research-article2014

JDR

93210.1177/0022034513512507

β

RESEARCH REPORTS Biological

Y. Tang1,2, X. Zhou2, B. Gao1, X. Xu1, J. Sun2, L. Cheng1,2, X. Zhou1,2, and L. Zheng1,2* 1

State Key Laboratory of Oral Diseases, West China Hospital of Stomatology, Sichuan University, Chengdu, China; and 2 West China School of Stomatology, Sichuan University, Chengdu, China; *corresponding author, zhenglw399@ hotmail.com

Modulation of Wnt/β-catenin Signaling Attenuates Periapical Bone Lesions

J Dent Res 93(2):175-182, 2014

Abstract Wnt/β-catenin signaling plays an important role in bone biology. The present study investigated the involvement of Wnt/β-catenin signaling in rat periapical bone destruction and whether lithium chloride (LiCl), a glycogen synthase kinase-3β (GSK-3β) inhibitor, promotes bone restoration. Rat bone marrow mesenchymal cells (BMMSCs) treated with Porphyromonas gingivalis lipopolysaccharide (Pg LPS) showed decreased osteogenic potential through inhibited Wnt/β-catenin signaling as quantified by Western blot, immunofluorescence, and luciferase reporter assay. Transient Wnt3a treatment in vitro partially restored mineralization and Runx2/Osx and osteocalcin expression in cultures with Pg LPS-induced osteogenic arrest. Prolonged Wnt3a treatment impaired osteogenic commitment. X-ray microtomography showed dramatically enhanced periapical bone formation in rats gavage-fed with LiCl for 2 wks, while continuous LiCl treatment for 4 wks impaired periapical bone healing. LiCl treatment also increased GSK-3β phosphorylation and osteocalcin expression in periapical tissue. Collectively, these results indicate that Wnt/β-catenin has dichotomous functions in bone homeostasis. Modulation of this signaling pathway by LiCl may be a potential therapeutic option for bone destruction in endodontic disease.

KEY WORDS: BMMSCs, LPS, osteoblast, osteogenesis, Wnt3a, β-catenin. DOI: 10.1177/0022034513512507 Received July 26, 2013; Last revision October 22, 2013; Accepted October 22, 2013 A supplemental appendix to this article is published electronically only at http://jdr.sagepub.com/supplemental. © International & American Associations for Dental Research

Introduction

J

aw bone loss through oral diseases such as periodontitis or periapical infections is highly prevalent among adults; however, there is no effective treatment to restore lost bone. The pathogenesis of alveolar bone loss surrounding the tooth root involves a multitude of bacterial- and host-related factors. Lipopolysaccharide (LPS) produced by bacteria plays a critical role in stimulating the inflammatory response (Okahashi et al., 2004), which induces bone destruction by promoting osteoclast differentiation and inhibiting osteoblast differentiation (Okahashi et al., 2004; Bonsignore et al., 2013). Bone remodeling requires a complex network of systemic hormones and local factors for osteoprogenitor lineage cells to progress through stages of differentiation. Constituents of the Wnt/β-catenin pathway are among these factors; this pathway increases bone mass through mechanisms including renewal of stem cells, stimulation of pre-osteoblast replication, induction of osteoblastogenesis, and inhibition of osteoblast apoptosis (Bodine, 2008; Maupin et al., 2013). These processes can be stimulated by the inhibition of glycogen synthase kinase-3β (GSK-3β), an enzyme that phosphorylates and degrades β-catenin in the cytoplasm (Cho et al., 2012). LiCl, a well-known GSK-3β inhibitor, is therefore an enticing target in the development of new pharmacological interventions for bone loss (Moon et al., 2004). In this study, we tested the involvement of Wnt signaling in the attenuation of rat bone marrow mesenchymal cells (BMMSCs) osteogenic differentiation by Pg LPS and studied the treatment potential of LiCl in bone lesion restoration of rat apical periodontitis.

Materials & Methods BMMSCs Isolation and Treatment Rat BMMSCs were isolated (refer to the online Appendix for details) and cultured in osteogenic media (αMEM containing 20% FBS, 10 mM β-glycerophosphate, 50 mg/mL ascorbic acid, and 10 nM dexamethasone), supplemented with 10 μg/mL Pg LPS. For the rescue experiment, 10 ng/mL Wnt3a recombinant protein was co-administered either chronically, between days 1 and 28, or briefly, for 5 days between days 9 and 14. (For details of RNA isolation, quantitative RT-PCR analysis, immunofluorescence staining, protein extraction, and Western blotting, please refer to the online Appendix.)

Alkaline Phosphatase Activity and Mineralization Assay Alkaline phosphatase (ALP) activity in cell lysates was measured by a standard spectrophotometric method with an ALP Reagent Kit (JianCheng Bioengineering

175

176 

Tang et al.

Institute, Nanjing, Jiangsu, China). ALP values are expressed as ALP specific activity (IU/mg protein). Mineralization was assayed by Alizarin red staining on day 28 to detect calcification. For quantification, cells were de-stained with ethylpyridinium chloride (Sigma, St. Louis, MO, USA), which was then quantified by spectrophotometry at 550 nm.

TOPflash Reporter Assay β-Catenin/Tcf-Lef transcriptional activity was assayed in BMMSCs by luciferase assay. Wild-type or mutant TOPflash β-catenin LUC reporter containing 8 TCF/LEF binding sites (Addgene, Inc., Cambridge, MA, USA), along with pRL-TK Renilla luciferase reporter plasmid (Promega, Madison, WI, USA), was transfected into cells. Twenty-four hrs after transfection, cells were either left untreated or treated with Wnt3a in the presence or absence of LPS. Luciferase activity was measured with the Dual-Luciferase Reporter Assay kit (Promega). All luciferase values were normalized to the Renilla transfection control.

Experimental Periapical Lesion and Lithium Dosing Regimen Sprague-Dawley rats were purchased from the Experimental Animal Centre of Sichuan University. The principles and protocols of laboratory animal care, approved by the West China School and Hospital of Stomatology Ethics Committee, were followed throughout the study. A periapical lesion was induced by occlusal exposure of the pulp in both mandibular first molars as described previously (Liu et al., 2012). The cavity was left open to the oral environment without any restoration for 28 days. Rats with experimental periapical lesions were gavage-fed either a daily dose of 200 mg/Kg LiCl solution (Sigma, St. Louis, MO, USA) or distilled water (vehicle) (ClementLacroix et al., 2005). Rats were randomly divided into: (1) a briefly treated group that received LiCl from 7 days to 21 days after periapical lesion induction; (2) a continuously treated group that received LiCl for 28 days during the entire experimental period; and (3) a control group that received distilled water only, with at least 10 rats in each group.

MicroCT Measurement Rats were sacrificed at the end of the treatment course. Mandibles were dissected and fixed with 4% paraformaldehyde at 4°C for 2 days. Samples were scanned by a compact fan-beam-type tomography (μCT80, Scanco MedicalAG, Bassersdorf, Switzerland). For each sample, approximately 150 microtomographic slices of 17-µm increments were acquired at a pixel matrix of 1024×1024, covering the entire medial-lateral width of the mandible. At a 3D level, detailed microCT analyses of bone volume (BV)/total volume (TV) ratios of the periapical area surrounding the distal root of the first molars were made as previously published (Gao et al., 2013).

Histology and Immunohistochemistry Mandibular tissues were transferred in 17% EDTA for demineralization at 4°C for 4 wks and embedded in paraffin. Frontal

J Dent Res 93(2) 2014 serial sections at 6-μm thickness were obtained. Sections showing the apical foramen were selected. Histological analyses, including H/E staining, TRAP staining, and immunohistochemistry, were performed as described in the online Appendix.

Statistics All experiments were performed at least 3 times. The results are given as means ± SEM. Differences were analyzed by Student’s t test or one-way analysis of variance (ANOVA) as appropriate. A difference of p < .05 was considered statistically significant.

Results Porphyromonas gingivalis LPS Inhibited Osteoblastic Differentiation of BMMSCs through Suppression of the Wnt/β-catenin Pathway To examine Wnt/β-catenin pathway involvement in periapical bone destruction, we added Pg LPS to rat BMMSCs culture in vitro. Western blotting showed that Pg LPS stimulation significantly reduced phosphorylated GSK-3β (Ser9) in BMMSCs within 1 hr (p < .05). In contrast, protein levels of phosphorylated GSK-3β (Tyr216) remained unaltered (Fig. 1A). Cytoplasmic β-catenin accumulation remained unchanged under Pg LPS stimulation (Fig. 1B). However, nuclear β-catenin began to decrease 2 hrs after Pg LPS stimulation (p < .05, Fig. 1C), suggesting that Pg LPS suppressed β-catenin nuclear translocation. These results were also supported by immunofluorescence staining, which showed directly reduced β-catenin in the nucleus after Pg LPS treatment (Fig. 1D). Results from TOPflash reporter assays also demonstrated that Pg LPS significantly inhibited β-catenin/Tcf-Lef transcriptional activity, while Wnt3a treatment reversed this reduction (Fig. 1E).

Wnt3a Abrogated LPS-attenuated BMMSCs Osteogenic Activity in vitro Wnt3a protein was utilized to mimic Wnt signaling activation. Pg LPS treatment significantly decreased ALP activity in BMMSCs (p < .05). Inhibition of GSK-3β by Wnt3a, however, significantly attenuated the reduction in ALP activity (p < .05) (Fig. 2A). ALP mRNA levels also showed a pattern similar to those reported in ALP enzyme activity studies (Fig. 2B). Mineralization of BMMSCs was assayed by Alizarin red staining after 28 days of incubation and was significantly reduced in response to Pg LPS. Transient exposure (5 days) of the Pg LPS-treated cultures to Wnt3a rescued mineralization (Fig. 2C), whereas chronic Wnt3a treatment (28 days) failed to counteract the effect of Pg LPS in reducing mineralization of BMMSCs (Fig. 2D). Osteocalcin and Runx2/osterix mRNA levels were significantly reduced in BMMSCs treated with Pg LPS (p < .05) (Fig. 2E). Transient exposure to Wnt3a achieved partial rescue (p < .05) (Fig. 2E). However, both were down-regulated in response to chronic (28 days) Wnt3a treatment. Combination of chronic treatment of Wnt3a with Pg LPS led to even lower mRNA levels

J Dent Res 93(2) 2014  177 Wnt/β-catenin Signaling and Periapical Bone Lesions

Figure 1.  Pg LPS suppressed Wnt/β-catenin signaling in bone marrow mesenchymal cells (BMMSCs) in vitro. (A) Treatment with Pg LPS significantly decreased GSK-3b phosphorylation (Ser9) within 0.5 hr. In contrast, protein levels of phosphorylated GSK-3β (Tyr216) remained unaltered. (B) β-catenin accumulation level remained unchanged under Pg LPS stimulation. (C) Nuclear β-catenin began to decrease 2 hrs after Pg LPS stimulation. (D) Immunolocalization of β-catenin in BMMSCs by fluorescent microscopy (×400). Cells were treated for 2 hrs with Pg LPS or Wnt3a. Weaker β-catenin immunoreactivity in nuclear areas in Pg LPS-treated cells, compared with intense immunoreactivity in Wnt3a-treated cells. (E) The luciferase reporter assay for the TCF/LEF activity in the presence of Pg LPS or Wnt3a. Cells were co-transfected with wild-type or mutant TCF/LEF binding-site luciferase constructs, and luciferase activities were measured. Values are reported as firefly luciferase activity to Renilla luciferase activity. The results are expressed as means ± SEM (n = 3). Differences were assessed by one-way analysis of variance (ANOVA).*p < .05 as compared with the 0 hr samples.

178 

Tang et al.

J Dent Res 93(2) 2014 of osteocalcin and Runx2/Osx compared with Pg LPS treatment alone.

GSK-3β Inhibitor LiCl Decreased Periapical Bone Lesion Size Photomicrographs of periapical lesions revealed that the vehicle group had significant periapical bone destruction surrounding the distal root of the mandibular first molar (Figs. 3A3, 3A4) compared with the normal/non-surgical group (Figs. 3A1, 3A2). We then verified whether modulation of Wnt/β-catenin signaling through abrogating GSK-3β activity by LiCl could alter periapical bone destruction in vivo. The transient/ brief LiCl treatment (14 days) group showed decreased bone lesions (Figs. 3A5, 3A6). Bone volume/total volume (BV/TV) ratios of the periapical region of the first molar distal root were significantly higher in the brief LiCl treatment group (p < .05) (Figs. 3B5, 3B6, 3C) when compared with the vehicle group (Figs. 3B3, 3B4, 3C). However, there was no significant difference between the continuously treated (28 days) group (Figs. 3B7, 3B8, 3C) and vehicle.

LiCl Decreased Bone Lesions by Promoting Bone Formation and Regeneration

Figure 2. Wnt3a rescued the bone marrow mesenchymal cells (BMMSCs) differentiation in vitro when incubated with Pg LPS (10 μg/mL). (A) ALP activity was detected at day 10 and was significantly lower in the sample treated with Pg LPS alone as compared with that in any of the other 3 samples. ALP activity results are shown as means ± SEM (n = 3). (B) ALP mRNA levels were significantly lower in the sample treated with Pg LPS alone than in the other 3 samples on day 9. ALP mRNA levels were significantly higher in the Pg LPS + Wnt3a sample than in the sample treated with Pg LPS alone, but were still lower than those in the control and Wnt3a samples. Real-time PCR results are presented as means ± SEM (n = 3). Differences were analyzed by one-way analysis of variance (ANOVA). (C) Wnt3a rescued mineralization. Pg LPS was chronically administered between days 1 and 28, and Wnt3a was administered between days 9 and 14. (D) Chronic Wnt3a treatment did not counteract the inhibitory effect of Pg LPS. Pg LPS and Wnt3a were administered chronically, and alizarin red staining was performed on day 28. The lower panels of C and D are representative bright-field micrographs (×100) of the cultures scanned in the upper panels of C and D, respectively. (E) Transient/ short-time exposure to Wnt3a significantly elevated the Runx2/osterix and osteocalcin mRNA levels and partially rescued Pg LPS decreased expression, while their mRNA levels were downregulated in response to chronic Wnt3a treatment. Real-time PCR results are presented as means ± SEM (n = 3). Differences were analyzed by one-way ANOVA. *p < .05.

We next determined whether LiCl decreased lesion size by increasing osteoblastogenesis and promoting bone regeneration in vivo. Immunohistochemical staining showed a strong signal of p-GSK-3β in periapical tissues in the LiCl treatment group and a weak signal in other groups (Figs. 4A4, 4A5). Cuboid cells adjacent to trabecular bone of the transient/brief LiCl treatment group displayed intense osteocalcin signal in the cytoplasm (Fig. 4B9). Flat cells on the trabecular bone surface of the vehicle group stained positive for osteocalcin (Fig. 4B7). In the chronic treatment group, cell layers adjacent to the trabecular bone displayed weak osteocalcin signal (Fig. 4B10). Histochemical staining indicated that the transient LiCl treatment significantly attenuated loss of trabecular bone due to promotion of osteoblast number (Ob.No/mm) (Fig. 4C) and osteoblast surface (Ob.S/BS) (Fig. 4C) in the periapical region. The increase in bone volume by transient

J Dent Res 93(2) 2014  179 Wnt/β-catenin Signaling and Periapical Bone Lesions lithium treatment may also be due to a decrease in osteoclastogenesis. TRAP staining showed fewer TRAP-positive cells in the transient LiCl-treated group, compared with the vehicle group and the chronic LiCl treatment group (Figs. 4E, 4F). This may be explained by the possibility that transient LiCl treatment can promote osteoblast differentiation and regulate RANKL/OPG expression, which indirectly inhibits osteoclastogenesis in vivo. Quantitative real-time PCR analysis revealed that the transient LiCl treatment group had significantly higher mRNA expression of Runx2, osterix, osteocalcin, bone sialoprotein (BSP) compared with the other three groups (p < .05) (Fig. 4D). Collectively, the results showed that modulation of Wnt/β-catenin signaling by transient LiCl treatment promoted bone formation and regeneration in periapical bone lesions, but chronic administration of LiCl impaired bone healing (Fig. 4G).

Discussion One major reason for jawbone loss is bacterial infection, and LPS has been wellrecognized as a contributor to the bone loss process (Nonnenmacher et al., 2005). Previous studies have proposed LPSFigure 3.  LiCl treatment decreased periapical bone lesions in vivo. (A) Hematoxylin and eosin (H&E) staining of sections and computed tomography analysis (B) of the mandibular first mediated inhibition of osteoblast differenmolars from 4 groups. Severe periapical bone loss is visible in the untreated vehicle group, tiation (Tang Y et al., 2011). A recent and periapical bone lesions were significantly decreased in the transient/brief LiCl treatment study showed that NF-κB inhibits osteogroup (from day 7 to day 21) compared with the vehicle (untreated) group and the chronic/ genic differentiation of mesenchymal continuous LiCl treatment group. (C) Quantification of distal apical bone volume (BV)/total stem cells by promoting β-catenin degravolume (TV) from samples of 4 groups.*p < .05. dation (Chang et al., 2013). Since LPS activates NF-κB through the Toll-like receptor, the underlying mechanism of effect of Wnt3a on osteogenic differentiation depends on treatLPS in attenuating BMMSCs osteogenic differentiation, and thus ment duration. In this study, we showed that transient exposure the strategy of rescue by modulating Wnt/β-catenin signaling, to Wnt3a rescued Pg LPS inhibition of osteoblast differentiawas to be elucidated. In the current study, our results indicated tion, while chronic exposure inhibited differentiation. This may that LPS attenuates osteogenic differentiation of BMMSCs by suggest that the effects of Wnt3a on osteoblast differentiation inhibiting Wnt/β-catenin signaling, while LiCl treatment rescued might be more complex than previously thought (Krause et al., this inflammation-stressed osteogenic differentiation in vivo. 2010). The negative outcome of continuous administration posGSK-3β plays a central role in the Wnt signaling pathway. sibly reflects required down-regulation of the Wnt signaling Previous studies reported different responses of GSK-3β to LPS pathway at very late stages of osteoblast differentiation (Li in different cells. Stimulation of macrophages (Kim et al., 2010) et al., 2005). and stem cells of apical papilla with LPS induces the phosLithium chloride (LiCl) is a mood-stabilizing drug that has phorylation and inactivation of GSK3-β (Wang et al., 2013). In been used for decades. It has been recently reported to increase contrast, bacterial infection decreased phosphorylated GSK-3β bone mass in patients (Zamani et al., 2009). Lithium administrain colonic epithelial cells (Duan et al., 2007). In this study, we tion in animals has been widely shown to stimulate Wnt/βshowed significant down-regulation of GSK-3β phosphorylacatenin-mediated transcription, leading to an increase in bone tion at Ser9 in BMMSCs in response to Pg LPS challenge. mass and improvement in bone lesion healing (Clement-Lacroix Wnt3a has been shown to be a potent regulator of osteogenic et al., 2005; Chen et al., 2007; Tang GH et al., 2011; Zeng et al., differentiation (Baksh et al., 2007; Leclerc et al., 2008). The

180 

Tang et al.

J Dent Res 93(2) 2014

Figure 4. LiCl activated Wnt/β-catenin signaling and promoted bone formation in periapical lesion. (A) Immunohistochemical staining of periapical lesion sections indicates strong positive staining of phosphorylated GSK-3 β (Ser9) (brown area) in periapical tissue from the LiCl-treated group (briefly and continuously) compared with the normal group and the vehicle group. (B) Immunohistochemical staining of the bone formation specific marker, osteocalcin. White arrow: Cuboidal and active osteoblastic cell layers adjacent to trabecular bone with brief LiCl treatment displayed intense osteocalcin expression. Black arrow: Flat osteoblast cells expressed osteocalcin immunoreactivities in the vehicle group, which indicated that these cells were in the quiescent period. Red arrow: Cell layers adjacent to trabecular bone displayed weak osteocalcin expression, which may be osteoblastic progenitor cells. (C) Histomorphometric analyses proved that bone area covered by cuboidal osteoblasts is significantly increased in the ‘brief LiCl treatment’ group, while this effect is not obvious in the continuous-treatment group. Ob.S/BS, osteoblast surface/bone surface; Ob.N/BS, osteoblast number/bone surface. (D) qRT-PCR results for the osteoblast differentiation marker proved that brief LiCl treatment rescued Runx2, osterix, osteocalcin, and BSP expression in periapical lesion, while continuous treatment did not have this effect. Data are presented as means ± SEM (n = 3). Differences were analyzed by one-way analysis of variance (ANOVA).*p < .05. (E) TRAP-stained paraffinembedded periapical tissue section photographed at 200×. (F) Mean values of osteoclast numbers (N.Oc/BS, n/mm). Values are expressed as mean ± SEM. *p < .05. Yellow arrow: Osteoclasts adjacent to periapical bone. (G) Schematic model of activating Wnt/β-catenin can rescue impaired osteogenic activity due to microbial challenge and inflammatory cytokines.

J Dent Res 93(2) 2014  181 Wnt/β-catenin Signaling and Periapical Bone Lesions 2013). The dosing regimen used in the present study was previously demonstrated to result in a serum lithium level of 0.4 to 0.5 mM, which activates β-catenin signaling and enhances new bone regeneration in rats (Clement-Lacroix et al., 2005). Our findings indicate that LiCl treatment plays a disparate role and acts differently during different phases of bone lesion repair. A significant increase in new bone volume was identified after brief LiCl administration starting from 7 days and ending 21 days after apical lesion induction. Chronic LiCl treatment resulted in significantly retarded bone formation. These results are consistent with our in vitro results. Likewise, early in vivo studies found that continuous treatment with lithium activates β-catenin, similar to the activating mutation resulting in constitutively active β-catenin, which leads to a block of osteoblast differentiation and a delay in bone-fracture healing (Chen et al., 2007). Vertical root fractures were observed in some of the proximal roots due to occlusion exposure, and transient LiCl treatment failed to increase the bone volume in the proximal apical region (Appendix Fig.). Vertical fractures may augment the bacteria challenge by facilitating the ingress of bacteria and associated irritants, while the mobility of root fragments accelerated bone resorption (Gao et al., 2012). This may further induce tissue destruction and bone loss in samples with proximal root fractures. Another consideration is that LiCl may have effects outside of the Wnt pathway. It has been recently reported that LiCl can attenuate BMP-2 signaling and inhibits in vitro osteogenic differentiation (Li et al., 2011); this should be considered in future studies. It should be noted that this study focused mainly on the effect of Wnt/β-catenin on osteoblasts in bone lesions. Prior studies have implicated canonical Wnt signaling in the regulation of bone metabolism at several levels, not only in modulating the osteoblast lineage, but also in inhibiting osteoclast differentiation (Glass et al., 2005; Jackson et al., 2005). Lithium is an effective drug for treating bipolar disorder. Major side-effects include extreme weight gain, diabetes insipidus, lithium toxicity leading to renal failure, dysarthria, and convulsions. It is not known to increase cancer risk, while many components of Wnt signaling are found to be overexpressed or mutated in cancer (Anastas and Moon, 2013). While lithium has been widely used in patients, its adverse effects remain a major concern. In summary, our findings demonstrate that promotion of osteoblastogenesis via modulation of Wnt/β-catenin signaling may be a viable therapeutic approach to accelerate healing of alveolar bone loss, which is a major concern for both patients and clinicians. Conversely, our results also show that bone healing is greatly retarded with continuously administered lithium. Thus, the risks and benefits of modulating Wnt/β-catenin signaling as a therapeutic aid for inducing bone formation and repair await comprehensive long-term animal studies and human trials.

Acknowledgments This study was supported by the National Program on Key Basic Research Project (973 Pilot Program, grant number 2011CB512108), by NSFC grants 81371136 and JCPT2011-9 to XDZ, and by NSFC grant 81200760 to LWZ. The authors

declare no potential conflicts of interest with respect to the authorship and/or publication of this article.

References Anastas JN, Moon RT (2013). WNT signalling pathways as therapeutic targets in cancer. Nat Rev Cancer 13:11-26. Baksh D, Boland GM, Tuan RS (2007). Cross-talk between Wnt signaling pathways in human mesenchymal stem cells leads to functional antagonism during osteogenic differentiation. J Cell Biochem 101:1109-1124. Bodine PV (2008). Wnt signaling control of bone cell apoptosis. Cell Res 18:248-253. Bonsignore LA, Anderson JR, Lee Z, Goldberg VM, Greenfield EM (2013). Adherent lipopolysaccharide inhibits the osseointegration of orthopedic implants by impairing osteoblast differentiation. Bone 52:93-101. Chang J, Liu F, Lee M, Wu B, Ting K, Zara JN, et al. (2013). NF-kappaB inhibits osteogenic differentiation of mesenchymal stem cells by promoting beta-catenin degradation. Proc Natl Acad Sci USA 110:9469-9474. Chen Y, Whetstone HC, Lin AC, Nadesan P, Wei Q, Poon R, et al. (2007). Betacatenin signaling plays a disparate role in different phases of fracture repair implications for therapy to improve bone healing. PloS One 4:e249. Cho YD, Kim WJ, Yoon WJ, Woo KM, Baek JH, Lee G, et al. (2012). Wnt3a stimulates Mepe, matrix extracellular phosphoglycoprotein, expression directly by the activation of the canonical Wnt signaling pathway and indirectly through the stimulation of autocrine Bmp-2 expression. J Cell Physiol 227:2287-2296. Clement-Lacroix P, Ai M, Morvan F, Roman-Roman S, Vayssiere B, Belleville C, et al. (2005). Lrp5-independent activation of Wnt signaling by lithium chloride increases bone formation and bone mass in mice. Proc Natl Acad Sci USA 102:17406-17411. Duan Y, Liao AP, Kuppireddi S, Ye Z, Ciancio MJ, Sun J (2007). betaCatenin activity negatively regulates bacteria-induced inflammation. Lab Invest 87:613-624. Gao B, Chen W, Hao L, Zhu G, Feng S, Ci H, et al. (2013). Inhibiting periapical lesions through AAV-RNAi silencing of cathepsin K. J Dent Res 92:180-186. Gao S-S, Zhang Y-R, Zhu Z-L, Yu H-Y (2012). Micromotions and combined damages at the dental implant/bone interface. Int J Oral Science 4:182-188. Glass DA 2nd, Bialek P, Ahn JD, Starbuck M, Patel MS, Clevers H, et al. (2005). Canonical Wnt signaling in differentiated osteoblasts controls osteoclast differentiation. Dev Cell 8:751-764. Jackson A, Vayssiere B, Garcia T, Newell W, Baron R, Roman-Roman S, et al. (2005). Gene array analysis of Wnt-regulated genes in C3H10T1/2 cells. Bone 36:585-598. Kim JS, Yeo S, Shin DG, Bae YS, Lee JJ, Chin BR, et al. (2010). Glycogen synthase kinase 3beta and beta-catenin pathway is involved in Toll-like receptor 4-mediated NADPH oxidase 1 expression in macrophages. FEBS J 277:2830-2837. Krause U, Harris S, Green A, Ylostalo J, Zeitouni S, Lee N, et al. (2010). Pharmaceutical modulation of canonical Wnt signaling in multipotent stromal cells for improved osteoinductive therapy. Proc Natl Acad Sci USA 107:4147-4152. Leclerc N, Noh T, Cogan J, Samarawickrama DB, Smith E, Frenkel B (2008). Opposing effects of glucocorticoids and Wnt signaling on Krox20 and mineral deposition in osteoblast cultures. J Cell Biochem 103:1938-1951. Li J, Khavandgar Z, Lin SH, Murshed M (2011). Lithium chloride attenuates BMP-2 signaling and inhibits osteogenic differentiation through a novel WNT/GSK3- independent mechanism. Bone 48:321-331. Li X, Liu P, Liu W, Maye P, Zhang J, Zhang Y, et al. (2005). Dkk2 has a role in terminal osteoblast differentiation and mineralized matrix formation. Nat Genet 37:945-952. Liu L, Zhang C, Hu Y, Peng B (2012). Protective effect of metformin on periapical lesions in rats by decreasing the ratio of receptor activator of nuclear factor kappa B ligand/osteoprotegerin. J Endod 38:943-947. Maupin K, Droscha C, Williams B (2013). A comprehensive overview of skeletal phenotypes associated with alterations in Wnt/β-catenin signaling in humans and mice. Bone Res 1:27-71.

182 

Tang et al.

Moon RT, Kohn AD, De Ferrari GV, Kaykas A (2004). WNT and betacatenin signalling: diseases and therapies. Nat Rev Genet 5:691-701. Nonnenmacher C, Dalpke A, Rochon J, Flores-de-Jacoby L, Mutters R, Heeg K (2005). Real-time polymerase chain reaction for detection and quantification of bacteria in periodontal patients. J Periodontol 76:1542-1549. Okahashi N, Inaba H, Nakagawa I, Yamamura T, Kuboniwa M, Nakayama K, et al. (2004). Porphyromonas gingivalis induces receptor activator of NF-κB ligand expression in osteoblasts through the activator protein 1 pathway. Infect Immun 72:1706-1714. Tang GH, Xu J, Chen RJ, Qian YF, Shen G (2011). Lithium delivery enhances bone growth during midpalatal expansion. J Dent Res 90:336340.

J Dent Res 93(2) 2014 Tang Y, Sun F, Li X, Zhou Y, Yin S, Zhou X (2011). Porphyromonas endodontalis lipopolysaccharides induce RANKL by mouse osteoblast in a way different from that of Escherichia coli lipopolysaccharide. J Endod 37:1653-1658. Wang J, Dai J, Liu B, Gu S, Cheng L, Liang J (2013). Porphyromonas gingivalis lipopolysaccharide activates canonical Wnt/beta-catenin and p38 MAPK signalling in stem cells from the apical papilla. Inflammation [Epub ahead of print 7/2/2013] (in press). Zamani A, Omrani GR, Nasab MM (2009). Lithium’s effect on bone mineral density. Bone 44:331-334. Zeng YT, Fu B, Tang GH, Zhang L, Qian YF (2013). Effects of lithium on extraction socket healing in rats assessed with micro-computed tomography. Acta Odontol Scand 71:1335-1340.