Dietary Salt Accelerates Orthodontic Tooth Movement by Increased Osteoclast Activity

Dietary salt uptake and inflammation promote sodium accumulation in tissues, thereby modulating cells like macrophages and fibroblasts. Previous studies showed salt effects on periodontal ligament fibroblasts and on bone metabolism by expression of nuclear factor of activated T-cells-5 (NFAT-5). Here, we investigated the impact of salt and NFAT-5 on osteoclast activity and orthodontic tooth movement (OTM). After treatment of osteoclasts without (NS) or with additional salt (HS), we analyzed gene expression and the release of tartrate-resistant acid phosphatase and calcium phosphate resorption. We kept wild-type mice and mice lacking NFAT-5 in myeloid cells either on a low, normal or high salt diet and inserted an elastic band between the first and second molar to induce OTM. We analyzed the expression of genes involved in bone metabolism, periodontal bone loss, OTM and bone density. Osteoclast activity was increased upon HS treatment. HS promoted periodontal bone loss and OTM and was associated with reduced bone density. Deletion of NFAT-5 led to increased osteoclast activity with NS, whereas we detected impaired OTM in mice. Dietary salt uptake seems to accelerate OTM and induce periodontal bone loss due to reduced bone density, which may be attributed to enhanced osteoclast activity. NFAT-5 influences this reaction to HS, as we detected impaired OTM and osteoclast activity upon deletion.


Introduction
Orthodontic tooth movement (OTM) is based on multicellular processes and is characterized by remodeling of the periodontal ligament and alveolar bone [1] due to the activity of bone-resorbing osteoclasts and bone-forming osteoblasts [2]. In contrast to osteoblasts, which are derived from mesenchymal stem cells [3], osteoclasts evolve from hematopoietic stem cells [4]. Therefore two factors are critically involved: the macrophagecolony-simulating factor (M-CSF) and receptor activator of NF-kB ligand (RANKL), which promotes the differentiation of osteoclast precursor cells to bone-resorbing osteoclasts [5]. The binding of RANKL to the RANK receptor on precursor cells is strictly controlled by the decoy receptor osteoprotegerin (OPG) [6,7]. The periodontal ligament is a fibrous connective tissue, which anchors the teeth in the alveolar bone and contains fibroblasts [1], which are the main cell population, as well as immune cells such as macrophages and T cells [8,9]. Periodontal ligament fibroblasts play a major regulating role during orthodontic tooth movement, not only by secretion of inflammatory cytokines and chemokines, but also by secretion of RANKL and OPG [10][11][12].

Impact of Salt on Osteoclast Activity
First, we investigated the impact of salt (sodium chloride, NaCl) on gene expression of tartrate-resistant alkaline phosphatase 5 (Acp-5) in osteoclasts differentiated from RAW264.7 macrophages. Under high salt conditions (HS; +40 mM NaCl) Acp-5 gene expression was significantly elevated (p < 0.0001; Figure 1a). Accordingly, we measured increased amounts of tartrate-resistant alkaline phosphatase 5 (TRAP) in the cell culture supernatant (p = 0.0062; Figure 1b). To further investigate the activity of osteoclasts under HS conditions, we tested for calcium phosphate (CaP) resorption and detected enhanced resorption with HS (p < 0.0001; Figure 1c), indicating that HS conditions favor enhanced osteoclast activity. and CaP resorption (c) with normal salt (NS) and high salt treatment (HS) in murine osteoclasts differentiated from RAW264.7 macrophages (n ≥ 6). Differentiation of osteoclasts was induced by treatment with M-CSF (30 ng/mL) and RANKL (50 ng/mL) for five days, followed by addition of 40 mM NaCl to the HS group. Symbols represent single data points, horizontal lines the arithmetic mean and vertical lines the standard error of the mean. AU: arbitrary units. Statistics: Two-tailed unpaired t-test. ** p < 0.01; *** p < (NS) and high salt treatment (HS) in murine osteoclasts differentiated from RAW264.7 macrophages (n ≥ 6). Differentiation of osteoclasts was induced by treatment with M-CSF (30 ng/mL) and RANKL (50 ng/mL) for five days, followed by addition of 40 mM NaCl to the HS group. Symbols represent single data points, horizontal lines the arithmetic mean and vertical lines the standard error of the mean. AU: arbitrary units. Statistics: Two-tailed unpaired t-test. ** p < 0.01; *** p < 0.001.

Impact of NaCl-Containing Diets on Expression of Genes Involved in Bone Remodelling
To investigate the role of salt-containing diets on orthodontic tooth movement (OTM), we fed mice either a low salt diet (LSD), a normal salt diet (NSD) or a high salt diet and induced orthodontic tooth movement by insertion of an elastic band. First, we tested the impact of different salt contents and OTM on the expression of genes involved in bone remodeling processes. We detected no effects of OTM (LSD: p = 0.9999; NSD: p = 0.9612; HSD: p = 0.6282) on bone-mineralization-associated alkaline phosphatase (Alpl; Figure 2a). Different nutritional salt contents also had no effect on Alpl gene expression compared to a normal salt diet. In line with this, we detected no changes in gene expression of runt-related transcription factor 2 (Runx-2; LSD: p = 0.8280; NSD: p = 0.7515; HSD: p > 0.9999), which is a key transcription factor for osteoblast differentiation, induced by OTM. Salt treatment did not affect Runx-2 gene expression in the periodontal ligament of mice ( Figure 2b). Prostaglandin endoperoxide synthase 2 (Ptgs-2) is involved in the synthesis of prostaglandin E2, which affects bone remodeling. We detected no effect of OTM on Ptgs-2 gene expression upon LSD treatment (p = 0.9634), whereas orthodontic treatment significantly increased Ptgs-2 gene expression under NSD (p = 0.0080) and HSD (p = 0.0019) conditions (Figure 2c). HSD elevated Ptgs-2 gene expression with OTM compared to LSD (p = 0.0104) and NSD (p = 0.0143). Osteoprotegerin (Opg) acts as receptor activator of NF-kB ligand (Rankl) decoy receptor and is thereby critically involved in osteoclastogenesis [6,7]. We detected no effects of orthodontic treatment (LSD: p > 0.9999; NSD: p = 0.9983; HSD: p = 0.9626) or the different salt-containing diets on Opg gene expression (Figure 2e). In contrast, Rankl gene expression was elevated due to OTM (LSD: p = 0.0134; NSD: p = 0.0493; HSD: p < 0.0001) in all tested diets. HSD further increased Rankl gene expression compared to LSD (p = 0.0001) and NSD (p = 0.0015; Figure 2e), indicating increased osteoclastogenesis upon HSD. Accordingly, we detected enhanced Acp-5 gene expression with NSD (p = 0.0041) and HSD (p < 0.0001) treatment, whereas this was not detectable with LSD (p = 0.7711; Rankl (e) and Acp-5 (f) under low salt (LSD), normal salt (NSD) or high salt diet (HSD) in dental-periodontal tissue at the first upper molar, assessed with RT-qPCR (n = 8). Symbols represent single data points, horizontal lines the arithmetic mean and vertical lines the standard error of the mean. AU: arbitrary units, OTM: orthodontic tooth movement, LSD: low salt diet, NSD: normal salt diet, HSD: high salt diet. Statistics: ANOVA followed by Holm-Sidak multiple comparison tests. * p < 0.05; ** p < 0.01; *** p < 0.001.

Effects of Different Salt Diets on Periodontal Bone Loss, Orthodontic Tooth Movement and Alveolar Bone Density
Next, we investigated the impact of salt-containing diets on periodontal bone loss. We measured significantly enhanced periodontal bone loss due to OTM upon NSD (p = 0.0346) and HSD (p < 0.0001), but not with LSD (p = 0.8900; Figure 3a). NSD (p = 0.0308) and HSD (p < 0.0001) further elevated periodontal bone loss induced by orthodontic treatment compared to LSD (Figure 3a). OTM was determined by measuring the distance between the orthodontically moved first (M1) and the second (M2) upper left molar. With all diets we detected an increased distance between M1 and M2 after OTM (LSD: p = 0.0177; NSD: p = 0.0003; HSD: p < 0.0001; Figure 3b). HSD potentiated the extent of OTM compared to LSD significantly (p = 0.0179), leading to accelerated tooth movement ( Figure 3b). To investigate the cause of this accelerated tooth movement, we investigated bone density between molars. We detected no significant effects of orthodontic treatment on bone density (LSD: p = 0.5196; NSD: p = 0.5196; HSD: p = 0.4724). HSD, however, reduced bone density compared to LSD with OTM treatment (p = 0.0251; Figure 3c), which could result in accelerated tooth movement.

Effects of Different Salt Diets on Periodontal Bone Loss, Orthodontic Tooth Movement and Alveolar Bone Density
Next, we investigated the impact of salt-containing diets on periodontal bone loss. We measured significantly enhanced periodontal bone loss due to OTM upon NSD (p = 0.0346) and HSD (p < 0.0001), but not with LSD (p = 0.8900; Figure 3a). NSD (p = 0.0308) and HSD (p < 0.0001) further elevated periodontal bone loss induced by orthodontic treatment compared to LSD (Figure 3a). OTM was determined by measuring the distance between the orthodontically moved first (M1) and the second (M2) upper left molar. With all diets we detected an increased distance between M1 and M2 after OTM (LSD: p = 0.0177; NSD: p = 0.0003; HSD: p < 0.0001; Figure 3b). HSD potentiated the extent of OTM compared to LSD significantly (p = 0.0179), leading to accelerated tooth movement ( Figure  3b). To investigate the cause of this accelerated tooth movement, we investigated bone density between molars. We detected no significant effects of orthodontic treatment on bone density (LSD: p = 0.5196; NSD: p = 0.5196; HSD: p = 0.4724). HSD, however, reduced bone density compared to LSD with OTM treatment (p = 0.0251; Figure 3c), which could result in accelerated tooth movement.

Impact of the Osmoprotective Transcription Factor NFAT-5 on Osteoclast Activity
The transcription factor NFAT-5 (nuclear factor of activated T cells 5) plays an important role in the osmoprotective adaption of cells and tissues to high salt conditions. Therefore, we investigated the role of NFAT-5 on osteoclast activity during salt treatment. Osteoclasts were differentiated from bone-marrow-derived monocytes (BMMs) derived

Impact of the Osmoprotective Transcription Factor NFAT-5 on Osteoclast Activity
The transcription factor NFAT-5 (nuclear factor of activated T cells 5) plays an important role in the osmoprotective adaption of cells and tissues to high salt conditions. Therefore, we investigated the role of NFAT-5 on osteoclast activity during salt treatment. Osteoclasts were differentiated from bone-marrow-derived monocytes (BMMs) derived from Nfat-5 ∆myel and control mice (WT). We determined increased NFAT-5 protein expression under HS conditions in osteoclasts from WT mice, whereas this effect was abolished in osteoclasts from Nfat-5 ∆myel mice (Figure 4a).
from Nfat-5 Δmyel and control mice (WT). We determined increased NFAT-5 protein expression under HS conditions in osteoclasts from WT mice, whereas this effect was abolished in osteoclasts from Nfat-5 Δmyel mice ( Figure 4a).

Discussion
In this study we observed increased osteoclast resorption activity upon salt treatment in vitro. In a murine model of tooth movement, increased Rankl and Acp-5 gene expression were associated with reduced bone density, elevated periodontal bone loss and acceleration of orthodontically-induced tooth movement upon HSD. Furthermore, the underlying increased osteoclast activity was associated with osmoprotective transcription factor

Discussion
In this study we observed increased osteoclast resorption activity upon salt treatment in vitro. In a murine model of tooth movement, increased Rankl and Acp-5 gene expression were associated with reduced bone density, elevated periodontal bone loss and acceleration of orthodontically-induced tooth movement upon HSD. Furthermore, the underlying increased osteoclast activity was associated with osmoprotective transcription factor NFAT-5, as deletion in myeloid-derived cells resulted in enhanced osteoclast activity under NS conditions, whereas an induction in control mice failed under HS conditions. In vivo this impairment mainly manifested itself in the form of reduced Rankl and Acp-5 gene expression and reduced orthodontic tooth movement (OTM) under HSD conditions in Nfat-5 ∆myel mice.
Salt-containing diets and other environmental challenges like inflammation have been associated with impaired sodium distributions in the body [16,20,26], promoting the reorganization of body metabolism [26,27]. Salt may affect osteoclast activity directly, as pumps acting as sodium exchangers, controlling the functionality of the hemi-vacuole involved in bone resorption [28][29][30]. In cell culture, however, the addition of sodium chloride is strictly associated with increased osmolality in the cell culture supernatant. Therefore, at this point of investigation, there is a possibility that the observed sodiuminduced results could be due to changes in osmolality, which requires further research in this area.
One key regulator of OTM is cells of the periodontal ligament, such as fibroblasts or immune cells, which regulate the extent of bone remodeling by producing inflammatory cytokines and proteins affecting osteoclastogenesis, thereby modulating OTM [2,10]. It is known that periodontal ligament fibroblasts react to additional salt with increased prostaglandin E2 and RANKL expression [18,31], which could explain our results concerning Rankl gene expression at the orthodontically treated jaw side under HSD in the murine model. Furthermore, it has been reported that salt affects the expression of genes involved in extracellular matrix remodeling and inflammatory responses and thereby influences the reorganization of the periodontal ligament [18].
Previous studies investigating the role of salt on osteoclastogenesis of murine osteoclast progenitor cells revealed a tremendous impact, as the addition of 40 mM salt prevented the differentiation of osteoclast progenitor cells to osteoclasts, whereas lower concentrations were reported to promote osteoclastogenesis [19,32]. In line with our data concerning the alveolar bone, bone density in the tibia was reduced with HSD in control mice, whereas this effect was not observed in Nfat-5 ∆myel mice [19]. The reduced bone density was caused by an increased number of osteoclasts, whereas the number of osteoblasts did not change with the diet [19]. The osmoprotective transcription factor NFAT-5 was shown to control the expression of bone-protective OPG in myeloid cells and osteoblasts [19]. However, we detected no changes in Opg gene expression in the periodontal ligament of Nfat-5 ∆myel mice.
Our data strongly indicate that salt-containing diets accelerate orthodontic tooth movement (OTM) and promote periodontal bone loss due to reduced bone density, which may be attributed to enhanced osteoclast activity. NFAT-5 seems to influence this reaction to HS, as we detected impaired OTM and osteoclast activity upon deletion. Dietary salt intake may affect the velocity of orthodontic tooth movement in patients. As according to the WHO, salt intake in Western societies is higher than recommended, and patients with high salt intake are expected to show increased osteoclastogenesis and tooth movement velocity. As periodontal bone loss has been observed to be a side effect of high-salt-containing diets, a reduction of salt intake during orthodontic treatment is recommendable in clinical practice based on the available data.

RNA Isolation from Cell Culture
We used 500 µL TriFast  The quantification was carried out in the Mastercycler ® realplex (Eppendorf, Hamburg, Germany). All primers were designed in accordance with the MIQE quality guidelines [34]. For normalization of the target genes, we used a combination of two validated reference genes ( Table 1). The relative gene expression was calculated with as 2 −∆Cq [35], with ∆Cq = Cq (target gene)-Cq (mean reference genes).

Tratrate-Resistant Acid Phosphatase (TRAP) Assay
The TRAP assay was performed with cell culture supernatants using a TRAP staining kit (PMC-AK04F-COS, Cosmo Bio, Tokyo, Japan), following the manufacturer's instructions, and staining was quantified at 540 nm with an ELISA reader after 3 h at 37 • C.  The plates were washed twice with sterile water (L0015; Biochrom, Berlin, Germany) and dried. To determine the resorption activity of calcium phosphate, the supernatant was removed and cells were washed with 1 mL prewarmed PBS twice, following addition of 200 µL of a 1 M NaCl (3957.1, Carl Roth, Karlsruhe, Germany), mixed with 0.2% Triton (T9284, Sigma Aldrich, St. Louis, MO, USA) for 10 min. The coated wells were washed twice with water to remove cells from the surface. This was followed by the addition of 200 µL 5% AgNO 3 (7908.1, Carl Roth, Karlsruhe, Germany) per well. The coated wells were illuminated with UV light at room temperature for 45 min and washed with water. They were photographed using a light microscope and resorption lacunae were evaluated using ImageJ software (ver. 1.47, Wayne Rasband, National Institutes of Health, Bethesda, MD, USA).

Micro-Computed Tomography (µCT)-Analysis
The upper jaw samples were kept in 5% formalin for 24 h, then formalin was diluted to 0.1% until µCT measurements. Measurements were performed with the Phoenix vltomelxs 240/180 device (GE Healthcare, Chicago, IL, USA), using the 180 kV NF tube under the following settings: voxel size: 10 µm, images: 1800, timing: 333 ms, voltage: 50 kV, current: 750 µA, Fastscan: Scan time 10 min. Image reconstruction and evaluation were performed using the software VG Studio Max (Volume Graphics, Heidelberg, Germany). A two-dimensional plane was selected for the various measurements running through the middle of the first molar at the level of the enamel-cementum boundary and aligned perpendicular to the palate and the chewing plane to enable reproducible measurements. Periodontal bone loss around the first molars was determined orally. For this purpose, the distance between the enamel-cementum boundary and the bony limbus alveolaris on the treated and the control side was measured. To determine orthodontic tooth movement (OTM), we measured the distance between the moved first (M1) and the second (M2) upper left molar. We determined the smallest distance between the crowns of M1 and M2 using the caliper function of the software. The measurement was carried out both on the orthodontically treated side and on the control side. To determine alveolar bone density, an interradicular, cube-shaped (edge length 0.35 mm) region of interest (ROI) was analyzed using the morphometric function of the software. Care was taken to ensure that this ROI did not reach into or overlap the periodontal ligament or the tooth roots.

Statistical Analysis
Normal distribution was tested with Shapiro-Wilk tests. Comparing two groups, Student's t-tests were performed. Analyzing more than two groups, either ordinary ANOVAs followed by the Holm-Sidak multiple comparison tests or Welch-corrected ANOVAs followed by Games-Howell multiple comparison tests were performed. The distance between M1 and M2 was analyzed using the Kruskal-Wallis test. Differences were considered significant at p < 0.05. Statistical analysis was performed with GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA).