IL‐33 guides osteogenesis and increases proliferation and pluripotency marker expression in dental stem cells

Abstract Objectives Soluble IL‐33 (interleukin (IL)‐1‐like cytokine) acts as endogenous alarm signal (alarmin). Since alarmins, besides activating immune system, act to restore tissue homeostasis, we investigated whether IL‐33 exerts beneficial effects on oral stem cell pull. Materials and Methods Clonogenicity, proliferation, differentiation and senescence of stem cells derived from human periodontal ligament (PDLSCs) and dental pulp (DPSCs) were determined after in vitro exposure to IL‐33. Cellular changes were detected by flow cytometry, Western blot, immunocytochemistry and semiquantitative RT‐PCR. Results IL‐33 stimulated proliferation, clonogenicity and expression of pluripotency markers, OCT‐4, SOX‐2 and NANOG, but it inhibited ALP activity and mineralization in both PDLSCs and DPSCs. Higher Ki67 expression and reduced β‐galactosidase activity in IL‐33‐treated cells were demonstrated, whereas these trends were more conspicuous in osteogenic medium. However, after 7‐day IL‐33 pretreatment, differentiation capacity of IL‐33‐pretreated cells was retained, and increased ALP activity was observed in both cell types. Results showed that IL‐33 regulates NF‐κB and β‐catenin signalling, indicating the association of these molecules with changes observed in IL‐33‐treated PDLSCs and DPSCs, particularly their proliferation, pluripotency‐associated marker expression and osteogenesis. Conclusions IL‐33 treatment impairs osteogenesis of PDLSCs and DPSCs, while increases their clonogenicity, proliferation and pluripotency marker expression. After exposure to IL‐33, osteogenic capacity of cells stayed intact. NF‐κB and β‐catenin are implicated in the effects achieved by IL‐33 in PDLSCs and DPSCs.

crevicular fluid 7 and gingival epithelium of chronic periodontitis patients. 8 Association of increased IL-33 level with periodontitis and alveolar bone resorption and loss was also reported. 8,9 Recent findings indicated that IL-33 expression in cells of periapical lesion and radicular cyst may be involved in periapical inflammation 10,11 which is caused by the pulpal infection. 12 Since the root apex and dental pulp are tightly interconnected tissues, communicating through periodontal pocket and apical foramen, 12,13 during periodontal disease IL-33 within crevicular fluid may influence dental pulp cells. However, despite the findings that indicate correlation of IL-33 expression and periodontal inflammatory diseases, the involvement IL-33 in regeneration and repair of oral tissues is not fully understood, particularly since there are still no data regarding the influence of IL-33 on oral stem cells nor mode of its action.
Oral and maxillofacial tissues present highly accessible sources of adult progenitor/stem cells which possess features assigned to in vitro-observed mesenchymal stem/stromal cell (MSC) properties, such as self-renewal and multilineage differentiation. 14,15 Regarding the heterogeneity within (craniofacial) oral stem cells populations, functional differences in vivo are reported. Dental MSCs, including exfoliated deciduous teeth stem cells (SCs), apical papilla SCs and dental pulp SCs (DPSCs), form dentine-like structures when transplanted in immunocompromised mice, in contrast to periodontal ligament (PDL) SCs and gingival SCs that in vivo form PDL-like structures. 16 Dental pulp (DP) forms dentin, whereas PDL is tooth-supportive connective tissue that ensures gently tooth anchorage to the alveolar bone, both providing tooth nutrition, protection and sensory perception, together contributing to the tooth longevity. [17][18][19][20] Since PDL and DP are soft, connective tissues surrounded by hard, mineralized tissues, regulation of mineralization level is the main physiological demand within these tissues in order to adapt functional changes. 18,21,22 While DPSCs contribute to replacement of damaged tissue and repair of complete tooth, PDLSCs have predominant role in tooth functions and development. 23 Resident DPSCs and PDLSCs respond to activation stimuli of dynamic microenvironment, governing tissue homeostasis, differentiation and regeneration. 18,21,22 Detailed understanding of functional behaviour of oral MSCs, both in vitro and in vivo, is still necessary regarding their potential use in cellular therapy and maxillofacial reconstruction.
As previous reports indicated different protein and gene expression patterns in human DPSCs and PDLSCs, 24,25 in this study, we evaluated the response of PDLSCs and DPSCs to IL-33, through the analysis of their proliferation and differentiation potential. Since regulatory proteins NF-κB and β-catenin are implicated in tissue immune homeostasis, osteogenesis and stemness maintaining, 26 we also analysed the role of NF-κB and β-catenin in IL-33-mediated effects in PDLSCs and DPSCs. matrix mineralization by Alizarin red staining were estimated as previously described. 29 Light microscope (Olympus, Japan) was used to capture cells, while the mineralization level was quantified by densitometry in NIH-ImageJ software (LOCI, University of Wisconsin, Madison, WI, USA).

| Cellular clonogenicity, proliferation and senescence in osteogenic differentiation settings
Colony-forming unit-fibroblast (CFU-F) assay was performed as we previously described for PDLSCs and DPSCs. 27,28 CFU efficiency was determined as the percentage of colonies relative to total number of seeded cells in each well.
To determine ALP + CFU-Fs, termed as CFU-osteoblasts (CFU-O), cells were seeded in 24-well plates at concentration 100 cells/well and cultivated in GM in standard conditions during 7 days. 30 Then, the OM was added and incubation was continued for the next 7 days with or without IL-33 (100 ng/mL) when the colonies were stained for ALP activity. Percentage of CFU-O was expressed as ratio of ALP-positive CFUs/total CFUs*100.
Following the IL-33 (100 ng/mL) during 7-day osteogenic induction cell number, expression of intracellular proliferation marker Ki67 and β-galactosidase activity were estimated. Cell number was determined by Trypan blue exclusion test.
Percentage of Ki67-positive cells was determined by flow cytometry equipment (as for surface marker detection). Cells were washed with PBS, fixed in 5% formaldehyde and permeabilized in 0.5% BSA/ PBS containing 0.1% Triton X-100. Following the nonspecific blocking (15 minutes in 0.5% BSA/PBS), cells were labelled with rabbit anti-Ki67 antibody (Abcam, UK) and secondary anti-rabbit antibody FITC (Sigma-Aldrich). Level of nonspecific binding was determined by FITC-conjugated isotype control antibodies.
For single-cell β-galactosidase staining, cells were seeded at concentration 2 × 10 3 cells/well in 96-well plates, adhered during 6 hours, and stained using Senescence Cells Histochemical Kit (Sigma-Aldrich). The single-stained cells were captured and counted under light microscope, and the percentage of stained cells was determined for several separated visual fields.

| Western blot
After cultivation with IL-33 (100 ng/mL) at different time points, total protein extracts were isolated using lysis buffer. Same amounts of protein samples (concentration determined by BCA assay, Serva, Germany) were separated by SDS-PAGE and electrotransfered onto nitrocellulose membrane Hybond ECL (AppliChem, Germany).
Statistical significances ( * P < 0.05) were determined by Student's t test. Data were analysed and graphed using GraphPad Prism 6 Software (San Diego, CA).

| Mesenchymal stem/stromal cell features of PDLSCs and DPSCs: Effects of IL-33
By using MTT test, we examined viability of PDLSCs and DPSCs in  Figure 1D).   F I G U R E 2 Cellular clonogenicity and ALP activity of cultivated PDLSCs and DPSCs in the presence of IL-33 in osteogenic differentiation settings. Cells were seeded in 24-well plates (100 cells/well) and grown in standard conditions during 14 d with or without IL-33 (100 ng/ mL) in GM or OM. A, Colony-forming unit-fibroblast (CFU-F) efficiency of PDLSCs and DPSCs. Representative images of CFU-F stained with Crystal violet are shown (Scale bars: 500 µm). For CFU-osteoblasts (CFU-O), capacity cells were seeded in 24-well plate (100 cells/well) and cultivated in GM in standard conditions during 7 d when the osteogenic medium (OM) was added and incubation was continued for the next 7 d with or without IL-33 (100 ng/mL). B, CFU-osteoblast (CFU-O) capacity of PDLSCs and DPSCs determined based on number of ALPpositive colonies. Representative images of CFU-O stained for ALP activity are shown (Scale bars: 500 µm). Results in graphs are presented as means of percentages ± SEM from at least three independent experiments. Statistically significant difference in comparison with GM in the absence of IL-33 by t test: *P < 0.05; **P < 0.01; or in comparison with OM in the absence of IL33: *P < 0.05 or between PDLSCs and DPSCs: ## P < 0.01; ### P < 0.001. (C,D), For osteogenic differentiation, detected by using ALP staining, cells were cultivated in GM or OM in the presence of IL-33 (100 ng/mL) during 7 d; or (F,G), pretreated with IL-33 (100 ng/mL) during 7 d and then induced for osteogenesis. (C,F), Representative images of osteogenic differentiation are shown (Scale bars: 20 µm). (F,G), Quantification of ALP staining. Results in graphs are presented as means ± SEM of four different samples (n = 4) from at least three independent experiments. Statistically significant difference in comparison with GM in the absence of IL-33 by t test: **P < 0.01; ***P < 0.01; or in comparison with OM in the absence of IL33: **P < 0.01; ***P < 0.01; or between PDLSCs and DPSCs: # P < 0.05; ## P < 0.01. E, For mRNA analysis, cells were cultivated in GM in the presence or absence of IL-33 (100 ng/mL) 24 h. As a gel loading control, GAPDH was used. Representative gels from three different samples (n = 3) are shown. Molecular weight standards are indicated in bp for PCR products. Results in graphs are presented as mean ± SEM from at least three independent experiments. Statistically significant differences between PDLSCs and DPSCs by t test: *P < 0.05; **P < 0.01

| Roles of NF-κB and β-catenin in IL-33mediated PDLSCs and DPSCs proliferation
To evaluate involvement of NF-κB and β-catenin in IL-33-elevated On the other side, PNU itself supported proliferation of DPSCs, but not altering IL-33-stimulated proliferation of DPSCs ( Figure 6B).

| Involvement of NF-κB and β-catenin in IL-33stimulated pluripotency-associated marker expression
Results obtained for PDLSCs showed that PDTC inhibitor abolished Except Oct4B mRNA which was decreased by PNU, inhibitors did not alter IL-33-increased pluripotency marker gene expression.

PDTC and PNU inhibitors restored IL-33-reduced ALP activity and
Ca deposition levels in both cell types, without affecting the basal osteogenic differentiation ( Figure 8A, B, D, E). Surprisingly, RT-PCR results did not reveal any significant effect of the presence of PDTC and PNU on Runx2 and ALP gene expression which stayed unaffected after IL-33 treatment in both cell type ( Figure 8C, F).

| D ISCUSS I ON
The goal of this study was to evaluate the in vitro effects of IL-33 on main functional properties of PDLSCs and DPSCs, by addressing their basal stemness-related features, such as colony-forming capacity, proliferation, expression of pluripotency-associated markers and differentiation, especially referring to osteogenesis.
The PDLSCs and DPSCs used in our study, as resident population of progenitor cells of periodontal ligament and dental pulp, share characteristics with the bone marrow mesenchymal stem/stromal cells (BM-MSC), such as their colony-forming capacity,in vitro trilineage differentiation and surface markers expression. 31 Although PDLSCs and DPSCs share close anatomical location, differences between these cells were observed in mineral composition generation, 32 stability after ectopic transplantation, 33 hard tissue formation and alkaline phosphatase activity, 34 proteome 24 and gene expression. 35 As PDLSCs and DPSCs respond to specific stimuli triggered by inflammation or injury, which affect cell recruitment and proliferation in dental regions, 36  significantly contributes to bone tissue remodelling phase. 46 Our study revealed predominantly Ki67 cytoplasmic staining in dental stem cells and similar was found in ASCs. 45 Since there are few evidences regarding Ki67 expression in dental stem cells, it would be reasonable to find out precise biological role of nuclear and cytoplasmic Ki67.
Activation of NF-κB by IL-33 has been shown in different cell types, such as fibroblast-like synoviocytes, 47 endothelial cells 48 and human HEK293RI cells. 49 Here, for the first time, we showed that IL-33 stimulated activation of NF-κB in PDLSCs and DPSCs. The observed IL-33 increased proliferation of DPSCs, achieved via NF-κB activation, when cultivated in the presence of osteogenic stimuli, is in agreement with previous report demonstrating that inflammatory cytokine IFN-γ-promoted DPSC proliferation via NF-κB. 50 Observed involvement of NF-κB activation in the inhibition of ALP activity and matrix mineralization achieved by IL-33 treatment were also in accordance with previously reported data for PDLSCs 51 and DPSCs. 50 The β-catenin, as part of canonical Wnt signalling, controls neural stem cell proliferation 52 and has important role in tooth biology. 53 Here, IL-33-elevated proliferation of PDLSCs was achieved by β-cat- on the other imply that the role of β-catenin signalling in osteogenesis is not fully comprehended. However, inhibition of osteogenic differentiation is mediated through NF-κB activation, which further on promoted β-catenin ubiquitination and degradation in human and mouse MSCs. 56 Therefore, it can be assumed that cooperation of NF-κB and β-catenin may exist and be involved in regulation of dental stem cell functions, too.
We found that IL-33 decreased β-galactosidase activity in PDLSCs and DPSCs simultaneously with stimulation of their proliferation capacity. The inhibition of β-galactosidase activity was additionally strengthened in OM, which can independently reduce β-galactosidase activity in both cell types. Important stimulator of osteogenesis in OM, ascorbic acid-2-phosphate increases proliferative capacity by attenuating senescence and expression of reactive oxygen species in human osteoarthritic osteoblast. 57 Whether the antioxidative effect is also mechanism of the action of IL-33 needs to be further investigated. Promotion of cell divisions and increased ALP activity in human and mouse BM-MSCs could be the mechanism through which reduction in senescence can be achieved. 58 However, our data indicated that the inhibitory effect of IL-33 treatment on osteogenesis was not achieved through antiproliferative or pro-ageing mechanism.
Important outcome from our study is the observation that IL-33 Human PDLSCs and DPSCs retain stemness genes and proliferation for prolonged time during in vitro cultivation, 38 and expression of stemness markers was often associated with their origin from neural crest (stem) cells which migrate and establish stem cell pools through the anatomical sites such as dental region or even bone marrow. 61,62 Our previous study demonstrated that DPSCs and PDLSCs express pluripotency markers in higher extent than other adult stem cells, such as adipose tissue progenitors. 63 The elevated pluripotency marker expression accompanied with IL-33-stimulated proliferation and clonogenicity of PDLSCs and DPSCs is in accordance with previous studies demonstrating overexpression of OCT-4A, 64 SOX-2 65 and NANOG 66 associated with increased proliferation of DPSCs or clonogenic capacity of human BM-MSCs. 67 Liu et al also reported that OCT-4A overexpression in DPSCs leads to upregulation of other pluripotency genes, such as Oct4B1, SOX-2, NANOG, Klf4 and c-Myc, thus indicating possible close mutual cooperation of these transcriptional factors. 64 Contrary to OCT-4A, the role of OCT4B in pluripotency is not clear. However, it may be assumed that OCT4B actually governs self-renewal and maintenance of tissue homeostasis within adult tissues. 68 Previous studies indicated that OCT4B might be involved in cellular stress response 69 and antiapoptotic activity in cancer cells. 70,71 Importantly, OCT4B may indirectly be increased in dental pulp tissues and dental pulp cells exposed to proinflammatory factors, where it has been suggested that OCT4B represents a antiapoptotic mediator and subsequently may be the important factor in dental regeneration. 72 Our results indicate that IL-33 increases expression of OCT4B mRNA expression in both PDLSCs and DPSCs. Whether IL-33-stimulated OCT4B has some regulatory role in dental stem cell phenotype and behaviour needs to be additionally investigated. Previously, the role of NF-κB signalling in maintenance of pluripotency of human iPSC was shown, as the augmented NF-κB activity correlated with increased expression of OCT-4A and NANOG, sustaining human iPSC in undifferentiated state. 73 Our analysis also showed that the activation of NF-κB is involved in IL-33-upregulated Oct4 and NANOG expression in PDLSCs and Oct4, SOX-2 and NANOG in DPSCs. Additionally, IL-33- In summary, our study brings new evidence about the effects of IL-33 on dental stem cell functions. It can be concluded that IL-33 treatment impairs PDLSC and DPSC osteogenesis, favouring their proliferation and clonogenicity, but not diminishing their differentiation potential. These findings suggest that IL-33 might delay dental stem cell pool exhaustion caused by activation stimuli, triggered by inflammation or stress, which initiate their functional response.