Follicle‐stimulating hormone impairs dental pulp stem cells odontogenic differentiation

Abstract In addition to bone, the dentin‐pulp complex is also influenced by menopause, showing a decreased regenerative capacity. High levels of follicle‐stimulating hormone (FSH) during menopause could directly regulate bone metabolism. Here, the role of FSH in the odontogenic differentiation of the dentin‐pulp complex was investigated. Dental pulp stem cells (DPSCs) were isolated. CCK‐8 assays, cell apoptosis assays, Western blotting (WB), real‐time RT‐PCR, alkaline phosphatase activity assays, and Alizarin Red S staining were used to clarify the effects of FSH on the proliferation, apoptosis and odontogenic differentiation of the DPSCs. MAPK pathway‐related factors were explored by WB assays. FSH and its inhibitor were used in OVX rats combined with a direct pulp‐capping model. HE and immunohistochemistry were used to detect reparative dentin formation and related features. The results indicated that FSH significantly decreased the odontogenic differentiation of the DPSCs without affecting cell proliferation and apoptosis. Moreover, FSH significantly activated the JNK signalling pathway, and JNK inhibitor partly rescued the inhibitory effect of FSH on DPSC differentiation. In vivo, FSH treatment attenuated the dentin bridge formation and mineralization‐related protein expression in the OVX rats. Our findings indicated that FSH reduced the odontogenic capacity of the DPSCs and was involved in reparative dentinogenesis during menopause.

potential tool for tissue engineering. 5 As a kind of somatic stem cells in dental pulp tissue, DPSCs can be efficiently obtained from medical waste (such as orthodontic teeth to be removed and impacted wisdom teeth) and have high proliferative, self-renewal and multi-directional differentiation ability. It is well known that DPSCs can differentiate into odontoblast-like cells after mineralization induction in vitro. 6,7 The characteristics of DPSCs that can form dentin-pulp complexes provide theoretical basis for the physiology, pathology and mechanism of dentin formation. 8 It also reveals potential and advantages of DPSCs in the clinical application prospects of preservation and regeneration of pulp. Researchers found that oestrogen deficiency can reduce the odonto/osteogenic differentiation of DPSCs and impair the odontogenic capacity of the dentin-pulp complex. [9][10][11] The oestrogen level is controlled by the hypothalamus-pituitary-ovary axis and is mainly regulated by follicle-stimulating hormone (FSH). FSH belongs to the hypothalamic hormone family and is a glycoprotein hormone secreted by the anterior pituitary.
Recently, the traditional view that FSH works solely as a gonadal hormone has changed. 12 More and more evidences showed that ovarian failure in late menopause is related to a sharp increase of serum FSH, which is consistent with the most rapid bone loss and the occurrence of visceral adiposity. [13][14][15] A series of studies have shown that FSH directly regulates bone mass in mice via a non-classical pathway by regulating osteoclast activity. 16 FSHR, a specific receptor of FSH, was expressed in mesenchymal stem

cells but not in osteoblasts and their precursors. A previous study
showed that blocking FSH β not only inhibited osteoclast activity and reduced the inflammatory response but also increased bone formation and osteoblast number. 17 This study suggested that FSH may play a role in inhibiting bone formation by inhibiting the differentiation of mesenchymal stem cells into osteoblasts. DPSCs/ odontoblasts are similar to BMMSCs/osteoblasts in many aspects.
Therefore, FSH may negatively regulate the odontogenic differentiation of DPSCs, impairing the dentinogenic capacity of the dentin-pulp complex. The purpose of our research is to examine the effects of FSH and uncover the possible mechanism of the mineralization ability of DPSCs.

| DPSC isolation, culture and passaging
With the informed consent, healthy human third molars were collected from patients (18)(19)(20)(21)(22)(23)(24)(25) years old) at the Oral and Maxillofacial Surgery Clinics of Shandong Provincial Stomatological Hospital for orthodontic needs. Ethical approval was obtained from the Ethics Committee of the School of Stomatology, Shandong University, China. As previously described, the teeth were rinsed by phosphate buffered saline (PBS) with 2% penicillin and streptomycin, and the pulp was completely removed in a sterile clean biological safety cabin. The pulp was then cut into small pieces and digested in enzyme mixture (4 mg/mL trypsin (Gibco, Life Technologies) +3 mg/mL collagenase type I (Gibco, Life Technologies, Grand Island, NY, USA) for 50 minutes at 37°C. Equal volumes of neutralization medium (10% FBS in α-MEM with 1% penicillin/streptomycin) were added to the digestive solution and mixed well to terminate the digestion. The dissociated-cell solution was then centrifuged at 1000 rpm for 5 minutes to collect the cell precipitates. Subsequently, the cell precipitates were cultured in α-MEM (Hyclone) containing 20% foetal bovine serum (FBS; BI technology), 100 U/mL penicillin and 100 μg/mL streptomycin in a 37°C incubator with 5% CO 2 . After reaching 80%-85% confluence, The DPSCs were subcultured. The cells at 3rd-5th passages were used in next studies.

| Cell counting kit-8 assay (CCK-8)
DPSCs (2 × 10 3 cells/well) were plated on 96-well plate. To detect the effect of FSH on the proliferation of DPSCs, cells were incubated with CCK-8 reagents (Dojindo, Kyushu Island, Japan) at 37°C for 2 hours at the indicated time points for CCK-8 analysis. Optical density (OD) at 450 nm was detected by a microplate reader (Bio-Tek, Winooski, VT).

| Alkaline phosphatase (ALP) activity
DPSCs were cultured in odontogenic medium with FSH and SP600125 for 7 days, and the medium was changed every other day. Then, they were washed 3 times with PBS and incubated with 50 μL of 0.2% Triton X-100 at 4°C overnight. After confirming that the cells were completely lysed under microscope, ALP activity were measured by corresponding kit (Jiancheng, Nanjing, China) referred to the manufacturer's instructions. Optical density (OD) was determined at 450 nm and normalized to total protein by a BCA kit (Beyotime, Shanghai, China).

| Mineralized nodule formation and detection of calcium content
The mineralization ability of the DPSCs treated with FSH at concen-  (Table S1). The quantitative PCR conditions were set as the manufacturer's instructions.
The mRNA expression levels were calculated by the 2 −ΔΔCt method.
All PCRs were carried out in triplicate.

| Ovariectomy for the rats and drug application
The rats were anesthetized via the intraperitoneal injection performed with 10% chloral hydrate (Qilu hospital, Jinan, China).
Bilateral ovariectomies were performed in the rats of the OVX group, while rats in the sham group were subjected to sham surgeries. After surgery, the rats of the corresponding groups were injected subcutaneously with LE, FSH or vehicle at relevant concentrations. LE is a sustained-release dosage form that maintains effective blood concentration and prevents circulating FSH levels increase for at least 4 weeks after injection. (Filicori et al, 1996).

| Direct pulp capping
One week after the ovariectomies and sham surgeries, all rats were anaesthetized as described earlier. The direct pulp capping of rats were performed as Tran et al reported. 19 Briefly, after chemically cleaned and disinfected by NaOCl (5%) and 0.1%, Chlorhexamed Fluid (Evonik-Degussa, Germany), the cavities were prepared on the mesial surface of maxillary first molars under sterile water cooling.
When it is close to the pulp, 10# K file was used to induce perforation of pulp chamber. The perforations were directly capped with white ProRoot MTA cement (Dentsply Tulsa Dental, Tulsa, OK, USA) and filled with GC Fuji IX glass ionomer cement (GC Corporation, Tokyo, Japan) following cleaning with sterile saline and paper points.

| Detection of serum hormone levels
Blood samples were collected from all rats on the 7, 14 and 21 days after pulp capping through the groin vein. The serum was separated from blood samples by centrifuging at 1000 g for 10 minutes under 4°C and measured within 48 hour to detect the levels of E2 and FSH according to previous reports. 20

| Tissue preparation
On day 21 after direct pulp capping, the rats were killed. The samples were dissected and fixed with 4% paraformaldehyde for 2 days at 4°C and demineralized in 10% EDTA for at 4°C for approximately 4 weeks. The specimens were dehydrated, clarified, embedded in paraffin and cut at a thickness of 5 μm.

| Histologic observation and analysis
Sections were stained for and observed under a light microscope (Eclipse E400; Nikon, Tokyo, Japan). Histological characteristics, such as pulp tissue disorganization (PTD), inflammatory cell infiltration (ICI) and dentin bridge formation (DBF), were evaluated according to modified versions of the ISO 10993 and 7405 standards as previously described that allows detailed evaluation of the changes in pulp tissue after direct pulp capping. 21,22 The scores of each sample were determined in a blinded manner by 3 observers.

| Immunohistochemistry
The immunohistochemistry procedures were performed following the manufacturer's instructions (ZSGB-Bio, Beijing, China).
Sections were routinely dewaxed to water. After antigen retrieval and serum blocking, primary antibodies against FSHR (Santa Cruz

| Statistical analysis
The results of all measurements are demonstrated as the mean value ± standard deviation. GraphPad Prism 5.0 for Windows (GraphPad Software, Inc) was used for statistical analysis by oneway ANOVA and post-Tukey or Bonferroni multiple comparison tests. P < 0.05 was considered significant.

| Characterization of DPSCs
As previously described, DPSCs have a fibroblast-like morphology.

| Influence of FSH on the proliferation and apoptosis of the DPSCs
The proliferative capacity in the control group and the groups

| FSH decreased the odontogenic differentiation of the DPSCs
To investigate whether FSH could directly act on the DPSCs, we performed real-time PCR to assess the expression of FSHR mRNA in the DPSCs and to confirm the identity of the PCR product performed with DNA sequencing (data not shown). Real-time PCR showed that 3, 10 and 30 ng/mL FSH significantly induced FSHR expression, especially 30 ng/ mL, while 100 ng/mL showed no significant changes compared with the negative control ( Figure 3B(c)). Western blotting showed that protein expression levels of FSHR were significantly up-regulated by FSH application, especially 30 ng/mL ( Figure 3A). The results demonstrated that the DPSCs could directly act as targets of FSH via the FSH receptor.
The influence of FSH on odontogenic differentiation was examined. Western blot analysis revealed that the protein levels of ALP, RUNX2, DSPP and OSX were the lowest in the 30 ng/mL FSH group ( Figure 3A). Quantitative analysis of the protein bands normalized to GAPDH is depicted in Figure 3C. Quantitative real-time PCR demonstrated that the expression levels of osteo/odontogenic-related genes, such as ALP and DSPP, were significantly lower in the DPSCs cultured in mineralization-inducing medium (MM) with FSH (30 ng/mL) for 3 days than in those cultured in control medium ( Figure 3B(a,b)). In addition, fewer calcified nodules were found in the 30 ng/mL FSH + MM (mineralization-inducing medium) group than in the MM group ( Figure 3E). CPC assay further presented that the calcium concentration in the 30 ng/mL FSH + MM group was lower than that in the MM group (P < 0.05; Figure 3D).

| FSH activated the JNK-MAPK pathway in DPSCs
To explore the role of the MAPK pathways in the FSH-mediated reduction of odontogenic differentiation of DPSCs, total proteins were extracted in the FSH-treated (30 ng/mL) cells at 0, 30, 60, 90 and 120 minutes. Western blotting were performed to detect the protein levels of ERK, p-ERK, P38, p-P38, JNK and p-JNK. In this study, FSH treatment did not affect the total ERK, P38 or JNK expression levels. ERK and p38 Phosphorylation did not change with time, and during FSH treatment, p-JNK was gradually increased at 60 minutes and reached the highest level at 90 minutes ( Figure 4A). The p-JNK/ JNK ratio indicated the activation of the JNK pathway after FSH treatment ( Figure 4B).

| Inhibition of the JNK-MAPK pathways upregulated odonto/osteogenic differentiation of the FSH-treated DPSCs
To further clarify the role of the JNK-MAPK signalling pathway in the odontogenic differentiation of the FSH-treated DPSCs, 10 μmol/L SP600125 (a specific inhibitor of JNK) was added to suppress the activity of JNK. Prior to FSH (30 ng/mL) treatment, the DPSCs were incubated with SP600125 for 90 minutes. Western blotting assay was performed to detect the efficiency of SP600125. After administration of SP600125, p-JNK was down-regulated compared to the 60 minutes group (Figure 5A,B). Real-time PCR presented that ALP, DSPP and RUNX2 mRNA expression levels were higher in FSH + SP600125 group compared with FSH group ( Figure 5C).
The ALP activity of the DPSCs in the FSH + SP600125 treatment group at day 7 was higher than that in the FSH treatment group ( Figure 5D). Alizarin Red S staining showed significantly increased mineralization after 2 weeks odontogenic differentiation induction in FSH + SP600125 group compared to the FSH group ( Figure 5F).
The calcium concentration by CPC assay further presented that in the FSH + SP600125 group was higher than that in the FSH group (P < 0.05; Figure 3D).

| FSH attenuated reparative dentin formation in rat molars
The E2 and FSH levels in the rat serum were recorded every week.
From 7 days after OVX operation, E2 levels in the serum decreased significantly in comparison with SHAM group(P < 0.01), with no significant changes found by use of LE and FSH(P > 0.05) ( Figure 6A). FSH levels were significantly increased in OVX group. In FSH + LE group, FSH levels significantly were significantly lower than that in OVX group (P < 0.05). After injection of FSH, its serum levels ascended significant than those without FSH injection (P < 0.05) ( Figure 6B).
The molars with filling loss and periapical lesions by X-ray were excluded from the analysis of pulpal response. The histologic analysis is presented in Figure 6C-J. On day 21, in the sham group, reparative dentin was directly observed at the injured region in the pulps capped with MTA. The adjacent pulp tissue appeared normal and free of inflammatory cells. (Figure 6C,G) Compared with those in the sham groups, the specimens in the OVX group exhibited less and poorly organized reparative dentin formation as well as stronger dental pulp inflammation ( Figure 6D,H). LE application rescued the reparative dentin formation and weakened the dental pulp inflammation in the OVX rats (Figure 6 E,I). Reparative dentin formation was the lowest in the FSH treatment group, and inflammation below the dentinal bridge was the most severe in the FSH group compared with all other groups ( Figure 6F,J). A summary of histopathological evaluation results is presented in Figure 6K-M.

| FSH weakened the expression of the mineralization-related markers
The IHC results showed that expression DSP, OCN and RUNX2 was stronger in the cell-rich zone of sham molars ( Figure 7A

| D ISCUSS I ON
Previous studies have mainly attributed the aetiology of diseases accompanying menopause solely to oestrogen deficiency. FSH is controlled by oestrogen levels through negative feedback. Traditionally, FSH was thought to be mainly targeting gonadal tissues, such as granulosa cells in the ovaries and Sertoli cells in the testis. Recently, these ideas are changing. 23 In recent years, studies have shown that FSH directly affects skeletal remodelling during menopause 16 and inhibition of FSHβ rescues bone loss post-ovariectomy. 17,24 FSH can increase the secretion of TNF-α by granulocytes and macrophages in bone marrow, while FSH can promote the production of IL-1, TNF-α and IL-6 by osteoclast precursor cells. 25

,26 Blocking
FSHβ can not only inhibit osteoclast activity and reduce the inflammatory response but also increase bone formation and osteoblast number. 17 The influence of oestrogen deficiency on DPSCs has been elucidated. Oestrogen deficiency can down-regulate the odonto/ osteogenic capability of DPSCs both in vitro and in vivo, with no effect on proliferation. 10 However, the effects of FSH on dentin and DPSCs have rarely been reported. Whether FSH could act on DPSCs remains unclear at present.
FSH exerts its biological action through its specific receptor, FSH receptor (FSHR), a type of G protein-coupled receptors. FSHR was traditionally considered to be specifically expressed in gonadal tissues. In recent decades, tissues in addition to gonads, including bone, have been found to express FSHRs. 27 In bone tissues, the F I G U R E 3 FSH decreased odonto/osteogenic differentiation of the DPSCs. A, Western blotting showed lower expression levels of ALP, Runx2, DSPP and OSX in the 30 ng/mL and 100 ng/mL FSH groups than the control group in odonto/osteogenic medium for 7 d. FSH significantly up-regulated the FSHR protein expression, especially in the 30 ng/mL group. GAPDH was the internal control. B, mRNA expression levels of ALP, DSPP and FSHR levels of the DPSCs in the negative control and FSH-treated groups (3-100 ng/mL) in odonto/ osteogenic medium for 3 d. Data are presented as the mean ± standard deviation, ***P < 0.0001. C, Grey scale analysis of the protein levels in the Western blotting results in A by ImageJ. Data are presented as the mean ± standard deviation, *P < 0.05, **P < 0.01, ***P < 0.0001. D, After odonto/osteogenic induction for 14 d, calcium contents in the 30 ng/mL and 100 ng/mL group were lower than those in the group without FSH treatment. Data are presented as the mean ± standard deviation. ***P < 0.0001. E, After odonto/osteogenic induction for 14 d, Alizarin Red staining of the mineralization nodules showed fewer calcified nodules in the 10-100 ng/mL FSH-treated groups, especially the 30 and 100 ng/mL groups, than in the control group F I G U R E 4 A concentration of 30 ng/ mL FSH activated the JNK pathway in the DPSCs. A, Protein expressions of ERK1/2, p-ERK1/2, p38, p-p38, JNK and p-JNK by Western blotting analysis. GAPDH was the internal control. B, Quantification analysis for the ratios of p-ERK1/2/ ERK1/2 (a), p-p38/pp8 (b) and p-JNK/ JNK (c) in A, Data are presented as the mean ± standard deviation,**P < 0.01, ***P < 0.0001 | 10631 QIAN ANd GUAN activation of FSHR in osteoclasts can promote osteoclast formation and enhance osteoclast function. FSHRs were expressed on bone marrow mesenchymal stem cells (BMSCs), while their expression was down-regulated in mature osteoblasts, indicating that FSHRs may negatively drive MSC differentiated into osteoblasts. 16 Application of an FSH antibody could induce osteoblast colonies formation. Similarly, osteoblast colonies cultured from FSHR -/mice were more than those in wild type. 24 DPSCs are important mesenchymal stem cells in dentin-pulp complex, possessing many similarities to BMSCs. In our study, we found that FSHR was expressed in the DPSCs. Application of FSH, especially at a concentration of 30 ng/mL, significantly up-regulated the expression levels of FSHR in the DPSCs. We also found that the FSHR expression levels declined during odonto-osteogenic differentiation of the DPSCs.
The proliferative and differentiated abilities of DPSCs are very important features in regeneration of dentin-pulp complex. In our study, a potential impact of FSH on the proliferation of DPSCs was investigated. We found that 3, 10, 30 and 100 ng/mL FSH did not affect the proliferation of DPSCs. No obvious changes in the effects of FSH on the apoptotic rate of the DPSCs were found. FSH (30 ng/ mL) significantly decreased the ALP activity, mineralization capacity and odonto/osteogenic potential of the cells. FSH could down-regulate the protein levels of osteo/odontogenic-related genes, including DSPP, ALP, RUNX2 and OSX. Runx2 belongs to the Runx family of transcription factors, which is a key regulator of osteoblast differentiation. 28 Runx2 is expressed in osteoblast lineage cells and chondrocytes. 29 In Runx2 −/− mice, intramembranous and endochondral bone formation are both completely interrupted. 30 Runx2 significantly stimulated calcium accumulation and alkaline phosphatase activity in dental pulp stem cells (DPSCs). 31 Dentin sialophosphoprotein (DSPP), a chimeric extracellular matrix protein, is expressed in teeth, predominantly by odontoblasts. 32 This protein is processed by proteases into three protein products: dentin sialoprotein (DSP), dentin glycoprotein (DGP) and dentin phosphoprotein (DPP). DSPP is important in dentin biomineralization, the mutations of which interrupt mineralization homeostasis during odontoblast differentiation. 33 Osx is an osteoblast-specific transcription factor required for bone formation and tooth development. 34 Osx has been identified to be associated with odontoblast differentiation. 35 MAPK is a broad-based serine/threonine protein kinase and plays an important role in signal transduction of cell proliferation and differentiation in DPSCs. 36 The ovariectomized (OVX) rat has been applied widely as menopausal model, as it has been validated to show oestrogen deficiency. 42 However, only ovariectomy model was unable to highlight the roles of FSH. Our group administered FSH and its inhibitor based on OVX model to study the independent role of FSH. 18,20 No obvious changes were found in E2 levels of OVX groups after FSH and LE application. Furthermore, significant changes were found in FSH levels in OVX + FSH and OVX + LE groups compared with OVX group. Based on this rat model, we combined a direct pulp-capping model to test the role of FSH in dental pulp healing. Herein, as previously observed, the formation of dentin bridges was observed beneath the perforation of dental pulps in the Sham group. 19 Histologically, higher rates of inflammation and less calcified bridge formation in the OVX groups. This condition was more severe in the FSH groups and alleviated in the LE groups. We also gained insights into the role of some mineralization-related proteins, such as DSP, OCN and RUNX2. In this study, the OVX group presented weaker expression of several odonto/osteogenic proteins in the dentin-pulp complex than the sham group, which was in accordance with a previous study. 11 Administration of FSH further weakened the expression of odonto/osteogenic proteins, and in the LE groups, their expression levels were stronger than those in the OVX groups.
Since the cell-rich zone in the dentin-pulp complex mainly contains pulp fibroblasts and DPSCs, the down-regulation of these odonto/ osteogenic proteins may indicate that the differentiation rate of the F I G U R E 5 Effects of the JNK inhibitor on odonto/osteogenic differentiation of DPSCs treated with the FSH (30 ng/mL). A, Protein levels of p-JNK and JNK at 60 min in the NC group, FSH group, SP600125 group and FSH + SP600125 group. GAPDH was the internal control. B, Grey scale analysis of the protein levels in the western blotting results in A by ImageJ. Data are presented as the mean ± standard deviation, ***P < 0.0001. C, The real-time PCR expression levels of ALP (a), DSPP (b) and RUNX2 (c) in the FSH + SP600125 group were up-regulated in comparison with the FSH group at day 3. Data are presented as the mean ± standard deviation, **P < 0.01, ***P < 0.0001. D, The ALP activities in the NC group, mineralization medium (MM) group, FSH group and FSH + SP600125 group were detected at day 7. Data are presented as the mean ± standard deviation, ***P < 0.0001. E, Calcium contents in the FSH + SP600125 group were higher than that in the FSH group at day 7. Data are presented as the mean ± standard deviation, ***P < 0.0001. F, Alizarin Red staining of the mineralized nodules in the NC group, mineralization medium (MM) group, MM + FSH group and MM + FSH + SP600125 group for 14 d. FSH, follicle-stimulating hormone; DPSCs, dental pulp stem cells; ALP, alkaline phosphatase; JNK, c-Jun N-terminal kinases DPSCs in the OVX groups was decreased compared with that in the sham groups. The FSH-treated groups and the LE groups presented the opposite results, indicating that FSH may inhibit the differentiation of the DPSCs. Therefore, these results suggested that FSH, independent of oestrogen, can impair the formation of reparative dentin at the injury site of direct pulp capping. F I G U R E 6 FSH impaired dental pulp healing after direct pulp capping in the OVX rats. A, The levels of E2 in the serum. Data are presented as the mean ± standard deviation. **P < 0.01 (Sham group vs OVX/OVX + LE/OVX + FSH group). B, Levels of FSH in the serum. Data are presented as the mean ± standard deviation. *P < 0.05 (Sham group vs OVX group; OVX group vs OVX + LE group), **P < 0.01 (OVX + FSH group vs Sham/ OVX/OVX + LE group). C-J, Representative pictures of histologic features in each group at day 21. (C,G) Sham, (D,H) OVX, (E,I) OVX + LE, (F,J) OVX + FSH. (Haematoxylin-eosin staining; original magnification, ×40 for C-F, ×200 for G-J). K-M Summary of different categories of histologic features according to the scores. Data are presented as the mean ± standard deviation, *P < 0.05, **P < 0.01, ***P < 0.0001 F I G U R E 7 Immunohistochemical observation of mineralization-related proteins and FSHR. A-P, present the immunostaining for DSP (A-D), OCN (E-H), RUNX2 (I-L) and FSHR (M-P) in the dentin-pulp complex. Q-T shows the quantitative analysis of their histologic expression levels in each group. Scale bars = 50 μm. Data are presented as the mean ± standard deviation, *P < 0.05, **P < 0.01, ***P < 0.0001 In summary, this study showed that FSH did not influence the proliferation of human DPSCs. FSH, especially at a concentration of 30 ng/mL, inhibited osteo/odontogenic differentiation of the DPSCs via modulating the JNK-MAPK pathway. Moreover, an in vivo study further confirmed that FSH could block the reparative dentine formation in the OVX rats. Although further studies are required, our findings suggest that FSH may play a role in the DPSCs as a negative regulator of mineralization, which may be a crucial factor for clinical pulp healing and regeneration during menopause.

ACK N OWLED G EM ENTS
This work was supported by The Youth Scientific Research Funds of School of Stomatology, Shandong University (2018QNJJ05).

CO N FLI C T O F I NTE R E S T
The authors deny any conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are openly available