TNF-α, IL-1B and IL-6 affect the differentiation ability of dental pulp stem cells

DPSCs were isolated from the impacted third molar teeth of three 22–30-year-old patients in the Marmara University Faculty of Dentistry Oral and Maxillofacial Surgery Department, Istanbul, Turkey. All patients provided informed consent, and the Ethics Committee of the Marmara University Clinical Researches in Istanbul, Turkey approved the study protocol (22.05.15–1). The extraction procedure was performed atraumatically and under sterile conditions. Extracted teeth were transported in Dulbecco’s phosphate-buffered saline (DPBS, Gibco, Grand Island, NY, USA) with 1% penicillin/streptomycin (Gibco, USA) within ice cubes within 4 h to the laboratory in the Department of Pediatric Allergy-Immunology, Marmara University Research Hospital, where all laboratory work was performed. The pulp was separated from the tooth by cracking the crown under sterile conditions. First, the pulp was broken down to 0.1–0.5-mm pieces mechanically with a sterile scalpel and then enzymatically treated with 2 mL of collagenase type I solution (3 mg/mL) (Gibco, USA); then, incubation for 45 min at 37 °C was carried out to digest the pulp tissue enzymatically. The enzymatic activation stopped with 2 mL of 1% penicillin/streptomycin and 10% fetal bovine serum (FBS) containing Dulbecco’s modified Eagle medium (DMEM, Gibco, USA), followed by centrifugation at 1500 rpm for 5 min. The supernatant was aspirated, and cell pellets were obtained and then suspended with 5 mL of DMEM and cultivated in T-75 flasks with a 5% CO₂ atmosphere under 37 °C for 7 days. The culture medium was changed every 2–3 days until the cells became confluent at 80%. Thereafter, adherent cells were cultured until the third passage to characterize and analyse specific surface markers. The third passage cells are used in the culture studies. These specific cellular determinations and analyses were performed using flow cytometry.

The cells from the third passage were used to analyse cell surface antigen expressions. Approximately 1 × 10⁶ cells were counted and homogenised in PBS and incubated with antibodies at room temperature in the dark for 15 min. After incubation, 0.1% sodium azide containing PBS was added and procedure followed by centrifugation at 1200 rpm for 5 min and the cell suspension analysed with FACSCalibur Flow Cytometry device with BD Cell Quest TM software (BD Biosciences, San Jose, CA, USA). CD29, CD105, CD146, CD73, and CD90 were determined as positive antibodies, whereas CD3, CD4, CD20, CD34, CD45, and HLA-DR were determined as negative antibodies.

After determining DPSCs with flow cytometry analysis for specific surface markers, DPSCs from three impacted third molar teeth were randomly divided into 5 main groups with 3 subdivisions for each group making a total of 15 groups. One group of unstimulated control and three groups treated with TNF-α (100 ng/mL) (R&D Systems, UK), IL-1β (100 ng/mL) (R&D Systems UK), or IL-6 (100 ng/mL) (R&D Systems UK) for 48 h [45].

After culturing DPSCs with inflammatory cytokines, each group was divided into three subgroups with approximately 100,000 cells to induce osteogenic, adipogenic, and chondrogenic differentiation, leading to 12 separate differentiation groups in total. The cells were incubated in a 5% CO₂ atmosphere under 37 °C for 7 days until they became 80%–90% confluent. For osteogenic (MesenCult, Stemcell Technologies, North America), adipogenic, and chondrogenic (Gibco, Grand Island, USA) differentiation, human MSC functional identification kits were used. For the differentiation procedure, the cells were plated in 6-well plates, and the differentiation medium was prepared following the manufacturer’s instructions. The differentiation medium changed every 3 days, and at the end of 21–28 days, the formed tissues were determined using different staining procedures.

After 28 days, osteogenic, chondrogenic, and adipogenic differentiations were determined by staining with Alizarin red, Alcian blue, and oil red, respectively. Staining solutions were prepared according to the manufacturer’s instructions, and after all groups were examined by the biological light microscope under 20 × objective. Histological analysis of osteogenic, chondrogenic, and adipogenic differentiated pellets was assessed following staining protocols using a modified version of the Bern Score proposed by Grogan et al. [46]. In brief, cell pellets were assessed using the following criteria: uniformity and intensity of staining and distance between cells/amount of matrix produced and cell morphology. Each of these three categories was scored from 0 to 3 (Table 1). The evaluation was performed by calculating the arithmetical means of the scores within three criteria, as given in Table 1.

DPSCs attached to the bottom of the culture flasks and showed a fibroblast-like morphology at the early days of incubation (Fig. 1(a)). DPSCs began to proliferate in 4–5 days and slowly formed colonies (Fig. 1(b)). Ten days after being plated in their very first cell passage, the DPSCs showed 70% confluency. Almost all the DPSCs showed a fibroblast-like spindle-shaped morphology in their later passages (Fig. 1(c), (d). Flow cytometry analysis revealed that DPSCs were positive for CD29, CD146, CD105, and CD73 and negative for CD3, CD4, and CD20 (Fig. 2).

The unstimulated DPSC control group exhibited even, dark staining with osteogenic colonies and the highest amount of Ca⁺⁺ deposits (Table 2, Fig. 3(a)). The TNF-α-stimulated group exhibited weak staining of the poorly formed matrix, no osteogenic colonies, and calcium deposits with spindle/fibrous cell morphology (P < 0.0001) (Table 2, Fig. 3(b)). The IL-1β-stimulated group exhibited moderately even staining, no osteogenic colonies with weak staining of calcium deposits, and mixed spindle/fibrous cell morphology with calcium deposits. (P < 0.05) (Table 2, Fig. 3(c)). The IL-6-stimulated group exhibited weak staining of the poorly formed matrix with osteogenic colonies, weak staining of calcium deposits, and spindle/fibrous cell morphology (P < 0.01) (Table 2, Fig. 3(d)). The DPSC group stimulated with all three cytokines showed weak staining of the poorly formed matrix with no osteogenic colonies, weak staining of calcium deposits, and spindle/fibrous cell morphology (P < 0.001) (Table 2, Fig. 3(e)). Stimulation with inflammatory cytokines significantly inhibited the osteogenic differentiation of DPSCs (P < 0.05) (Fig. 6(a)).

The unstimulated DPSC control group exhibited even, dark staining with high cell density, moderate distance between cells, and majority cartilage forming/chondrogenic cell morphology (Table 2, Fig. 4(a)). The TNF-α-stimulated group exhibited moderately even staining, low cell density with little intercellular matrix, and spindle/fibrous cell morphology (P < 0.001) (Table 2 Fig. 4(b)). The IL-1β-stimulated groups showed weak staining of the poorly formed matrix, low cell density with no intercellular matrix, and condensed/necrotic bodies (P < 0.0001) (Table 2, Fig. 4(c)). The IL-6-stimulated group exhibited weak staining of the poorly formed matrix, low cell density with little intercellular matrix, and condensed/necrotic cell bodies (P < 0.0001) (Table 2 Fig. 4(d)). The DPSC group stimulated with all three cytokines showed weak staining of the poorly formed matrix, low cell density with no intercellular matrix, and condensed/necrotic cell bodies (P < 0.0001) (Table 2, Fig. 4(e)). Stimulation with inflammatory cytokines significantly inhibited the chondrogenic differentiation of DPSCs (P < 0.05) (Fig. 6(b)).

The unstimulated DPSC control group exhibited moderately even staining, adipocytes with a high amount of oil droplets, and adipogenic cell morphology as most oil droplets (Table 2, Fig. 5(a)). The TNF-α-stimulated group exhibited weak staining of the poorly formed matrix, adipocytes with weak staining of oil droplets, and spindle/fibrous cell morphology (P < 0.001) (Table 2, Fig. 5(b)). The IL-1β-stimulated group exhibited moderately even staining, adipocytes with weak staining of oil droplets, and spindle/fibrous cell morphology (P < 0.01) (Table 2, Fig. 5(c)). The IL-6-stimulated group exhibited moderately even staining, adipocytes with weak staining of oil droplets, and mixed spindle/fibrous cell morphology with oil droplets (P < 0.01) (Table 2, Fig. 5(d)). The DPSC group stimulated with all three cytokines showed weak staining of the poorly formed matrix, no adipocytes, no oil droplets, and spindle/fibrous cell morphology (P < 0.0001) (Table 2, Fig. 5(e)). Stimulation with inflammatory cytokines significantly inhibited the adipogenic differentiation of DPSCs (P < 0.05) (Fig. 6(c)).

MSCs are a promising source of stem cells and have been isolated from various tissues with different techniques [2, 32, 47]. After Friedenstein et al. [48] described MSCs in 1968, MSCs were defined in further studies as expressing CD73, CD90, and CD105 and not expressing CD11b, CD14, CD34, and CD45. Additionally, MSCs must be adherent to plastic surfaces and able to differentiate into certain cells in vitro as osteoblasts, chondroblasts, and adipocytes [49]. In our study, we used DPSCs, which we thought had the advantage of relatively easy access. The stem cell sources were the impacted third molar teeth, which were extracted during dental health care procedures, resulting in no ethical debate. Furthermore, a large proliferation capacity, keeping up their cellular phenotype for an extended period, and promising cell lines for regenerative applications were also considered among the great advantages of DPSCs [50, 51]. The two most commonly used techniques for isolating DPSCs from dental pulp tissue are the explant method and enzymatic digestion. Hilkens et al. [52] reported no difference in the tissue differentiation potential of DPSCs regarding the isolation method. In our study, the enzymatic digestion method used to isolate DPSCs was based on previous studies [9, 10]. After isolation, the cells displayed fibroblast-shaped morphology and were adherent to plastic Petri plates. Flow cytometry analysis declared that the cells showed positive expression for CD29, CD105, CD146, and CD73 markers and negative for CD3, CD4, and CD20 markers, which agree with the criteria of the International Society of Cellular Therapy [48]. Osteogenic, chondrogenic, and adipogenic differentiation procedures were performed on characterised DPSCs, and Alizarin red, Alcian blue, and oil red stains were used, respectively, to determine their differentiation into cell lines in accordance with previous studies [2, 53, 54]. Tarte et al. [55] also used staining procedures to compare the proliferation and differentiation of SHEDs and PDLSCs, but used von Kossa staining instead of Alcian blue to determine chondrogenic differentiation. MSCs can differentiate in vitro spontaneously or by the induction of biologically active molecules [56]. DPSCs can also proliferate and differentiate, as can other stem cells of dental origin [2, 9, 10, 51].

Cytokines modulate immune and inflammatory responses and are markers of inflammation [21, 57]. In many situations, certain tissues need to be regenerated due to injury. Whether the tissue injury is caused by microorganisms (e.g. pulpitis) or trauma (e.g. bone fractures), proinflammatory cytokines are the superior markers of inflammatory responses [32, 58]. Both positive and negative impacts of cytokines on MSC differentiation and tissue healing have been reported [32, 47]. The present study determined how DPSCs might behave in an inflammatory environment set up with some key proinflammatory cytokines TNF-α, IL-1β, and IL-6. Many previous in vitro and in vivo studies have evaluated their roles in osteogenic and chondrogenic differentiation of MSCs. Kondo et al. [59]. indicated that in the early stages, TNF-α, IL-1β, and IL-6 contribute to fracture healing and bone remodelling. Xie et al. also [54] demonstrated that IL-6 promotes osteogenic differentiation in BM-MSCs in vitro. In vitro studies demonstrating the effects of TNF-α, IL-1β and IL-6 have mostly involved osteogenesis with MSCs other than dental origin and have not directly compared their effects on differentiation [32, 47]. Contrary to our results, Liu et al. [51] demonstrated that TNF-α promoted the osteogenic differentiation of DPSCs in vitro. Similarly, Feng et al. [25] demonstrated that TNF-α activates the NF-κB pathway and promotes osteogenic differentiation of DPSCs in vitro. Another in vitro study showed increased calcium deposits following IL-1β pretreatment when culturing BM-MSCs in osteogenic medium.

On contrary, Kondo et al. [59] also reported bone resorption can be induced under IL-6 stimulation. Lacey et al. [32] compared the effects of TNF-α and IL-1β on the osteogenic capacity of murine MSCs and found that these cytokines inhibited MSC differentiation to osteoblasts, which agrees with our findings. Liu et al. [51] investigated osteogenic differentiation of DPSCs promoted by TNF-α; this was similar to our study with the difference of evaluating transcriptome changes. Additionally, relatively long-term exposure to inflammatory mediators were reported to suppresses DPSC differentiation ability [60].

Considering the chondrogenic differentiation of BM-MSCs, Mumme et al. demonstrated the most intense staining for cartilage with low-dose IL-1β (10 and 50 pg/mL) [41]. The discrepancies in the outcomes among these studies and our study can be explained by the differences in the concentrations of proinflammatory cytokines and in the origin of the stem cells.

A limitation of the present study was that it was an in vitro analysis and not in vivo. The cytokine concentrations used in the study were used as the highest possible concentrations to assess the differentiation potential of DPSCs into the inflammatory niche. These concentrations may not be similar in vivo. In addition, the differentiation potential of DPSCs may vary in the presence of anti-inflammatory drugs such as anti-TNF-α, anti-IL6, which are used in some autoimmune or inflammatory diseases. Nonetheless, our findings might be useful for further studies for understanding the mechanisms and outcomes of DPSC differentiation with specific cytokine modulation both in vitro and in vivo because the functions and expressions of proinflammatory cytokines during certain tissue differentiations remain unclear in vivo [54]. In addition, stem cells of dental origin are expected to be preferred more frequently in future research because they are easy to obtain. Future studies should be designed to include different concentrations of inflammatory cytokines, evaluation of gene expression, and use of dental stem cells with different origins.