hPDLCs were isolated from healthy premolar teeth using a previously described method (23). Healthy subjects <20 years old undergoing orthodontic treatment were recruited with the understanding and consent for the isolation of hPDLCs. All the experimental protocols used were approved by the Ethics Committee of Sun Yat-sen University (Guangdong, China). Briefly, fresh PDL tissues were collected rapidly following the tooth extraction. Under sterile conditions, the PDL tissue was washed with sterile phosphate-buffered saline (PBS) and scraped from the middle-third of the root with a scalpel. The tissue was subsequently chopped into smaller pieces and soaked in fresh Dulbecco's modified Eagle's medium [DMEM/high glucose; Hyclone Biochemical Product (Beijing) Co., Ltd., Beijing, China] containing 20% fetal bovine serum (FBS) (Biological Industries Israel Beit-Haemek, Ltd., Kibbutz Beit-Haemek, Israel) and 2% (v/v) penicillin/streptomycin (Invitrogen Life Technologies, Carlsbad, CA, USA). An explant culture method was used in the primary culture. The medium was changed every 3 days. Cells typically emerged from the tissues within 1–2 weeks of being explanted in a humidified atmosphere of 5% CO₂ at 37°C. When the cells growing from the explants reached ~80% confluency, they were subcultured at a 1:3 ratio through trypsinization (trypsin/EDTA; Invitrogen Life Technologies). Only the fifth passage of hPDLCs was used in subsequent experiments.

Recombinant Human FGF-basic (FGF-2) (Peprotech, Inc., Rocky Hill, NJ, USA) was added to the defined growth culture medium and osteogenic inductive culture medium. There were four treatment groups: i) hPDLCs + growth medium (GM) (DMEM containing 10% FBS and 2% penicillin/streptomycin); ii) hPDLCs + GM + 20 ng/ml FGF-2 (GM + FGF-2); iii) hPDLCs + osteogenic medium (OM) [DMEM with 10% FBS, 2% (v/v) penicillin/streptomycin, 10⁻⁸ M DEX, 50 µg/ml ascorbic acid and 8 mM β-GP]; and iv) hPDLCs + OM + 20 ng/ml FGF-2 (OM + FGF-2).

hPDLC viability was monitored using a Cell Counting kit-8 (CCK-8) (Beyotime, Shanghai, China). Briefly, hPDLCs at passage 5 were seeded in 96-well plates at an initial density of 3×10³ cells/well in medium, with five replicates for each group. The medium was replaced every 3 days. After 1, 2, 3, 5 and 7 days of subculture in medium treated differently, 20 µl of CCK-8 solution was added to every well and incubated with the cells for 4 h in a humidified atmosphere of 5% CO₂ at 37°C. The absorbance of each sample was determined at the wavelength of 450 nm on a 96-well plate reader.

hPDLCs were seeded in 24-well plates in triplicate at a density of 2×10⁴ cells/well and cultured in defined media subsequent to reaching ~80% confluency. Following 3, 7, 14 and 21 days in culture, an ALP activity assay kit (Nanjing Jiancheng Biotechnology Co., Ltd., Jiangsu, China) was used to detect ALP activity of hPDLCs, following the manufacturer's instructions. In brief, the cell layers in plates were washed three times with 10 mM sterile PBS, and subsequently a small amount (1 ml) of cold 10 mM Tris-HCl buffer (pH 7.4) containing 0.1% Triton X-100 was added to the culture well, and the mixture was lysed through two freeze (−20°C)/thaw cycles. The supernatant (50 µl/well) was placed into 24-well plates for the ALP activity assay using 50 µl of an ALP substrate solution (16 mM p-nitrophenyl phosphate and 2 mM MgCl₂). After a 30-min incubation at 37°C, the reaction was stopped by the addition of 50 µl of 0.2 M NaOH, and the absorbance of the reaction product (p-nitrophenol) was read at 520 nm.

After 7 and 14 days of culture, the expression levels of selected osteoblastic and fibroblastic genes were determined in triplicate by RT-qPCR. Total RNA was extracted from hPDLCs using TRIzol Reagent (Invitrogen Life Technologies) and was quantified in a spectrophotometer by absorbance readings at 260 nm. The extracted total RNA (2 µg) was treated with 1 unit DNase I for cDNA synthesis in each RT-qPCR using the RevertAid™ First Strand cDNA Synthesis kit (K1622; MBI Fermentas, Inc., Burlington, ON, Canada). PCR was carried out using a combination of 2.5 µl of diluted cDNA and 12.5 µl of SYBR-Green Real-time PCR Master mix (QPK-201; Toyobo Co., Ltd., Osaka, Japan) in a Chromo4 System (Bio-Rad, Hercules, CA, USA). The sequential reaction conditions were as follows: 95°C for 1 min denaturation followed by 40 cycles of denaturation at 95°C for 15 sec, annealing at 60°C for 15 sec and extension at 72°C for 30 sec. The specificity of the PCR products was confirmed by melting curve analysis (60–93°C, read every 0.4°C, hold 1 sec). GAPDH was used as the reference gene for data normalization. The number of cDNA copies was calculated with the apparent cycle threshold (CT). The ΔCT value expresses the difference between the CT of the target gene and the CT of GAPDH from the same sample: ΔCT = the target gene CT - GAPDH CT, which can be expressed as a percentage of GAPDH. The relative expression level of the target gene (fold of the reference gene) was obtained by transforming the logarithmic values to absolute values using 2^−ΔCT. The primer sequences are shown in Table I.

After 7 and 14 days of culture, Runx2 and osteocalcin (OCN) protein expression was measured using western blotting. In brief, hPDLCs were washed three times with pre-cooling PBS. The constructs were subsequently homogenized in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer, and cellular protein was extracted in lysis buffer [50 mM Tris-HCl, 0.5% Triton X-100, 2 mM EDTA and 150 mM NaCl (pH 7.5)] containing phenylmethylsulfonyl fluoride. The protein samples (30 µg) were electrophoresed through 10% SDS-PAGE and transferred to a polyvinylidene fluoride (PVDF) membrane (Millipore, Bedford, MA, USA) in a wet blotting apparatus for 1.5 h at 300 mA. The PVDF membrane was subsequently blocked with 5% bovine serum albumin (Wuhan Boster Biological Technology, Ltd., Hubei, China) for 1 h at room temperature, and the blot was incubated with 1:1,000 dilution of anti-human Runx2 antibody (Cat. no. AF2006; R&D Systems, Minneapolis, MN, USA) and 1:500 of anti-human OCN (Cat. no. ab10911; Millipore, Billerica, MA, USA) for 2 h, washed with Tris-buffered saline Tween 20 (TBST) four times and incubated with a 1:2,500 dilution of horseradish peroxidase (HRP) AffiniPure goat anti-rabbit secondary antibody (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) for 1 h, and further washed four times with TBST. HRP detection was accomplished via chemical luminescence exposed to X-ray film with an ECL western blotting detection system.

The calcium content of the cells was measured by a colorimetric quantitative method using a Calcium Assay kit (Shanghai Genmed Gene Pharmaceutical Technology Co., Ltd., Shanghai, China) following the manufacturer's instructions. Cells were seeded at a cell density of 5×10⁴ cells/well in 12-well tissue culture plates. After 14 and 21 days of culture, deposition of calcium in the cell layer was performed in triplicate. The absorbances of dyes were read at a wavelength of 595 nm using a UV/Visible light Spectrophotometer (Varian Cary 50; Varian Australia Pty, Ltd., Mulgrave, VIC, Australia).

After 7, 14 and 21 days of culture, the mineralized matrix nodules were detected using an Alizarin Red S Staining kit (Shanghai Genmed Gene Pharmaceutical Technology Co., Ltd.) following the manufacturer's instructions. Prior to staining, the cell cultures were washed five times with 10 mM sterile PBS and fixed using 10% (v/v) neutral buffered formalin for 30 min. Subsequently, the samples were stained for 5 min and the excess stain was rinsed with PBS. An Inverted Phase Contrast Microscope (Olympus IX41; Olympus Corp., Tokyo, Japan) was used to observe the stained cells.

All the values are reported as the mean ± standard deviation. Statistical analysis of the data was performed using the SPSS 17.0 software package (SPSS, Inc., Chicago, IL, USA). One-way analysis of variance and the least significant difference test were used for parameter estimation and hypothesis testing. P<0.05 was considered to indicate a statistically significant difference.

At days 1 and 2, FGF-2 had no significant effect on the cell proliferation rate (P>0.05). From day 3 onwards, hPDLCs cultured in the GM with added FGF-2 exhibited the maximal levels of proliferation, and those in the OM exhibited minimum levels (P<0.05) (Fig. 1). This suggested that supplementing growth media with FGF-2 enhanced proliferation of hPDLCs in the presence and absence of osteogenic inducers.

The specific ALP activity progressively increased with the duration of culture time in all the groups. At days 3 and 7 of culture, FGF-2 inhibited the ALP activity. However, FGF-2 promoted the ALP activity at days 14 and 21. There were, however, no significant differences between FGF-2-treated and -untreated groups at any time-points (P>0.05) (Fig. 2).

In the cultures without osteogenic inducers, Runx2 mRNA expression was significantly upregulated under the stimulation of FGF-2 at days 7 and 14 (P<0.05). However, in the OM, there was no evident effect on Runx2 expression (P>0.05) (Fig. 3A).

In the western blot analysis, Runx2 expression was increased at days 7 and 14 in the medium without osteogenic inducers under the stimulation of FGF-2. In the OM samples, there was no increase in Runx2 protein levels at day 7, whereas at day 14 there was a significantly increased protein expression stimulated by FGF-2 (Fig. 4A). However, the protein expression of OCN showed no evident changes at day 7 and 14 (Fig. 4B).

There was a significant decrease in Col1a1 mRNA levels in the GM + FGF-2 samples (P<0.05), whereas in the OM samples the gene expression was downregulated at day 7 (P>0.05) under the stimulation of FGF-2 (Fig. 3B). OCN and EGFR expression was upregulated in GM + FGF-2 (P<0.05); but downregulated to various degrees in the OM + FGF-2 samples (P>0.05) (Fig. 3C and D). Compared with the GM condition, OM culture stimulated OCN and EGFR expression in the absence of FGF-2.

After 14 and 21 days of culture, the calcium content in the cultures was measured. The calcium content in the FGF-2-treated groups was higher than the non-FGF-2 groups (P>0.05). However, in the OM + FGF-2 group there was a significant reduction of the calcium content (P<0.05) (Fig. 5A).

Alizarin Red S staining indicated that mineralized nodes were observed only in the OM, and mineralized matrix deposition gradually increased with time in culture. The addition of FGF-2 resulted in reduced calcium deposition as evaluated at days 7, 14 and 21 (Fig. 5B).