Effects of sodium chloride on the gene expression profile of periodontal ligament fibroblasts during tensile strain

During orthodontic tooth movement, pressure and tension zones develop in the periodontal ligament, and periodontal ligament fibroblasts (PDLF) become exposed to mechanical strain. Enhanced salt (NaCl) concentrations are known to modulate responses of PDLF and immune cells to different stimuli like mechanical strain. Here, we investigated the impact of tensile strain on the gene expression profile of PDLF under normal (NS) and high salt (HS) conditions. After preincubation under NS or HS (+40 mM NaCl in medium) conditions for 24 h, PDLF were stretched 16% for 48 h using custom-made spherical cap silicone stamps using an established and published setup. After determination of cell number and cytotoxicity, we analyzed expression of genes involved in extracellular matrix reorganization, angiogenesis, bone remodeling, and inflammation by quantitative real-time polymerase chain reaction (RT-qPCR). Tensile strain did not affect the expression of genes involved in angiogenesis or extracellular matrix reorganization by PDLF, which however modulate inflammatory responses and bone remodeling in reaction to 16% static tensile strain. Salt (NaCl) treatment triggered enhanced extracellular matrix formation, expression of cyclooxygenase 2 and bone metabolism in PDLF during tensile strain. Salt (NaCl) consumption may influence orthodontic tooth movement and periodontal bone loss via modulation of extracellular matrix and bone metabolism. Excessive salt intake during orthodontic therapy may cause adverse effects regarding periodontal inflammation and bone resorption.


Introduction
During orthodontic tooth movement, pressure and tension zones develop in the periodontal ligament (PDL), a connective tissue responsible for the attachment of the tooth to the alveolar bone [25]. Periodontal ligament fibroblasts (PDLF) make up the majority of cells within the periodontal ligament and are the first cells exposed to mechanical stimuli occurring during orthodontic tooth movement. The main function of PDLF is the maintenance of tissue homeostasis and production of collagenous structural proteins and glycosaminoglycans. Furthermore, they sustain regulatory functions in innate immune defense [16,25].
After an orthodontic force is applied to the tooth, PDLF are subjected to mechanical strain and play an important regulatory role in orthodontic tooth movement. As an early response to compressive forces, PDLF increase prostaglandin production via enhancement of cyclooxygenase 2 (COX-2) activity, resulting in enhanced expression of RANKL (receptor activator of NF-κB ligand) [16,38,48], while simultaneously secretion of RANKL decoy receptor osteoprotegerin (OPG) is reduced [38]. This promotes osteoclastogenesis in pressure areas of the PDL resulting in alveolar bone resorption. In contrast, in tension areas bone formation by osteoblastic activity is promoted [25].
Malocclusions have a negative impact on oral-health-related quality of life and mental well-being of children leading to reduced self-confidence [8,42]. Often these psychological effects triggered by malocclusions are the reason to improve esthetics through orthodontic treatment [42]. Therapy of malocclusions is of distinct medical importance, as recent studies associated malpositioned teeth with the development of caries or periodontitis [2,37,41]. Therefore, orthodontic corrections are helpful to prevent the development and progression of these oral diseases. Despite the importance of orthodontic treatment for oral health, many aspects of orthodontic therapy remain uninvestigated, and many problems that arise during an ongoing orthodontic therapy are still unresolved.
Possible influences of nutrition on orthodontic treatment have hardly been investigated. Since orthodontic tooth movement comprises a local sterile-inflammatory process, numerous possibilities exist for the immune system and general metabolism to modulate these processes [18,53]. Components of nutrition are reported to influence not only chronic diseases like hypertension [22,23] or osteopenia [44], but also have an impact on the oral microflora [52] and periodontal bone loss [29]. In Western societies electrolytes such as sodium are consumed to a high degree as food supplements in the form of salt (sodium chloride, NaCl) and are known to modulate tissue response to different stimuli by their local tissue concentration [28]. It can be assumed that about 70-80% of the salt intake comes from "hidden" salts in processed foods such as cheese, bread, and readyto-eat meals [6]. Most people in Europe consume significantly more salt (8-11 g/day) than recommended by the German Nutrition Society (1.5 g/day) [6,51]. Therefore, it can be assumed that salt-related effects on orthodontic tooth movement in industrialized nations are relevant to the vast majority of patients.
Consumption of high salt diet or inflammation induces Na + accumulation in various tissues, thereby, modulating the activity of cells of the mononuclear phagocytic system [3,14,22,23]. Furthermore, Na + accumulation may also occur in tissues of the oral cavity, especially the gingiva, the oral mucous membrane, and the periodontal ligament [39]. As salt (sodium chloride) is known to have a direct impact on the activity of cells of the mononuclear phagocytic system [14,23,32] and osteoclasts [54], it may also have an effect on the response of PDLF to tensile strain.

In vitro cell culture experiments
All experiments were performed in accordance with the relevant guidelines and regulations. Approval for the collection and use of PDLF was obtained from the ethics committee of the University of Regensburg, Germany (approval  number 12-170-0150). We obtained informed consent from all patients or their legal guardians. Periodontal ligament fibroblasts (PDLF) were isolated from periodontal ligament tissue scraped off the middle third of human teeth. For these experiments we used a pool of PDLF from six different patients (3 male, 3 female, age range 17-27 years). Isolation and characterization was performed as previously described [17,38]. For our experiments, we used PDLF of the 3rd to 6th passage. We determined cell number with a Beckman Coulter Counter Z2™ (Beckman Coulter GmbH, Krefeld, Germany) according to the manufacturer's instructions.

Assessment of cell number
After 48 h of isotropic tensile strain, the medium was removed and PDLF were scraped off the bioflex membrane in 1 ml phosphate buffered saline (PBS) for cell number determination. Cells were counted automatically using a Beckman Coulter Counter Z2™ (Beckman Coulter GmbH, Krefeld, Germany) according to the manufacturer's instructions.

LDH cytotoxicity assay
To test for cytotoxicity, we performed lactate dehydrogenase (LDH) assays (04744926001, Roche, Mannheim, Germany) using cell culture supernatants according to the manufacturer's instructions.  [30]. Relative gene expression was calculated as 2 -Cq with Cq = Cq (target gene) -Cq (geometric mean TBP/PPIB) [21]. All primers (Eurofins MWG, Huntsville, AL, USA) are listed in Table 1 and were constructed using NCBI PrimerBLAST according to MIQE guidelines as previously described [17,19,30]. For each primer pair and RT-qPCR run, a no-templatecontrol without cDNA was performed. Experiments were repeated two to three times (N = 2-3) with at least two to three biological replicates (n = 2-3).

Statistical methods
One-way analysis of variance (ANOVA) followed by Games-Howell post hoc tests was performed using Graph-Pad Prism version 8.0.0 for Windows (GraphPad Software, San Diego, CA, USA). Welch tests were performed in case of heterogeneity of variance. The significance level was set at p ≤ 0.05.

Impact of tensile strain and sodium chloride on cell number and cytotoxicity
First, we assessed the PDLF number and possible cytotoxic effects after 48 h of tensile strain and 72 h of sodium chloride treatment. Tensile strain significantly increased cell numbers under normal salt treatment (NS, p = 0.035), while high salt treatment (HS; p = 0.036) with an additional 40 mM NaCl in the medium had no effect on cell number (Fig. 1b). Neither tension treatment nor addition of 40 mM NaCl showed any cytotoxic effects as no significantly increased LDH activity could be detected (p = 0.889) in the cell culture supernatant throughout the tested conditions (Fig. 1c).

Effects of tensile strain and sodium chloride on extracellular-matrix-forming genes
Next

Effects of tensile strain and sodium chloride on angiogenetic and bone-forming genes
The vascular endothelial growth factor A (VEGFA) is involved in tissue neoformation and responsible for growth of blood vessels. Surprisingly, neither tensile strain (NS: p = 0.170, HS: p = 0.262) nor addition of NaCl (without tension: p = 0.972, tension: p = 0.979) showed a significant effect on gene expression of VEGFA (Fig. 3a), although an increase of VEGFA gene expression under tensile strain was observed by tendency. The alkaline phosphatase (ALP) gene is involved in bone formation. We observed a significant increase in ALP gene expression after tensile strain under NS conditions (p = 0.002, Fig. 3b). Treatment of PDLF with 40 mM NaCl also increased ALP gene expression sig- nificantly (p < 0.001). However, we no longer detected an increased ALP gene expression due to tensile strain under HS conditions (p = 0.312, Fig. 3b).

Effects of tensile strain and HS on proinflammatory genes
Next we focused on the gene expression of proinflammatory genes like cyclooxygenase 2 (COX-2) and interleukin 6 (IL-6). Tensile strain resulted in a significant increase of COX-2 gene expression (p = 0.017, Fig. 4a), whereas IL-6 gene expression was significantly reduced (p = 0.019, Fig. 4b) under NS conditions. Salt treatment enhanced COX-2 gene expression (p < 0.001, Fig. 4a) and also reduced IL-6 gene expression (p = 0.031, Fig. 4b) compared to NS conditions.

Effects of cell stretching and salt treatment on the RANKL/OPG ratio
The RANKL/OPG ratio plays a major regulating role during bone resorption. Tensile strain reduced gene expression of RANKL significantly (p = 0.008, Fig. 5a), whereas OPG gene expression was not affected by PDLF stretching (p = 0.355, Fig. 5b). This resulted in a significantly reduced RANKL/OPG ratio (p = 0.044, Fig. 5c) during tensile strain under NS conditions. HS conditions increased RANKL gene expression without (p = 0.023) and with additional tensile strain (p < 0.001). OPG gene expression was also increased under HS conditions without tensile strain (p = 0.042, Fig. 5b), whereas OPG expression was not affected under HS conditions with tension treatment (p = 0.815). This led to a significantly increased RANKL/OPG ratio under HS conditions during tensile strain compared to NS conditions with cell stretching (p = 0.033, Fig. 5c).

Discussion
In this study, we investigated the effects of tensile strain and salt (sodium chloride) on the expression levels of genes involved in extracellular matrix reorganization, angiogenesis, bone remodeling, and inflammation in PDLF. We could show that application of tension resulted in a reduced RANKL/OPG ratio, which was accompanied by enhanced ALP gene expression indicating elevated bone formation. Tensile strain increased COX-2, but concurrently reduced IL-6 gene expression. Surprisingly, we detected no effects of tension on genes involved in extracellular matrix remodeling or angiogenesis, whereas salt had a significant impact. During orthodontic tooth movement, PDLF are suspected to be involved in extracellular matrix remodeling especially in the formation and breakdown of collagen fibrils [25]. For that reason, we analyzed expression of genes involved in collagen formation (COL1A2, P4HA1) and degradation (MMP8), but also fibronectin 1 (FN1), which interacts with collagen and other molecules of the extracellular matrix like heparin sulfate and serves as an adhesion molecule [36]. Furthermore, FN1 has already been associated before with extracellular matrix remodeling during orthodontic tooth movement [1]. We detected no changes in expression of genes involved in collagen synthesis after application of tension. Howard et al. reported increased FN1 expression after 10% cyclic tensile strain; however, COL1 expression was not changed by PDLF stretching [11], supporting our data. A recent study investigated collagen and fibronectin expression in histological samples. They observed a downregulation of FN1, while COL1 was upregulated at the tension side [26]. Compressive force treatment was reported to affect collagen formation within 24 h because of increased gene expression of COL1A2 and P4HA1 [38]. In contrast to genes involved in collagen formation, FN1 was not affected by compressive force treatment [38]. MMPs are proteolytic enzymes, which degenerate different components of the extracellular matrix [4]. In this study we investigated the gene expression of MMP8, which acts as collagenase and is expressed by PDLF [46]. Contrary to our results it was reported that collagenases like MMP8 were upregulated by tensile forces with the strength of the tensile strain playing a crucial role [5,13,31], whereas MMP8 was downregulated after compressive force treatment in PDLF [39,43]. In the current study, salt (sodium chloride) treatment affected gene expression of COL1A2, FN1 and MMP8 in PDLF, as reported previously [39]. Salt consumption has already been reported to be involved in extracellular matrix reorganization, as it impacts on glycosaminoglycan sulfatation [45,49].
Application of orthodontic forces changes the blood flow in the surrounding tissue. To avoid hypoxic conditions, vascular endothelial growth factor A (VEGFA) expression is induced in the periodontal ligament (PDL) due to mechanical strain. VEGFA is a growth factor involved in the reshaping of blood vessels and angiogenesis [7]. Increased VEGFA expression was reported at compression and tension areas of the PDL in histological samples after tooth movement [27]. In this study, however, we detected no significant effect of tensile strain or salt on VEGFA gene expression. As increased VEGFA gene expression, however, occurred quite early after the onset of mechanical strain [38], our timing of detection might have been too late, as we analyzed VEGFA gene expression not earlier than after 48 h of tensile strain.
According to the common pressure-tension theory during orthodontic tooth movement, bone resorption happens at the pressure areas, while bone formation occurs at tension areas of the PDL [25]. Alkaline phosphatase (ALP) activity is elevated in the periodontal ligament compared to other connective tissues and is associated with bone formation [10]. In line with our data, static and cyclic tensile strain increased ALP expression dependent on the applied magnitude of tensile strain [12,30,31,50], which may enhance the osteoblastic phenotype of PDLF and prompt bone formation [12,55]. Furthermore, increased ALP levels were observed in human crevicular fluid after orthodontic treatment [15,24,35]. Salt treatment enhanced ALP gene expression, suggesting that NaCl promotes an osteoblastic phenotype of PDLF.
PDLF modulate the expression of proinflammatory genes in reaction to orthodontic forces [16,30,38]. In this study gene expression of COX-2 was increased after 48 h of tensile strain. This was in line with Shimizu et al. who reported that enhanced COX-2 expression was accompanied by increased PG-E2 levels in PDLF after stimulation with cyclic stretching [40]. As already reported, salt treatment of PDLF also enhanced COX-2 gene expression [39]. As it is well established that NaCl increases expression of the osmoprotective transcription factor NFAT5 [23,33] and that COX-2 is an NFAT5 target gene [9], this is not by surprise. In contrast to COX-2, gene expression of IL-6 was reduced with tension treatment and NaCl addition. This is in line with prior publications reporting simultaneous increase of COX-2 and reduction of IL-6 expression upon tensile strain [30,39]. IL-6 modulates the extent of immune responses during inflammation [34] and can influence osteoclastogenesis [20]. Reduction of IL-6 expression after stretching and NaCl could contribute to the osteoblastic phenotype of PDLF.
Bone metabolism strongly depends on the interaction of RANKL (receptor activator of NF-κB ligand) and OPG (osteoprotegerin) [47]. While binding of RANKL to the RANK receptor on osteoclast precursor cells is critical for osteoclast formation and activation, secretion of the decoy receptor OPG inhibits this interaction [47]. In contrast to pressure application [38,39], tensile strain resulted in reduced RANKL expression, while OPG gene expression remained unaffected. This resulted in a shifting of the RANKL/OPG ratio towards OPG suggesting less bone resorption. As already reported, salt treatment increased gene expression of RANKL and OPG in PDLF without tensile strain [39]. In contrast to the normal salt-treated PDLFs, we observed a reduction of OPG gene expression under high salt treatment with additional tensile strain, resulting in an increased RANKL/OPG ratio upon salt treatment with stretching. Therefore salt, that is sodium chloride, may modulate bone metabolism at the tension site as well.
For our in vitro experiments, we used salt concentrations (40 mM) corresponding to the local Na + accumulation measured under high salt diet in the murine mandible including the associated mucosa [39] to maximize transferability of results to the in vivo situation within the PDL and surround-ing alveolar bone. The addition of 40 mM NaCl to the cell culture medium did have an impact on the expression of genes involved in extracellular matrix and bone remodeling as well as prostaglandin synthesis supporting previous results [39]. A high salt environment in combination with force application affected the RANKL/OPG ratio under tensile strain as well as during compressive force treatment, indicating a stimulating role of salt on osteoclastogenesis and thereby bone resorption [39]. To further investigate the role of salt on orthodontic tooth movement, in vivo experiments with animals receiving low, normal, and high salt diets with and without orthodontic tooth movement are required. Based on our in vitro results, we surmise that increased sodium concentrations due to high salt intake or possibly a local therapeutic injection into the periodontal ligament may accelerate orthodontic tooth movement due to an increase in osteoclastogenesis in pressure areas as well as elevated osteoblastic activity in tensile areas, but this might also have detrimental effects such as periodontal bone loss or dental root resorptions, which needs to be clarified in further in vivo studies.

Conclusions
Salt (NaCl) treatment has an impact on extracellular matrix formation, expression of proinflammatory cytokines and bone metabolism during tensile strain in PDLF (periodontal ligament fibroblasts). Additional NaCl exposure increased ALP (alkaline phosphatase) expression by PDLF and could thereby promote bone formation at tension areas of the PDL (periodontal ligament). Excessive salt intake during orthodontic therapy may cause stimulatory effects on periodontal inflammation and bone resorption, possibly leading to increased tooth movement, but also periodontal bone loss and dental root resorptions. Tensile strain did not affect expression of genes involved in angiogenesis or extracellular matrix reorganization in PDLFs. PDLFs modulate inflammatory responses and bone remodeling in reaction to static tensile strain.