https://orcid.org/0000-0001-5801-5019
Summary
This study investigated the expression of extracellular matrix metalloproteinase inducer (EMMPRIN) in the compression area during orthodontic relapse in rat molars.
Thirty Wistar rats (6 weeks old) underwent orthodontic tooth movement (OTM) of the left first maxillary molar for 21 days, followed by removal of the force device. The contralateral maxillary molar served as a control with no mechanical force stimuli. Animals were sacrificed at 0, 1, 3, 7, and 14 days of relapse after force withdrawal. Tooth relapse and alveolar bone parameters were measured using microcomputed tomography (micro-CT). Maxilla sections were obtained for haematoxylin and eosin (HE), immunohistochemical staining [EMMPRIN, nuclear factor kappa B ligand (RANKL) and vascular endothelial growth factor (VEGF)] and tartrate-resistant acid phosphatase (TRAP). Correlation analyses were then performed.
After force removal, nearly 79.88% of the total relapse occurred within the initial 3 days. The number of osteoclasts clearly increased while the alveolar bone density decreased on the pressure side on Day 3 of relapse. Moreover, the EMMPRIN expression level significantly increased on Day 1, peaked up on Day 3 and decreased on Days 7 and 14. Statistically, a strong positive correlation was found between EMMRPIN expression and the osteoclast number and RANKL and VEGF expression.
EMMPRIN was highly expressed on the pressure side during the orthodontic tooth relapse, which could be involved in osteoclastogenesis and alveolar bone resorption in association with RANKL and VEGF expression.
Introduction
Orthodontic teeth, which are teeth that are moved under force stimuli, tend to easily regress to their original position after orthodontic treatment. This progress is known as relapse and has adverse effects on clinical treatment due to the high rate of occurrence. The precise cellular mechanism underlying this event remains poorly understood. A previously reported theory (1) suggests that orthodontic relapse involves cellular patterns similar to orthodontic tooth movement (OTM). Generally, it is accepted that alveolar bone remodelling, such as osteoclast-induced alveolar bone resorption adjacent to the compression area, is one of the main causes in these processes (2). Thus, factors that are involved in osteoclastogenesis and bone resorption on the pressure side of alveolar bone are likely to influence post-treatment relapse.
Extracellular matrix metalloproteinase inducer (EMMPRIN), also known as CD147, is a highly glycosylated transmembrane protein belonging to the immunoglobulin superfamily. In 1980s, Biswas et al. first discovered CD147 in hepatocellular carcinoma (HCC) and its ability to induce matrix metalloproteinases (MMPs) (3–5). Recently, EMMPRIN has been detected in healthy periodontal tissues and been found to be highly elevated in periodontitis and periapical areas (6, 7). In 2010, Liu et al. (7) discovered that EMMPRIN is distributed in gingival connective tissues, gingival epithelium, and alveolar bone, contributing to periodontitis development.
Several lines of evidences have shown that EMMPRIN plays a pivotal role in osteoclastogenesis and bone resorption. By administrating EMMPRIN antibodies in rats, Yang et al. demonstrated that EMMPRIN participates in alveolar bone resorption in periodontitis. They found that EMMPRIN inhibition remarkably reduces alveolar bone loss and osteoclastogenesis (8). Additionally, EMMPRIN contributes to several bone destructive pathologies, such as rheumatoid arthritis and bone metastasis of malignant tumour, by promoting osteoclast differentiation and function (9, 10).
The process of EMMPRIN-induced osteoclast formation and activation has been reported to be associated with nuclear factor kappa B ligand (RANKL) and vascular endothelial growth factor (VEGF), which are both deemed to directly stimulate bone resorption and activate mature osteoclasts, respectively (11, 12). Numerous researches on cancer bone metastasis have revealed that EMMPRIN expression, which is induced by RANKL, upregulates downstream VEGF expression and therefore enhances osteoclastic activity and tumour-induced bone resorption (10, 13). In OTM and relapse, RANKL and VEGF play indispensable roles by promoting bone resorption. Studies on orthodontic force-induce bone remodelling have indicated that mechanical forces up-regulate RANKL and VEGF expression (14–16). Increasing the expression level of RANKL or VEGF via gene transfection has been shown to enhance bone resorption and osteoclastogenesis on the compression side and accelerate OTM (16, 17). Furthermore, suppression of either RANKL or VEGF expression has been demonstrated to decrease the number of osteoclasts in pressure areas and thus results in a reduction in the amount of OTM and tooth relapse after force withdrawal (18–20).
The association between EMMPRIN and bone resorption in orthodontic tooth relapse has not yet been investigated. The aim of this study was to analyse the expression of EMMPRIN on the compression side during orthodontic tooth relapse and assess its possible correlation with RANKL and VEGF expression during this process.
Materials and methods
Animals
Thirty male 6-week-old Wistar rats (151.93 g) were used in this study. The rats were purchased from the Experimental Animal Center of Hubei province (Wuhan, Hubei, China) and housed under specific pathogen-free conditions with a 12 hours light/dark cycle, at a temperature of 22 ± 2°C and relative humidity of 65 ± 5%. The rats received water and standard rat chow pellets ad libitum and were kept under observation for 1 week before starting the experiment. All the management experimental procedures were approved by the Ethics Committee for Animal Research, School and Hospital of Stomatology, Wuhan University (No.00275268).
Orthodontic appliance for tooth movement and relapse
A split-mouth design was used in this study (21). After the rats were anesthetized by isoflurane inhalation, a closed nickel-titanium coil spring (Tomy B284, Japan) was ligated between the first maxillary left molar (M1) and the incisors, and the contralateral side served as a control (Figure 1B). The devices exerted 50 g of force, which was monitored with a force gauge (HANDPI HP-2, China), to move the M1 mesially based on previous evidence that 50 g of force provides substantial tooth movement (21). Throughout the study, the body weight of animals were evaluated weekly, and device breakage and gingival or other soft tissue inflammation were evaluated twice per week.
(A) Experimental study design. After 21 days of teeth movement, the force device was removed and the left first molars (M1) were allowed to relapse. Tooth relapse, bone density, and histologic analysis (HE, TRAP, and immunohistochemistry staining) were analysed at 0, 1, 3, 7, and 14 days after force device withdrawal. (B) Force device in rats. (C) Distances ± standard deviation (SD) between M1 and second molar (M2) at 0, 1, 3, 7, and 14 days in experimental groups. **P < 0.01, ***P < 0.001. (D) Relapse distances. (E) Relapse ratio (relapse distance / original distance × 100%). Original distance corresponds to the distance between M1 and M2 after 21 days tooth movement.
The force device was in place for 21 days and six rats were sacrificed immediately after force withdrawal (R0; n = 6). The others were sacrificed at 1 (R1; n = 6), 3 (R3, n = 6), 7 (R7, n = 6), 14 (R14, n = 6) days during orthodontic relapse (Figure 1A). All the animals were sacrificed by cervical dislocation under isoflurane inhalation anaesthesia.
Micro-computed tomography examination
The maxillae were removed, immediately fixed in 10% formalin for 48 hours at 4°C and placed in 70% ethanol for scanning using a polychromatic cone beam micro-CT system (SkyScan 1176, Bruker, Germany) with a 18-µm image pixel size. Images were obtained at 20–90 kV and 25 W. 3-D digital images were reconstructed using software compatible with the micro-CT and were rotated into a standard orientation (Figure 2A–2C) to measure the distance between the midpoint (a) of the distal-edge ridge of M1 and the midpoint (b) of the mesio-edge ridge of the second molars (M2) from the occlusal view (Figure 2B). Thus, the relapse distance and the ratio of relapse (Figure 1D and 1E, ratio of relapse = distance of relapse / distance of original tooth movement × 100%) could be obtained, as previously described (22). The alveolar bone parameters bone mineral density (BMD, mg/cm3), bone volume to tissue volume (BV/TV, %), trabecular thickness (Tb.Th, mm) and trabecular separation (Tb.Sp, mm) were evaluated. The region of interest (ROI) was defined as a cube (290 µm × 290 µm × 290 µm) distal to the cervical third of the mesial palatal root (MPR), which was on the pressure side during relapse (Figure 2D).
Three-dimensional micro-CT images and bone density. (A) Standard orientation adjustment in the axial view was conducted by adjusting the palatal suture to be tangent to the blue line. The red arrow indicates the mesial palatal root (MPR) of the first molars (M1). (B) The distance between the midpoint (a) of the distal-edge ridge of M1 and the midpoint (b) of the mesio-edge ridge of M2. (C) Standard orientation adjustment in the coronal view was conducted by making the red line parallel to the lower portion of the palatal bone. (D) The ROI is the red square distal to the cervical third of the MPR. (E–H) The BMD, BV/TV, Tb.Th and Tb.Sp at 0, 1, 3, 7, 14 days and in the control group. (I) Correlations between BMD and EMMPRIN expression (r = −0.869; P < 0.0001; r2 = 0.755). MR, mesial root; MBR, mesial buccal root, DBR, disto-palatal root; DPR, disto-palatal root. *P < 0.05, ****P < 0.0001.
An independent investigator, who was blind to the study, made the measurements. Each measurement was repeated three times and no variation was observed in measured values.
Histology
Following micro-CT scanning, the maxillae were dissected into halves, decalcified with 10% ethylenediaminetetraacetic acid (EDTA) for 8 weeks, dehydrated through a gradient ethanol and embedded in paraffin for histological analysis. Five-micron sagittal sections parallel to the long axis of palatal root of M1 were obtained and subjected to haematoxylin and eosin (HE), tartrate-resistant acid phosphatase (TRAP), and immunohistochemical (EMMPRIN, RANKL, and VEGF) staining. Areas distal to the cervical third of MPR, consisting of alveolar bone and periodontal ligament that were on the pressure side of relapse, converted from tension side of tooth movement, were evaluated in the stained sections (Figure 3A).
(A) Analysis of histological section focused on the area distal to the cervical third of MPR of the first molar, consisting of alveolar bone and periodontal ligament parallel to the root (yellow square area). (B) HE staining of regularly arranged collagen fibres in control groups. (C) On Day 0, PDL width increased and collagen fibres were stretched due to mesial forces. (D) On Day 3, PDL width decreased and collagen fibres showed a broken and disorganized pattern. Osteoclasts accumulated along the bone surface. Dark arrows indicate osteoclasts. (E) On Day 14, PDL width and collagen fibres recovered to normal patterns similar to those in the control groups. (F) PDL width measurements. MPR, mesial palatal root; DPR, disto-palatal root; PDL, periodontal ligament; ab, alveolar bone; The red arrow indicates relapse direction. A, scale bar = 200 μm; B–D, scale bars = 50 μm.
Immunohistochemistry
Immunohistochemical staining was performed using the Streptavidin-Peroxidase (SP) method. Antigen retrieval was performed using pepsin 30 minutes at 37°C. The sections were incubated with rabbit monoclonal EMMPRIN antibody (Abcam, United Kingdom, ab 108317; 1:200 dilution), mouse monoclonal RANKL antibody (Santa Cruz Biotechnology, USA, sc-52950; 1:50 dilution) and rabbit monoclonal VEGF antibody (Santa Cruz Biotechnology, USA, sc-152; 1:50 dilution) for 10 hours at 4°C. For visualization, the sections were incubated in the avidin–biotin complex and then stained in 3,3-diaminobenzidine. As a negative control, the same procedure was performed on sections without the addition of primary antibodies.
Digital images of ROI were obtained at magnification of ×400 using a camera (OLYMUS BX51, Japan) under same settings. Aperio ImageScope 12.3.2 software (Leica Biosystems, Germany) was used to quantify the positively immunostained area, which was calculated by a positive pixel count algorithm in the software (positivity (%) = the total number of positive pixels (Np) / the total number of pixels (Ntotal); Np = the number of weak-positive pixels + the number of positive pixels + the number of strong-positive pixels, and Ntotal = Np+ the number of negative pixels).
TRAP staining
For osteoclast analysis, sections were stained with a TRAPstaining kit (387A-1KT, Sigma–Aldrich, USA) according to the manufacturer’s instructions. Cells were defined as osteoclasts if they were TRAP-positive, multinucleated, and located on the bone surface or in Howship’s lacunae. Cell counts in the distal part of the cervical third of the MPR in each section were blindly analysed by an investigator using microscopy under high magnification (×400) and three sections at 35-μm intervals were analysed for each rat.
Root histologic analysis
TRAP staining of five roots from rats in the control groups was performed. The mesial root (MR) in the experimental groups was analysed via immunohistochemical and TRAP staining.
Statistical analysis
The data are presented as the means and standard deviations. At each timepoint, the means were compared between groups using one-way analysis of variance (ANOVA) followed by Tukey’s test. Pearson correlation coefficients were used to examine the association between EMMPRIN and RANKL, VEGF expression, and the number of osteoclasts. An r value >0.4 was recognized as a moderate correlation and r > 0.7 was recognized as a strong correlation. A P value <0.05 was considered statistically significant. All the results were processed with IBM SPSS Statistics 21.0 software (SPSS, Inc., Chicago, IL, USA).
Results
Animal weight
The initial weight of the animals was 151.93 ± 11.0 g and the weight at the time of orthodontic device removal was 237.58 ± 12.2 g. The mean weight at the end of the relapse phase was 289.08 ± 12.5 g. There was no difference (P > 0.05) in weight gain between each group during the OTM and relapse phases.
Relapse distance and ratio
Orthodontic teeth began to relapse distally in experiment groups. There was a significant change between Day 0 and Days 3, 7, and 14 (Figure 1C). The distance of OTM on Day 21 was 975.78 μm (Figure 1C). The total relapse distance on Day 14 was up to 334.44 μm (Figure 1D), which was 34.27% of the OTM. The relapse distance on Day 3 was 267.11 μm (Figure 1D), accounting for nearly 79.88% of the total relapse distance. Relapse ratio of Day 3 was 27.37% (Figure 1E). Then, the rate of relapse gradually decreased until Day 14.
Micro-CT
To investigate alveolar bone remodelling, alveolar bone parameters on the distal side of the MPR were examined by micro-CT. There was no significant difference of bone parameters between the control groups as relapse over time. From Day 1 to Day 7, BMD, BV/TV, Tb.Th significantly decreased and Tb.Sp increased, indicating bone resorption. Moreover, BV/TV, Tb.Th, and Tb.Sp on Day 14 returned to levels similar to those in the control group (Figure 2E–2H). Furthermore, statistical analysis showed a negative correlation (r = −0.869; P < 0.0001; r2 = 0.755) between BMD and EMMPRIN expression (Figure 2I)
HE staining
Pathologic events, such as inflammation, which may influence the presence of EMMPRIN, were not observed in any of the groups. Compared with control groups, collagen fibres showed stretching patterns on Day 0 that changed to broken and disorganized patterns on Day 3 and finally recovered to regular arrangement on Day 14. In addition, periodontal ligament (PDL) width increased during OTM time on Day 0 but decreased on Days 1 and 3. At the end of relapse on Day 14, the PDL width recovered to normal (Figure 3B–3F). It was suggested that the area distal to the cervical third of the MPR of M1 was converted from the initial tension side during OTM to the pressure side after force withdrawal.
Immunohistochemistry
Immunohistochemistry examination showed that EMMPRIN were slightly distributed in the control groups (Figure 4A) and no significant differences were observed between various timepoints. On Day 0 of relapse, the expression level of EMMPRIN was relatively low (Figure 4B) and exhibited no significant difference compared with the control group. However, EMMPRIN level increased as the molars relapse over time (P < 0.05), which was mainly observed in osteoclasts, fibroblasts on the pressure side (Figure 4C–4F, Table 1). The number of positively stained cells increased from Day 1, reached a maximum on Day 3, and decreased on Days 7 and 14 (Figure 4a).
The percentage (%) of different cells’ immunostained positivity accounting for total immunostained positivity on Day 3 of relapse.
| EMMPRIN | RANKL | VEGF | |
|---|---|---|---|
| Osteoclasts | 48.68 ± 7.49 | 40.01 ± 5.64 | 38.38 ± 4.34 |
| Fibroblats | 34.34 ± 8.38 | 29.82 ± 4.06 | 37.45 ± 8.12 |
| EMMPRIN | RANKL | VEGF | |
|---|---|---|---|
| Osteoclasts | 48.68 ± 7.49 | 40.01 ± 5.64 | 38.38 ± 4.34 |
| Fibroblats | 34.34 ± 8.38 | 29.82 ± 4.06 | 37.45 ± 8.12 |
The percentage (%) of different cells’ immunostained positivity accounting for total immunostained positivity on Day 3 of relapse.
| EMMPRIN | RANKL | VEGF | |
|---|---|---|---|
| Osteoclasts | 48.68 ± 7.49 | 40.01 ± 5.64 | 38.38 ± 4.34 |
| Fibroblats | 34.34 ± 8.38 | 29.82 ± 4.06 | 37.45 ± 8.12 |
| EMMPRIN | RANKL | VEGF | |
|---|---|---|---|
| Osteoclasts | 48.68 ± 7.49 | 40.01 ± 5.64 | 38.38 ± 4.34 |
| Fibroblats | 34.34 ± 8.38 | 29.82 ± 4.06 | 37.45 ± 8.12 |
(A–F) EMMPRIN expression and localization in the compression area during the relapse phase. EMMPRIN was weakly expressed in control groups (A) and at Day 0 (B). The overexpressed EMMPRIN was mainly localized in osteoclast cells and fibroblasts (D). Significant differences in EMMPRIN expression were observed on Days 1, 3, and 7 in the experimental groups compared with control groups (a). (G–R) Immunohistochemical analysis of RANKL and VEGF on the pressure site during relapse. (b, c) Semiquantitative analysis of RANKL and VEGF. (d, e) Correlation analysis revealed that EMMPRIN expression was strongly positively correlated with RANKL expression (r = 0.87; P < 0.0001; r2 = 0.757;) and VEGF expression (r = 0.93; P < 0.0001; r2 = 0.861). OC, osteoclasts; F, fibroblasts. Scale bars 50 μm.
Similarly, RANKL and VEGF distribution showed a pattern similar to that of EMMPRIN distribution (Figure 4G–4R, 4b and 4c). Correlation analysis showed that EMMPRIN expression had strong positive correlations with RANKL (r = 0.87; P < 0.0001; r2 = 0.757) and VEGF (r = 0.93; P < 0.0001; r2 = 0.861) (Figure 4d and 4e).
TRAP
In the control and Day 0 groups, osteoclast activity was relatively low, and only 0–2 TRAP-positive cells were observed (Figure 5A and 5B). Nevertheless, on Day 3 in the experimental group, osteoclast counts significantly increased up to 5–6 (P < 0.05) and decreased on Days 7 and 14 (Figure 5C–5E). Additionally, the data showed a strong correlation (r = 0.947; P < 0.0001; r2 = 0.897) between osteoclast counts and EMMPRIN expression (Figure 5F).
TRAP-stained histological sections from the distal cervical third of MPR. (A) Control groups. (B) Day 0 of the relapse phase. (C) TRAP-positive multinucleated osteoclasts located on the alveolar bone crest surface on Day 3 (×200 magnification). (D) Osteoclasts on Day 3 of relapse phase (×400 magnification). (E) The number of osteoclasts at 0, 1, 3, 7, and 14 days of relapse. (F) Correlation analysis between osteoclast counts and EMMPRIN expression. OC, osteoclast. Red arrow indicates direction of relapse. A, C, scale bar 50 = μm; B, D scale bar = 20 μm.
Root histologic analysis
The numbers of osteoclasts in the distal cervical third of the mesial buccal root (MBR), disto-buccal root (DBR), and disto-palatal root (DPR) were higher than those in the MPR and MR in the control groups, but the differences were not significant (Figure 6A). In the experimental groups, EMMPRIN expression and osteoclast number were relatively low in the MR compared with the MPR (Figure 6B and 6C).
Histological analysis of the distal cervical third of different M1 tooth roots. (A) The number of osteoclasts in the control groups. (B) EMMPRIN expression in the MPR and MR in the experimental groups. (C) The number of osteoclasts in the MPR and MR in the experimental groups.
Discussion
Analysis of tooth relapse in the root of interest
Orthodontic tooth relapse is mainly caused by alveolar bone remodelling after cessation of OTM (2) and has biological responses similar to OTM, characterized as bone resorption on the compression side and formation on the tension side (23). We selected the area distal to the cervical third of MPR as the evaluation region to analyse bone density and histology for the following reasons. (a) The selected area was converted into the compression side from tension side after force removal and previous studies have demonstrated that 40–60 g of force exerts tipping movement of the M1 in rats (24). (b) Physically distal drifts of maxillary molars were common phenomena in rats (25), which explained the activation of osteoclasts on the distal parts of the five roots from the control groups. The number of osteoclasts in the MBR, DBR, and DPR in control groups was greater than that in the MPR, although the difference was not significant. (c) In the experimental groups, EMMPRIN expression and the activity of osteoclasts were relatively low in the MR (the largest) compared with the MPR. Additionally, the MBR (the smallest) was too small to obtain the complete root. Overall, to reduce distal drift influence and easily obtain complete roots, we focused on the distal part of the MPR. In this study, a nickel-titanium closed coil spring was used to exert 50 g of forces on theM1 in rats because compelling evidence has identified that nickel-titanium springs produce persistent forces (26). The M1 of rats moved mesially with application of mechanical force for 21 days and relapsed distally following removal of the force device. The total relapse ratio on Day 14 was 34.27%. The molar experienced rapid relapse within the initial 3 days, accounting for nearly 79.88% of the total relapse, which showed a trend similar to that reported in previous researches (23, 27). However, relapse ratio data are not comparable with those from other studies because of variations in the duration of OTM or relapse and differences in the force magnitude applied (21, 22).
EMMPRIN in osteoclastogenesis and bone resorption promotion
EMMPRIN promotes osteoclastogenesis and bone resorption in periodontitis (8) and periapical diseases (6). However, its role in bone remodelling during tooth relapse is unknown. In this study, we demonstrated that EMMPRIN was slightly expressed in periodontal tissues at basal levels in control groups and significantly upregulated on the distal parts after forces disappear. The expression level of EMMPRIN gradually increased from Day 1 after force application removal, peaked on Day 3 and decreased on Days 7 and 14, which was consistent with the changes in number of osteoclasts in the same area, suggesting EMMPRIN may contribute to osteoclast activation during tooth relapse. Furthermore, morphometric analysis showed that bone parameters (BMD, TV/BV, Tb.Th) on the pressure side in the initial 3 days of relapse decreased and there was a close negative correlation (r =−0.869; P < 0.0001; r2 = 0.755) between EMMPRIN expression and BMD. One plausible explanation for these findings is that EMMPRIN promotes bone resorption by recruiting circulating monocytes and macrophages (28) and by augmenting osteoclastogenesis.
EMMPRIN expression has been shown to be closely associated with RANKL and VEGF expression in several bone resorption process (9, 10). This could explain why the high EMMPRIN expression level on the pressure side during the relapse period was accompanied by simultaneous RANKL and VEGF expression in the same area. In addition, EMMPRIN expression was significantly correlated with RANKL and VEGF expression, respectively at different timepoints. Consistently, previous studies have reported that after force device removal, the production of RANKL and VEGF were enhanced on the pressure side for bone resorption (15,20). Therefore, it is possible that the participation of EMMPRIN in alveolar bone resorption during tooth relapse could be influenced by RANKL and VEGF expression. The exact mechanism underlying bone remodelling mediated by EMMPRIN requires further exploration.
Day 3 of relapse
Interestingly, we observed that on Day 3 of relapse, alveolar bone parameters reached lowest levels, whereas osteoclast counts exhibited an opposite trend and reached the maximum levels. Additionally, the expression levels of EMMPRIN, RANKL, and VEGF similarly peaked at the same timepoint and area. Overall, Day 3 of relapse may be the most active period of tooth relapse, indicating that relapse was rapidly initiated after the force withdrawal. Consistently, previous studies have reported similar relapse activity timepoints (1).
EMMPRIN and collagen turnover
In addition to alveolar bone remodelling, periodontal ligament collagen turnover plays a crucial role during orthodontic relapse. Collagen recovery and synthesis contribute to early relapse (29). Furthermore, collagen deposition reduces the extent of this process by decreasing mechanical strength of collagen fibers, which consequently repress tooth relapse (30). EMMPRIN regulates extracellular matrix (ECM) synthesis and deposition through MMPs induction (3–6). Since collagen is the main structural component of ECM in PDL, the possible role of EMMPRIN in mediating collagen turnover during tooth relapse needs to be further investigated.
Strengths and limitations
To the best of our knowledge, related studies have not investigated EMMPRIN expression in orthodontic tooth relapse in rats. This is the first study to explore the association between EMMPRIN and osteoclastogenesis and bone resorption factors in reverse tooth movement. However, we only examined EMMPRIN protein expression through immunohistochemistry. EMMPRIN inhibition experiments should be conducted in further in-depth studies.
Conclusion
Orthodontic relapse occurs rapidly after force removal. EMMPRIN expression, which could be associated with RANKL and VEGF expression, could be involved in osteoclastogenesis and bone resorption in the compression area during tooth relapse.
Funding
This work was supported by Foundation of Wuhan Municipal Science and Technology Bureau [2015060101010052], China.
Acknowledgement
We thank Guo Ning for critical reading of the manuscript.
Conflicts of interest
None to declare.
