CRT Silencing Inhibits ECM Synthesis in HGFs Cultured on Three-Dimensional PLGA Scaffolds Via Inhibiting The CaN/NFAT3 Signalling Pathway

The stability of orthodontic treatment is believed to be greatly affected by the compression and contraction of gum tissue, but the underlying molecular mechanism remains unclear. The aim of the current study was to explore the effects of mechanical forces on CRT, CaN, NFAT3, p-NFAT3 and COL-I expression in human gingival broblasts (HGFs) cultured on three-dimensional (3D) poly(lactic-co-glycolic acid) (PLGA) scaffolds. A mechanical force of 25 g/cm 2 was applied to HGFs for 0, 6, 24, 48, and 72h. The expression of CRT, CaN, NFAT3, p-NFAT3, and COL-I were examined by reverse transcription-quantitative polymerase chain reaction and western blotting analysis. The application of mechanical force on HGFs cultured on the 3D PLGA scaffolds led to a signicant increase in CRT, CaN, and COL ‐ I expression, as well as a reduction in p-NFAT3 expression. The mechanical force effects were reversed by silencing CRT. By lowering the CRT expression, p-NFAT3 was upregulated, while CaN and COL ‐ I were downregulated in HGFs. These ndings suggested that downregulation of CRT inhibited extracellular matrix synthesis, potentially via CaN/NFAT3 signalling pathway. Therefore, CRT may serve as a potential therapeutic target against gingival brosis. which can synchronously imitate the inuences of mechanical force and hypoxic environment, can afford a 3D HGFs culture model that is more analogous to the tissue-like environment of natural ECM 4 . In the current study, we aimed to establish a 3D co-culture model of HGFs and PLGA scaffolds to explore the role of CRT, CaN, NFAT3, and phosphorylated NFAT3 (p-NFAT3) in HGFs mechanotransduction and to investigate the role of CRT in ECM deposition in HGFs. Our research offers novel insights into the mechanism of gingival reconstruction during orthodontic tooth movement.


Introduction
During orthodontic treatment, we apply mechanical forces to orthodontic appliances, which are transmitted to the periodontal tissues such as the alveolar bone, gingiva, periodontal membrane to move the teeth 1,2 . In contrast to alveolar bone and periodontal ligament, which can easily be assimilated, gingival tissue generally tends to accumulate after correction of twisted teeth rotation, buccal and labial tooth movement, and closing of the extraction space. The main reason for the accumulation of gum tissue is contraction and compression. This kind of gum tissue, like compressive rubber, has a tendency to relapse to the pretreatment position, which may cause the gaps to reappear at the extraction site or the teeth to return to their original position 3 . It has been reported that orthodontic force can disrupt the balance between collagen synthesis and degradation of gingival tissue, which is very important for maintaining homeostasis, and also increases the multiplication of human gingival broblasts (HGFs) and the synthesis of type I collagen (COL-I), which is the main structural component in the extracellular matrix (ECM) of the gingival tissue 4,5 . Such persistent ECM deposition in the gum tissue is the main driver of recurrence after orthodontic treatment 6 . However, the mechanisms by which HGFs perceive and respond to orthodontic forces, that result in ECM deposition are still unclear.
Calreticulin CRT , located in the endoplasmic reticulum (ER) lumen is a 46-kDa protein that regulates cellular reactions to stress through its roles in its chaperone activation and the unfolded protein reaction 7,8 . Recent literature suggests that CRT adjusts the transcription of multifarious ECM proteins in calcium-dependent manners and plays a posttranscriptional role in collagen transport and matrix assembling 9,10 . Interestingly, CRT expression was shown to be overexpressed in many diverse brosis models, including bleomycin-induced pulmonary brosis, unilateral ureteral obstruction renal brosis model, and chronic brosis proliferative diseases [11][12][13] . Although the mechanisms was not yet clear, heart-speci c upregulation of CRT during upgrowth led to interstitial brosis 14 . The expression level of CRT was enhanced by factors which were well-known to cause brotic diseases and ER stress, including oxidative stress, pneumatorexis, and hyperglycaemia 11,13,15,16 . Therefore, we speculated that CRT might play a role in gingival brosis under mechanical force, although the role and mechanisms of CRT in the occurrence and development of gingival brosis still remain unknown.
The calcineurin(CaN)/nuclear factor of activated T cells 3 (NFAT3) signalling is an important intracellular signalling pathway that can participate in the process of cell brosis by mediating a variety of cytokines [17][18][19] . Calreticulin is an essential protein serine/threonine phosphatase that is regulated by calcium ion/calmodulin (CaM) 20 . NFAT3 is the main substrate of CaN and is an important nuclear transcription factor that mediates intracellular signal transduction 21,22,23 . Additionally, CRT plays an important role in the regulation of downstream Ca 2+ -dependent CaN/NFAT signalling pathway 8 . In the absence of calreticulin, the release of ER Ca2+ is compromised, results in downstream Ca2+/CaN signal transduction defects, as well as decreasing NFAT and myocyte enhancer factor 2C(MEF2C) nuclear translocation 24,25 . Calreticulin-de cient mouse could be saved from embryonic death via constitutive active CaN, which causes nuclear translocation of MEF2C and NFAT offering proof that CRT is an upriver regulator of CaN signalling.Therefore, we speculated that CRT might play a role in ECM synthesis via regulating the activation of CaN/NFAT3 signalling under mechanical force.
In comparison with two-dimensional (2D) culture, three-dimensional (3D) culture can more accurately re ect the morphological characteristics of cells in tissues. For example , HGFs cultured on 3D nanoporous ber scaffolds exhibit a spindle shape instead of being a clustered plane on 2D model 26 . The poly lactide-co-glycolide (PLGA) biological scaffold, which can synchronously imitate the in uences of mechanical force and hypoxic environment, can afford a 3D HGFs culture model that is more analogous to the tissue-like environment of natural ECM 4 . In the current study, we aimed to establish a 3D co-culture model of HGFs and PLGA scaffolds to explore the role of CRT, CaN, NFAT3, and phosphorylated NFAT3 (p-NFAT3) in HGFs mechanotransduction and to investigate the role of CRT in ECM deposition in HGFs. Our research offers novel insights into the mechanism of gingival reconstruction during orthodontic tooth movement.

3D culture of HGFs
Gingival tissue specimens were obtained from healthy premolar teeth of patients (age 12-25 years) who had planned to undergo orthodontic treatment at the a liated stomatology hospital of Guangxi Medical University. The patients had no systemic diseases, gum in ammation, or gum hyperplasia. This research was reviewed and approved by the Medical Ethics Committee of Guangxi Medical University (approval number: 20150304-22) and patients and their families were fully informed and offered written consent.
The research was performed according to the principles of the revisory Declaration of Helsinki (World Medical Association General Assembly, 2008).
The gingival tissue was scraped from the root surface and cut into small tissue blocks (1 mm 3 ). These tissue pieces were cultured in Dulbecco's modi ed Eagle's medium (DMEM; Invitrogen; Thermo Fisher Scienti c,Inc., MA, USA ), supplemented with 20% fetal bovine serum (FBS; Gibco; Thermo Fisher Scienti c, Inc., MA, USA) and 1% (v/v) penicillin and streptomycin (Gibco; Thermo Fisher Scienti c,Inc., MA, USA) and maintained in a humidi ed atmosphere at 37°C with 5% CO 2 . The PLGA scaffold (size, 1cm × 1cm × 300 µm; molar ratio of lactic acid to glycolic acid, 75:25; pore size, 100-120 µm; porosity, 85%) was synthesised by solvent casting/particle leaching technology. The 4th generation (1 × 10 5 ) HGFs in good growth conditions in the logarithmic growth phase with a single cell suspension were inoculated on a single sheet of PLGA scaffold to establish a co-culture model of HGFs and PLGA scaffolds.

Application of mechanical force
The above HGFs-PLGA co-culture model was placed into a sterile well plate and incubated for 24 h and then moved to a new well plate. The cells in the PLGA co-culture model were subjected to a compressive force of 25 g/cm 2 by placing a sterilized cylindrical bottle containing lead granules for 6, 24, 48, and 72 h.
The stress value was determined based on previous studies, under a force of 25g/cm 2 , the proliferation index of HGFs was at the maximum 4 . In addition, the stress value was also in line with the level reported by Schwarz 46 , who suggested that the proper force for moving teeth should be within the scope of capillary pressure, which is 7-26 g/cm2 46 . The cells in the control group were cultured under the same conditions without applying a mechanical force. In another experiment, the 3D-cultured cells were infected with CRT-siRNA lentivirus and NC lentivirus (GeneChem, Shanghai, China). After 3 d of infection, green uorescent protein (GFP) expression was observed by uorescent microscopy, and a mechanical force of 25 g/cm 2 was applied to the cells for 24 h. Quantitative real-time PCR and western blot analysis were performed to determine the knock-down e ciency, and a dose-response assay using the Cell Counting Kit-8 (CCK8; Dojindo, Japan) was performed to assess cytotoxicity.

Laser scanning usingconfocal microscope
The 3D-cultured cells that were treated with compressive force were washed twice with phosphatebuffered saline (PBS). Then, 4% paraformaldehyde was applied for cell xation for 30 min and washed with PBS three times for 5 minutes each time. The complex was treated with DAPI (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) for 10 min at 37°C. Immuno uorescence micrographs (Olympus Corporation, Japan) were obtained using a laser confocal microscope (light wavelength of 405nm).
Quantitative real-time PCR Total RNA was extracted from the HGFs-PLGA Complex using the Trizol kit (Aidlab Biotechnology Co., Ltd., Beijing, China) in each group. The obtained RNA was reverse-transcribed to cDNA using the ExScript RT kit (Takara Bio, Inc., Otsu, Japan), RT-qPCR was performed based on the (SYBR) manufacturer's instruction. The 2 −ΔΔCt was considered to be the relevant expression of the target gene in each group. Each experiment was repeated three times, and the average value was recorded.
The primers used had the following sequences:

Statistical analysis
Statistical analyses were performed using SPSS 20.0 (IBM, Chicago, IL, USA) statistical software. The results were expressed as mean ± standard deviation (SD), while t-test methods were used for comparison between two groups, with one-way analysis of variance used for comparison among multiple groups. The inspection level of α=0.05, P<0.05 were considered to be statistically signi cant.

Results
Laser scanning using confocal microscope As shown in Figure 1, HGFs grew densely, crisscrossing and interconnecting with each other in all directions on the 3D HGFs-PLGA scaffold model under a confocal laser microscope (Olympus Corporation, Tokyo Metropolis, Japan). The samples were scanned in layers, and it could be seen that there were cells at different levels.  (Figure 2c, 2d). In addition, with respect to the levels of NFAT3 phosphorylation, the p-NFAT3/NFAT3 ratio was signi cantly decreased within the rst 24h and then increased over time (Figure 2e). These results illustrated that mechanical force was a su cient stimulus to induce CRT and CaN/NFAT3 activation in HGFs.
At the same time, we detected COL-I expression in the HGFs in response to compressive force. In comparison to the control group, the mRNA and protein expression levels of COL-I were signi cantly increased, and the highest expression level was detected at 24 h (Figure 2f, 2g). Therefore, the mechanical force stimulus caused higher collagen expression in the gingival tissue. In addition, when the compressive force was applied for 24 hours, the expression levels of CRT, CaN, and p-NFAT3 all had signi cant changes, and thus was selected for subsequent experiments.
Reversal of the mechanical force effects by blocking CRT As shown in Figure Figure 3a, b, c, d, after infection, the e ciencies were greater than 80% for both CRT-siRNA lentivirus and NC lentivirus, indicating that the CRT-siRNA lentivirus and NC lentivirus were highly infectious to HGFs. Regarding the function of CRT on cell growth, knockdown of CRT expression had no signi cant effect on the proliferation of HGFs (Figure 3e). The cultures infected with the CRT-siRNA lentivirus had signi cantly lower levels of CRT mRNA and protein compared to levels in cultures infected with the NC lentivirus (all P 0.05, Figure 3f, 3g).
According to our analysis, the expression level of CaN was signi cantly decreased (all P 0.05, Figure 4a, 4b), and p-NFAT3 expression was increased when HGFs were transfected with si-CRT (all P 0.05, Figure  4c). It is noteworthy that CRT blockade decreased the expression of COL-I in HGF (all P 0.05, Figure 4d, 4e). These results suggested that CRT is an effective mediator in increasing CaN/NFAT3 activation and the production of COL-I in HGFs under mechanical force.

Discussion
Increasing evidence has showed that, compared with 2D culture systems, 3D cell culture systems could more accurately response the microenvironment of cells in the tissue, and thereby the behaviour of the cells cultured on 3-dimension system was more consistent with the actual in vivo celluar reaction 27 . The PLGA is polymerized by polyglycolic acid (PGA) and polylactic acid (PLA), which has not only the advantages of PGA in the aspect of cell stickiness, but also that of PLA with respect to physical stage 28,29 .In addition, it is more in accordance with the periodontal membrane tissue, and can more accurately simulate the anoxic environment 30 . When mechanical forces are applied to teeth, the periodontal cells on the compression side are simultaneously stimulated by mechanical and hypoxic factors due to vascular atresia, which act upon each other and accelerate the formation of osteoclasts, thus promoting the absorption of alveolar bone on the compression side. The combination of these stimuli has been shown to be bene cial for orthodontic tooth movement 31,32 . In this research, due to the limited diffusion after the application of mechanical force, the cells in the central part of the complex were in a state of hypoxia and nutrient de ciency. So, the PLGA 3D model imitated the in uences of mechanical forces and low-oxygen environments at the same time, furnishing a rm base for the followup experiments.
Converting the mechanical energy of a stimulus signal into a chemical signal is called mechanical transduction 33 . When mechanical force acts on the cell surface and the mechanical signal is transmitted to the cytoskeleton, ER, and nucleus, the stimulation of mechanical force can cause the ER stress response and upregulate the expression of CRT 34 . Calreticulin is an important ER Ca2+-buffering protein participating in adjusting ER Ca2+ reserve and discharge 35,36 , acts as a molecular chaperone, which can help proteins fold correctly [37][38][39] . So far, the majority of studies have concentrated upon the expression and function of CRT in pulmonary and renal brosis. In brotic disorders studies, researche found that CRT was unusually expressed in bleomycin-induced pulmonary brosis and unilateral ureteral obstruction (UUO) renal brosis 11,12 . However, the expression and function of CRT in gingival brosis had not been studied. Therefore, we rst detected the expression level of CRT in HGFs under mechanical forces. CRT was more highly expressed in HGFS than that in the group without mechanical force, indicating the application of mechanical force resulted in an increase in CRT expression, suggesting that CRT might be involved in the process of gingival brosis. To some extent, we also found that CRT expression increased with a trend, that is, with the extension of the application time of mechanical force, the CRT expression continued to rise to the highest value, and then gradually decreased. This may be due to the fact that with the extension of the application time, the sensitivity of the cells to the stimulus gradually decreases.Additionally, CRT plays an important role in regulating downriver cellular Ca2+-dependent signalling through CaN and NFAT 8,40 . Calreticulin-lacking mice die during embryogenesis because of damaged heart development; thus, CRT-lacking mice could be saved from embryonic lethality if CaN was constitutively activated during development 24 , suggesting the importance of the role of CRT in adjusting the ER Ca2+ discharge and downriver CaN/NFAT signalling.
Under the stimulation of mechanical force, collagen synthesis in the HGFs increased, which was consistent with the results of our previous study and other research 4 . Recent literature has reported that broblasts were sensitive to the mechanics of their microenvironment, and abnormal broblast mechanosensing can drive the brotic program 42 . In recent years, numerous studies have reported the key role of the CaN/NFAT3 signalling pathway in ECM accumulation in the process of tissues brosis, including liver brosis and cardiomyocyte broblasts 19,43 . The deposition of ECM was obviously inhibited when an CaN or NFAT inhibitor was used in these brotic tissues 17,18,44 . However, the expression and function of CaN/NFAT3 signalling pathway in gingival brosis had not been studied. Thus, we rst explored the expression levels of CaN, NFAT3, p-NFAT3 in HGFs under mechanical forces and found that a signi cant increase in CaN expression, as well as a decrease in p-NFAT3 expression, indicating the application of mechanical force signi cantly increased the activation of the CaN/NFAT3 signalling pathway. To some extent, we also found that the activity of the CaN/NFAT3 signalling pathway and the expression level of CRT showed the same trend of change, suggesting that the expression of the CaN/NFAT3 signalling pathway and CRT may have rhythmic characteristics. In order to assess CRT and CaN/NFAT3 signalling function in ECM synthesis in HGFs, we infected HGFs with CRT-siRNA lentivirus or control lentivirus. Compared to controls, CRT-siRNA treated cells showed decreased CaN and COL-I expression, increased p-NFAT3 expression. These ndings suggested that downregulation of CRT inhibited gingival collagen synthesis, potentially via CaN/NFAT3 signalling pathway. Additionally, Owusu proposed that factors such as high glucose levels, nutrient deprivation, and oxidative stress trigger ER stress. Endoplasmic reticulum stress increases CRT levels; in turn, CRT stimulates Ca2+ release from the ER, leading to increased cytosolic Ca2+, that results in activation of CaN, which dephosphorylates cytoplasmic NFAT, leading to NFAT activation and nuclear translocation, inducing the transcription of related target genes, promoting collagen secretion and procollagen processing into the ECM 45 . Nevertheless, the effects of CRT on HGFs under mechanical forces should be further demonstrated using a more rigorous approach that includes genetic knockouts and overexpression vectors. This would be performed in future studies.

Conclusion
Our ndings demonstrated activation of the CaN /NFAT3 signalling pathway, a signi cant increase in CRT expression and COL-I synthesis in the HGFs cultured on 3D PLGA scaffolds under the mechanical force. Speci cally, blocking CRT could inhibit CaN/NFAT3 activation and collagen synthesis. These ndings ultimately offer novel insights into the mechanisms of mechanical conduction and ECM synthesis in HGFs, and may be conducive to the progress of targeted therapeutics by accelerating gingival tissue remodeling.