Methacryloyl gelatin (GelMA, the amino substitution degree was 60%), Lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP), and the portable curing light source facility (It emits visible light at 405 nm) were purchased from EFL-Tech Co., Ltd, Suzhou, China; Ginsenoside Rg1 was purchased from Beijing Solarbio Science & Technology Co., China; Human gingival fibroblasts (hGFs) were obtained from Oral & Maxillofacial Reconstruction and Regeneration of Key Laboratory, of which the identification has been completed in previous experiments of our research group (Huang et al., 2019). All media and reagents related to cell experiments were purchased from Thermo Fisher Scientific Inc., United States; The Cell Counting Kit-8 (CCK-8) assay kit was purchased from Dojindo Laboratories, Japan; The rats were obtained from the Experimental Animal Center of Southwest Medical University (Luzhou, China). All the reagents were of analytical grade and used as received.

Photoinitiator LAP (0.05 g) was put into a dark brown glass bottle, added 20 mL of phosphate-buffered saline (PBS, 1×, pH 7.2–7.4, 0.01M; this specification of PBS was used in all subsequent experiments unless otherwise stated), and heated in a water bath at 40–50°C for 15 min, shocking it while heating. After this step is completed, the mixture is divided into two dark centrifuge tubes of 10 mL each; 0.6 g and 0.3 g of GelMA were placed, respectively, in centrifuge tubes containing 10 mL of the mixture. The mixture was heated in a water bath at 40–50°C for 30 min in the dark. During the period, it was repeatedly shaken to prepare a light-cured hydrogel precursor solution with GelMA concentrations of 3% w/v and 6% w/v, respectively. The solution was immediately filtered for subsequent experiments with a sterile needle filter (pore size of 0.22 μm). The precursor solution could be photo-polymerized under 405 nm visible light irradiation for 10 s to form the hydrogel.

The hydrogel material with better physicochemical properties was selected in previous experiments. Dissolve 20 mg of Rg1 in 2 mL of PBS solution to obtain a highly concentrated Rg1-PBS mixture for subsequent use. The drug-laden hydrogel precursor solution with Rg1 concentrations of 500 μg/mL and 1,000 μg/mL can be obtained by adding 50 μL and 100 μL of Rg1-PBS mixture to each 1 mL of hydrogel precursor solution using a high-precision micropipette, respectively. The Rg1-laden hydrogel precursor solution was converted into gel under 405 nm visible light irradiation, and we named them Rg1-500-Gel and Rg1-1000-Gel, respectively.

The Fourier transform infrared spectroscopy (FTIR) analysis (Jasco, 6300, Tokyo, Japan) of Rg1 and the hydrogel sample was performed in the range of 400–4500 cm⁻¹ to investigate any possible interaction between Rg1 and the raw material of the hydrogel (Song et al., 2020).

The release of Rg1 from Rg1-500-hydrogel and Rg1-1000-hydrogel was determined by high-performance liquid chromatography (HPLC). 1.5 mL of precursor solution of Rg1-500-hydrogel and Rg1-1000-hydrogel were placed into cylindrical molds with a diameter of 20 mm, respectively, and demoulded after light curing as described previously. The samples in each group were prepared for 3 replicates. The sample was placed in a 6-well plate, 5 mL of PBS solution was added to each well, and the plate was placed in a 37°C incubator, extracted 2 mL of the soaking solution every 24 h and supplemented with 2 mL of fresh PBS solution. The concentration of Rg1 in the extract was determined by HPLC (Agilent II 1260 Infinity LC, United States). The chromatographic column used in this experiment was Diamonsil C18 (4.6 × 250 mm, 5 μm), and the detection wavelength was 203 nm. (Xu et al., 2023).

The cytotoxicity of the photocurable hydrogel with or without ginsenoside Rg1 towards human gingival fibroblasts (hGFs) was evaluated by the CCK-8 assay. The same size hydrogel (Φ10 × 10 mm) was immersed in 6 mL of complete growth medium (including 10% FBS, 1% penicillin-streptomycin solution, and basal DMEM medium) per well and placed in a cell incubator (37°C, 5% CO²). After 48 h, the hydrogel soaking solution was collected and filtered with a sterile needle filter (pore size 0.22 μm) for subsequent use. The hGFs were seeded into 96-well plates at a density of 1 × 105 cells/well and then treated with different soaking solutions. After 24, 48, and 72 h of incubation, the CCK-8 mixed medium was added to each well and measured at 450 nm on a microplate reader (SynergyH1, BioTek, United States).

All data were presented as the mean ± standard deviation. Comparisons between groups were performed by applying one-way ANOVA under Tukey’s test using GraphPad Prism (GraphPad Software, United States). Differences were considered statistically significant at *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

At present, there exists an abundance of hydrogel materials that are utilized in the field of wound healing. Some of them promote tissue repair through the hydrogel material components themselves, such as chitosan (Hamedi et al., 2018), hyaluronic acid (Graça et al., 2020), silk fibroin (Gholipourmalekabadi et al., 2020), etc., while others reduce peri-wound inflammation and promote tissue repair by loading biologically active ingredients, such as antibiotics, nanoparticles, stem cells, and growth factors (Ferreira et al., 2018; Fan et al., 2019; Liang et al., 2022). The previous study conducted by our research group demonstrated that ginsenoside Rg1 exhibits notable antioxidant and anti-apoptotic properties (Chu et al., 2023). Furthermore, it has been confirmed that Rg1 plays a distinct function in facilitating vascular regeneration (Chen et al., 2019). Based on the above information, it may be inferred that Rg1 has significant potential for application in the field of tissue regeneration. However, the majority of these investigations are focused on in vitro experiments. At the moment, we intend to use a single bioactive component, ginsenoside Rg1, piggybacked on a basic hydrogel scaffold for the restorative treatment of oral mucosal defects in rats, in order to preliminarily explore the potential application of Rg1 in the field of tissue regeneration. Excellent physical and mechanical properties are critical to the selection of hydrogel scaffold materials for carrying therapeutic components (Atia et al., 2023). In order to screen out better hydrogel scaffolds carrying the target drug ginsenoside Rg1, we evaluated the performance of hydrogel scaffolds with GelMA concentrations of 3% w/v and 6% w/v, respectively. We synthesized the 3% and 6% of the scaffold materials (hereinafter referred to as 3%-Gel and 6%-Gel) as shown in Figure 1A and observed their microscopic morphology using scanning electron microscopy, which showed that both concentrations of scaffold materials exhibited an interconnected pore structure. As shown in Figure 1B, for the porous structure of 3%-Gel, the pore wall is thin and easily broken, and many pore wall fragments can be seen in the electron microscope image. For 6%-Gel, the pore wall was thick and regular, and no debris was found. The ImageJ (ImageJ 2.3.0 software, United States) was used to analyze three random images collected by SEM. It was found that the pore diameter of 3%-Gel was about 0–282.9 μm, and that of 6%-Gel was about 0–200.7 μm. With the concentration of GelMA increasing from 3% to 6%, the average pore diameter decreased from 196.3 ± 42.9 μm to 164.8 ± 33.3 μm (Supplementary Figure S1). The 6%-Gel scaffold is expected to withstand a greater biomechanical load and provide physical space for the growth of cells from the surrounding tissues, as shown by the compression curve in Figure 1C, where the compressive strength of the 6%-Gel scaffold is significantly higher than that of the 3%-Gel scaffold (He et al., 2000).

Superior swelling property is a crucial factor for assessing the application potential of hydrogel scaffold materials, as stable swelling performance can prolong the release of encapsulated pharmaceuticals (Wang et al., 2020). As demonstrated in Figure 1D, the 3%-Gel has a higher swelling rate than the 6%-Gel, but it reaches its maximum mass between 24 and 48 h and the mass reduction occurs before 48 h, whereas the mass of the 6%-gel begins to decline after 48 h, showing a more stable swelling performance. The freeze-dried hydrogel scaffold is shown in Supplementary Figure S2A. During 24 h of immersion in PBS at 37°C, the diameters of both 3%-Gel and 6%-Gel increased due to water absorption. However, compared with 3%-Gel, the shape of 6%-Gel is more regular and stable (Supplementary Figures S2B, C). After 24 h, 3%-Gel begins to degrade. Overall, the 6%-Gel displayed an excellent balance between its physical properties and swelling behavior. Therefore, all of the following experiments were conducted with 6%-Gel.

The hydrogel scaffold with GelMA concentrations of 6% was selected to carry ginsenoside Rg1 with concentrations of 500 μg/mL and 1000 μg/mL (hereinafter referred to as Rg1-500-Gel and Rg1-1000-Gel), respectively (Figure 2A). In order to further explore the chemical structure and interaction between ginsenoside Rg1 and hydrogel scaffold materials, the pure extract of ginsenoside Rg1, blank hydrogel, and Rg1 loaded hydrogel were measured and compared by the Fourier transform infrared spectroscopy (FTIR). The molecular structure diagram of ginsenoside Rg1 and the raw materials of hydrogel scaffolds (GelMA and LAP) are shown in Figures 2B–D. As demonstrated in Figure 2E, the FTIR spectra of ginsenoside Rg1 exhibited characteristic absorption peaks at 3,369 cm⁻¹ and 1390 cm⁻¹ (trough position), corresponding to the O-H stretching, which may be related to the high O-H content of Rg1 (Salarian et al., 2016). In the spectra of Blank Gel, characteristic peaks appeared near 1700 cm⁻¹ and 871 cm⁻¹, corresponding to the characteristic functional groups C=O and N-H of GelMA, and the peaks located in the range of 1454 cm⁻¹ to 1505cm^-1 were attributed to the characteristic structure of the benzene ring and functional group P=O (1261cm⁻¹) of the photoinitiator LAP (Xu et al., 2022; Zhou et al., 2023). After loading ginsenoside Rg1 into the hydrogel scaffold, the spectra of Rg1-Gel revealed that the addition of Rg1 had no effect on the overall structure of Blank Gel. Possibly due to the concentration of Rg1 being relatively low, the absorption of Rg1 was obscured by the absorption of hydrogel scaffold functional groups, resulting in the variation trend of the Rg1-Gel spectrum exhibiting essentially the same characteristics as those of Blank Gel, with the exception of a slight increase in absorption intensity. Therefore, it can be concluded that the incorporation of Rg1 into the hydrogel did not alter its fundamental structure.

To determine whether Rg1-loaded hydrogels can deliver the bioactive drug Rg1 in a sustained and efficient manner, we analyzed the Rg1 concentration in the soaking solution of the hydrogels using high-performance liquid chromatography (HPLC). In our 7-day drug release assay, hydrogels loaded with two different concentrations of Rg1 showed a sustained release of Rg1 for more than 4 days, as illustrated in Figure 3A. Based on the outcomes of our group’s prior in vitro investigations concerning ginsenoside Rg1, we selected the hydrogel containing Rg1 at a concentration of 1,000 μg/mL (Rg1-1000-Gel) for the follow-up experiments. This particular formulation was chosen due to its ability to release a drug concentration roughly comparable to the previously determined optimum biological concentration of Rg1 (Chu et al., 2023). Under a scanning electron microscope (SEM), we observed Rg1-1000-Gel and determined that Rg1 had no significant effect on the porous structure of the hydrogel scaffold; however, particle deposition was observed on the pore wall. Compared to pure Rg1 powder, its morphology was found to be consistent (Figure 3B). We therefore speculated that the particles attached to the pore wall could be a part of the ginsenoside Rg1 loaded onto the hydrogel scaffold, while the other part of the Rg1 was encapsulated inside the hydrogel material. This inference also corresponds to the rapid release of Rg1 on the first day and slow release on the following days in the release experiments of Rg1 (Figure 3A). Excellent cytocompatibility is an essential requirement for the in vivo use of biomaterials (Oladipo et al., 2023). The Figure 3C shows the cell viability of gingival fibroblasts treated with the soaking solution of blank hydrogels and hydrogels loaded with different concentrations of Rg1 (Rg1-500-Gel and Rg1-1000-Gel). At 24, 48, and 72 h, there was no significant difference in cell viability (p > 0.05) between the hydrogel scaffold groups with or without Rg1. This result suggests that both the hydrogel scaffold and the Rg1-loaded hydrogel have good biocompatibility, providing a credible premise for subsequent in vivo experiments.