. 3A and Fig. S5 in Supporting information). In this process, HGF cells were stained with the Calcein-AM/propidium iodide (PI) cell viability probe, where calcein-AM colors live cells green and PI dyes dead cells red. There was a significant increase in green fluorescence after 5 days compared to 1 day, implying a significant increase in cell number. There was minimal cell death during the process, and the fluorescence of dead cells was barely visible. Therefore, GIL2 is considered a balanced formulation with good mechanical properties and cell compatibility, making it the optimal choice for future studies on oral mucosal repair. Therefore, in the follow-up work, we focused on the feasibility and superiority of GIL2 for oral ulcer restoration.

  Figure 3

  

  Figure 3.  Evaluation of toxicity, hemolytic, and ROS scavenging activity of G-gel, GIL1, GIL2 and GIL3 hydrogels. (A) Calcein-AM/PI staining of HGFs post-incubation with hydrogels. Scale bar: 100 µm. (B) CCK-8 assay outcomes for cytotoxicity: absorbance results of G-gel, GIL1, GIL2, and GIL3 hydrogels on HGF cells. (C) Hemolysis tests for GIL series with corresponding photos inset. (D) DPPH and PTIO scavenging of G-gel, GIL1, GIL2 and GIL3 hydrogels. (E) Relative DCFH-DA fluorescence ratios of HGFs without H₂O₂ and hydrogel treatment (control), H₂O₂ treatment (H₂O₂), and H₂O₂ and hydrogel co-treatment (G-gel or GIL2). (F) Assessment of intracellular ROS reduction by hydrogels, using DCFH-DA (green fluorescence). Scale bar: 100 µm. (G) Assessment of GIL hydrogels’ antibacterial properties. (a) Images showing E. coli and S. aureus. (b) Typical CLSM photos of E. coli and S. aureus stained with SYTO9 (green) and PI (red) (scale bar: 10 µm). (c) SEM photographs of E. coli (scale bar: 1 µm) and S. aureus (scale bar: 500 nm) post-incubation with PBS and GIL2 hydrogels (1 mL, 1.0 × 10⁸ CFU/mL). Data are presented as the mean ± standard deviation (n = 4).

  DownLoad: Full-Size Img PowerPoint

  Limited differentiation of oral keratinocytes and pro-inflammatory responses accelerate wound healing in oral ulcers [47]. However, the intense inflammatory response disrupts the body's oxidative-antioxidant balance, which in turn triggers the formation of free radicals and causes cellular damage. Some patients with oral ulcers have impaired enzymatic antioxidant defense systems [48–50]. Thus, oral ulcer repair gels need to have antioxidant capacity [51]. DPPH and PTIO probes were employed to evaluate the antioxidant activity of the GIL specimens (Fig. 3D). GIL demonstrated enhanced radical scavenging abilities compared to G-gel, potentially due to the presence of dynamic borate ester bonds [26]. The GIL hydrogels were further evaluated for their antioxidant activity at the cellular level by employing a DCFH-DA fluorescent probe to observe the production of intracellular ROS in response to H₂O₂. As depicted in Figs. 3E and F, the mitochondrial ROS content in healthy cellular environments was nearly negligible in HGFs. In contrast, exposure to H₂O₂ resulted in a substantial enhancement in the level of mitochondrial ROS in HGFs. G-gel and GIL2 hydrogels both demonstrated effective suppression of intracellular ROS, with cells subjected to GIL2 treatment exhibiting the faintest green fluorescence and minimal intracellular ROS levels (Figs. 3E and F). The observation indicates that GIL2 has significant ROS scavenging performance and can shield cells from oxidative stress-induced peroxidative harm.

  The lamina propria is richly supplied with blood. Therefore, the hydrogel is expected to induce minimal or no hemolysis upon contact with bleeding mucosal wounds. Hemolysis tests evaluate the hemocompatibility of these tailored hydrogels. As shown in Fig. 3C, solutions from all hydrogel groups appeared colorless and transparent, similar to the PBS group. The hemolysis rate for each hydrogel was under 5%, showcasing the exceptional hemocompatibility of both G-gel and GIL.

  The healing of mouth ulcers is susceptible to bacterial contamination. Investigating the antimicrobial properties of GIL2 hydrogel constitutes significant research. For assessing its antibacterial activity, model strains Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus) were utilized. Fig. 3G-a demonstrates that, compared to the control, colonies of S. aureus and E. coli were nearly undetectable in the medium with GIL2 hydrogel, suggesting an antibacterial rate approaching 100%. Data from the live/dead bacterial staining corroborated the plate findings, showing that the group treated with the gel displayed a higher amount of red fluorescence, indicative of dead bacteria, in contrast to the control group (Fig. 3G-b). Next, changes in the morphology of E. coli and S. aureus were observed by SEM (Fig. 3G-c). Standard E. coli and S. aureus were spherical and rod-shaped, respectively. The smooth and undamaged surfaces of the membranes before treatment were rough and irregular after GIL2 treatment, indicating that the membranes were severely damaged [52]. The above phenomenon proves that GIL2 hydrogel is effective in disrupting the bacterial membrane of S. aureus and E. coli so that part of their contents can escape and thus play a bactericidal role. The auto-antimicrobial effect of GIL hydrogels mainly comes from PIL. The modified phenylboronic acid group on GelMA can form reversible covalent bonds with the diol molecules on bacterial glycoproteins for the purpose of trapping bacteria. In addition, the electrostatic interaction of the positively charged imidazole moiety with the negatively charged bacterial membrane can serve as an effective means of killing the bacteria. The powerful antimicrobial effect is realized through the dual action of capturing and killing.

  A rat severed tail model was established for further validation of the hemostatic properties of the hydrogel (Fig. 4A). The Institutional Animal Care and Use Committee at Wenzhou Medical University approved the animal studies according to its regulations and guidelines under the approval code wydw2022–0682. In the experiment, rats receiving PBS treatment served as the blank control group. Meanwhile, the efficacy of GIL material was contrasted with that of a commonly utilized hemostatic substance, gauze. Fig. 4B and Fig. S6 (Supporting information) illustrate the efficacy of GIL2 (0.72%) in halting bleeding, with the bleeding area's percentage being roughly a third of that in the gauze group (2.56%) and a fifth compared to the blank group (3.53%). Hydrogel adheres to oral wounds, rapidly absorbs oozing blood and tissue fluids and acts as a physical barrier. Secondly, coagulated fibrinogen, plasma proteins, erythrocytes and platelets are brought together by the copious protonated amine groups in the polymers, thus triggering the coagulation process at the onset of hemostasis [53,54]. The outcomes of the in vivo hemostasis examination align with those of the in vitro hemolysis assessment.

  Figure 4

  

  Figure 4.  The oral ulcer healing of GIL hydrogels in rats. (A) Illustrative schematic showing the construction process for a hemostasis model in a rat tail. (B) Images depicting the results of hemostasis in rats post-treatment with various groups. (C) The progression of oral ulcer healing in rats over time following treatment with PBS, lidocaine, G-gel, and GIL2, respectively. (D) In vivo experimental procedure for the modeling and treatment of oral ulcers in rats (scale bar: 5 mm). (E) Food intake of rats during the experiment. (F) Analysis of ulcer wound area. (G) Histologic examination of areas of oral ulcers in rats: Photographic representation of H&E (a) and Masson (b) staining (scale bar: 1 mm), along with immunochemical staining for inflammatory markers TNF-α (c) and angiogenesis factors (CD31; d) (scale bar: 100 µm) at the ulcer region in rats after 7-day treatment with various materials. (H) Analysis of the collagen index quantitatively in stained image representations. (I) TNF-α and CD31 measured in tissues seven days after treatment. Data are presented as the mean ± standard deviation (n = 3).

  DownLoad: Full-Size Img PowerPoint

  A thorough assessment of GIL2 hydrogel's characteristics revealed that this hydrogel possesses outstanding stability in volume (minimal swelling), properties for antimicrobial hemostasis, and is biocompatible. Therefore, GIL2 has great potential for the treatment of oral ulcers. A rat oral ulcer model was established using chemical cautery (40% acetic acid). The effects of PBS (control set), commercial hydrogel (lidocaine set) and GIL antibacterial hydrogels (G-gel and GIL2 sets) on ulcer surface healing and daily food intake of rats in each group were assessed throughout the experiment (Figs. 4C and E). Initially, on day 0, rats’ oral mucosa across all groups exhibited a consistently smooth and pink hue. By day 1, each group's rats developed oral ulcers, either round or oval, measuring about 0.5 mm across. These ulcers were coated with a yellow pseudomembrane and featured a sunken center, distinct edges, and mucosa that appeared reddish and swollen. Statistical analysis revealed no significant differences between the sets. Following the establishment of the model, group-based treatments were administered to the rats. Over time, the wound areas of each group gradually decreased (as shown in Fig. 4D). The ulcer wound areas after treatment were quantitatively analyzed, and the results are shown in Fig. 4F. On the 4^th day, the ulcer wound areas in the PBS group (0.514 cm²), lidocaine group (0.683 cm²), and G-gel group (0.631 cm²) were considerably larger than that in the GIL2 set (0.431 cm²). In the GIL2 group, there was a notable decrease in the area of ulcer wounds and a faster rate of re-epithelialization between days 5 and 7. On day 7 of treatment, the ulcer was almost completely healed (0.035 cm²). Therefore, GIL2 proves to be more efficacious in facilitating the healing process of oral ulcers.

  We used mouse wound tissue sections for hematoxylin-eosin (H&E) staining and immunohistochemical analysis to further explore the biological mechanism involved in the healing process. Four days after ulcer treatment, microscopic images of H&E-stained tissue sections showed that neoepithelial formation was more complete on the surface of G-gel and GIL2-treated oral mucosal wounds (defects of 1.57 mm and 0.37 mm, respectively), whereas H&E staining of the PBS group (3.65 mm) versus the lidocaine group (2.20 mm) still showed obvious oral mucosal defects (Fig. 4G-a), suggesting that GIL2 use positively promotes the regeneration of oral ulcer epithelium. Furthermore, collagen is an essential component that promotes cell adhesion, motility, proliferation and deposition on neoplastic connective tissue. Collagen induces fibroblast chemotaxis and accelerates tissue repair [55]. Therefore, following four days of treating the ulcerated area, the excised tissues underwent Masson's trichrome staining. This process was essential for determining collagen levels, serving as a measure to evaluate the therapeutic impact (Figs. 4G-b and H). Blue staining intensity and coverage revealed minor variations in collagen accumulation among the experimental groups. However, the most substantial collagen deposition occurred in the GIL2 group, registering at 35.54%. This percentage surpassed that of the prevalent commercial oral ulcer dressing, lidocaine, which stood at 32.75%. These findings indicate the superior efficacy of GIL2 in enhancing the healing of oral ulcers. In addition, GIL2-treated wounds showed denser collagen fibers in the extracellular matrix, which contrasted with other samples showing loose collagen fibers. Mouth ulcers, a type of inflammatory disease, progress due to several factors, including tumor necrosis factor-alpha (TNF-α) [56]. This element triggers chemokines release from adjacent fibroblasts, subsequently attracting lymphocytes to the affected area. Such activity results in the mucosal epithelium's detachment, culminating in the formation of ulcers.

  Immunohistochemical staining was performed to detect the level of inflammation-associated cytokine TNF-α, and the results are shown in Figs. 4G-c and I. Relative to the control group (143 pg/mL), TNF-α expression levels showed a reduction across other groups, exhibiting the most significant decline in GIL2. GIL2 hydrogel exhibits a potent anti-inflammatory impact that relieves the inflammatory action accompanied by the onset of mouth ulcers. Staining and quantitative analysis of endothelial cell adhesion molecule-31, an angiogenic marker [57], showed that GIL2 promoted the generation of CD31 in traumatized tissues (745%), with a significant increase in horizontal vascular density, compared with the almost non-expression (100% and 184%) in the first two groups (Figs. 4G-d and I). The observed experimental results demonstrate that combining GelMA hydrogel with ionic liquid effectively lessens inflammation and accelerates re-epithelialization, presenting a unique method for enhancing the healing of oral ulcers.

  In conclusion, this study proposes a rapidly gelatable and in situ reinforced injectable hydrogel for treating oral ulcers. The high-performance new ionic liquid monomers are introduced into highly bioactive GelMA through a safe and easy photopolymerization and esterification reaction. The resulting GelMA-ionic liquid (GIL2) quickly forms a gel in just 3 s and exhibits excellent injectability, making it well-suited for irregular oral ulcer wounds. GIL2 hydrogel exhibited strong adhesive strength (63.38 kPa) and high toughness (the tensile strength reached 56.83 kPa). The hydrogel also demonstrated satisfactory antimicrobial activity against both E. coli and S. aureus strains. Blood compatibility tests, live/dead cell staining, and CCK-8 experiments have established the superior biocompatibility of GIL2 hydrogel. The in vivo assay indicated GIL2 hydrogel's effectiveness in eliminating bacteria, diminishing inflammation, fostering healthy epithelial regeneration, and greatly improving the healing of oral ulcer wounds. Notably, the GIL2 group demonstrated accelerated healing of oral ulcers compared to both the PBS group and the commercial lidocaine hydrogel patch while exhibiting no signs of inflammatory reactions. Collectively, the preparation of GIL2 hydrogel involves a straightforward and pragmatic technique, ensuring the effective defense of the oral mucosa against harmful actions, including saliva flushing and movements within the mouth. Such findings encourage the thoughtful development of dressings tailored for enhancing the healing of oral ulcers.

  Declaration of competing interest

  The contributors have disclosed no significant financial conflicts or personal connections that might seem to affect the outcomes presented in this work.

  CRediT authorship contribution statement

  Mengyu Chen: Data curation, Conceptualization. Qinglin Zhou: Formal analysis, Data curation. Tianyun Qin: Investigation. Ningyao Sun: Methodology. Yuxi Chen: Methodology. Yuwei Gong: Resources. Xingyi Li: Validation. Jinsong Liu: Supervision.

  Acknowledgments

  This research received funding from the National Natural Science Foundation of China (Nos. 82071170 and 82371016) and the Zhejiang Provincial Science and Technology Project for Public Welfare (No. LGF21H140004).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi: 10.1016/j.cclet.2024.110441.

  
  
• 1. [1]

     C. Cui, L. Mei, D. Wang, P. Jia, et al., Nat. Commun. 14 (2023) 7707.

  2. [2]

     A. Gasmi Benahmed, S. Noor, A. Menzel, et al., Arch. Razi Inst. 76 (2021) 1155–1163.

  3. [3]

     Y. Zeng, Y. Gao, L. He, et al., Mater. Today Bio 22 (2023) 100782.

  4. [4]

     X. Chen, X. Zhang, S. Yang, et al., Chin. Chem. Lett. 36 (2025) 110021. doi: 10.1016/j.cclet.2024.110021

  5. [5]

     X. Liang, C. Huang, H. Liu, et al., Chin. Chem. Lett. 35 (2024) 109442.

  6. [6]

     G. Liu, Y. Zhou, Z. Xu, Z. Bao, et al., Chin. Chem. Lett. 34 (2023) 107705.

  7. [7]

     S. Guo, Y. Ren, R. Chang, et al., ACS Appl. Mater. Interfaces 14 (2022) 34455–34469. doi: 10.1021/acsami.2c08870

  8. [8]

     S. Cheng, H. Wang, X. Pan, et al., ACS Appl. Mater. Interfaces 14 (2022) 11144–11155. doi: 10.1021/acsami.1c25014

  9. [9]

     S. Hu, X. Pei, L. Duan, et al., Nat. Commun. 12 (2021) 1689.

  10. [10]

      K. Zhang, Q. Feng, Z. Fang, et al., Chem. Rev. 121 (2021) 11149–11193. doi: 10.1021/acs.chemrev.1c00071

  11. [11]

      H. Yuk, C.E. Varela, C.S. Nabzdyk, et al., Nature 575 (2019) 169–174. doi: 10.1038/s41586-019-1710-5

  12. [12]

      B. Kong, Y. Chen, R. Liu, et al., Nat. Commun. 11 (2020) 1435.

  13. [13]

      L. Wang, L. Chen, J. Wang, et al., Chin. Chem. Lett. 33 (2022) 1956–1962.

  14. [14]

      S. Chen, Y. Wang, J. Lai, et al., Biomacromolecules 24 (2023) 2928. doi: 10.1021/acs.biomac.3c00302

  15. [15]

      Y. Wang, C. Zhang, S. Han, et al., Chin. Chem. Lett. 35 (2024) 109578.

  16. [16]

      Y. Yang, J. Yuan, Y. Ni, et al., Composites Part B 243 (2022) 110119.

  17. [17]

      B. Lv, L. Lu, L. Hu, et al., Theranostics 13 (2023) 2015– 2039. doi: 10.7150/thno.80615

  18. [18]

      X. Gan, C. Li, J. Sun, et al., RSC Adv. 13 (2023) 1558–1566. doi: 10.1039/d2ra06101e

  19. [19]

      F. Jahanmard, A. Khodaei, J. Flapper, et al., J. Control. Release 358 (2023) 667–680.

  20. [20]

      Z. Li, S. Zhang, Y. Chen, H. Ling, et al., Adv. Funct. Mater. 30 (2020) 2003601.

  21. [21]

      L. Zhang, Z. Dong, S. Yu, et al., Chin. Chem. Lett. 35 (2024) 109101.

  22. [22]

      J. Wang, X. Ge, Y. Xiang, et al., Chin. Chem. Lett. 36 (2025) 109819. doi: 10.1016/j.cclet.2024.109819

  23. [23]

      X. Wu, Q. Zhu, Z. Chen, et al., J. Control. Release 338 (2021) 268–283.

  24. [24]

      X. Yao, S. Zhang, L. Qian, et al., Adv. Funct. Mater. 32 (2022) 2204565.

  25. [25]

      X. Pan, R. Li, W. Li, et al., Nat. Commun. 15 (2024) 2651.

  26. [26]

      Y. Ma, J. Zhang, Y. Tian, et al., Nat. Commun. 14 (2023) 4958.

  27. [27]

      H. Shirahama, B.H. Lee, L.P. Tan, N.J. Cho, Sci. Rep. 6 (2016) 31036.

  28. [28]

      K. Feng, L. Liu, B. Tang, et al., ACS Appl. Mater. Interfaces 8 (2016) 11516–11525. doi: 10.1021/acsami.6b02248

  29. [29]

      Y. Hao, Z. Li, N. Ding, et al., Spectrochim. Acta Part A 268 (2022) 120642.

  30. [30]

      H. Wang, X. Nie, W. You, et al., ACS Appl. Mater. Interfaces 13 (2021) 56838–56849. doi: 10.1021/acsami.1c15868

  31. [31]

      F. Vidal, S. Smith, C.K. Williams, J. Am. Chem. Soc. 145 (2023) 13888–13900. doi: 10.1021/jacs.3c03261

  32. [32]

      K. Elkhoury, M. Morsink, Y. Tahri, et al., Int. J. Biol. Macromol. 183 (2021) 918–926.

  33. [33]

      K. Yue, X. Li, K. Schrobback, et al., Biomaterials 139 (2017) 163–171.

  34. [34]

      Y. Huang, S. Qian, J. Zhou, et al., Adv. Funct. Mater. 33 (2023) 2213549.

  35. [35]

      T. Ghosh, A.K. Das, Coord. Chem. Rev. 488 (2023) 215170.

  36. [36]

      L. Pang, T. Xue, H. Cong, et al., Colloids Surf. B 188 (2020) 110807.

  37. [37]

      X. Gao, Q.-H. Hu, Y.-Z. Shi, et al., Chem. Eng. J. 468 (2023) 143838.

  38. [38]

      J. Qian, X. Xu, J. Su, et al., Compos. Part B 231 (2022) 109563.

  39. [39]

      Y.-C. Chen, R.-Z. Lin, H. Qi, et al., Adv. Funct. Mater. 22 (2012) 2027–2039. doi: 10.1002/adfm.201101662

  40. [40]

      J. Deng, R. Bai, J. Zhao, et al., Angew. Chem., Int. Ed. 62 (2023) e202309058.

  41. [41]

      Y. Mao, Z. Xu, Z. He, et al., Chin. Chem. Lett. 34 (2023) 107461.

  42. [42]

      W. Peng, C. Liu, Y. Lai, et al., Adv. Sci. 10 (2023) 2301427.

  43. [43]

      U.S. Malik, Q. Duan, M.B.K. Niazi, et al., Chin. Chem. Lett. 34 (2023) 108071.

  44. [44]

      S. Cheng, M. Pan, D. Hu, et al., Chin. Chem. Lett. 34 (2023) 108276.

  45. [45]

      H. An, M. Zhang, Z. Huang, Y. Xu, et al., Adv. Mater. 36 (2023) 2310164.

  46. [46]

      L. Hu, Y. Wang, Q. Liu, et al., Chin. Chem. Lett. 34 (2023) 108262.

  47. [47]

      R. Iglesias-Bartolome, A. Uchiyama, A.A. Molinolo, et al., Sci. Transl. Med. 10 (2018) eaap8798.

  48. [48]

      N. Sardaro, F.D. Vella, M.A. Incalza, et al., In Vivo 33 (2019) 289–296. doi: 10.21873/invivo.11474

  49. [49]

      Y. Yan, J. Song, D. Liu, et al., Chin. Chem. Lett. 35 (2024) 109736.

  50. [50]

      G. He, X. Yan, Z. Miao, et al., Chin. Chem. Lett. 31 (2020) 1807–1811.

  51. [51]

      J. Sun, T. Chen, B. Zhao, et al., ACS Appl. Mater. Interfaces 15 (2023) 416– 431. doi: 10.1021/acsami.2c17424

  52. [52]

      Y. Wang, C. Zhang, H. Zhang, et al., Chin. Chem. Lett. 33 (2022) 4605–4609.

  53. [53]

      G. Chen, Y. Yu, X. Wu, et al., Adv. Funct. Mater. 28 (2018) 1801386.

  54. [54]

      S. Pourshahrestani, E. Zeimaran, N.A. Kadri, et al., Adv. Healthc. Mater. 9 (2020) 2000905.

  55. [55]

      X. Shuai, N. Kang, Y. Li, et al., Oral Dis. 30 (2023) 1286–1295.

  56. [56]

      B.V. Oliveira, P.G. Barros Silva, J. de Santiago Nojosa, et al., J. Appl. Oral Sci. 24 (2016) 278–290. doi: 10.1590/1678-775720150481

  57. [57]

      C. Xu, Y. Kang, S. Guan, et al., Chin. Chem. Lett. 34 (2023) 107825.

• 

  Scheme 1  Development and utilization of GIL2 hydrogel in treating bacterial-infected oral ulcers.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Preparation and characterization of GIL precursor and hydrogel dressing. ¹H NMR spectra of PIL (A) and GelMA (B). (C) Infrared spectrogram of G-gel, GIL1, GIL2 and GIL3 hydrogels featuring varying PIL concentrations. (D) XPS full spectrum of G-gel, GIL1, GIL2 and GIL3 hydrogels, each with different levels of PIL. (E) B 1s and (F) N 1s XPS signal of GIL2. (G) Synthesis reaction of GIL. (H) Photocrosslinking of GIL. (I) Rheology characterizations of GIL. (J) EDS elemental mapping of GIL2 and SEM images of GIL hydrogels.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Performance of adhesion and mechanical characteristics of GIL hydrogels. (A) Conventional testing approach for evaluating adhesive strength properties in lap-shear through tension loading (ASTM F2255). (B) Adhesion of GIL2 to various biological tissues and inorganic materials. (C) Swelling conditions of G-gel, GIL1, GIL2 and GIL3 hydrogels in PBS buffer. (D) Force-displacement curves and (E) shear strength of G-gel, GIL1, GIL2 and GIL3 hydrogels. (F) Tensile strength and (G) Water uptake ratio of G-gel, GIL1, GIL2 and GIL3 hydrogels. Data are presented as the mean ± standard deviation (n = 3).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Evaluation of toxicity, hemolytic, and ROS scavenging activity of G-gel, GIL1, GIL2 and GIL3 hydrogels. (A) Calcein-AM/PI staining of HGFs post-incubation with hydrogels. Scale bar: 100 µm. (B) CCK-8 assay outcomes for cytotoxicity: absorbance results of G-gel, GIL1, GIL2, and GIL3 hydrogels on HGF cells. (C) Hemolysis tests for GIL series with corresponding photos inset. (D) DPPH and PTIO scavenging of G-gel, GIL1, GIL2 and GIL3 hydrogels. (E) Relative DCFH-DA fluorescence ratios of HGFs without H₂O₂ and hydrogel treatment (control), H₂O₂ treatment (H₂O₂), and H₂O₂ and hydrogel co-treatment (G-gel or GIL2). (F) Assessment of intracellular ROS reduction by hydrogels, using DCFH-DA (green fluorescence). Scale bar: 100 µm. (G) Assessment of GIL hydrogels’ antibacterial properties. (a) Images showing E. coli and S. aureus. (b) Typical CLSM photos of E. coli and S. aureus stained with SYTO9 (green) and PI (red) (scale bar: 10 µm). (c) SEM photographs of E. coli (scale bar: 1 µm) and S. aureus (scale bar: 500 nm) post-incubation with PBS and GIL2 hydrogels (1 mL, 1.0 × 10⁸ CFU/mL). Data are presented as the mean ± standard deviation (n = 4).

  下载: 全尺寸图片 幻灯片
  

  Figure 4  The oral ulcer healing of GIL hydrogels in rats. (A) Illustrative schematic showing the construction process for a hemostasis model in a rat tail. (B) Images depicting the results of hemostasis in rats post-treatment with various groups. (C) The progression of oral ulcer healing in rats over time following treatment with PBS, lidocaine, G-gel, and GIL2, respectively. (D) In vivo experimental procedure for the modeling and treatment of oral ulcers in rats (scale bar: 5 mm). (E) Food intake of rats during the experiment. (F) Analysis of ulcer wound area. (G) Histologic examination of areas of oral ulcers in rats: Photographic representation of H&E (a) and Masson (b) staining (scale bar: 1 mm), along with immunochemical staining for inflammatory markers TNF-α (c) and angiogenesis factors (CD31; d) (scale bar: 100 µm) at the ulcer region in rats after 7-day treatment with various materials. (H) Analysis of the collagen index quantitatively in stained image representations. (I) TNF-α and CD31 measured in tissues seven days after treatment. Data are presented as the mean ± standard deviation (n = 3).

  下载: 全尺寸图片 幻灯片
•