Lithium carbonate accelerates the healing of apical periodontitis

Apical periodontitis is a disease caused by bacterial invasions through the root canals. Our previous study reported that lithium chloride (LiCl) had a healing effect on apical periodontitis. The aim of this report is to investigate the healing properties and mechanism of lithium ion (Li+) for apical periodontitis using rat root canal treatment model. 10-week-old male Wistar rat’s mandibular first molars with experimentally induced apical periodontitis underwent root canal treatment and were applied lithium carbonate (Li2CO3) containing intracanal medicament. Base material of the medicament was used as a control. Subject teeth were scanned by micro-CT every week and the periapical lesion volume was evaluated. The lesion volume of Li2CO3 group was significantly smaller than that of the control group. Histological analysis showed that in Li2CO3 group, M2 macrophages and regulatory T cells were induced in the periapical lesion. In situ hybridization experiments revealed a greater expression of Col1a1 in Li2CO3 group compared with the control group. At 24 h after application of intracanal medicament, Axin2-positive cells were distributed in Li2CO3 group. In conclusion, Li2CO3 stimulates Wnt/β-catenin signaling pathway and accelerate the healing process of apical periodontitis, modulating the immune system and the bone metabolism.

Moreover, the clinical use of LiCl was questionable in terms of its safety. In this report, we applied lithium carbonate (Li 2 CO 3 ) into root canals instead of LiCl aiming to confirm the safety of Li + in clinical application and the healing ability of Li + for apical periodontitis. This is because Li 2 CO 3 has already been used as a primary therapeutic agent for bipolar disorder. Furthermore, rats were used in this study to determine if the healing ability of Li + can be also observed in another animal. Because of the larger size of mandibular first molar of rats than that of mice, rubber dam could be used for root canal treatment, enabling treatment sterility. Alternatively, another objective of our study was to analyze the mechanism of the healing ability of Li + because the mechanism remained unclear in our previous study. To elucidate this mechanism, histological analysis was performed on rat mandibular tissue targeting immune cells, osteoblast, and the Wnt/β-catenin signaling pathway.

Results
Verify the safety during application of Li 2 CO 3 into root canal. We applied a 12% Li 2 CO 3 paste into the root canals to verify the safety for periapical tissue. The experimental procedure is shown in Fig. 1a. The hematoxylin and eosin (H&E) staining images showed that there was no difference in the periapical tissues between the control group and the 12% Li 2 CO 3 group (Fig. 1b). In detail, there was no inflammatory cell infiltration around the apical foramen and the area that came into contact with the intracanal medicament in both the control and 12% Li 2 CO 3 groups. In addition, there was no pathological alveolar bone and root resorption in these two groups. Further, to evaluate the systemic effects of the application of 12% Li 2 CO 3 paste into root canals, we monitored the blood concentration of Li + for 72 h. The intraperitoneal group showed the transient increase in the blood concentration of Li + up to around 2 mM at 1 h after the administration (Fig. 1c). After that, the blood concentration of Li + was gradually decreased until 72 h. On the other hand, in cases of the application of 12% Li 2 CO 3 into the root canals, there was no increase in the blood concentration of Li + throughout the observation period. These results demonstrated that applied Li + did not diffuse into the blood (Fig. 1c); therefore, indicating that the Li 2 CO 3 was safe to apply into the root canal. Li 2 CO 3 reduced the volume of periapical lesions. To evaluate the applicability of Li 2 CO 3 for the treatment of apical periodontitis, we applied a 12% Li 2 CO 3 paste into root canals using a rat root canal treatment model. Figure 2a shows the experimental procedure. The results of H&E staining at 28 d after intracanal medication showed that the size of periapical lesions of the 12% Li 2 CO 3 group was much smaller than that of the control group (Fig. 2b,c). In the 12% Li 2 CO 3 group, there was hardly any infiltration of inflammatory cells. However, in the control group, many inflammatory cells were observed inside the periapical lesion. Additionally, some alveolar bone areas in the 12% Li 2 CO 3 group appeared to have undergone healing of the lesions with bone tissue.
Next, we analyzed periapical lesion volumes using micro-CT. Throughout the study, we were able to check whether the intracanal medicament reached the apex of the root canal because of the X-ray contrast given to the base material of the control and 12% Li 2 CO 3 paste (Fig. 2b,c). Periapical lesion volumes were not significantly different between the control and the 12% Li 2 CO 3 groups from 0 (p = 0.932) to 7 d (p = 0.075). In contrast, the periapical lesion volume of the 12% Li 2 CO 3 group was significantly smaller than that of the control group Li 2 CO 3 has the healing ability for apical periodontitis through the stimulation of Wnt/β-catenin signaling pathway. To elucidate the mechanism of the ameliorative effect of Li 2 CO 3 on apical periodontitis, histological experiments were performed on periapical tissues. Though many CD86-positive cells were observed in the control group at 7, 14, and 21 d, there were very few CD86-positive cells in the 12% Li 2 CO 3 group at the same time points (Fig. 3). At 28 d, there were few CD86-positive cells in both the control and 12% Li 2 CO 3 groups. In contrast, many CD68/CD206-double positive cells were detected in the 12% Li 2 CO 3 group compared with the control group from 0 to 28 d (Fig. 4). Furthermore, many Foxp3-positive cells were detected in the 12% Li 2 CO 3 group at 7, 14, and 21 d, while there were very few Foxp3-positive cells in the control group at those time points (Fig. 5a-c). At 28 d, Foxp3-positive cells were observed in both groups (Fig. 5d). In situ hybridization experiments revealed a greater expression of Col1a1 in the 12% Li 2 CO 3 group compared with the control group from 7 to 28 d (Fig. 6). At 24 h after application of intracanal medicament, Axin2-positive cells were distributed in the 12% Li 2 CO 3 group, but not in the control group (Fig. 7).

Discussion
Our recent study demonstrated that a SNP in LRP5 was associated with the development of apical periodontitis 10 . According to the results of the SNP analysis, we hypothesized that the development of apical periodontitis might be related with the Wnt/β-catenin signaling pathway. To clarify the role of this pathway in the development of apical periodontitis, we practiced root canal treatment on mice. In this previous experiment, the application of LiCl, which was reported to regulate this signaling pathway, into the murine root canals promoted the healing of apical periodontitis. This result implied that Li + regulates apical periodontitis development and healing. However, the clinical practices of root canal treatments on mice were difficult because of the teeth size. Even when the root canal treatment was successfully practiced, subject teeth were often broken during follow-up due to their fragility. In the current study, we applied Li 2 CO 3 that released Li + to confirm the role of Li + in the regulation of apical periodontitis development. Considering future clinical applications for humans, we used Li 2 CO 3 , because Li 2 CO 3 has been already used as a primary therapeutic agent for bipolar disorder for a long time 18 . However, because a high concentration of Li + induces side effects, such as lithium toxicity, bradycardia, and renal symptoms, patients who take Li 2 CO 3 need Therapeutic Drug Monitoring of their Li + blood concentrations. To ensure the safety of Li 2 CO 3 application into root canals, we continually monitored the blood Li + concentrations of the rat experimental model. As seen in previous reports 19  www.nature.com/scientificreports/ increases in Li + concentration in the blood (Fig. 1c). Comparatively, there was no increase in Li + concentration in the blood (Fig. 1c). These results implied that patients who were administered Li 2 CO 3 into their root canals would not require Therapeutic Drug Monitoring of blood Li + levels.
To evaluate the healing ability of Li 2 CO 3 for apical periodontitis, we practiced a recently developed rat root canal treatment model [22][23][24] . From H&E staining experiments, the size of the periapical lesion of the 12% Li 2 CO 3 group at 28 d appeared smaller than that of the control group (Fig. 2b,c). To quantify the periapical lesion volume, the volume of radiolucent area around the root apex was analyzed using micro-CT. The periapical lesion volume in the 12% Li 2 CO 3 group was significantly smaller than that in the control group (Fig. 2d). Since the intracanal medicament applied in the control group contained only the base material without medicinal properties, healing of the periapical lesion was dependent on the rats' inherent healing ability due to the removal of bacteria in the root canal. On the other hand, in the 12% Li 2 CO 3 group, in addition to the removal of bacteria, the Li + released from the intracanal medicament could diminish the periapical lesion volume compared to the control group.
Next, we analyzed the mechanism of action of Li 2 CO 3 for improving apical periodontitis with histopathological procedures. The healing of inflammatory diseases, including apical periodontitis, involves anti-inflammatory effects that occur at the inflammation site, such as the polarization of M1 to M2 macrophages and the induction of regulatory T cells 25,26 . Furthermore, previous reports have shown that the Wnt/β-catenin pathway is associated with the orchestration of anti-inflammatory effects [27][28][29][30] . At the early stage of inflammation, M1 macrophages are predominant and are induced by LPS and inflammatory cytokines, such as TNF-α and IFN-γ. M1 macrophages secrete pro-inflammatory cytokines and regulate the differentiation of Th1 and Th17 cells 25 . Whereas M1 macrophages are predominant in the early stage of inflammation, M2 macrophages increase in the middle to late  31,32 . Regulatory T cells have the capacity to downregulate all T cell-mediated immune responses [33][34][35] . Our histological data showed that there were many CD86-positive cells from 7 to 21 d in the control group and very few at 28 d (Fig. 3). CD68/CD206-double positive cells were observed from 28 d in the control group (Fig. 4). These results implied that the control group was in a proinflammatory state from 7 to 21 d and in an anti-inflammatory or healing state after 28 d. In contrast, there were very few CD86-positive cells in the 12% Li 2 CO 3 group at 7 d (Fig. 3). CD68/CD206-double positive cells were observed from 7 to 28 d in the Li 2 CO 3 application group (Fig. 4). These results showed that Li 2 CO 3 induced polarization from M1 to M2 macrophages, and the Li 2 CO 3 group was already in an anti-inflammatory or healing state at 7 d. Furthermore, though regulatory T cells were observed only after 28 d in the control group, they were already observed after 7 d in the 12% Li 2 CO 3 group (Fig. 5). Regulatory T cells induced by Li 2 CO 3 might downregulate immune responses in periapical lesions. Previous studies have reported that regulatory T cells secrete anti-inflammatory cytokines such as IL-10, IL-35 and TGF-β, and transfer the polarization from M1 to M2 macrophages 36,37 . Thus, regulatory T cells induced by Li 2 CO 3 may suppress M1 macrophages' differentiation and enhance M2 macrophages' differentiation at the healing stages of apical periodontitis. Li 2 CO 3 application was considered to activate the Wnt/β-catenin signaling pathway. This activation then induced the suppression of M1 macrophages, and the induction of M2 macrophages and regulatory T cells 15,38 . We employed in situ hybridization for Col1a1 to evaluate osteoblast differentiation in the periapical lesion. In the control group, some level of Col1a1 expression on the alveolar bone surface was observed throughout the entire experiment (Fig. 6). In contrast, in the 12% Li 2 CO 3 group, Col1a1 expression was higher than in the www.nature.com/scientificreports/ control group throughout the experiment. Since Col1a1 is known as a marker gene of osteoblast differentiation, a strong expression of Col1a1 was indicative of accelerated bone healing in the periapical lesions of the 12% Li 2 CO 3 group. According to the above results, regulatory T cells induced by Li 2 CO 3 might support alveolar bone healing 39 . Finally, we confirm whether Li 2 CO 3 activates the Wnt/β-catenin signaling pathway. Finally, we performed the immunohistochemical staining for Axin2 to investigate the behavior of the canonical Wnt/β-catenin signaling pathway. Axin2 is well known as a target-gene product of canonical Wnt signaling. At 24 h after 12% Li 2 CO 3 application into the root canals, sections were stained with an anti-Axin2 antibody, and many Axin2-positive cells were distributed in the periapical lesion of 12% Li 2 CO 3 group (Fig. 7). This result demonstrated that Li 2 CO 3 application induces the stimulation of the Wnt/β-catenin signaling pathway.
In conclusion, the role of Li 2 CO 3 and its mechanism in the promotion of healing responses in apical periodontitis were determined. Li 2 CO 3 application first activates the Wnt/β-catenin signaling pathway. This activation then induces Axin2 to regulate immune responses, such as the suppression of M1 macrophages, and the induction of M2 macrophages and regulatory T cells. Li 2 CO 3 medication may change periapical lesions' inflammatory states to healing states at early stages of inflammation. Moreover, regulatory T cells induced by Li 2 CO 3 may support alveolar bone healing through osteoblast differentiation and reduce the volume of periapical lesions. These results propose that Li 2 CO 3 could be a bioactive medicament in root canal treatments.   Root canal treatment model of rat. Male Wistar rats (10 weeks old) were intraperitoneally anesthetized using Domitor (0.3 mg/kg: Nippon Zenyaku Kogyo Co., Fukushima, Japan), Dolmicam (4 mg/kg: Astellas Pharma Inc., Tokyo, Japan), and Betorphale (5 mg/kg: Meiji Seika Pharma, Tokyo, Japan). Experimental periapical lesion formation was performed following previous reports 40,41 . To induce apical periodontitis, the pulp chambers of the mandibular first molars were opened by a #1/2 round bur equipped with an electric engine (VIVA MATE G 5: NSK, Tochigi, Japan). Next, the root canals were penetrated with a #08 K-file (Dentsply  Germany) and exposed to the oral cavity. At 28 d after pulp exposure, the cleaning of root canals was practiced as described below. The tooth was isolated using a custom-made rubber dam clamp (YDM, Tokyo, Japan) and a rubber dam sheet (Heraeus Kulzer, South Bend, USA). Necrotic coronal pulp and infected dentin were taken away using a #1/2 round bur. Infected dentin in the pulpal floor was taken away with a microexcavator (OK Micro-exca: Seto, Ibaraki, Japan) to avoid perforation. Root canal enlargement was performed to the level of 0.5 as indicated by an electric root canal meter (Root ZX: J Morita, Tokyo, Japan) using K-files (Dentsply Maillefer) up to a #20 file. Root canals were irrigated with 2.5% sodium hypochlorite (Neo Dental Chemical Products, Tokyo, Japan) using 30-gauge needles (NaviTip, Ultradent Products, South Jordan, UT). After root canal enlargement, the root canal was dried with a sterile paper point. Intracanal medicaments, which was prepared as paste filled in syringe, were applied into the mesial root canals under the operating microscope. After treatment with a bonding system (CLEARFIL Universal Bond Quick: Kuraray Noritake Dental, Tokyo, Japan), the pulp chamber was filled with the flowable composite resin (MI FLOW: GC, Tokyo, Japan).

Evaluation of Li 2 CO 3 medication safety.
To verify the safety of Li 2 CO 3 medicament, we modified the rat root canal treatment model described above. Rats without induced apical periodontitis underwent pulpectomy (n = 4). Pulp tissue in coronal area was taken away using a #1/2 round bur. Root canal enlargement was practiced in the same procedure as described above. After root canal irrigation and drying, 12% Li 2 CO 3 was applied into the mesial root canal. 12% was the maximum concentration of Li 2 CO 3 that can be contained while maintaining adequate fluidity of the intracanal medicament. The control group were filled with base material (barium sulfate, aluminum oxide, titanium oxide, purified water) of the medicament. Finally, the coronal cavity was capped with the flowable composite resin. At 28 d after medication, rats were sacrificed for the histological analysis. At least two randomly selected samples from each group were used. Further, to monitor the blood concentration of Li + , rats were classified into two groups. Rats in the first group underwent application of 12% Li 2 CO 3 into their mesial root canals (n = 4). The rats in the second group underwent an intraperitoneal administration of Li 2 CO 3 (74 mg/kg), which was dissolved in saline (n = 4). Peripheral blood was collected from the subclavian vein at 1, 3, 6, 12, 24, 48, and 72 h after intracanal medication, and then centrifuged at 1000 × g for 20 min to collect the serum. Collected serum was stored at − 80 °C until used in experiments. The concentration of Li + in the serum was measured using the quantification reagent (LI01M; Metallogenics, Chiba, Japan) and the microplate reader (Wallac 1420 ARVO MX: PerkinElmer, Waltham, MA, USA). ics, Tokyo, Japan) was used to scan the area around the mandibular first molar. Rats were sacrificed and scanned every week. The imaging conditions were adjusted as follows: 160 μA tube current, 90 kV tube voltage, and 5 μm slice width. The obtained images were analyzed by Simple Viewer software (Science Mechatronics). According to the methods described Yoneda et al. 24 and Kalatizis-Sousa et al. 42 , the volume of periapical lesion was defined as the volume of radiolucent area around the root apex and calculated using the bone morphometrics software (TRI 3D-BON: RATOC, Osaka, Japan). Then, the lesion volume was compared between experimental groups (n = 4 each) as previously reported 22 .
Sample preparation for the histological analysis. After the rats were subjected to the above experiments, they were perfused with 4% paraformaldehyde solution. The mandibles of the rats were collected and immersed in 4% paraformaldehyde solution and fixed for 24-48 h, followed by decalcification with Kalkitox (Fujifilm, Tokyo, Japan) for 14 d with gentle agitation. For H&E staining, the tissues were dehydrated with ascending ethanol series, penetrated with xylene, and finally embedded in paraffin. Samples were then sliced to 7 µm thickness. For immunostaining and in situ hybridization, decalcified mandibular tissues were embedded in O.C.T compound (Sakura Finetek, Tokyo, Japan), frozen at − 80 °C, and sliced to 14 µm thickness. At least two randomly selected samples from each group were used. In situ hybridization. Frozen sections were rinsed with 10 mM PBS at room temperature for 10 min, fixed with 4% PFA for 30 min at 37 °C, and washed again with distilled water. Then, sections were treated with 0.2% HCl for 10 min, followed by the reaction with 1 μg/ml proteinase K (Takara Bio, Shiga, Japan) for 10 min at 37 °C. After rinsing with PBS, sections were washed with G-Wash (Genostaff, Tokyo, Japan). Hybridization was performed in a humid chamber at 50 °C overnight using digoxigenin-labeled RNA probes, Col1a1 (NM_053304, nt3946-4887) diluted with G-Hybo (Genostaff). After the hybridization, the sections were washed with 50% formamide in G-Wash for at least 30 min at 50 °C and rinsed with TBST. Blocking was then performed with G-Block (Genostaff) for 15-30 min. After reaction with alkaline phosphatase (AP)-labeled anti-digoxigenin antibody (1:2000, Roche, Basel, Switzerland) at room temperature for 1 h, the sections were rinsed with TBST, followed by rinse with distilled water. The sections were reacted with BM Purple AP (Roche) as a substrate for 1 h at room temperature, after which they rinsed with PBS and were stained with Nuclear Fast Red solution as a counter staining. The sections were finally sealed with G-Mount (Genostaff). Periapical lesion and alveolar bone were observed under the optical microscope.