All chemicals were purchased from FUJIFILM Wako Pure Chemical Corporation unless stated otherwise. A canonical Wnt signaling pathway specific inhibitor LF3 and a glycogen synthase kinase-3 (GSK-3) specific inhibitor 6-bromoindirubin-3'-oxime (BIO) were from Cayman Chemical (Ann Arbor). For Western blotting, the primary antibodies against β-defensin 4 (BS60360, Bioworld Technology, St. Louis Park, MN, USA) and β-actin (PM053-7, MBL Life Science), and the secondary antibodies, horse radish peroxidase (HRP)-conjugated against mouse (#7076S, Cell Signaling Technology) and rabbit (#7074S, Cell Signaling Technology) were used. A Mouse Beta-defensin 4 ELISA kit was obtained from BT LAB.

AGE powder was prepared as previously described (16). The powder was dissolved in deionized water (DW) to obtain the AGE stock solution (20 mg/ml). The stock solution was stored at -20˚C until use.

Five weeks old male ddY mice were purchased from Japan SLC Inc. (Hamamatsu, Shizuoka, Japan) and kept at 23±3˚C and 50±10% humidity, under a 12 h light-dark cycle in the animal facility at Wakunaga Pharmaceutical Co., Ltd. Food (CE-2; CLEA Japan Inc.) and water were provided ad libitum. Mice were allowed to acclimate for 1 week, and then at 6 weeks of age, randomly divided into the DW-treated (control) group and the AGE-treated group. The control and AGE-treated groups were given DW and AGE, respectively, by oral gavage administration (10 ml/kg body weight) using a disposable feeding needle (Fuchigami). We used the dose 2 g/kg/day of AGE that has been shown to be safe and sufficiently effective in our previous studies (17-19). Gingival tissues were dissected out after the mice were euthanized by exsanguination under anesthesia with 2.5% isoflurane for induction and maintenance. Animal experiments were approved by the Wakunaga Pharmaceutical Company Institutional Animal Care and Use Committee (approval no. 360).

Mouse gingival epithelial GE1 cells (RCB1709, RIKEN Bioresource Research Center) were cultured in Minimum Essential Medium alpha (MEMα) with 10% fetal bovine serum, penicillin-streptomycin (x1), and 10 ng/ml recombinant murine epidermal growth factor (PeproTech) at 37˚C and 5% CO₂ in a humidified atmosphere. GE1 cells were seeded at a density of 13,000 cells/cm² and grown to confluent monolayer. After confluency, the culture medium was replaced by fresh MEMα with 1% fetal bovine serum, penicillin-streptomycin (x1), and 10 ng/ml recombinant murine epidermal growth factor (20).

Total RNA was extracted from mouse gingiva with acid guanidinium thiocyanate-phenol-chloroform extraction using RNAiso plus (Takara Bio Inc.). Complementary DNA was synthesized from total RNA using a PrimeScript RT reagent kit with a genomic DNA Eraser (RR047A, Takara Bio Inc.), and amplified on a CFX96 real-time PCR detection system (Bio-Rad Laboratories) with KAPA SYBR fast qPCR master mix (KAPA Biosystems). The PCR primers (Integrated DNA Technologies, Inc.) are listed in Table I. The fold change in the mRNA level relative to β-actin was calculated based on the ΔΔCt method (21).

Gingival tissues and GE1 cells were lysed in a radio-immunoprecipitation assay buffer (Merck) with 1X PhosSTOP™ (Sigma-Aldrich) and 1X cOmplete™ protease inhibitor cocktail (Roche, Basel, Switzerland) to obtain total protein. For nuclear protein extraction, two extraction buffers were used as follows; Buffer A, 10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM ethylenediaminetetraacetic acid (EDTA), 1 mM dithiothreitol (DTT), 1X PhosSTOP™ and 1X cOmplete™ protease inhibitor cocktail; Buffer B, 20 mM HEPES, 400 mM NaCl, 1 mM EDTA, 1 mM DTT, 1X PhosSTOP™ and 1X cOmplete™ protease inhibitor cocktail. GE1 cells were incubated in Buffer A on ice for 15 min, and then added 1/10 volume of 10% Nonidet P-40 substitute (Nacalai Tesque, Kyoto, Japan). Lysates were centrifuged at 20,000 x g for 5 min at 4˚C. The supernatant was removed, and the resultant pellet was washed twice with Buffer A. The washed pellet was resuspended in Buffer B, incubated on ice for 30 min, and subsequently centrifuged at 20,000 x g for 30 min at 4˚C. After centrifugation, the supernatant was used as the nuclear protein fraction. Each extracted protein was diluted to a protein concentration of 1 mg/ml with 4X sample buffer containing 250 mM Tris-HCl (pH 6.8), 8% sodium dodecyl sulfate, 40% glycerol, 2% bromophenol blue, and 400 mM DTT, and then boiled at 98˚C for 5 min. Protein extracts (20 µg) were separated on 4-20% Mini-PROTEAN TGX™ Gel (Bio-Rad Laboratories) and transferred onto Trans-Blot Turbo nitrocellulose membranes (Bio-Rad Laboratories) using a Trans-Blot Turbo Transfer System (Bio-Rad Laboratories). The membranes were treated with the primary antibody against β-defensin 4 (1:500) or β-actin (1:2,000) overnight at 4˚C, and then the HRP-conjugated secondary antibody (1:20,000) for 1 h at room temperature. Immunoreactive proteins were visualized with Armasham ECL Prime peroxidase solution (Cytiva) or ImmunoStar™ LD by using ChemiDoc™ MP (Bio-Rad Laboratories). The density of each immunoreactive band was analyzed using Band/Peak Quantification Tool in ImageJ 1.54i (22).

Quantification of β-defensin 4 in culture medium was performed using a Mouse Beta-defensin 4 ELISA kit according to the manufacturer's protocol. Culture medium was collected and centrifuged at 13,200 x g for 10 min at 4˚C. The supernatants were stored at -80˚C until use. The colorimetric absorbance was measured at a test wavelength of 450 nm using Multiskan GO Microplate Spectrophotometer (Thermo Scientific).

The lysis of gingival tissues resected from mice after 6 h treatment with AGE was performed by incubating with tris(2-carboxyethyl)phosphine hydrochloride for 1 h at 55˚C, alkylating with iodoacetamide for 30 min at room temperature, and digesting overnight with Pierce™ Trypsin Protease MS-Grade (Thermo Fisher Scientific) at a trypsin-protein ratio of 1:50 (w/w). Phosphorylated peptides were enriched using Titansphere Phos-TiO Tip (GL Sciences Inc.). Residual detergents and salts in the samples were removed using a HiPPR Detergent Removal Spin Column Kit (Thermo Fisher Scientific) and GL-Tip SDB columns (GL Sciences Inc.), respectively. The clean-up peptides were analyzed on a Q-Exactive Mass Spectrometer equipped with a Vanquish Neo LC System (Thermo Fisher Scientific). Phosphorylated peptides were identified using the Proteome Discoverer software (Thermo Fisher Scientific), and then compared with the Uniport curated M. musculus proteome database (release 2023.6). As a result, only ‘Annotated Sequence’ passing a cut-off of 5% false discovery rate (FDR Confidence: ‘Medium’) was considered for further analysis. Functional enrichment analysis was performed by Gene Ontology (GO) biological process database using GeneCodis 4(23).

Data analyses and graphical visualization were performed using KyPlot Free ver. 6.0.2 (KyensLab Inc.). Data are expressed as mean ± standard deviation. Unpaired Student's t test, Welch's t test, Mann-Whitney U test (for 2 groups) or one-way analysis of variance (for more than 3 groups), followed by post hoc Dunnett's test or Holm-Bonferroni test, were used to assess statistical significance. Differences at P<0.05 were considered statistically significant.

We administered AGE (2 g/kg/day) to mice for 2 weeks and examined the production of antimicrobial peptides, specifically β-defensin 1, β-defensin 4, β-defensin 14, and cathelicidin, in gingiva. As shown in Fig. 1A, the mRNA level of Defb4 was significantly increased in AGE-treated mice compared to DW-treated (control) mice. In contrast, the mRNA level of other epithelial antimicrobial peptides, including Defb1, Defb14, and Cramp, remained unchanged (Fig. S1). We next performed Western blot analysis to examine the effect of AGE on the protein level of β-defensin 4, and found that AGE induced the significant increase (Fig. 1B).

We next gave a single administration of AGE and examined the change of the β-defensin 4 during 24 h. The data obtained by reverse transcription-quantitative PCR analysis indicated that AGE induced a transient and significant increase in the mRNA level of Defb4 at 6 h (Fig. 2A). In addition, the protein level of β-defensin 4 was significantly elevated in the AGE group at 16 h compared to the control group (Fig. 2B).

To explore the underlying mechanism of AGE-induced increase of β-defensin 4, we performed phosphoproteomics analysis using mouse gingival tissues treated with AGE for 6 h. Our phosphoproteomics analysis identified a total of 2,298 phosphopeptides, revealing 79 up-regulated and 52 down-regulated phosphopeptides in the AGE group compared to the control group (Fig. 3A). In addition, among the 131 phosphopeptides, the phosphorylated amino acid residues were distinctly defined in 47 up-regulated and 28 down-regulated phosphopeptides (Fig. 3A). We subsequently performed GO biological process enrichment analysis of proteins with differentially phosphorylated peptides in the AGE group using GeneCodis 4(23). The top 5 significantly enriched GO terms in the biological process, based on the number of proteins with up-regulated phosphopeptides (adjusted P-value <0.05), were axon extension, neurofilament bundle assembly, virion attachment to host cell, telomerase holoenzyme complex assembly, and intermediate filament bundle assembly (Fig. 3B). Similarly, the top 5 significantly enriched GO terms for proteins with down-regulated phosphopeptides were negative regulation of canonical Wnt signaling pathway, protein stabilization, ovulation from ovarian follicle, regulation of translational initiation, and response to cocaine (Fig. 3B). The GO term ‘negative regulation of canonical Wnt signaling pathway’ included proteins that are involved in forkhead box protein O3 (S7), glycogen synthase kinase-3 alpha (GSK-3α) (Y279), and catenin delta-1 (S252) (Table II). Non-targeted proteomics analysis revealed a decrease in axin-1 (adjusted P-value=0.047) among the 125 up-regulated and 73 down-regulated proteins (Fig. S2), indicating the involvement of the canonical Wnt pathway.

The proteomics analysis on mouse gingiva suggested the possible involvement of the canonical Wnt signaling pathway in AGE-induced β-defensin 4 production. This hypothesis was supported by two key observations: (1) the suppressed phosphorylation level of GSK-3α, a negative regulator of the Wnt/β-catenin pathway (Fig. 3B; Table II), and (2) the decreased protein level of axin-1 (adjusted P-value=0.047), a component of the β-catenin destruction complex along with GSK-3α. Since β-catenin functions as a transcription factor downstream of this pathway and its localization to the nucleus is essential for exerting its transcriptional effects (24), we investigated the mechanism by which AGE induces β-defensin 4 in gingiva, using mouse gingival epithelial GE1 cells in culture.

As shown in Fig. 4A, treatment with AGE at 2 mg/ml for 24 h significantly increased the amount of β-defensin 4 in culture medium. We next examined β-catenin localization to assess the involvement of the Wnt/β-catenin pathway. We found that β-catenin protein accumulated within the nucleus of cells treated with AGE (2 mg/ml) for 3 h (Fig. 4B). To further examine the involvement of the Wnt/β-catenin pathway, we used two specific inhibitors of this pathway, LF3 and BIO. Simultaneous treatment of GE1 cells for 24 h with LF3 (30 µM), a specific inhibitor of β-catenin on canonical Wnt signaling, significantly suppressed the AGE-induced increase in the β-defensin 4 protein production (Fig. 4C). Moreover, treatment with a GSK-3 specific inhibitor BIO (0.1 and 1 µM) alone resulted in a statistically significant increase of the β-defensin 4 production (Fig. S3). These results suggested the involvement of the Wnt/β-catenin pathway in the β-defensin 4 production induced by AGE in mouse gingival epithelial cells.