Effect of BMP- 2, -4, and -7 on Proliferation and Osteogenic Differentiation in Cultured Human PDLSCs

Aim To investigate the effects of bone morphogenetic proteins (BMPs) 2, 4, and 7 on proliferation and osteogenic differentiation in human periodontal ligament stem cells (PDLSCs). Methods PDLSCs were isolated by an immunomagnetic method. Expression of cell surface antigens CD146, CD44, and CD34, and pluripotency (osteogenic and adipogenic) were measured. Cultured PDLSCs were treated, in dose- and time-dependent experiments, with single BMPs, with 1:1 combinations, and with a mix of all three BMPs (1:3 each). For dose-dependent experiments, PDLSCs were incubated for 12 d with media containing BMPs at 0, 10, 25, 50, and 100 ng/ml. For time-dependent experiments, PDLSCs were treated with media containing 50 ng/ml BMPs for 0, 3, 6, 12, and 24 d. Cell growth and alkaline phosphatase activities were measured by MTT and an enzyme kit. Immunohistochemistry and western blotting were used to detect osteogenic differentiation-related proteins, i.e., osteocalcin, bone sialoprotein, collagen type I, and collagen type III. Results PDLSCs displayed CD146 (93%) and CD44 (91.2%) positive expression; CD34 (1.8%) showed negative expression. All cells exhibited osteogenic and adipogenic potential. The proliferation and alkaline phosphatase activities of PDLSCs treated with the aforesaid single and combined BMPs increased in a dose- and time-dependent manner; proliferation and alkaline phosphatase activity were greater with the BMP combinations. Compared with the control group, the levels of osteogenic differentiation-related proteins increased markedly in PDLSCs treated with 50 ng/ml BMPs for 12 d, whereas no signicant differences were observed between the different BMP treatments. Conclusion BMP-2, -4, and -7, singly and in combination, promoted development and osteogenic differentiation of PDLSCs, and both cellular outcomes were more pronounced with BMP combinations.


Introduction
Periodontitis is a widespread chronic in ammatory disease characterized by irreversible disruption of the supporting periodontal tissues. If untreated, periodontitis may cause irreversible destruction of periodontal ligament, gingival tissue and supporting alveolar bone. Ultimately, tooth loss occurs (1), which may seriously affect an individual's physical and mental health. The goal of periodontitis treatment is complete reconstruction of disrupted periodontium to its original shape and function. Current conventional therapies, such as scaling and root planning (2), can prevent in ammatory progression and reconstruction of lost attachment to some extent; however, alveolar bone loss is not restored satisfactorily. In 2004, Seo et al. extracted human periodontal ligament stem cells (PDLSCs) that, for the rst time, exhibited self-renewal and differentiation into various cell types (3). These cells have been considered a highly promising stem cell population for regeneration therapy in periodontium (4). However, the proliferation and osteogenic potential of PDLSCs were controlled by a variety of growth factors, extracellular matrix components, and transcription factors. Thus, there is still a challenge to simplify and improve induction of PDLSC growth and osteogenic differentiation for periodontal regeneration.

Page 3/17
The bone morphogenetic proteins (BMPs), a 20-member family of secreted extracellular matrixassociated proteins, are widely expressed in PDLSCs while their physiological levels are insu cient for appropriate regeneration during periodontal disease (5). Among all the BMPs, BMP-2 is demonstrated to possess the greatest potential to promote bone formation (6,7). However, exogenous BMP-2 protein is unstable and can lose bioactivity rapidly. In order to maintain the su cient BMP-2, overdose BMP-2 was usually used in clinical practice, while a growing clinical complications have emerged, including postoperative in ammation, ectopic bone formation, and osteoclast-mediated bone resorption (8).
Besides, another problem associated with the clinical application of BMP-2 is its high cost due to the need for high doses. BMP-4 has been shown to regulate bone and cartilage formation, morphogenesis, cell proliferation and apoptosis of a wide variety of tissue and cells (9). For clinical applications, recombinant human BMP-4 (rhBMP-4) is produced in Chinese hamster ovary (CHO) cells, while achieving a higher titer of functionally active rhBMP-4 in CHO cell cultures remains a challenge (10,11). BMP-7 has been approved for clinical practice to promote bone formation in the United States, Europe, and Australia (12). However, the clinically effective doses of BMP-7 are extremely high, which leads to a heavy economic burden for patients and causes many potential side effects, such as overstimulation of osteoclastic differentiation and topical bone formation at unintended sides (13). Therefore, in the present study, we examined the effects of BMP-2, -4 and -7, singly and combined, on the proliferation and osteogenic differentiation of cultured PDLSCs to nd a more effective method for periodontium regeneration.

PDLSCs isolation and culture
Human periodontal ligament (PDL) tissues were collected from the mid-root surface of third molars extracted for orthodontic reasons from six patients (3 male, 3 female, ages 18-25 years). The tissues were digested with collagenase Type I to generate single-cell suspensions. The cells were cultured in Dulbecco's modi ed Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) and 1% Penicillin-Streptomycin Solution at 37℃ in a 5% CO 2 humidi ed atmosphere. The cells from one tooth were seeded onto a 60 mm plastic tissue culture dish. The medium was replaced every 2-3 days to enable further growth. The cells grown to 70% con uence were de ned as passage zero (P 0 ). Later passages were named accordingly. P 3 PDLSCs were isolated by an immunomagnetic method using the marker stromatin-1 (STRO-1, R&D Systems, Minneapolis, MN, USA) (14). Brie y, 3×10 7 cells/ml suspensions were incubated with mouse anti-STRO-1 antibody (1:200 dilution) at 4℃ for 30 min. The cells were washed with PBS and incubated with rat anti-mouse IgM-conjugated magnetic beads (Dynal Biotech, Bromborough, Wirral, UK.) on a rotary mixer at 4℃ for 1 h. The bead-bound cells were isolated by a magnet. After washing, the bead-bound cells were selected using a magnetic particle concentrator. Finally, STRO-1-positive cells were counted and harvested for further analysis.
PDLSCs identi cation Surface marker evaluation: P 3 PDLSCs (n=6) were harvested and resuspended in culture medium at 1×10 6 cells/ml. One ml of cell suspension was mixed with 10 μl of antibodies against CD146, CD44, or CD34 (1:100 dilution, R&D Systems, Minneapolis, MN, USA) and incubated at room temperature for 1 h. After washing with PBS, the cells were treated with FITC-conjugated goat anti-mouse IgG secondary antibody (1:100 dilution) for 30 min in the dark at room temperature. Finally, the cells were analyzed with a FACS Calibur ow cytometer (Becton Dickinson, Franklin Lakes, NJ).

Western blotting
Total protein (n=6) was collected with RIPA buffer, and the concentration was measured with a BCA TM protein assay kit (Thermo Scientific Pierce, Rockford, IL, USA). Equal protein amounts (15 μg) were separated in 10-15% SDS-PAGE and transferred to PVDF membranes (Millipore, Billerica, MA, USA). Following blocking with 5% skim milk, the membranes were incubated overnight with anti-osteocalcin, anti-bone sialoprotein, anti-collagen type I, anti-collagen type III, and anti-β-actin antibodies (1:100 dilution) at 4℃. The membranes were then treated with horseradish peroxidase-conjugated anti-rabbit or anti-mouse antibody (1:100 dilution) for 1 h at 37℃. The protein bands were developed with an enhanced chemiluminescence kit (Millipore, Billerica, MA, USA).

Statistical analysis
Data analysis was performed with SPSS 19.0 software and the data were expressed as the mean ± SEM.
Differences between groups were analyzed by Student's t test, and one-way ANOVA with Tukey's test was applied for more than two groups. A value of P < 0.05 was considered statistically signi cant.

Results
Morphology, characterization, osteogenic, and adipogenic differentiation of PDLSCs We cultured human PDL cell suspensions. On day 5, the cells attached to the asks displayed typical broblastic morphology ( Figure 1A). About 14 days was required to reach 70-80% con uence at P 0. . We used the marker STRO-1 and immunomagnetic beads to puri ed the PDLSCs at P 3 ( Figure 1B).
Subsequently, we used ow cytometry to characterize PDLSC surface markers. The cells expressed mesenchymal stem cell markers CD146 and CD44, with the positive rates were 93% and 91.2%, respectively. However, PDLSCs were negative for CD34 (1.8%) ( Figure 1C, D and E). We used alizarin red and oil red O staining to assess osteogenic and adipogenic differentiation of PDLSCs. The PDLSCs had formed calci ed deposits and developed into lipid-laden fat cells ( Figure 1F and G).

Effect of BMPs on the ALP activity in PDLSCs
We performed concentration and time course experiments to assess the effect of BMP-2, -4 and -7 on ALP activity (ALP is an early marker of osteogenic differentiation). All BMP treatments produced dosedependent increases in PDLSCs ALP P<0.05 . Compared with single BMP treatments, ALP activity was greater with the combined BMP treatments at 50 ng/ml and 100 ng/ml P<0.05 . At 10 ng/ml and 25 ng/ml, the ALP activity was signi cantly increased only with the BMP-2+4+7 treatment P<0.05 Figure  3A . ALP activity also increased in a time-dependent manner in PDLSCs treated with all BMPs P<0.05 . Compared with single BMP treatments, ALP activities for all BMP combination treatments were higher at 6, 12 and 24 d P<0.05 whereas, at 3 d, only the BMP-2+4+7 group exhibited increased ALP activity (P<0.05) Figure 3B .

Effect of BMPs on osteogenic differentiation-related proteins in PDLSCs
We used immunohistochemistry and western blotting to assess the effects of BMP-2, -4, -7 on PDLSCs osteogenic differentiation-related proteins. Compared with the controls, the levels of osteocalcin, bone sialoprotein, collagen type I, and collagen type III treated with 50 ng/ml BMPs for 12 d were all increased P<0.05 , and we did not observe any signi cant differences between the BMP treatment groups Figure 4 and 5 .

Discussion
Periodontitis may cause irreversible destruction of tooth attachment and the surrounding bone, eventually leading to tooth-loss. Complete and predictable reconstruction of periodontal tissue destroyed by periodontal diseases is still a major challenge. Ideal periodontal reconstruction demands consideration of many factors related to periodontal development, including the use of a stem cell population, signaling molecules, and scaffold materials in ordered, temporal, and spatial sequences (15). PDLSCs, which were implicated in the generation of alveolar bone, cementum and periodontal ligament, have shown selfrenewal and the potential to differentiate into multiple cell types that could produce a periodontal ligament-like structure and cementum similar to natural periodontal complex. However, successful PDLSC proliferation, migration, and maturation depended on inclusion of growth factors and full contact with extracellular matrix to control gene expression. Conversely, healing may be compromised and occur by repair instead of regeneration (16,17). In this study, we isolated and characterized PDLSCs from the periodontal ligament of human third molars. The cells were positive for mesenchymal stem cells markers CD146 and CD44, and displayed osteogenic and adipogenic differentiation potential. Therefore, we con rmed that the cells were PDLSCs, and they could be used for subsequent experiments.
One of the most promising approaches in periodontal regeneration has been the application of growth factors, especially BMPs. Most BMPs stimulate osteogenesis, with BMP-2, BMP-4 and BMP-7 being the most potent. Investigators have found that BMP-2, BMP-4 and BMP-7 separately may stimulate proliferation and osteoblast differentiation of PDLSCs (18)(19)(20)(21). However, there were few studies that reported relationships between PDLSCs proliferation, osteogenic differentiation, and effects of BMP combinations. Thus, we rst examined the effects of BMP-2, -4 and -7 on proliferation and ALP (an early marker of osteogenic differentiation) activity. Proliferation and ALP activity increased, in a dose-and time-dependent manner, for all single and combined BMP treatments. Notably, growth and ALP activity increased to greater extents with BMP combinations.
Stem cell osteogenic differentiation and maturation of a mineralized bone matrix can be divided into early, middle, and late stages. Makers of the middle and late stages, such as collagen type I, collagen type III, osteocalcin, and bone sialoprotein, are often used to evaluate the effect of growth factors on osteogenic differentiation (22,23). Thus, we further investigated the effect of BMPs on the levels of collagen type I, collagen type III, osteocalcin, and bone sialoprotein proteins in PDLSCs. Compared with the controls, all the maker proteins exhibited increased abundances in PDLSCs treated with 50 ng/ml BMPs for 12 d; further, we did not observe any signi cant differences between the BMP treatment groups. However, Qing et al. (24) reported that forced BMP-2 and BMP-7 activation with lentiviral vectors in rat adipose stem cells signi cantly increased expression of osteocalcin and osteopontin, and enhanced the rate of new bone formation. We speculate that the differences between our results and the ndings of Qing et al. may have been due to the use of different types of stem cells. In addition, another interesting nding in our study was the difference in magnitude of response between early, middle, and late stage markers of osteogenic differentiation with combined BMP stimulation. This nding indicated that combined BMP stimulation of PDLSCs may be more important for inducing early markers compared with middle and late stage markers of osteoblastic differentiation. However, the mechanism of this effect requires further investigation.
Aoki et al. reported that combined BMP-4 and BMP-6 proteins had a synergistic effect on C2C12 myoblasts (25), which agrees with our results that BMP-2, BMP-4, and BMP-7 had a synergistic effect on PDLSCs proliferation and early osteoblastic differentiation, while Qing et al. showed in vitro that recombinant BMP-2 and BMP-7 enhanced osteoblastic differentiation under combined BMP gene transfer (24). The question of whether combined BMP gene transfer versus combined BMP protein have enhanced effects due to a similar mechanism should be a fruitful area of investigation that may yield important understanding of how BMP therapy may be improved.
In conclusion, we demonstrated that BMP-2, BMP-4, BMP-7, singly and in combination, promoted proliferation and osteogenic differentiation of PDLSCs, and the increases in both aspects of cellular behavior were greater with the various BMP combinations. Although the abundances of collagen type I, collagen type III, osteocalcin, and bone sialoprotein increased with all BMP treatments, there were no signi cant differences in protein expression between the treatment groups. However, all the experiments were conducted in vitro in the present study, and experiments in vivo are needed in the further study to con rm our results.

Declarations
Ethics approval and consent to participate The Ethics Committee of Xi'an Jiaotong University granted written approval to obtain human periodontal ligament samples, and all patients in this study provided written informed consent.
Consent for publication All Authors have agreed to publish the manuscript in BMC Oral Health.
Competing interests The Authors declare that no con ict of interest exists.
Funding This study was funded by the Key Science and Technology Program of Shaanxi Province, China (2014K11-03-08-05).
Authors' Contributions This study is a product of the intellectual efforts of entire team; all members have contributed to various degrees. The contributions of each author are listed as follows: Juedan Li (study design, data acquisition and analysis, statistical analysis, manuscript draft); Min Wang (data acquisition and analysis, statistical analysis, manuscript revision for important intellectual content); Min Cui and Cheng Chen (data acquisition and analysis, statistical analysis); Zheng Cheng (study conception and design, nal approval of the article, obtained funding, overall supervision). All authors read and approved the nal manuscript.

Figure 4
Immunohistochemistry of osteocalcin, bone sialoprotein, collagen type I and collagen type III proteins in PDLSCs treated with 50 ng/ml BMPs for 12 d (×200). Osteogenic differentiation-related proteins were detected by immunohistochemistry. A E: osteocalcin B F: bone sialoprotein C G: collagen type I D H: collagen type III. *P 0.05, compared with the control group; NS: no signi cance; Data is presented as the mean ± SEM. Figure 5 Western blotting of osteocalcin, bone sialoprotein, collagen type I and collagen type III proteins in PDLSCs treated with 50 ng/ml BMPs for 12 d. Osteogenic differentiation-related proteins were detected by western blotting. β-actin was used as a loading control. A: osteocalcin B: bone sialoprotein C: collagen type I D: collagen type III. *P 0.05, compared with the control group; NS: no signi cance; Data are presented as the mean ± SEM.