Pro-Inflammatory Cytokine TNF-α Attenuates BMP9-Induced Osteo/ Odontoblastic Differentiation of the Stem Cells of Dental Apical Papilla (SCAPs) Open Access

While occurring in all ages periapical periodontitis of deciduous tooth can be caused by caries, tooth trauma or fracture of abnormal central cusp and is one of the most prevalent oral diseases in children. The inflammatory injury of the periapical tissue may lead to bone defects and disturb the root development. Bone defects can aggravate the immature teeth looseness and increase the treatment difficulty. The limitations of current treatments have attracted increased interests in exploring the new treatment methods [1, 2]. In recent years, tissue regenerative engineering has been a focus of tooth tissue repair. Dental stem cells have multiple differentiation potential to regenerate intended tissue types and are the basis of any successful tissue engineering.

The stem cells from apical papilla (SCAPs) are one of the five types of dental stem cells, which are considered as a population of mesenchymal stem cells (MSCs)-like cells. Several studies have demonstrated that SCAPs possess the osteo-/odontogenic potential and researchers believed that SCAPs could be valuable seed cells for the periapical tissue regeneration [3‒6]. However, the existing studies were mostly performed under minimal inflammation condition. During the inflammation phase of periapical periodontitis chemokines and pro-inflammatory cytokines are largely released by activated macrophages. Growing evidence suggested that the osteogenic differentiation of MSCs may be regulated by molecular signals from the extracellular microenvironment, and the pro-inflammatory cytokines in the cell microenvironment [7‒10].

Tumor necrosis factor-α (TNF-α) is one of largest quantity of pro-inflammatory cytokines during the inflammation phase of periapical periodontitis, which was considered as a major contributor to bone pathophysiology due to its stimulation of bone resorption [11‒13]. In mature osteoblasts, TNF-α inhibits the expression of genes necessary for bone formation and stimulates the expression of genes that induce osteoclasts [14, 15]. It was shown that high dose of TNF-α suppressed the osteogenic differentiation of MSC-like cells by decreased levels of alkaline phosphatase (ALP), osteocalcin (OCN), and Runx2 [16‒18]. On the other hand, several studies suggest that TNF-α may have an inhibitory effect on BMP-induced osteoblast differentiation of MSCs or preosteoblastic cells [19‒21].

BMP9 (also known as growth and differentiation factor 2, or GDF2) is one of the most potent factors that can induce osteogenic differentiation of MSCs [22‒24]. We have recently established the reversibly immortalized mouse stem cells of apical papilla tissue of mouse lower incisor teeth (iSCAPs), which were shown to retain the ability to differentiate into multiple lineages [25, 26]. We also demonstrated that BMP9 can effectively induce osteo/odontoblastic differentiation of the iSCAPs [25, 26]. However, considering that tissue regeneration may occur under the inflammation phase of periapical periodontitis, the effect of high dose of TNF-α on BMP9-induced osteogenic differentiation of iSCAPs remains unclear.

In this study, we established a rat model of periodontitis of the immature permanent teeth with periapical lesions to detect TNF-α expression during the inflammatory phase in vivo. Furthermore, we investigated the effects of high doses of TNF-α on BMP9-induced osteoblastic differentiation of iSCAPs. Our results demonstrate that TNF-α plays an important role in in the induction of periapical bone defect during the inflammatory phase and inhibits BMP9-induced osteoblastic differentiation of iSCAPs in periapical bone regeneration. Therefore, our results strongly suggest that BMP9 may be further explored as a potent osteogenic factor to improve the efficacy of osteoblastic and odontogenic regeneration in tooth engineering under chronic inflammation conditions.

The use and care of all animals in this study was approved by the Experimental Animal Ethics Committee of Chongqing Medical University, Chongqing, China. Thirty-five Sprague-Dawley rats (male, 4-week-old) were purchased from the Laboratory Animal Center of Chongqing Medical University (Chongqing, China) and were randomly divided into seven groups. Animals were anaesthetized by intraperitoneal injection of pentobarbital (30mg/kg). The pulp of maxillary first molar of each rat was exposed to the oral environment. Five rats of each group on days 0 (control), 3, 7, 14, 21, 28, and 35 after lesion induction were sacrificed for microCT and histologic analysis.

The rat maxillae were retrieved and fixed with 4% paraformaldehyde at 4°C for 48h, and then taken in uniform position to hold the same angulation to the film [27]. Digital images were taken using µCT (vivaCT 40, SCANCO) and analyzed the images by calculating the area of mesial root apical lesion using the µCT V6.1 software.

After microCT imaging, the retrieved samples were demineralized with 17% EDTA (pH 7.3) solution for 2 weeks, followed by paraffin embedment. The embedded samples were serially sectioned, deparaffinized, rehydrated and subjected to immunohistochemical staining using the primary antibody (Polyclonal rabbit anti-TNF-α, 1: 100; Abcam, UK; no.ab6671) for 24 h at 4˚C. The primary antibody was then removed by washing the slides thrice with PBS for 5 min. The sections were incubated with secondary antibody kits (SAB, USA; no.L3012-2) according to the manufacturer's protocols. The slides were stained for 30-40s with diaminobenzidine (DAB) and the nucleus was counterstained with hematoxylin for 10s. Images were obtained using a microscope with a camera (Leica, Germany).

The iSCAPs which were previously characterized [25].The 293pTP line was used for adenovirus amplification. Both 293pTP and iSCAPs were maintained in complete Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal bovine serum (FBS, GIBCO,CA), 100 units of penicillin and 100 mg of streptomycin at 37°C in 5% CO₂.Recombinant adenoviruses expressing BMP9 and/or GFP were constructed as previously reported [25, 26]. Polybrene (10µg/ml) was used to enhance infection efficiency for all adenoviral infections [28]. Recombinant mouse protein TNF-α was purchased from Sino-Bio (Beijing, China).

The iSCAPs were seeded in 24-well plates and infected with AdBMP9 or AdGFP. The experiment groups were added with 0.1, 1.0, 10.0, 100ng/ml TNF-α. Alkaline phosphatase (ALP) staining and quantitative assays were done on day 3 and day 5. ALP staining was performed according to the protocol of the NBT/BCIP staining kit (Beyotime-Bio, China). ALP activity was calculated the absorbance value at 405nm based on the rate of conversion of p-nitrophenyl phosphate disodium salt hexahydrate to p-nitrophenol (Sigma, USA). Each sample was performed in triplicate. ALP activity was normalized by calculating the total protein content of the samples.

Total RNA was isolated using the Trizol reagent (Invitrogen), and subjected to reverse-transcription using the cDNA Reverse Transcription Kit (Takara, Japan). The qPCR analyses were performed in the ABI Prism 7500 Real-Time PCR System with the SYBR Green PCR master mix reagent (Takara) in a 40-cycle PCR. PCR cycling program was 95 °C for 30 s, 95 °C for 5 s, 60 °C for 34 s and 95 °C for 15 s. The target gene expression was normalized relative to the level of D-glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The qprimer sequences were: for mouse OPN, 5’-CCT CCC GGT GAA AGT GAC-3’ and 5’-CTG TGG CGC AAG GAG ATT-3’; for mouse OCN, 5’-CCT TCA TGT CCA AGC AGG A-3’ and 5’-GGC GGT CTT CAA GCC ATA C-3’; and for mouse Gapdh, 5’-ACC CAG AAG ACT GTG GAT GG-3’ and 5’-CAC ATT GGG GGT AGG AAC AC-3’.

The iSCAPs were seeded into 24-well plates and infected with AdBMP9 or AdGFP. After grown to 80% confluence, cells were incubated in mineralization medium (normal culture medium containing 50 mg/ml ascorbic acid, 10 mM β-glycerophosphate) supplemented with different concentrations of TNF-α. After 2 weeks, the cultures were washed 3 times in PBS and fixed in 4% polyoxymethylene for 30 minutes, washed with deionized water 3 times, and then stained with 0.4% Alizarin Red S (Sigma-Aldrich, USA) for 30 minutes at room temperature. The excess dye was removed by washing 3 times with deionized water. PBS was added to each well to prevent the cells from drying before the images were acquired.

Stem cell implantation-mediated ectopic bone formation was performed as described [25, 26, 29]. Briefly, the iSCAP cells were infected with AdBMP9 or AdGFP, and treated with TNF-α (10ng/ml) condition for 36h, collected and resuspended in PBS with TNF-α (10ng/ml) for subcutaneous injection (5×10⁶ cells/injection) into the flanks of athymic nude (nu/nu) mice (5 animals per group, 4-6 week old, female, acquired from the Laboratory Animal Center, Chongqing, China). At 4 weeks post implantation, animals were sacrificed, and the implantation sites were retrieved for µCT imaging, histologic evaluation, and special stains.

The retrieved tissues were fixed, decalcified in 17% EDTA solution, and paraffin embedded. Serial sections of the embedded specimens were deparaffinized and stained with hematoxylin & eosin (H & E). Trichrome staining was also carried out as previously described [25, 26, 29].

The quantitative assays were performed in triplicate and/or repeated three times. Data were expressed as mean ± SD. Statistical significances were determined by one-way analysis of variance and the student's t test. A value of p< 0.05 was considered statistically significant.

In the process of establishing the rat model of periapical periodontitis of immature permanent teeth with periapical lesions, we found that the periapical alveolar bone loss after 3-days pulp exposed (Fig. 1A). Lesions area was gradually increased from day 3 to day 14, and then sharply increased until 28 days post exposure. The lesion area kept rather stably from day 28 to day 35. The bone loss area of mesial root apical alveolar was also calculated. Quantitatively, the rapid phase of bone absorption was from day 14 to day 28 (p< 0.01) (Fig. 1B).

Histologic evaluation showed that the apical foramens were not closed and large quantities of inflammatory cells were observed from day 7 to day 28. The bone boss area was filled with connective tissue, and the trend of lesions area development was consistent with that observed by imaging analysis (Fig. 2A). The immunohistochemical analysis indicated that the expression of TNF-α was gradually increased from day 3 to day 7. The expression of TNF-α was significantly elevated from day 14 to day 28 and descended from day 28 to day 35. Thus, the expression of TNF-α was seemingly correlated well with the area of periapical bone defects. These results suggest that TNF-α may be primarily released during the development of periapical lesions and positive correlated with the area of periapical bone defects (Fig. 2B).

We found that the addition of TNF-α suppressed the ALP activity in BMP9-induced iSCAPs on day 3 and day 5 (p< 0.01) (Fig. 3A-B). Furthermore, higher concentrations of TNF-α (10ng/ml and100ng/ml) significantly suppressed the ALP activity, compared to that at the lower concentrations (0.1ng/ml, and 1ng/ml) (Fig. 3A-B), suggesting that the inhibition capability of TNF-α on BMP9-induced ALP activity may be dose-dependent. The qPCR analysis showed that the expression of the late osteoblastic markers OCN and OPN on day 5 was significantly reduced in the presence of TNF-α (10ng/ml, 100ng/ml) (p< 0.01) (Fig. 3 C). Furthermore, Alizarin Red S staining was also carried out to test the late stage mineralization and found that the TNF-α treatments profoundly inhibited in vitro mineralization, compared with the control groups (Fig. 3 D). Collectively, these results strongly suggest that TNF-α may inhibit BMP9-induced osteoblast differentiation of iSCAPs in a dose-dependent manner.

We further investigated if the increase in the titers of AdBMP9 could overcome the inhibitory effect of osteogenic activity by a high concentration of TNF-α (10ng/ml). When iSCAPs were transduced with the increasing titers of AdBMP9, we found that there was a trend of increasing ALP activities at both days 3 and 5 (p< 0.01) (Fig. 4A-B) although ALP activities were not fully recovered, compared to that of the control group (p< 0.01) (Fig. 4A-B). Accordingly, we found the expressions of OCN and OPN were gradually up-regulated with the increasing AdBMP9 titers while at the highest titers the relative expression of OCN and OPN was still lower than that of the control’s(Fig. 4 C). Furthermore, Alizarin red S staining results were consistent with the ALP and qPCR analysis findings (Fig. 4 D). Thus, these results suggest that higher levels of BMP9 may partly overcome the inhibitory effect of high concentration of TNF-α on osteogenesis.

Using our previously established stem cell implantation assay [25, 26, 29], we tested the in vivo effect of high dose TNF-α (10ng/ml) on BMP9-induced ectopic bone formation of iSCAPs. Cells were transduced with AdBMP9, or AdGFP, and incubated with or without TNF-α (10ng/ml) and injected subcutaneously into the flanks of athymic nude mice for 4 weeks. No detectable masses were retrieved from the AdGFP or TNF-α alone group. Robust bony masses were retrieved from BMP9 transduced iSCAPs group, while smaller masses were recovered from the BMP9 plus TNF-α groups as the µCT imaging results revealed the distinct size differences among these samples (Fig. 5A, panel a). Quantitative analysis of the total bone volumes indicates that BMP9 transduction induced significantly more robust bone formation than that transduced with BMP9 in the presence of TNF-α (10ng/ml) (p< 0.01) (Fig. 5A, panel b), suggesting that TNF-α (10ng/ml) may inhibit BMP9-induced bone formation. Accordingly, H & E histological evaluation revealed that BMP9-transduced cells formed apparent trabecular bone, which was significantly reduced in the presence of TNF-α (10ng/ml) (Fig. 5B, panels a vs. b). Trichrome staining confirmed that iSCAPs transduced with BMP9 formed more mature and well-mineralized bone matrices, while the maturity and mineralization were significantly diminished in BMP9 with TNF-α group (Fig. 5B, panels a vs. b). Taken together, these in vivo results strongly suggest that high dose of TNF-α (e.g., 10ng/ml) may significantly inhibit the BMP9-induced bone formation.

During therapeutic bone regeneration, the defective or injured tissues are frequently associated with the presence of abnormal inflammatory and protein mediators [30]. Thus, tissue regeneration may be compromised under chronic inflammation [31]. We and others previously demonstrated that BMP9 is one of the most potent of osteogenic differentiation in mesenchymal stem cells and in iSCAPs [18, 22, 23, 32, 33]. However, it remains unclear if BMP9-induced bone formation will be affected under inflammatory conditions. Here, we investigated the expression of TNF-α in periapical lesions in vivo. Furthermore, we used exogenous TNF-α to simulate the inflammatory microenvironment and evaluated its effects on BMP9-induced osteogenic differentiation of iSCAPs. Our results demonstrate that BMP9-induced osteogenic differentiation can be inhibited by TNF-α in a dose-dependent fashion.

TNF-α is one of the most potent pro-inflammatory cytokines during the inflammation phase. We established an animal model of immature permanent teeth with periapical lesions to observe TNF-α expression during the development of periapical lesions. Our results showed that TNF-α expression was highly correlated with development of periapical periodontitis. Similarly, it was reported that the expression of TNF-α was positively correlated with periapical damage development during the inflammatory phase in rat models [34]. In addition, our results were also consistent with recent clinical studies in which the levels of TNF-α were shown to increase significantly in teeth with periapical pathosis, from smaller to larger lesions [35, 36]. These results suggest that TNF-α may play an active role in periapical bone destruction progress during periapical periodontitis.

It was reported that TNF-α can inhibit the osteogenic differentiation and bone formation [9, 37, 38]. It was shown that high concentrations of TNF-α suppressed the expression of osteogenic indicators in MSCs [9, 17, 39]. TNF-α was shown to inhibit osteoblasts differentiation by suppressing the expression of Runx2 in pre-osteoblastic cells [40‒43]. However, the effects of TNF-α on osteogenic differentiation of SCAPs remain unclear. Here, we demonstrate that osteogenic differentiation of SCAPs was effectively repressed in the presence of high dose of TNF-α. Thus, TNF-α should be considered as an important negative regulator of bone regeneration under inflammatory microenvironment. As BMP9 has been proven to be one of the most potent osteogenic factors to promote bone defect repair and regeneration [22‒24], it is conceivable that BMP9 may be explored as an agent to combat the inhibitory effect of TNF-α on osteogenic differentiation and induction of bone lesions during chronic inflammation.

Our results suggest that high doses of TNF-α (10ng/ml, 100ng/ml) significantly inhibit BMP9-induced osteogenic differentiation of iSCAPs. Similarly, it was reported that the BMP2-induced osteoblastic differentiation of MSCs was inhibited by TNF-α through activation of MAPK signaling pathway [19, 44, 45], while other studies suggest that TNF-α inhibited BMP/Smads signaling through the activation of NF-κB signaling pathway[46‒48]. Nonetheless, the molecular mechanisms underlying such inhibition remain to be thoroughly investigated.

Interestingly, in present study we found that an increase in the titers of AdBMP9 in presence of a constant high concentration of TNF-α (10ng/ml) can partially overcome the inhibitory effect exerted by TNF-α in osteoblastic differentiation of iSCAPs. Similarly, an earlier study reported that TNF-α and BMP2 had opposing effects on osteoblastic differentiation of C2C12 and MC3T3-E1 cells [19]. While the exact mechanisms need to be understood, it offers a promise that exogenous BMP9 may be used to promote the bone defect repair and bone regeneration in the presence of high doses of TNF-α or under chronic inflammation conditions.

In summary, our results strongly suggest that TNF-α may play an important role in periapical bone defect during the inflammatory phase. High doses of TNF-α inhibit the BMP9-induced osteoblastic differentiation of iSCAPs, and the inhibitory effect of TNF-α can be partially reversed by an overexpression of BMP9. Therefore, our results strongly suggest that BMP9 may be further explored as a potent osteogenic factor to improve the efficacy of osteoblastic and odontogenic regeneration in tooth engineering under chronic inflammation conditions.

The reported work was supported in part by research grants from Chongqing Municipal Commission on Science & Technology (#cstc2014jcyjA10010 to HZ; and #cstc2015shmszx00012 to HZ), Chongqing Yubei District Commission on Science & Technology (#2014-#15 HZ), and the National Natural Science Foundation of China (#81400493 to FZ) , Program for Innovation Team Building at Institutions of Higher Education in Chongqing in 2016 (#CXTDG201602006). Funding sources were not involved in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.

The authors declare no competing financial interests.