Introduction

Orthodontic treatment is the main method to correct malocclusions by applying force to teeth. The resulting orthodontic tooth movement (OTM) is controlled by a dynamic balance between osteoclastic bone resorption on the compression side and osteoblastic bone formation on the tension side1. Osteoclasts are bone-resorbing cells derived from hematopoietic stem cells. Macrophage-colony stimulation factor (M-CSF) and receptor-activator of nuclear factor-κB ligand (RANKL) are essential factors for osteoclast differentiation2. Moreover, the proinflammatory cytokine tumor necrosis factor-α (TNF-α) induces osteoclast differentiation both in vitro3 and in vivo.4 In humans, various cytokines, including TNF-α, are expressed in the periodontal ligament during OTM5. Previously, we have shown that osteoclast formation and tooth movement are decreased in TNF receptor type 1- and 2-deficient (TNFRsKO) mice during OTM6. Therefore, TNF-α is an important cytokine for tooth movement.

Osteocytes act as orchestrators of bone remodeling and coordinate osteoclasts and osteoblasts7. For decades, osteoblasts have been considered to be responsible for inducing osteoclast formation by expressing RANKL8. As osteocytes, similar to osteoblasts, also express RANKL and induce osteoclast formation, osteocytes came into the focus of bone metabolism research9. OTM is disturbed in osteocyte-ablated mice10 and mice lacking RANKL in osteocytes11. In our previous studies, TNF-α enhanced RANKL expression in osteocytes, inducing osteoclast formation during OTM12,13. Therefore, understanding the role of osteocytes is essential for elucidating the detailed mechanisms underlying alveolar bone resorption during tooth movement.

Previously, cell death was classified based on the morphological appearance into three different forms: apoptosis, autophagic cell death, and necrosis14. Currently, cell death is according to the Nomenclature Committee on Cell Death (NCDD) broadly divided into regulated cell death and accidental cell death, and various types of cell death are distinguished based on genetic, biochemical, pharmacological, and functional evidence15. Recent studies have identified novel types of regulated cell death, including pyroptosis, ferroptosis, and necroptosis15. Necroptosis resembles necrosis in morphology and is induced mainly by TNF-α stimulation with inhibited apoptosis16. Necroptosis research began with the discovery of TNF-α-induced necrotic cell death17. Necroptosis was identified in 2005 following the inhibition of TNF-α-dependent necrotic cell death by necrostatin-1 (Nec-1), an inhibitor of receptor-interacting protein 1 (RIP1)18. Additionally, RIP3 and mixed lineage kinase domain-like protein (MLKL) were identified as proteins essential for inducing necroptosis19,20. Thus, it is generally accepted that TNF-α-induced sequential phosphorylation of RIP1, RIP3, and MLKL induces necroptosis15,16. Necroptosis involves the rupture of the plasma membrane and the subsequent release of damage-associated molecular patterns (DAMPs)21. This release of DAMPs from necroptotic cells mediates various pathological conditions16,21. In bone diseases, necroptosis is, for instance, associated with rheumatoid arthritis22 and osteoarthritis23. Additionally, the recognition of DAMPs induces sterile inflammation, leading to inflammatory diseases such as metabolic disorders, neurodegenerative diseases, autoimmune diseases, and cancer24. Although alveolar bone resorption during OTM is perceived as sterile inflammation, the effects of DAMPs on tooth movement have not yet been investigated. Several studies observed during OTM osteocyte death in the alveolar bone on the compression side25,26,27.  A recent report suggested that osteocyte necroptosis may be associated with bone loss in postmenopausal osteoporosis28. However, the effects of osteocyte necroptosis on osteoclastogenesis during OTM remain unclear.

This study aimed to investigate the effects of TNF-α-induced osteocyte necroptosis on osteoclastogenesis during OTM. We hypothesized that osteocyte necroptosis involves TNF-α expressed in periodontal ligaments during OTM and that DAMPs released from necroptotic osteocytes affect osteoclastogenesis. Using a mouse model of OTM, we detected TNF-α-induced osteocyte necroptosis in the alveolar bone on the compression side. In vitro, necroptosis of primary osteocytes enhanced osteoclastogenesis. Collectively, this study provides insights into the mechanism of tooth movement and suggests the importance of osteocyte necroptosis in the release of DAMPs to enhance osteoclastogenesis.

Results

During OTM, osteocytes undergo cell death on the compression side

To understand the dynamics of tooth movement, we performed OTM in wild-type (WT) mice using a nickel-titanium (Ni-Ti) coil spring and measured the distance of tooth movement (Fig. 1a, b, Supplementary Fig. S1a). The body weight of the mice was properly monitored and maintained throughout the period of OTM (Fig. S1b). After the initial displacement, tooth movement stagnated until day 6 but restarted on day 8 (Fig. 1c). Histological analyses using tartrate-resistant acid phosphatase (TRAP) staining revealed that osteoclasts were induced on the compression side, resulting in alveolar bone resorption (Fig. 1d). The number of osteoclasts on the compression side significantly increased from day 6 to day 12 of OTM (Fig. 1e).

Fig. 1
figure 1

Time-course analysis of osteoclast formation in the mouse model of orthodontic tooth movement (OTM). a Left image: schematic representation of the experimental method in the mouse model of OTM. A Ni-Ti coil spring is ligated to the maxillary left first molar (M1) and the hole near the incisors. Right image: hematoxylin and eosin (H&E) staining of the sagittal section of the maxillary left molars on day 6 after OTM. Scale bar = 200 μm. b Representative images of maxillary left molars and dental impressions on day 0, 6, and 12 after OTM. c Distance of tooth movement from day 0 to day 12 (n = 6). d Tartrate-resistant acid phosphatase (TRAP) staining of horizontal sections with periodontal tissue and distobuccal root of the maxillary left first molar after OTM in wild-type (WT) mice. Scale bar = 100 μm. e Quantification of TRAP-positive cells on the compression side after OTM (n = 6). Black arrows indicate the direction of force. M1, first molar; M2, second molar; M3, third molar; M, mesial side; D, distal side; a, alveolar bone; p, periodontal ligament; r, root. **p < 0.01, ***p < 0.001, ****p < 0.0001.

Full size image

To assess when osteocyte death was observed in OTM mice, osteocyte death was analyzed using hematoxylin and eosin (H&E) staining at seven time points for up to 12 days of OTM. Many dying osteocytes and empty lacunae were observed in the alveolar bone on the compression side compared to fewer on the tension side (Fig. 2a). The percentage of osteocyte death was the highest on the compression side on day 6 of OTM (Fig. 2b). To further characterize osteocyte death, transmission electron microscopy (TEM) images were analyzed on day 6 of OTM. Apoptotic osteocytes with nuclear fragmentation and condensation, necrotic osteocytes with loss of the plasma membrane, and empty lacunae were observed (Fig. 2c).

Fig. 2
figure 2

Time-course analysis of osteocyte death in the mouse model of orthodontic tooth movement (OTM). a Hematoxylin and eosin (H&E) staining of horizontal sections with periodontal tissue and distobuccal root of the maxillary left first molar after OTM in WT mice. Upper left image: the rectangles of 400 μm × 200 μm indicate the region of interest (ROI) on the compression and tension sides. Black arrows indicate the direction of force. Scale bar = 100 μm. Other images: high-magnification images of ROIs on the compression and tension sides. White arrowheads indicate normal osteocytes, orange arrowheads indicate dying osteocytes, and red arrowheads indicate empty lacunae. Scale bar = 50 μm. b Percentage of normal osteocytes (white), dying osteocytes (orange), and empty lacunae (red) (n = 6). c Transmission electron microscopy (TEM) analysis on days 0 and 6 after OTM in WT mice. Yellow arrows indicate empty lacunae. Scale bar = 2 μm. M, mesial side; D, distal side; a, alveolar bone; p, periodontal ligament; r, root.

Full size image

TNF-α induces osteocyte necroptosis on the compression side during OTM

We next examined whether TNF-α induced osteocyte death during OTM. Because TNF-α is expressed in the periodontal ligament during OTM5we induced OTM using WT and TNFRsKO mice. According to our analysis of osteocyte death on the compression side on day 6 after OTM, the percentage of dying osteocytes and empty lacunae in TNFRsKO mice was significantly lower than that in WT mice (Fig. 3a, b). To examine the presence of osteocyte necroptosis, we used immunofluorescence staining to detect the necroptosis markers phosphorylated (p)-RIP3 and p-MLKL in osteocytes of WT and TNFRsKO OTM mice. p-RIP3-positive osteocytes were detected on the compression side of WT mice on day 6 (Fig. 3c, d). On day 6 of OTM, WT mice had the highest percentage of p-RIP3-positive osteocytes in the alveolar bone on the compression side, whereas p-RIP3-positive osteocytes were almost undetectable in TNFRsKO mice (Fig. 3e). Similarly, p-MLKL-positive osteocytes were observed on the compression side on day 6 after OTM in WT mice (Fig. 3f, g). The highest percentage of p-MLKL-positive osteocytes was also found on the compression side in WT mice on day 6, whereas these cells were rarely detected on the compression side in TNFRsKO mice or on the tension side in either mouse model (Fig. 3h). These results suggest that TNF-α induces osteocyte necroptosis on the compression side during OTM.

Fig. 3
figure 3

Detection of tumor necrosis factor-α (TNF-α)-induced osteocyte necroptosis in the mouse model of orthodontic tooth movement (OTM). a Hematoxylin and eosin (H&E) staining of the region of interest (ROI) on the compression side in wild-type (WT) mice and TNF receptor type 1- and 2-deficient (TNFRsKO) mice. Scale bar = 50 μm. b Percentage of dying osteocytes (orange) and empty lacunae (red) in WT and TNFRsKO mice (n = 6). c Immunofluorescence staining for phosphorylated (p)-RIP3 (red) and DAPI staining (blue) of ROIs on the compression and tension sides in WT and TNFRsKO mice. Scale bar = 50 μm. d High-magnification images of the compression side on day 6 after OTM in WT mice. White arrows indicate p-RIP3-positive osteocytes. Scale bar = 50 μm. e Percentage of p-RIP3-positive osteocytes on the compression and tension sides of WT and TNFRsKO mice (n = 6). f Immunofluorescence staining for p-MLKL (red) and DAPI staining (blue) of ROIs on the compression and tension sides in WT and TNFRsKO mice. Scale bar = 50 μm. g High-magnification images of the compression side on day 6 after OTM in WT mice. White arrows indicate p-MLKL-positive osteocytes. Scale bar = 50 μm. h Percentage of p-MLKL-positive osteocytes on the compression and tension sides in WT and TNFRsKO mice (n = 6). The dotted line indicates the border between the alveolar bone and periodontal ligament. a, alveolar bone; p, periodontal ligament. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Full size image

TNF-α induces necroptosis of primary osteocytes

To isolate primary osteocytes with high purity, Topaz-positive cells were collected from Dmp1-Topaz mice using fluorescence-activated cell sorting (FACS) (Fig. 4a). Topaz-positive cells expressed green fluorescent protein and had a stellate-shaped morphology with extended dendrites, characteristic of osteocytes (Fig. 4b). In our previous study, we showed that Topaz-positive cells express osteocyte-specific genes such as Dmp1 and SOST.12,29 Therefore, Topaz-positive osteocytes were used as primary osteocytes in the current study. To characterize TNF-α-induced necroptosis, primary osteocytes were stimulated with TNF-α, SM-164, and zVAD (TSZ), a well-established method for inducing necroptosis30, for up to 12 h. Immunofluorescence staining confirmed the presence of p-RIP3-positive osteocytes after the addition of TSZ (Fig. 4c). Nec-1 (a RIP1 inhibitor) and GSK872 (a RIP3 inhibitor) inhibited RIP3 phosphorylation, whereas necrosulfonamide (NSA) (an MLKL inhibitor) did not (Fig. 4d). The percentage of p-RIP3-positive osteocytes increased at approximately 6 h after TSZ treatment (Fig. 4e). The percentage of p-RIP3-positive osteocytes was decreased by Nec-1 and GSK872 treatment (Fig. 4f). Furthermore, we examined MLKL phosphorylation in primary osteocytes. We found that p-MLKL-positive osteocytes temporarily increased 6 h after the addition of TSZ (Fig. 4g). MLKL phosphorylation was inhibited by Nec-1, GSK872, and NSA (Fig. 4h). The percentage of p-MLKL-positive osteocytes was highest 6 h after the addition of TSZ (Fig. 4i). Treatment with Nec-1, GSK872, and NSA decreased the percentage of p-MLKL-positive osteocytes (Fig. 4j).

Fig. 4
figure 4figure 4

Necroptosis of primary osteocytes induced by tumor necrosis factor-α (TNF-α) stimulation. a Isolation of Topaz-positive cells with fluorescence-activated cell sorting (FACS). b Morphology of Topaz-positive cells under a fluorescence microscope. Scale bar = 50 μm. c Immunofluorescence staining for phosphorylated (p)-RIP3 (red) and DAPI staining (blue) in primary osteocytes treated with TNF-α, SM-164, and zVAD (TSZ) for the indicated time. White arrows indicate p-RIP3-positive osteocytes. Scale bar = 100 μm. d Immunofluorescence staining for p-RIP3 (red) and DAPI staining (blue) in primary osteocytes treated with TSZ, Nec-1, GSK872, and NSA for 6 h. White arrows indicate p-RIP3-positive osteocytes. Scale bar = 100 μm. e, f Percentage of p-RIP3-positive osteocytes (n = 3). g Immunofluorescence staining for p-MLKL (red) and DAPI staining (blue) in primary osteocytes treated with TSZ for the indicated time. White arrows indicate p-MLKL-positive osteocytes. Scale bar = 100 μm. h Immunofluorescence staining for p-MLKL (red) and DAPI staining (blue) in primary osteocytes treated with TSZ, Nec-1, GSK872, and NSA for 6 h. White arrows indicate p-MLKL-positive osteocytes. Scale bar = 100 μm. i, j Percentage of p-MLKL-positive osteocytes (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001.

Full size image

TNF-α-induced osteocyte necroptosis enhances osteoclastogenesis

We hypothesized that factors, including DAMPs from osteocyte necroptosis, are involved in osteoclastogenesis. To confirm our hypothesis, osteoclast precursors obtained from TNFRsKO mice were cultured in medium conditioned by necroptotic osteocytes with M-CSF and RANKL (Fig. 5a). Osteoclast precursors were generated from TNFRsKO mice to eliminate the effects of TNF-α on osteoclastogenesis because the osteocyte-conditioned medium contained TNF-α to induce osteocyte necroptosis. The conditioned medium from living osteocytes had no effect on osteoclastogenesis, whereas the conditioned medium from osteocytes undergoing necroptosis enhanced osteoclastogenesis (Fig. 5b, c). This result indicates that factors released during TNF-α-induced osteocyte necroptosis enhance osteoclastogenesis, and these factors may include DAMPs that promote osteoclastogenesis.

Fig. 5
figure 5

Tumor necrosis factor-α (TNF-α)-induced osteocyte necroptosis enhances osteoclastogenesis. a Schematic diagram showing the treatment of osteoclast precursors derived from TNF receptor type 1- and 2-deficient (TNFRsKO) mice with medium conditioned by necroptotic primary osteocytes. Representative images (b) and quantification of TRAP-positive multinucleated (≥ 3 nuclei) cells (c) (n = 4). TSZ, TNF-α (T) + SM-164 (S) + zVAD (Z); c.m., conditioned medium. Scale bar = 200 μm. **p < 0.01, ***p < 0.001, ****p < 0.0001.

Full size image

Discussion

Osteocytes are cells embedded in bone tissue and function as key components of bone remodeling by regulating osteoclasts and osteoblasts via signaling molecules such as the RANKL/OPG and Sost/Dkk1/Wnt axes7. In osteocyte-ablated mice exposed to OTM, the number of osteoclasts and the resorption of alveolar bone are significantly reduced on the compression side10. Immunohistochemical staining has shown that osteocytes express RANKL on the compression side during OTM13. Reduced tooth movement and decreased osteoclast formation in osteocyte-specific RANKL-deficient OTM mice demonstrated the crucial role of osteocyte-derived RANKL11. During OTM, osteocytes also promote osteoclastogenesis via autophagy-mediated RANKL secretion on the compression side31. As highly sensitive mechanosensory cells, osteocytes play a central role in coordinating bone remodeling processes32. These studies highlight the importance of osteocytes in the precise regulation of osteoclastogenesis during OTM. Further investigations on the role of osteocytes are required to elucidate the detailed mechanisms of tooth movement. In this study, we demonstrated that osteocyte necroptosis enhances osteoclastogenesis during OTM, providing novel concepts for biological responses to control tooth movement.

Osteocyte death can promote osteoclast-mediated bone resorption. Targeted ablation of osteocytes with diphtheria toxin in mice results in an osteoporotic bone phenotype33. A previous study showed that during OTM, osteocytes in the alveolar bone undergo cell death on the compression side25. Thus, osteocyte death due to orthodontic forces may promote alveolar bone resorption on the compression side. In the present study, our time-course analysis of OTM revealed that osteocyte death was observed from day 2, whereas osteoclast formation occurred from day 6. Thus, when orthodontic forces are applied to the tooth, osteocyte death occurs first, followed by the appearance of osteoclasts. Additionally, the percentage of osteocyte death had markedly decreased on day 12. This reduction is likely due to osteoclastic resorption of alveolar bone that contained dying osteocytes and empty lacunae, which may have led to the removal of these cells from the bone tissue. The patterns of dying osteocytes and empty lacunae were mainly consistent with our previous results and those of other previous reports25,26,27. Moreover, our TEM analysis morphologically identified various types of osteocyte death. In addition to apoptotic osteocytes with nuclear fragmentation and condensation, necrotic osteocytes with ruptured plasma membranes and empty lacunae were observed. Apoptosis is considered a weak inducer of inflammation and is non-immunogenic because the cell membrane is not ruptured but instead fragmented to form apoptotic bodies21. Therefore, necrotic rather than apoptotic osteocytes may be involved in osteoclastogenesis and alveolar bone resorption on the compression side during OTM.

TNF-α is a proinflammatory cytokine that is a key molecule for inflammatory osteoclastogenesis and bone destruction34. Our research group has found that TNF-α is an important cytokine for tooth movement, as the number of osteoclasts and the distance of tooth movement were reduced in TNFRsKO OTM mice compared to WT OTM mice6. TNF-α was also found to induce inflammatory responses by mediating various types of cell death35. In the present study, the percentages of dying osteocytes and empty lacunae were significantly reduced in TNFRsKO OTM mice. These findings suggest that TNF-α is related to osteocyte death on the compression side during OTM. Necroptosis is a type of TNF-α-induced cell death that can be confirmed by the phosphorylation of RIP3 and MLKL as specific biomarkers36. In the present study, p-RIP3 and p-MLKL were detected by immunofluorescence staining of osteocytes on the compression side of WT OTM mice, whereas these biomarkers were barely detectable in osteocytes on the compression side of TNFRsKO OTM mice. Morphologically, necroptosis resembles necrosis with ruptured plasma membranes and the release of DAMPs21,37. DAMPs promote sterile inflammation, which is important for tissue repair and regeneration24. Macrophages sense intercellular DAMPs to induce local sterile inflammation, promote cellular self-renewal, and recruit a large number of immune cells38. Collectively, these findings suggest that the DAMPs released by necroptotic osteocytes regulate sterile inflammation during OTM and control osteoclastogenesis on the compression side.

Regarding in vitro experiments with osteocytes, the use of osteocyte-like cell lines or primary osteocytes must be considered. Osteocyte-like cell lines, such as MLO-Y4 and IDG-SW3, are useful for analyzing biological responses and functions39,40. However, because osteocyte-like cell lines behave substantially differently from osteocytes in vivo, the reliability of results from cell line experiments must be carefully evaluated. Although various methods have been devised to isolate primary osteocytes, the low purity of isolated osteocytes remains an issue41. To isolate primary osteocytes with high purity, a new method was established using FACS from mice with osteocyte-specific expression of green fluorescent protein9. We successfully isolated primary osteocytes with high purity from Dmp1-Topaz mice using this method, with minor modifications12,29. In the current study, we induced necroptosis with TNF-α, SM-164, and zVAD (TSZ) in these primary osteocytes. Immunofluorescence staining revealed that the necroptosis biomarkers p-RIP3 and p-MLKL were detected approximately 6 h after TSZ stimulation. To verify necroptosis in primary osteocytes, the RIP1 inhibitor Nec-1, the RIP3 inhibitor GSK872, and the MLKL inhibitor NSA were used. Nec-1 and GSK872 inhibited RIP3 and MLKL phosphorylation. NSA did not inhibit RIP3 phosphorylation but inhibited MLKL phosphorylation because MLKL activation is downstream of RIP3 phosphorylation in the necroptotic signaling pathway. These data suggest that the three inhibitors suppressed necroptosis in primary osteocytes. Similar to our findings in primary osteocytes, TNF-α-induced necroptosis in MLO-Y4 cells is inhibited by Nec-142. Our results showed that TNF-α can induce necroptosis in osteocytes which may release intracellular contents as DAMPs.

DAMPs are endogenous molecules produced during cellular damage or stress that play an important role in inflammation and inflammatory diseases43. It has been previously suggested that DAMPs released during osteocyte necrosis reach the bone surface through canaliculi, thereby promoting there osteoclastogenesis and bone resorption44. In bone diseases associated with osteocyte death, such as bone fracture, arthritis, and osteonecrosis, DAMPs released from necrotic osteocytes induce osteoclast-mediated bone loss45. In the present study, we cultured osteoclast precursors in conditioned medium from necroptotic osteocytes to assess the effect of DAMPs on osteoclastogenesis. To eliminate the effects of TNF-α on osteoclastogenesis, TNFRsKO mice-derived osteoclast precursors were used in this experiment. As expected, medium conditioned by necroptotic osteocytes enhanced osteoclastogenesis, suggesting that factors, including DAMPs, released during osteocyte necroptosis promote osteoclastogenesis.

A major limitation of this study is that we were unable to confirm that DAMPs are the strongest factors that enhance osteoclastogenesis following osteocyte necroptosis. Osteocytes escaping necroptosis may release other factors promoting osteoclastogenesis by TNF-α stimulation. The expression of cytokines such as CXCL8 and CXCL1 is upregulated in a cell-autonomous manner during necroptosis30. Additionally, we were unable to identify the DAMPs released during osteocyte necroptosis-regulated osteoclastogenesis. Some DAMPs have proinflammatory effects, whereas others have anti-inflammatory effects43. For example, upon cell death, the DAMP prostaglandin E2 negatively regulates the immune response46. Moreover, we reported that IL-33, one of the DAMPs, inhibits TNF-α-induced osteoclastogenesis and bone resorption47. To identify which DAMPs released from necroptotic osteocytes enhance osteoclastogenesis, comprehensive approaches such as proteomics are required. However, our method of isolating primary osteocytes by FACS resulted in low cellular yields, limiting the experiments and analyses that could be performed.

In summary, our data indicate that TNF-α induces osteocyte necroptosis on the compression side during OTM. Factors, including DAMPs, released from osteocytes undergoing TNF-α-induced necroptosis enhance osteoclastogenesis and contribute to alveolar bone resorption during OTM (Fig. 6). Because the regulation of inflammatory responses by DAMPs is important for bone remodeling, targeting the TNF signaling pathway and osteocyte necroptosis might be a novel treatment strategy for controlling tooth movement during orthodontic treatment.

Fig. 6
figure 6

Schematic diagram explaining the role of osteocyte necroptosis on the compression side during orthodontic tooth movement (OTM).

Full size image

Materials and methods

Mice

This study was approved by the Institutional Animal Care and Use Committee of the Tohoku University Environmental & Safety Committee (Approval Number: 2018DnA-028-06). All procedures were conducted in accordance with the Regulations for Animal Experiments and Related Activities at Tohoku University. We also complied with ARRIVE guidelines. All mice were housed in specific pathogen-free conditions under a 12-hour light/dark cycle with ad libitum access to feed (Labo MR Stock, Nosan Corporation, Kanagawa, Japan). The mouse OTM model was provided with powdered feed (CE-2, CLEA Japan, Tokyo, Japan) to make it easier to feed even with an attached orthodontic appliance. B6;129 S-Tnfrsf1a1 (p55, TNF receptor type 1-deficient) Tnfrsf1btm1Imx/J (p75, TNF receptor type 2-deficient) mice and C57BL/6-Tg (Dmp1-Topaz) 1lkal/J mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). C57BL/6J (WT) mice were purchased from CLEA Japan.

Reagents and antibodies

Recombinant mouse TNF-α was purchased from R&D Systems (Minneapolis, MN, USA). SM-164 and z-VAD were purchased from Selleck Chemicals (Houston, TX, USA). Nec-1, GSK872, and NSA were purchased from Sigma-Aldrich (St. Louis, MO, USA). Recombinant mouse RANKL was purchased from PeproTech (Rocky Hill, NJ, USA). Recombinant mouse M-CSF was obtained using the M-CSF-expressing CMG14-12 cell line48. The following antibodies were used in this study: phospho-RIP3 rabbit mAb (#91702), phospho-MLKL rabbit mAb (#37333), rabbit mAb IgG Isotype Control (#3900), Alexa Fluor 594 anti-rabbit IgG (#8889), and Alexa Fluor 555 anti-rabbit IgG (#4413; all Cell Signaling Technology, Danvers, MA, USA).

Orthodontic tooth movement

A mouse model of OTM using a Ni-Ti closed-coil spring was established as previously described6. Following the anesthesia via intraperitoneal injection of medetomidine, midazolam, and butorphanol of 8-week-old male WT and TNFRsKO mice, a Ni-Ti closed-coil spring (TOMY, Tokyo, Japan) was attached between the maxillary left first molar and incisor (Fig. 1a). To tightly fix the orthodontic appliance, a hole was created in the alveolar bone near the incisor by drilling with a round carbide bur; then the appliance was ligated with 0.1 mm stainless steel wire (Unique Medical, Tokyo, Japan) through this hole. According to the manufacturer’s instructions, a 10 g force was applied to the maxillary left first molar for mesial movement. Tooth movement was carried out for up to 12 days.

Distance of tooth movement

After tooth movement was completed at seven time points (days 0, 2, 4, 6, 8, 10, and 12), the mice were anesthetized via intraperitoneal injection of medetomidine, midazolam, and butorphanol, subsequently euthanized by cervical dislocation, and then the Ni-Ti closed-coil spring was removed. Dental impressions of the maxillary teeth were obtained using hydrophilic vinyl polysiloxane impression material (EXAMIXFINE, GC, Tokyo, Japan). The maxillary teeth and dental impressions were observed under a stereomicroscope (Leica M165 FC; Leica Microsystems, Wetzlar, Germany), and the distance of tooth movement was measured using cellSens imaging software (Olympus, Tokyo, Japan). The central fossae of the first and second molars in the dental impression were connected by a line, and the gap between the distal surface of the first molar and the mesial surface of the second molar was defined as the distance of tooth movement (Supplementary Fig. S1).

Histological staining

The dissected maxillae were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 24 h and decalcified with 14% ethylenediaminetetraacetic acid (EDTA, pH 7.2) for 1 month. After dehydration using an automatic tissue processor (Leica TP1020; Leica Biosystems, Wetzlar, Germany), the maxillary bone tissues were embedded in paraffin. Tissues were sectioned in the horizontal plane at a thickness of 4 μm. Histological analysis was performed using alveolar bone around the distobuccal root of the first molar as the region of interest (ROI; Supplementary Fig. S2a). To confirm osteoclast formation on the compression side, TRAP was stained at five section levels (100, 140, 180, 220, and 260 μm from the branch point of tooth roots) using a TRAP/ALP stain kit (Fujifilm, Osaka, Japan; Supplementary Fig. S2b). Hematoxylin was used for nuclear counterstaining. To evaluate osteocyte death, H&E staining was performed at sections approximately 150 μm from the branch point of the tooth roots. Rectangles of 400 μm × 200 μm on the compression and tension sides adjacent to the periodontal ligament were set up in the ROI to evaluate osteocyte death. These sections were observed under a light microscope (Leica DMRBE, Leica Microsystems).

Transmission electron microscopy (TEM)

After decalcification of the maxillae with 14% EDTA, the specimens were trimmed into small pieces containing only the molar region and post-fixed with 1% osmium tetroxide in PBS for 2 h. The specimens were dehydrated in graded concentrations of ethanol and propylene oxide and embedded in Epon 812 resin (TAAB Laboratories Equipment, Berks, UK). Semi-thin sections were prepared using an ultramicrotome (Leica EM UC7, Leica Microsystems). Semi-thin sections were cut with a glass knife in the horizontal plane until approximately 150 μm from the branch point of tooth roots. The observation area was confirmed by toluidine blue staining of the semi-thin sections. Thin sections were cut using diamond knives (Syntek, Tokyo, Japan) to a thickness of 70 nm and placed on reticular copper grids (JEOL, Tokyo, Japan). For TEM, sections were double-stained with uranyl acetate and lead citrate. These alveolar bone sections were imaged using a transmission electron microscope (JEM-1400, JEOL) at 80 kV.

Immunofluorescence staining of histological sections

To detect osteocyte necroptosis during OTM, immunofluorescence staining was performed using sections at approximately 150 µm from the branch point of the tooth roots. The paraffin sections were deparaffinized and rehydrated in xylene and graded concentrations of ethanol. The sections were treated in 0.2% Triton X-100 (Sigma-Aldrich) for 15 min at room temperature. After blocking with 10% goat serum (Abcam, Cambridge, UK) in PBS for 1 h at room temperature, the sections were incubated with phospho-RIP3 rabbit mAb (1:400 dilution) or phospho-MLKL rabbit mAb (1:1600 dilution) at 4°C overnight. As an isotype control, rabbit mAb IgG was diluted to the same concentration as the specific antibody. The sections were subsequently incubated with Alexa Fluor 594 anti-rabbit IgG (1:500 dilution) for 1 h at room temperature and mounted using ProLong™ Gold Antifade Mountant with 4’, 6-diamidino-2’-phenylindole (DAPI; Thermo Fisher Scientific, Waltham, MA, USA). Rectangles of 400 μm × 200 μm on the compression and tension sides adjacent to the periodontal ligament were set up in the ROI for evaluating osteocyte necroptosis (Supplementary Fig. S2c). The samples were observed under a confocal laser scanning microscope (LSM 780; Zeiss, Oberkochen, Germany).

Isolation of primary osteocytes

Primary osteocytes were isolated as described previously12,29. Briefly, calvariae were aseptically dissected from newborn Dmp1-Topaz mice whose osteocytes express the green fluorescent protein. The calvariae were incubated using 0.2% collagenase solution (Fujifilm) for 20 min and 5 mM EDTA solution for 15 min at 37 °C with agitation, then each digestion solution was collected as fraction F. Incubation and fractionation were performed in the order collagenase (F1), EDTA (F2), collagenase (F3), collagenase (F4), and EDTA (F5). F2–F5 were cultured in alpha-modified minimal essential medium (α-MEM, Fujifilm) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin (P/S) at 37 °C and 5% CO2. After 24 h, non-attached cells were removed, and adherent cells were detached from the culture dishes with trypsin-EDTA (Thermo Fisher Scientific). The cell suspension was strained in preparation for FACS (FACSAria II, BD Biosciences, Franklin Lakes, NJ, USA) and analyzed using FlowJo software (BD Biosciences). Isolated Topaz-positive cells were used as primary osteocytes.

Immunofluorescence in primary osteocytes

Primary osteocytes were cultured in α-MEM supplemented with 10% FBS and 1% P/S at 37 °C and 5% CO2. Osteocytes were seeded at a density of 5.0 × 103 cells/well into multi-well glass bottom dishes (MATSUNAMI, Osaka, Japan). To induce necroptosis, osteocytes were treated with TNF-α (100 ng/mL), SM-164 (100 nM), and zVAD (20 µM; combined referred to as TSZ) for up to 12 h. The RIP1 inhibitor Nec-1 (20 µM), RIP3 inhibitor GSK872 (20 µM), and MLKL inhibitor NSA (20 µM) were used for necroptosis inhibition. The osteocytes were fixed with 4% paraformaldehyde for 15 min and permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) in PBS for 15 min at room temperature. After blocking with 1% bovine serum albumin in PBS for 1 h at room temperature, osteocytes were incubated with phospho-RIP3 rabbit mAb (1:400 dilution) or phospho-MLKL rabbit mAb (1:1600 dilution) at 4 °C overnight. Rabbit mAb IgG was used as the isotype control. Antibodies were detected using Alexa Fluor 555 anti-rabbit IgG (1:500 dilution) for 1 h at room temperature. DAPI (Thermo Fisher Scientific) was used for nuclear counterstaining. The images were acquired using a confocal laser scanning microscope (LSM 780, Zeiss).

Osteoclast differentiation

Bone marrow cells were isolated from 8-week-old male TNFRsKO mice by flushing their femora and tibiae. Bone marrow cells were seeded into Petri dishes and incubated in α-MEM containing 10% FBS, 1% P/S, and M-CSF (100 ng/mL) at 37 °C and 5% CO2 to obtain bone marrow macrophages as osteoclast precursors. After 3 days, non-attached cells were removed, and osteoclast precursors were collected from the culture dishes using trypsin-EDTA (Thermo Fisher Scientific). Subsequently, osteoclast precursors were cultured in 96-well plates at a density of 5.0 × 104 cells/well in osteoclast medium containing M-CSF (100 ng/mL) and RANKL (100 ng/mL). To examine the effects of factors released from osteocyte necroptosis on osteoclastogenesis, primary osteocytes were incubated with TNF-α (100 ng/mL), SM-164 (100 nM), and zVAD (20 µM; TSZ) at a density of 1.0 × 105 cells/well in 12-well plates for 24 h, and the supernatants were collected afterward. Subsequently, BMMs were stimulated with this supernatant as a conditioned medium. The medium was changed every 2 days. After 4 days, the differentiated osteoclasts were fixed with 4% paraformaldehyde for 15 min and permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) in PBS for 15 min at room temperature. A TRAP/ALP staining kit (Fujifilm) was used to visualize osteoclasts. TRAP-positive cells with three or more nuclei were considered osteoclasts, and the number of osteoclasts was counted under a light microscope (IX71, Olympus).

Statistical analysis

One-way analysis of variance (ANOVA) with the Tukey–Kramer test was used for post-hoc analyses. All data were analyzed using JMP Pro 17 software (JMP Statistical Discovery, Cary, NC, USA). All data are expressed as the mean ± standard error of the mean (SEM). Statistical significance was set at p < 0.05.