For cell culture, a-Minimum Essential Medium (a-MEM), M-199 and L-glutamine were purchased from Gibco (Gibco; Thermo Fisher Scientific, Inc.). Trypsin, Triton X-100, dimethyl sulfoxide and β-mercaptoethanol were obtained from Sigma-Aldrich (Sigma-Aldrich; Merck KGaA). BBR hydrochloride was acquired from Wako (Wako; Wako Pure Chemical Industries, Ltd). Natural AGE protein (cat. no. ab51995) was purchased from Abcam. XAV-939 (cat. no. S1180) and CHIR-99021 (cat. no. S1263) were obtained from Selleck Chemicals.

Primary and secondary antibodies were purchased from the following commercial sources: An antibody specific to glycogen synthase kinase 3β (GSK-3β) was purchased from Cell Signaling Technology, Inc. (Cell Signaling Technology); antibodies specific to β-actin, and horseradish peroxidase (HRP)-conjugated anti-mouse and anti-rabbit immunoglobulin G (IgG) were purchased from Sigma-Aldrich (Sigma-Aldrich; Merck KGaA); antibodies specific to Runt-related transcription factor 2 (Runx2), osteopontin (OPN), bone sialoprotein (BSP), osteocalcin (OCN), wnt3a, β-catenin, CD73, CD90, CD105 and vimentin were purchased from Abcam; antibodies specific to CD34 were purchased from BD Biosciences; antibodies specific to cytokeratin-19 (CK-19) were purchased from Santa Cruz Biotechnology, Inc.; Alexa Fluor^® 488-conjugated-goat anti-rabbit secondary antibodies were purchased from Invitrogen (Invitrogen; Thermo Fisher Scientific, Inc.). Other chemicals were of the highest grade available commercially.

hPDLSCs used in the present study were provided by the Oral Stem Cell Bank of Beijing (Beijing Tason Biotech Co., Ltd.). The cells were plated in 75 cm² polystyrene tissue culture flasks (Corning Inc.) at 37°C in a humidified atmosphere consisting of 95% air and 5% CO₂, and then subcultured at 3×10⁴ cells/cm² into 6-well plates (Corning Inc.) if required. The cells were cultured in a-MEM containing 10% fetal bovine serum (FBS; Biological Industries), 2 mmol/l L-glutamine and antibiotics with 100 U/ml penicillin (HyClone; GE Healthcare Life Sciences) and 100 µg/ml streptomycin (HyClone; GE Healthcare Life Sciences). Following culturing with complete α-MEM for 24 h, the culture medium was replaced with osteogenic induction medium [α-MEM supplemented with 10% FBS, 2 mmol/l L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, 10 mmol/l β-Glycerophosphate disodium salt hydrate (Biological Industries), 50 µg/ml ascorbic acid (Biological Industries), 10 nmol/l dexamethasone (Biological Industries)]. The medium was changed every 48 h throughout the experiment. Cells were cultured for 7, 14 or 21 days, according to the particular experimental requirements. Passages 2–4 were used in experiments. Each experiment was repeated at least three times.

hPDLSCs were cultured in osteogenic induction medium containing 200 µg/ml AGEs with or without 1 µmol/l XAV-939, or 200 µg/ml AGEs + 1 µmol/l BBR with or without 1 µmol/l CHIR-99021 according to the different experimental requirements. The concentrations of the drugs used in the present study were selected based on the Cell Counting Kit-8 (CCK-8) method described below, and were consistent with their use in other studies (25–27). Throughout the study, in experiments involving XAV-939 or CHIR-99021, these were added to the culture medium 30 min prior to the addition of other drugs. All drug treatments were performed at 37°C unless otherwise stated.

hPDLSCs were identified through cell surface marker analysis by flow cytometry. The specific experimental procedures were as follows: Cells were cultured in 25 cm² polystyrene tissue culture flasks for 5 days in α-MEM, then they were washed twice carefully in heat-inactivated PBS, scraped off from the culture flasks, digested with 0.25% trypsin, and centrifuged at 300 × g for 5 min at 4°C. The cells were fixed with 4% paraformaldehyde (PFA) for 30 min at 4°C, permeabilized with 0.1% Triton X-100 for 10 min at room temperature, and blocked with a mixture of 0.5% bovine serum albumin (BSA) and 0.02% Tween-20 for 20 min at room temperature. Cells were then incubated at 37°C for 30 min with the following primary monoclonal and polyclonal antibodies: Mouse anti-CD73 (1:50; cat. no. ab91086), anti-CD105 (1:200; cat. no. ab114052) and anti-CD34 (1:50; cat. no. 555820); rabbit anti-CD90 (1:50; cat. no. ab133350) and anti-vimentin (1:200; cat. no. ab92547); and goat anti-CK-19 (1:100; cat. no. SC-33119). Cells were washed twice with PBS, then treated with Alexa Fluor 488-conjugated fluorescent secondary antibodies (1:400; cat. no. A-11029; mouse, 1:400; cat. no. A-11034; rabbit and 1:400; cat. no. A-11055; goat) for 30 min at 4°C in the dark. Cold PBS (200 µl) was added to each tube following two washes in cold PBS. Finally, flow cytometric analysis was performed using a Beckman Coulter Epics XL (Beckman Coulter, Inc.). The results were analyzed using the EXPO 32 Analysis Software (Beckman Coulter, Inc.). Negative control experiments were performed according to the same protocol without the use of primary antibodies.

All subsequent experiments were performed in the context of osteogenic induction; therefore, throughout the study, hPDLSCs were first seeded in 75 or 25 cm² polystyrene tissue culture flasks, or 6/24/96-well plates, in α-MEM at 37°C in a humidified atmosphere consisting of 95% air and 5% CO₂ overnight, and then the culture medium was replaced by osteogenic induction medium. The culture medium was replaced every 2 days. In this study, cells at passage 2–4 were used, and cultured at 37°C with 5% CO₂ for 5, 7, 14 or 21 days, depending on the subsequent experiment performed.

In the present study, various drugs, including AGEs, BBR, XAV-939 and CHIR-99021, were used to treat hPDLSCs. The selection of appropriate concentrations was determined via a cell viability assay using Cell Counting Kit-8 (CCK-8). The specific protocol was: hPDLSCs were cultured in 96-well plates at a density of 5×10³ cells/well, with ~150 µl cell suspensions evenly distributed in each well. The culture medium was replaced with osteogenic induction medium after 24 h. Then, cells in different 96-well plates were treated with various drugs or drug combinations, including AGEs (0, 100, 200 and 400 µg/ml), BBR (0, 0.1, 1, 3 and 10 µmol/l), XAV-939 (0, 0.1, 1 and 10 µmol/l) and CHIR-99021 (0, 0.1, 1 and 10 µmol/l) for 0, 1, 3, 5 and 7 days. The culture medium was replaced three times per week, as required. Subsequently, 15 µl CCK-8 reagent (Dojindo Molecular Technologies, Inc.) was added to each well. Following incubation for 3 h at 37°C, the optical density (OD) was detected at an excitation wavelength of 450 nm on a microplate reader (Thermo Fisher Scientific, Inc.). The concentrations that did not significantly affect cell viability in these assays (data not shown), were selected for subsequent experiments: 200 µg/ml AGEs, 1 µmol/l XAV-939, 1 µmol/l BBR, and 1 µmol/l CHIR-99021.

Depending on the experimental requirements, hPDLSCs were divided into the following groups: osteogenic induction medium (OSTEO); OSTEO + AGEs; OSTEO + AGEs + BBR; OSTEO + AGEs + XAV-939; OSTEO + AGEs + BBR + CHIR-99021. Cells were seeded in 6-well plates (3×10⁴ cells/cm²); following 14 days of osteogenic induction, ALP staining was performed using a Cell Alkaline Phosphatase Activity Stain kit (Genmed Scientifics, Inc.) according to the manufacturer's protocols. ALP staining was observed under an inverted microscope (Olympus Corporation) and a total of 5 views per field were acquired.

According to the manufacturer's protocols, the ALP activity of cells was measured using a Cell Alkaline Phosphatase Activity Colorimetric Assay kit (Genmed Scientifics, Inc.). Cells were divided into groups as aforementioned. The specific operation procedures were as follows: hPDLSCs were seeded in 24-well plates (3×10⁴ cells/cm²) following culturing for 14 days with the various drug treatments. Cells were collected with cell scrapers after they were washed in Genmed reagent A and then mixed with Genmed reagent A again prior to centrifugation at 300 × g for 5 min at 4°C. The cell culture supernatant was discarded, and Genmed reagent B was added to lyse cells (15 sec), with subsequent incubation on ice for 30 min. Then, cells were subjected to additional centrifugation (300 × g, 5 min, 4°C). Finally, cell culture supernatants were collected, with one portion used for protein concentration detection (bicinchoninic acid assay) and the remainder frozen at −70°C for further ALP analysis. The ALP activity of each cell was determined from the OD value detected by a microplate reader at 405 nm, using p-nitrophenol as a standard, according to the manufacturer's protocols. The ALP activity of each sample was normalized to its protein concentration and presented as OD405/min/mg protein. The experiment was repeated three times.

Cells were divided into treatment groups as aforementioned. Cells were seeded in 6-well plates (3×10⁴ cells/cm²); following 21 days of osteogenic induction, alizarin red staining was performed using a Cell Alizarin Red Stain kit (Genmed Scientifics, Inc.), according to the manufacturer's protocols. Bright red stained areas indicated calcified nodules. Alizarin red staining was observed under an inverted microscope (Olympus Corporation) and a total of 5 views per field were acquired.

Cells were divided into treatment groups as aforementioned. Cells were seeded in 96-well plates (3×10⁴ cells/cm²); following 14 days of osteogenic induction, the cell culture supernatant in each well was collected to quantify the levels of osteoblast-associated protein human collagen type I (Collagen I). Each sample was analyzed in duplicate. An ELISA was performed to detect the expression levels of Collagen I in each group using a Human Collagen type IELISA kit (cat. no. CSB-E08082h; Cusabio Technology LLC), according to the manufacturer protocols. This experiment was repeated three times.

The mRNA expression levels of Runx2, Osterix, ALP, OPN, Collagen I, OCN, β-catenin, low-density lipoprotein receptor-related protein 5 (LRP5), dickkopf-1 (DKK-1) and GSK-3β were determined by RT-qPCR analysis. Cells were divided into treatment groups as aforementioned. hPDLSCs in each group were harvested from 25 cm² polystyrene tissue culture flasks following 7 days of osteogenic induction. Total RNA of hPDLSCs was extracted and DNase-treated using a RNeasy Mini kit (Qiagen GmbH) according to the manufacturer's protocols, and then an ultraviolet spectrophotometer was used to assess the purity and quantity of the RNA preparation. RNA (2 µg) was reverse transcribed into complementary DNA (cDNA) using Promega Reverse Transcription System (cat. no. Promega A3500; Promega Corporation) with the following parameters: 60 min at 42°C, 5 min at 95°C, and 3 min at 4°C. 40 µl cDNA was used for DNA amplification. Gene expression levels were quantified via qPCR analysis with SYBR Green Realtime PCR Master Mix (Toyobo Life Science), according to the manufacturer's protocols, in an ABI PRISM 7500 system (Applied Biosystems; Thermo Fisher Scientific, Inc.). qPCR was conducted as follows: 60 sec at 95°C, then 40 cycles of 15 sec at 94°C, 15 sec at 60°C and 30 sec at 72°C, and a final melt curve stage of 15 sec at 95°C, 60 sec at 60°C, 30 sec at 95°C and 15 sec at 60°C. β-actin was used as an internal reference; the mRNA levels of target genes were normalized to β-actin. The relative mRNA levels were quantified using the 2^−∆∆Cq method (28). The primer sequences used in this study for β-actin, Runx2, Osterix, ALP, OPN, Collagen I, OCN, β-catenin, LRP5, DKK-1 and GSK-3β are listed in Table I. All experiments were performed three times.

Cells were divided into treatment groups as aforementioned. hPDLSCs in each group were harvested from 75 cm² polystyrene tissue culture flasks following 7 days of osteogenic induction prior to protein extraction. Protein extraction was conducted on ice or at 4°C. In brief, cells were washed twice with cold PBS prior to disruption in SDS lysis buffer (cat. no. KGP706; KeyGen Biotech Co., Ltd.). Then, cell lysates were collected, placed at 4°C for 30–60 min and then centrifuged at 4°C for 5 min (13,800 × g). Cell supernatants were collected for protein concentration determination using a bicinchoninic acid assay, according to the manufacturer's protocols. Then, protein denaturation was conducted by boiling in 6X SDS sample buffer for 5 min prior to protein (40 µg) separation via 10% SDS-PAGE. The separated proteins were then transferred onto PVDF membranes (EMD Millipore). Membranes were blocked with 5% skim milk for 1 h at room temperature, and subsequently incubated with primary antibodies at 4°C overnight. The primary antibodies were: Anti-β-actin (1:2,500; cat. no. A5316); anti-Runx2 (1:460; cat. no. ab23981); anti-BSP (1:200; cat. no. ab52128); anti-OPN (1:2,000; cat. no. ab91655); anti-OCN (1:500; cat. no. ab13420); anti-β-catenin (1:1,000; cat. no. ab32572); anti-GSK3β (1:1,000; cat. no. 12456) and anti-wnt3a (1:500; cat. no. ab28472). Membranes were then incubated with HRP-conjugated rabbit anti-mouse IgG (1:10,000; cat. no. A9044) and HRP-conjugated goat anti-rabbit IgG (1:20,000; cat. no. A0545) secondary antibodies at room temperature for 1 h. The labeled protein bands were visualized using Immobilon^® enhanced chemiluminescence western HRP substrate (EMD Millipore). All experiments were repeated three times.

To detect nuclear translocation of β-catenin, IF staining was performed following 7 days of osteogenic induction of hPDLSCs. Cells were divided into treatment groups as aforementioned. Cells were cultured in 96-well plates at a density of 5×10³ cells/well, and then rinsed twice in warm PBS. Then, cells were fixed with 4% PFA for 20 min at room temperature, permeabilized with 0.1% Triton X-100 for 8–10 min at 4°C and blocked with a mixture of 0.2% BSA and 0.02% Tween-20 for 20 min at room temperature. Cells were then incubated with rabbit anti-β-catenin monoclonal antibody (1:250; cat. no. ab32572) overnight at 4°C, followed by Alexa Fluor 488-conjugated-goat anti-rabbit secondary antibody (1:400; cat. no. A-11034) for 35 min at 37°C in the dark. Following incubation with DAPI (1:100) for 2 min at room temperature in the dark for cell nucleus identification, cells were washed twice for 5 min in PBS. Cells were observed, and images were captured and analyzed under an inverted immunofluorescence microscopy (Olympus Corporation). A total of 5 views per field were acquired.

Statistical analysis in this study was performed using GraphPad Prism 7 (GraphPad Software, Inc.). Comparisons between groups were performed using one-way analysis of variance followed by Tukey's multiple comparison test, and all values in this study were presented as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

All of the hPDLSCs cultured in the present study positively expressed mesenchymal stem cell (MSC)-associated surface markers, including CD73, CD90 and CD105, but were negative for the hematopoietic marker CD34 (Fig. 1). Additionally, hPDLSCs expressed vimentin but not CK-19 (Fig. 1). The results indicated that the hPDLSCs used in the study were pure mesenchymal cells that were derived from the mesoderm, and not the ectoderm.

Cells underwent notable morphological alterations following osteogenic induction. Prior to osteogenic induction, hPDLSCs grew by static adherence and remained in the form of a monolayer. The cells established a typical long spindle shape with oval nuclei comprising 2–3 nucleoli, as observed under an inverted phase contrast microscope (Fig. 2A). Following osteogenic induction, the cells exhibited adherent growth characteristics and diverse morphology. They also exhibited noticeably increased local aggregation. Similar to osteoblasts, they appeared as thickened fusiform, polygonal or irregular triangular cells with pseudopodia (Fig. 2B).

To determine the effects of AGEs on the osteogenic differentiation ability of hPDLSCs, ALP activity assays, ALP staining, alizarin red staining, ELISA, and RT-qPCR and western blot analyses were performed following treatment with osteogenic induction medium for various numbers of days. The OSTEO group served as the control. It was demonstrated that the osteogenic potential of hPDLSCs significantly decreased in the OSTEO + AGEs group compared with the control group (Fig. 3). The ALP activity of hPDLSCs in the OSTEO + AGEs group was significantly decreased compared with the control group following 14 days of osteogenic induction (P<0.01; Fig. 3A). The degree of ALP staining was markedly decreased in the AGEs-enriched microenvironment following 14 days of osteogenic induction compared with the control (Fig. 3B). Following 21 days of culture in the osteogenic induction medium, the hPDLSCs in the two groups formed mineralized nodules that exhibited positive alizarin red staining; however, the cells in the OSTEO+AGEs group formed fewer mineralized nodules compared with the control group (Fig. 3C). Following 14 days of osteogenic induction, ELISA revealed that the levels of Collagen I in the OSTEO + AGEs group were significantly decreased compared with the control (P<0.01; Fig. 3D). Additionally, RT-qPCR and western blot analyses revealed that the expression levels of osteogenesis-associated genes Runx2, Osterix, ALP, OPN, Collagen I and OCN were significantly downregulated in the OSTEO + AGEs group compared with the control (Fig. 3E). Finally, the expression levels of osteogenesis-associated proteins Runx2, OPN, BSP and OCN were markedly downregulated in the OSTEO + AGEs group (Fig. 3F). These findings suggested that the administration of AGEs reduced the osteogenic differentiation ability of hPDLSCs.

The effect of BBR was then examined on the AGE-induced reduction in the osteogenic potential of hPDLSCs. The results revealed that BBR could partially rescue the AGE-induced inhibition of the osteogenic differentiation ability of hPDLSCs (Fig. 3). As presented in Fig. 3A, the ALP activity of hPDLSCs in the OSTEO + AGEs + BBR group was significantly increased compared with the OSTEO + AGEs group (P<0.01). Additionally, as presented in Fig. 3B and C, BBR notably promoted the expression of ALP and formation of mineralized nodules in hPDLSCs. Collagen I levels were significantly upregulated in the OSTEO + AGEs + BBR group compared with the OSTEO + AGEs group (P<0.01; Fig. 3D). In addition, RT-qPCR and western blot analyses demonstrated that the expression levels of osteogenesis-associated genes and proteins in the OSTEO + AGEs + BBR group were significantly upregulated following osteogenic induction for 7 days compared with the OSTEO + AGEs group (Fig. 3E and F). Collectively, the results suggested that administration of BBR in an AGE-enriched microenvironment attenuated the AGE-induced decrease in the osteogenic differentiation ability of hPDLSCs.

To investigate the mechanism by which AGEs reduce the osteogenic differentiation ability of hPDLSCs, RT-qPCR and western blot analyses were conducted following 7 days of osteogenic induction to determine the expression levels of canonical Wnt/β-catenin pathway-related genes and proteins, including β-catenin, LRP5, GSK-3β, DKK-1 and wnt3a. It was revealed that AGE administration significantly upregulated the mRNA expression levels of β-catenin and LRP5, and downregulated those of GSK-3β and DKK-1 (Fig. 4A). Additionally, the protein expression levels of wnt3a and β-catenin were notably increased, while those of GSK-3β were decreased, in the OSTEO + AGEs group compared with the OSTEO group (Fig. 4B). IF staining was then conducted to evaluate the nuclear translocation of β-catenin, as a marker of Wnt/β-catenin pathway activation. The results indicated that AGEs promoted the nuclear translocation of β-catenin compared with control (Fig. 4C).

To determine whether activation of the canonical Wnt/β-catenin pathway was required in the AGE-induced suppression of the osteogenic potential of hPDLSCs, the small molecule inhibitor XAV-939 was used. As presented in Fig. 5A-D, addition of XAV-939 resulted in attenuation of the effects of AGEs on ALP levels, nodule formation and Collage I levels. Furthermore, XAV-939 reversed the decreased expression of osteoblast markers in hPDLSCs compared with the OSTEO + AGEs group, with most genes and proteins tested being upregulated following XAV-939 treatment except for OPN (Fig. 5E and F). The results revealed that AGEs decreased the osteogenic differentiation ability of hPDLSCs partially by activating the canonical Wnt/β-catenin pathway.

The mechanism by which BBR rescued the inhibited osteogenic potential of AGEs-treated hPDLSCs was investigated. As presented in Fig. 6A and B, a significant decrease in the gene expression levels of β-catenin and LRP5 (P<0.01), and notable decrease in the protein expression levels of wnt3a and β-catenin was observed in the OSTEO + AGEs + BBR group compared with the OSTEO + AGEs group. Additionally, the gene expression levels of GSK-3β and DKK-1, and the protein expression levels of GSK-3β were markedly increased in the OSTEO + AGEs + BBR group compared with the OSTEO + AGEs group (Fig. 6A and B). IF staining revealed that BBR decreased the nuclear translocation of β-catenin compared with the OSTEO + AGEs group (Fig. 6C).

To further examine whether the canonical Wnt/β-catenin pathway was involved in the BBR function in hPDLSCs, a small molecule activator CHIR-99021 was used. As presented in Fig. 7A-F, the addition of CHIR-99021 reversed the effects of BBR on ALP activity, nodule formation and Collage I levels in AGE-treated hPDLSCs, and resulted in a significant decrease in the gene expression of all osteoblast markers of hPDLSCs compared to with OSTEO + AGEs + BBR group. Of note, the activity levels of ALP, the levels of Collagen I in the supernatant, and the mRNA expression levels of Runx2 and Osterix remained significantly increased compared with in the OSTEO + AGEs group (P<0.01; Fig. 7A, D and E). The present results indicated that BBR promoted the osteogenic differentiation ability of hPDLSCs in an AGE-enriched microenvironment at least in part by inhibiting the canonical Wnt/β-catenin signaling pathway.