Penicillin/streptomycin, Minimum Essential Medium-α, trypsin and fetal bovine serum were provided by Biological Industries. Lipopolysaccharide (LPS) and Oroxylin A were purchased from Beijing Solarbio Science & Technology Co., Ltd. RNAiso Plus, TB Green™ Premix Ex Taq™ II and PrimeScript™ RT reagent kit with gDNA Eraser were provided by Takara Bio, Inc. PCR primers were purchased from Sangon Biotech Co., Ltd.

The 8-week-old male Wistar rats weighing 250–300 g were purchased from Charles River Laboratories, Inc. A total of 30 rats were housed in the Experimental Animal Center of the Affiliated Hospital of Qingdao University and maintained in a specific-pathogen-free environment at 25°C, 40% humidity, and a 12-h light/dark cycle. The animals had free access to food and water. Periodontitis was induced by tying silk ligatures around the maxillary second molars and high sugar diet, as previously described (17).

Rats were randomly divided into the following groups (n=10/group): Sham operation (PBS, no periodontitis, no silk ligation), periodontitis (PBS+silk ligation) and treatment group (silk ligation+Oroxylin A). General anesthesia was induced by intraperitoneal injection of pentobarbital sodium (40 mg/kg, 2%). Periodontitis was induced by placing a silk thread between the right maxillary first and second molars. Rats in the treatment group were injected with 8 µl Oroxylin A solution with a final concentration of 0.5 µg/µl at the gingival site every 48 h for 2 weeks, while the sham group and the periodontitis group were injected with the same volume of PBS solution. A total of 4 weeks later, the rats were sacrificed by intraperitoneal injection of sodium pentobarbital (200 mg/kg) followed by rapid decapitation; death was confirmed by loss of respiration and heartbeat. The gingival tissue was collected to extract total RNA and total protein for detecting the expression of COX −2, TNF-α, RANKL and OPG and paraffin sections were prepared. All animal studies were approved by the Research Ethics Committee of the Affiliated Hospital of Qingdao University (Qingdao, China, AHQU-MAL20210326). The expression level of COX-2 was observed by immunohistochemistry. The gingival tissue specimens were formalin-fixed and paraffin-embedded at room temperature (fixative concentration, 4%; duration 12 h). Sections of 3 µm thickness were examined with immunohistochemistry. Antigen retrieval was performed by heating the specimens at 97°C, followed by washing with xylene and gradual rehydration with a descending ethanol series. Sections were blocked with the bovine serum albumin(BSA,3%) for 30 min at room temperature. Immunohistochemistry sections were incubated in 3% H₂O₂ in methanol for 30 min to block endogenous peroxidase activity. Primary antibody against COX-2 (1:1000, Abcam, ab151571) was incubated at 37°C for 1 h. Goat anti mouse IgG conjugated with horseradish peroxidase (1:1000, Abcam, ab6789) used as the secondary antibodies which was incubated at 37°C for 30 min. Chromogen detection was performed using DAB for 1 min at room temperature. Add sufficient amount of hematoxylin solution to the tissue slices to completely cover and incubate for 1 min at room temperature. Images are acquired at 40× magnification using a light microscope (Olympus, BX43). Image-Pro Plus version 6.0 software (Media Cybernetics, Inc., USA) was used for analysis.

After anesthesia, rat gingival tissue obtained from the attached gingiva of Wistar rats were cut into pieces and cultured in complete Minimum Essential Medium-α containing 10% fetal bovine serum and 1% penicillin/streptomycin, in a humidified incubator of 5% CO₂ at 37°C (18). When the cells grew out from the edge of gingival tissue, and the fusion rate reached 80%, the cells were digested, centrifuged, and passaged. Third-to-fifth passage cells were used for subsequent analysis. RGFs were identified by immunofluorescence staining with anti-vimentin, anti-fibronectin, anti-cytokeratin antibodies and PBS for control group (19). Cells were divided into the following groups: Control (PBS), LPS (10 pg/ml) and 50, 100, 200 and 400 µg/ml Oroxylin A+LPS (10 pg/ml). All treatments were performed at 37°C for 24 h. Cells were formalin-fixed at 4°C (fixative concentration, 4%; 20 min), and blocked with normal goat serum(10%, Absin) for 1 h at 37°C. Cells were incubated with primary antibodies as follows: Anti-vimentin (1:1000, Abcam, ab8978); anti- fibronectin (1:1,000, Abcam, ab2413), anti-cytokeratin (1:1000, Abcam, ab53280) at 37°C for 2 h. Goat anti mouse IgG conjugated with horseradish peroxidase (1:1000, Abcam, ab6789) used as the secondary antibodies which was incubated at room temperature for 2h. Nucleus was stained with DAPI. Images were captured with fluorescent microscopy (Keyence, BZ-X800). Image-Pro Plus version 6.0 software (Media Cybernetics, Inc.) was used for analysis.

RGFs were transfected with small interfering (si)RNA (Gibco; Thermo Fisher Scientific, Inc.) at 37°C. The sequences used were as follows: Rat HO-1 forward, 5′-ACAAGCAGAACCCAGUCUA-3′ and reverse, 5′-UAGACUGGGUUCUGCUUGU-3′ and negative control (NC) forward, 5′-UUCUCCGAACGUGUCACGUTT-3′ and reverse, 5′-ACGUGACACGUUCGGAGAATT-3′. RGFs (5×10⁵ cells/well) were transfected with either HO-1 (20 nM) or NC siRNA (20 nM) using the RNAimax-transfection system (Invitrogen; Thermo Fisher Scientific, Inc.). Following transfection for 36 h, cells were treated with complete medium in the presence of baicalein (50 µM; 37°C, 5%CO₂) for 12 h (for PCR) or 24 h (for western blotting) (20).

Cell viability was analyzed by CCK8 (Abcam) according to the manufacturer's protocol Cells were seeded and cultured for 3 h at a density of 5×10³/well in 100 µl medium into 96-well microplates(37°C). Then, the cells were treated with Oroxylin A at 37°C for 24 h. 10 µl CCK-8 reagent was added to each well and then cultured at 37°C for 2 h. All experiments were performed in triplicate. The absorbance was analyzed at 450 nm using a microplate reader.

TRIzol (Thermo Fisher Scientific, Inc.)was used to isolate total RNA from RGFs. RT-qPCR was then performed using RT-qPCR kits (SYBR Premix Ex Taq II, DRR820A, Takara), the temperature and duration were according to the manufacturers protocol. RNA was reverse-transcribed into cDNA using a Light Cycler LC480 (Roche Diagnostics). Cycling conditions consisted of initial incubations, followed by 40 cycles of denaturation, annealing, and extension. The thermocycling conditions of the qPCR steps were as follows: activation (temperature: 50°C; duration: 2 min), Dual-Lock DNA polymerase (temperature: 95°C; duration: 2 min), denaturation (temperature: 95°C; duration: 15 sec), and annealing/extension (temperature: 60°C; duration: 1 min). The mRNA expression levels were normalized to the level of GAPDH, a housekeeping gene. The 2-ΔΔCq method was used to determine relative expression of target genes. The primers used were as follows: GAPDH forward, 5′-ACCACAGTCCATGCCATCAC-3′ and reverse, 5′-TCCACCACCCTGTTGCTGTA-3′; TNF-α (GenBank: NM_013091.2) forward, 5′-ATGGGCTCCCTCTCATCAGT-3′ and reverse, 5′-AAATGGCAAATCGGCTGACG-3′; COX-2 (GenBank: NM_017232.4) forward, 5′-CTCAGCCATGCAGCAAATCC-3′ and reverse, 5′-GGGTGGGCTTCAGCAGTAAT-3′; RANKL (GenBank: NM_057149.2) forward, 5′-CCTGTACTTTCGAGCGCAGA-3′ and reverse, 5′-AGTCGAGTCCTGCAAACCTG-3′; OPG (GenBank: NM_012870.2) forward, 5′- GAATGTGAGGAAGGGCGCTA-3′ and reverse, 5′-CTTCGCACAGGGTGACATCT-3′ and HO-1 (GenBank: NM_012580.2) forward, 5′-ATGCCCCACTCTACTTCCCT-3′ and reverse, 5′-TGTGTGGCTGGTGTGTAAGG-3′.

TRIzol (Thermo Fisher Scientific, Inc.) was used to isolate total protein from RGFs. Protein concentrations were estimated via BCA assay. 30 µg total protein was loaded per lane. 12% separation gel and 5% stacking gel were used. Proteins were transferred to PVDF membranes. Membranes were blocked with 5% skimmed milk for 1.5 h at room temperature prior to western blot analysis. Incubate the blot with the primary antibody at 4°C overnight followed by secondary antibody for 2 h at room temperature. Primary antibodies against COX −2 (ab15191, 1:1000), TNF-α (ab9579,1:1000), OPG (ab73400, 1:1000), RANKL (ab93719, 1:1000) and HO-1 (ab13248, 1:1000) were all purchased from Abcam. Secondary horseradish peroxidase-linked antibodies against rabbit IgG (7074, 1:1,000) and anti-β-actin were from Cell Signaling Technology. Finally, protein bands were observed by ECL (Abcam, ab133406) and quantified using GraphPad Prism (version 8; Dotmatics) and ImageJ 2 (National Institutes of Health).

Statistical analysis was performed using GraphPad Prism (version 8; Dotmatics). Data are presented as mean ± standard deviation of three independent experimental repeats. One-way analysis of variance followed by Tukey's post hoc test was used to compare groups. P<0.05 was considered to indicate a statistically significant difference.

After being cultured for 3–7 days, the periodontal tissue fragment adhered to the wall and some cells grew from the edge of the tissue fragment (Fig. 1A). After 2–3 weeks of culture, Cells proliferated toward the center of the tissue fragment where they form a swirl pattern. Cell morphology was mostly fusiform, the nucleus was aggregated in the center and the protrusion formed by the cytoplasm radiated outwards. After 3–4 weeks, the fusion rate of the cells reached 80% (Fig. 1B).

Immunofluorescence results showed positive expression of fibronectin and vimentin antibodies (Fig. 1C and D), while negative expression of keratin antibodies (Fig. 1E). Fig. 1F shows a negative control, and the primary antibody was replaced by PBS without any expression. The outcome of positive expression of fibronectin and vimentin antibodies, while negative expression of keratin antibodies showed that periodontal membrane stem cells were derived from mesenchyma.

Preliminary CCK-8 results showed that Oroxylin A (50, 100, 200M, 400M) has no cytotoxicity (Fig. S1). RT-qPCR showed that compared with the blank control, the mRNA expression levels of COX-2, TNF-α and RANKL in the LPS-stimulated group decreased with the increase of Oroxylin A concentration (Fig. 2A-C). LPS inflammatory stimulation did not significantly increase the expression of OPG, however, preconditioning with Oroxylin A increased the expression of OPG mRNA in LPS-stimulated RGFs in Oroxylin A dose-dependent manner (Fig. 2D). The RANKL/OPG ratio was significantly increased in the LPS group and showed a downward trend with the increase of the dose of Oroxylin A (Fig. 2E).

Western blotting showed that LPS could significantly induce the protein expression levels of COX-2, TNF-α and RANKL and Oroxylin A could downregulate their expression in a concentration-dependent manner (Fig. 2F-I). Oroxylin A could induce OPG protein expression in a concentration-dependent manner (Fig. 2J), while the RANKL/OPG ratio was increased by LPS stimulation and decreased with the dose of Oroxylin A (Fig. 2K).

RT-qPCR results showed that the expression of HO-1 mRNA in normal RGFs was increased following Oroxylin A treatment and the degree of increase was positively associated with the concentration of Oroxylin A (Fig. 3A). The protein expression of HO-1 was increased following by Oroxylin A treatment (Fig. 3B). The expression of HO-1 increased following Oroxylin A treatment in a dose-dependent manner (Fig. 3C)

Following LPS induction, the mRNA and protein expression of HO-1 in RGFs were significantly higher than those in blank control group without inflammation induction, and HO-1 was upregulated with the increase of Oroxylin A concentration (Fig. 3D-F).

Following the transfection of RGFs with fluorescence-labeled siRNA, peripheral cells showed high brightness under microscope (Fig. 4A). The results of fluorescence microscopy showed that the fluorescence was evenly distributed in the cells (Fig. 4B). RT-qPCR and western blotting showed no significant differences in the HO-1 mRNA and protein expression levels of scrambled group and the control group (Fig. 4C-E).

RGFs were induced by LPS inflammation following HO-1 gene silencing, and RT-qPCR and WB showed that the mRNA and protein expression of COX-2, TNF-α and RANKL were significantly increased, and there was no statistical difference in the OPG group (Fig. 5A-K). mRNA and protein expression levels of COX-2, TNF-α, RANKL and OPG as well as the RANKL/OPG ratio were not significantly decreased by different concentrations of Oroxylin A (Fig. 5A-K).

Immunohistochemical results showed that COX-2 was expressed in the periodontal membrane of rats and the positive cells were yellowish brown. Expression of COX-2⁺ cells in every group was significantly different. The expression of COX-2 in the periodontal tissues of the control and the Oroxylin A treatment group was significantly lower than that of the periodontitis group (Fig. 6A-F) and the expression of COX-2, TNF-α and RANKL/OPG ratio in the periodontal tissue of the Oroxylin A treatment group was slightly higher than that of the control, but the expression level decreased with the increase of the dose (Fig. 6G-L).