This article is mentioned in:

Introduction

Periodontitis is a globally prevalent human inflammatory disease that affects approximately 15% of the adult human population and is characterized by periodontal tissue destruction (1). Patients afflicted with periodontitis commonly exhibit clinical symptoms, such as the formation of deep periodontal pockets, extensive alveolar bone loss, the development of a necrotic or exposed cementum and inflammation of the connective tissues (2). Uncontrolled periodontal tissue destruction eventually leads to tooth migration, increased tooth mobility, tooth loss and impairments in chewing function, phonetics and esthetics. Regeneration of destroyed alveolar bone is one of the ultimate goals of periodontal therapy (1). Human periodontal ligament cells (hPDLCs) have multipotential characteristics and have been regarded as potentially useful sources for the regeneration of periodontal tissue, including bone, cementum and the periodontal ligament (2–5).

Calcium phosphate ceramics were introduced over 30 years ago as bone substitutes and have been used as bone-filling materials to provide scaffolds for cell migration and rapid bone formation due to their close chemical and crystal resemblance to bone minerals (6). The most widely used forms of calcium phosphate ceramics are hydroxyapatite (HA, Ca10[PO4]6[OH]2) and tricalcium phosphate (TCP, Ca3[PO4]). The physicochemical properties of these two synthetic calcium phosphates vary widely (7). β-tricalcium phosphate (β-TCP) ceramics are typical representatives that have been extensively investigated regarding their applications for bone engineering due to their osteoconductivity and bioresorbability (8–10). However, β-TCP ceramics have been reported to have limited osteoinductivity when mixed with hPDLCs (11–13). Additionally, it remains controversial as to whether β-TCP possesses intrinsic osteoinductive activity.

Demineralized bone matrix (DBM) is an osteoconductive and osteoinductive commercial biomaterial and an approved medical device for use in bone defects that has a long track record of clinical uses in diverse forms (14). Unlike synthetic materials and mineralized allografts, DBM is well known for being osteoinductive and capable of stimulating bone formation in both heterotopic and orthotopic implant sites due to its content of osteogenic factors, including bone morphogenetic proteins (BMPs) and other osteogenic non-collagenous proteins (15). Numerous studies have demonstrated the potent effects of DBM on the differentiation of osteoprogenitor cells into osteoblasts (14). However, only a few studies have documented the effects of DBM on the proliferation and osteogenic differentiation of hPDLCs in periodontal regeneration (16,17).

Thus, the objective of the present study was to examine the effects of β-TCP ceramics and DBM on the proliferation and osteogenic differentiation of hPDLCs.

Materials and methods

Cell culture

hPDLCs were obtained from healthy premolar teeth that were extracted for orthodontic reasons from subjects under the age of 20 years. The experimental protocol was approved by the Ethics Committee of Sun Yat-sen University, Guangzhou, China and informed consent was obtained from all subjects. The hPDLCs were isolated from fresh periodontal ligament (PDL) tissues using the protocols described in our previous study (18). Briefly, PDL tissue was removed from the middle-third of the root using a sterile scalpel and then dissected into small sections and soaked in fresh Dulbecco's modified Eagle's medium (DMEM/high glucose; HyClone, Beijing, China) containing 20% fetal bovine serum (FBS; BioInd, Kibbutz Beit Haemek, Israel) and 2% (v/v) penicillin/streptomycin (Invitrogen, Grand Island, NY, USA). The cells that migrated out of the tissue samples were cultured at 37°C in a humidified atmosphere of 5% CO2. Cells at passages 3–5 were used in the subsequent experiments.

Scaffold processing and cell seeding

The porous β-TCP ceramics were obtained from the National Engineering Research Center for Biomaterials (Chengdu, China). The mean pore sizes ranged from 200 to 500 μm, and the total porosity was >60% (Fig. 1A). The ceramics were machined into cubes with dimensions of 5×5×5 mm. Prior to use, all scaffolds were sterilized for 15 min (121°C, 15 bar pressure) by autoclaving.

Figure 1 Scanning electron microscopy (SEM) observations of human periodontal ligament cell (hPDLC) growth within the β-tricalcium phosphate (β-TCP) and demineralized bone matrix (DBM) scaffolds. SEM images showing the inner morphologies of the (A) β-TCP and (B) DBM. Images taken on day 3 showing that the cells had successfully adhered to the (C and E) β-TCP and the (D and F) DBM scaffolds. The cells cultured on the β-TCP appeared spindle-shaped, and those within the DBM were triangle-shaped.

DBM collagen scaffolds were prepared from the spongy bone of humans. The spongy bone was provided by Osteolink Biomaterials Inc. (Wuhan, China). The demineralization process was similar to that described in the studies by Urist (19) and Reddi and Huggins (20), with the addition of several washing steps to remove the residual chemicals. Briefly, the spongy bone was separated and soaked in ethyl ether for 24 h to remove the fatty components. Subsequently, 0.6 M HCl was used to demineralize the spongy bone, and this step was followed by a complete washing in phosphate-buffered saline (PBS) and subsequent freeze-drying. Prior to use, all the samples were cut into 5×5×5-mm cube sizes and sterilized by Co60 radiation (Fig. 1B).

Prior to cell seeding, all scaffolds were pre-wet and incubated in DMEM supplemented with 10% FBS and 2% penicillin/streptomycin for 4 h. Subsequently, the hPDLCs were seeded on the scaffolds at two different densities. The cell-scaffold complexes were transferred onto 24-well plates.

For the cell proliferation assay, 20 μl of cell suspension (1×105 cells/ml) was used to seed the cells on each scaffold. The cell-scaffold complexes were then cultured in DMEM growth culture medium supplemented with 10% FBS and 2% penicillin/streptomycin.

For the cell differentiation assay, 100 μl of cell suspension (3×106 cells/ml) was seeded on each scaffold. The cell-scaffold complexes were then cultured in growth medium or osteogenic culture medium containing 10 mM β-glycerophosphate (β-GP), 10−8 M dexamethasone (Dex), 50 μg/ml L-ascorbic acid and 50 μg/ml gentamycin. The culture medium was refreshed every 3 days.

Scanning electron microscopy (SEM) analysis

On day 3 after cell seeding, the cell-scaffold complexes were washed in PBS twice and fixed in 2.5% glutaraldehyde for 4 h. Then, the samples were dehydrated in a graded series of ethanol and air-dried in tetramethylsilane (Merck, Darmstadt, Germany). After gold sputtering, the specimens were examined under a scanning electron microscope (JSM 5600; JEOL, Tokyo, Japan) at 15 kV.

Cell proliferation assay

The proliferation of the hPDLCs cultured on the scaffolds was measured in the growth culture medium using the Cell Counting kit-8 (CCK-8; Beyotime, Shanghai, China). In brief, the hPDLCs on the scaffolds were cultured in growth medium for 1, 3, 5, 7, 9 or 11 days with 4 repeats for each group. The cell-scaffolds complexes were cut into small sections, and the cells were removed from the scaffolds by trypsinization. Ten microliters of CCK-8 solution were placed into each well. The absorbance of the supernatant was read using an enzyme-labeled meter (Infinite 200; Tecan, Grödig, Austria) at 490 nm. A standard curve relating cell number to absorbance was established to determine the cell numbers. The scaffolds with the cells in the same media were used as the negative controls. The monolayer cultures of hPDLCs in the microplates served as the positive controls.

Alkaline phosphatase (ALP) activity

Following 3, 7, 14, 21 and 28 days of culture, ALP activity was measured using an ALP activity assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) following the manufacturer's instructions. All analyses were performed for 3 separate experiments. Briefly, the cell-scaffold complexes were cut into small sections, and the proteins from the cell-scaffold complexes were extracted by the addition of a cell lysis buffer containing 0.1% Triton X-100 to the samples. Aliquots (50 μl in each well) of these supernatants were placed into 24-well plates containing 50 μl of an ALP substrate solution (2 mM MgCl2 and 16 mM p-nitrophenyl phosphate). Following incubation at 37°C for 30 min, the reaction was terminated by the addition of 50 μl of 0.2 M NaOH, and the liberated p-nitrophenol was measured on a plate reader at 520 nm.

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

The expression levels of Runx2 and osteocalcin (OCN) were confirmed by RT-qPCR on day 21. Total RNA from the cells was prepared from the crushed sections of each scaffold (n=3) and then harvested using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The total RNA was used as a template for the synthesis of cDNA with oligo(dT) and RevertAid™ M-MuLV reverse transcriptase (MBI/Fermentas, Burlington, ON, Canada). The following PCR amplification reaction employed Taq polymerase and specific primers. Each PCR was duplicated with the same total RNA. The relative gene expression levels of Runx2 and OCN were normalized to the housekeeping gene, glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The primers for the selected genes are listed in Table I.

Table I Primer sequences used for RT-qPCR.

Table I

Primer sequences used for RT-qPCR.

Genes	Primers	Size (bp)
Runx2	F: CCAACCCACGAATGCACTATC R: TAGTGAGTGGTGGCGGACATAC	91
OCN	F: CACTCCTCGCCCTATTGGC R: GCCTGGGTCTCTTCACTACCT	148
GAPDH	F: CATGTTCCAATATGATTCCACC R: GATGGGATTTCCATTGATGAC	88

[i] F, forward; R, reverse; OCN, osteocalcin.

Statistical analysis

All experiments were performed 3 or 4 times, and each treatment was conducted in triplicate. The means (Ms) and standard deviations (SDs) were calculated, and the statistical significances of the differences between groups were examined with one-way ANOVA followed by the least significant difference (LSD) test. The SPSS 17.0 program was employed for all statistical analyses, and differences were considered significant when the P-value was <0.05.

Results

Scanning electron microscopy (SEM)

On day 3 after cell seeding, the hPDLCs had successfully adhered to both scaffolds. The cell morphologies of the hPDLCs on the β-TCP ceramics and DBM surfaces exhibited some differences. The cells cultured on the β-TCP surfaces appeared spindle-shaped and much more spread out than the cells cultured on the DBM surfaces, and the hPDLCs within the DBM exhibited a triangular shape (Fig. 1C–F).

Cell proliferation

Compared with the monolayer group (positive control), the cell proliferation levels in the scaffolds were significantly inhibited from days 3 to 7, but significantly elevated from days 9 to 11 (P<0.05). A comparison of the growth of the cells in the scaffolds revealed that the cell numbers in the DBM from days 1 to 11 were lower than those in the β-TCP ceramics. However, the difference in cell proliferation between these 2 groups was not statistically significant (P>0.05) with the exception of day 3 (Fig. 2).

Figure 2 Proliferation of human periodontal ligament cells (hPDLCs) cultured on β-tricalcium phosphate (β-TCP) and demineralized bone matrix (DBM) scaffolds. The rates of hPDLC proliferation in the β-TCP were slightly greater than those in the DBM from days 1 to 11, but these differences were not significant with the exception of day 3. *P<0.05, β-TCP compared with the control; #P<0.05, DBM compared with the controls, ΔP<0.05, DBM compared with β-TCP.

ALP activity

In both the growth culture medium and the osteogenic culture medium, the ALP activity gradually increased from days 3 to 28. The hPDLCs cultured in the osteogenic medium exhibited greater ALP activity than did those (same scaffoled different medium) cultured in the growth medium. The ALP activity of the hPDLCs cultured on the DBM was significantly greater than that of the cells cultured on β-TCP at all time points in both the growth medium and the osteogenic culture medium (Fig. 3).

Figure 3 Alkaline phosphatase (ALP) activity of the human periodontal ligament cells (hPDLCs) that were cultured on the β-tricalcium phosphate (β-TCP) and demineralized bone matrix (DBM) scaffolds. In both the growth culture medium (GM) and the osteogenic culture medium (OM), the ALP activity of the cells that were cultured on the DBM was significantly increased compared with that of the cells cultured on β-TCP at all time points with the exception of day 3. *P<0.05, compared with the controls.

Expression of osteogenic differentiation genes

After 21 days of culture, the hPDLCs cultured on the DBM exhibited significantly greater levels of Runx2 and OCN both in the growth medium and the osteogenic culture medium (P<0.05) compared with those cultured on β-TCP. The mRNA expression of Runx2 was increased in the presence of the osteogenic inducers to a greater extent than in the absence of the osteogenic inducers. Runx2 expression was induced to a greater extent with osteogenic inducers only with DBM. It was not as high with β-TCP with osteogenic inducers. The mRNA expression of OCN exhibited no detectable difference between the different culture media (Fig. 4).

Figure 4 RT-qPCR of osteogenic differentiation markers in human periodontal ligament cells (hPDLCs). After 21 days of culture, the hPDLCs within the demineralized bone matrix (DBM) exhibited significantly greater levels of Runx2 and osteocalcin (OCN) in both the growth medium (GM) and the osteogenic culture medium (OM) compared with those in the β-TCP. *P<0.05, compared with the control.

Discussion

Scaffolding materials should be biocompatible and bioactive to allow specific cells to attach, proliferate, and differentiate on the scaffolding materials to enable tissue repair and regeneration (4–8). β-TCP is a porous bioceramic material with biological properties that include non-reactivity and resorbability. β-TCP can serve as a scaffold for bone regeneration as it is progressively degraded and replaced by bone. Due to its osteoconductivity and bone replacement capability, β-TCP is highly promising for use in numerous dental and craniofacial procedures, including the reconstruction of frontal sinus cavities, the augmentation of craniofacial skeletal defects and the repair of periodontal tooth and bone defects (21–24). DBM products were clinically introduced in 1991 and have since served as effective osteoconductive scaffolds (25,26). The osteoinductivity of DBM can be attributed to a series of low-molecular-weight glycoproteins that include the BMPs. The decalcification of cortical bone exposes these osteoinductive growth factors that are buried within the mineralized matrix and thereby enhances the bone formation process (27). DBM has a number of additional advantages that make it an attractive bone graft alternative. DBM is cost-effective and readily available from tissue banks. The demineralization process destroys the antigenic materials in the bone, which makes DBM less immunogenic than mineralized allografts (25,28).

The present study investigated the effects of β-TCP ceramics and DBM, which are commercially available bone graft substitutes, on the proliferation and osteogenic differentiation of hPDLCs. A CCK-8 assay revealed that both the β-TCP scaffolds and DBM were able to promote the late, but not the early proliferation of hPDLCs. SEM observations revealed that the hPDLCs attached to the surfaces of the β-TCP scaffolds and the DBM by spreading themselves out. Such responses are indicative of good cytocompatibility and the close interactions of the scaffolds with the hPDLCs. The cells on the β-TCP surfaces appeared spindle-shaped and much more spread out than those on the DBM surfaces and the hPDLCs within the DBM became triangle-shaped. These different cellular morphologies may be associated with the differentiation of the hPDLCs cultured on DBM toward the osteogenic lineage.

ALP is an early biomarker of osteogenic differentiation and plays a key role in bone mineralization through the initiation and/or promotion of the formation of hydroxyapatite crystals in the matrix vesicles of osteoblasts (29). The present study revealed that the hPDLCs cultured on the DBM exhibited greater ALP activity than did the cells cultured on the β-TCP scaffolds. These results may be related to the different chemical compositions of the scaffolds. β-TCP is biodegradable and can release various concentrations of calcium and phosphate ions into the culture media or body fluids (30–32). In a previous study, the concentrations of calcium and inorganic phosphate ions released from β-TCP ceramics after 3 days of culture were 1.76 and 0.96 mmol/l, respectively (33). In a previous study of ours, we demonstrated that this release of free calcium and inorganic phosphate ions significantly reduces the ALP activities of hPDLCs (18). However, the DBM was prepared by acid extraction from human bone sources and thus retained type I collagen, other proteins and BMPs. BMPs are obviously able to stimulate ALP activity (34,35).

The differentiation of hPDLCs is one of the key processes of bone regeneration. The mRNA expression levels of Runx2 and OCN were selected as markers of the osteogenic phenotype. β-TCP ceramics possess osteoconductive properties, but they do not have intrinsic osteoinductive capacities. DBM has been shown to have osteoconductive and osteoinductive properties (36). In the present study, compared to the β-TCP ceramics, the DBM scaffolds upregulated the expression levels of these genes in both the growth and osteogenic culture media on day 21. These results are consistent with those of a previous study by Kasten et al (37) who reported that DBM increased OCN gene expression in human bone marrow stromal cells. Runx2 has been identified as a ‘master gene’ in the control of osteogenic differentiation (38,39). Thus, the significant upregulation of Runx2 gene expression led to elevated mRNA expression levels of OCN in the hPDLCs within the DBM. Runx2 is an early-stage transcription factor that activates osteoblastic differentiation (36,37), and its gene expression was significantly increased in the DBM substrates (1.9- and 2.6-fold in the growth and osteogenic media, respectively). OCN is a late-stage osteogenic differentiation marker and is synthesized only by mature osteoblasts, odontoblasts and cementoblasts (38,39). The results from RT-qPCR revealed that OCN gene expression was slightly increased in the DBM substrates (1.5- and 1.3-fold in the growth and osteogenic media, respectively). Thus, we conclude that DBM promoted the osteogenic differentiation of hPDLCs.

β-TCP is a biodegradable inorganic bone substitute that can release varying amounts of calcium and phosphate ions into culture media or body fluids (30,31,40). These releases of free calcium or phosphate ions significantly affect the proliferation and osteogenic differentiation of hPDLCs (18). DBM is an allogenous material and thus contains bone morphogenic and matrix proteins that ceramic materials do not (14). The results of the present study may be due to the sum of the effects of the dissolved ions and matrix proteins and the direct effects of the scaffold architecture and surface chemistry. Notably, the chemical composition, surface topography, surface roughness and other characteristics of scaffolds have been reported to affect cell activities, such as adhesion, proliferation and differentiation (41).

In conclusion, the results of the present study revealed that in addition to the induction of cell proliferation in the absence of an osteogenic inducer, DBM scaffolds exhibited greater effects on the osteogenic differentiation of hPDLCs than did porous β-TCP ceramics. The organic elements of bone have more prominent effects on the osteogenic differentiation of hPDLCs than do the inorganic bone elements. It may thus be possible to develop novel periodontal tissue engineering scaffolds by combining the organic (e.g., collagen, other proteins and BMPs) and inorganic elements (e.g., calcium and phosphate ions) of bone in various concentrations and ratios.

Acknowledgments

This study was supported by the Guangdong Natural Science Foundation (grant no. S2012040007041).

References