Background
Achieving co-occurrence of tooth movement acceleration and root protection has been a fascinating challenge. This study investigated how interleukin-33 (IL-33)-superoxide dismutase 3 (SOD3) crosstalk affects the alveolar bone and cementum during orthodontic treatment.
Methods
The effects of IL-33-SOD3 signaling on the immortalized murine cementoblast cell line 30 cells, bone marrow stromal cells, and RAW 264.7 cells were evaluated. In vivo, the distance of orthodontic tooth movement and the volume of root resorption were quantified in mice treated with soluble suppression of tumorigenicity 2 (sST2, a decoy receptor for IL-33) and adenovirus-mediated sod3 .
Results
Endogenous SOD3 was found to enhance the osteogenic differentiation and mineralization of cementoblast-like cells, whereas IL-33-inhibited SOD3 expression and the SOD3-mediated promineralization effect on these cells. In bone marrow stromal cells, endogenous SOD3 promoted osteoblast differentiation and mineralization, although to a lesser extent than in the immortalized murine cementoblast cell line 30 cells, with a lack of effect of IL-33 on SOD3 expression. SOD3 attenuated osteoclastic activity in RAW 264.7 cells, to a lesser extent, compared with the IL-33-mediated suppression of osteoclastogenesis. Combined administration of sST2 and adenovirus-mediated sod3 gene transfer resulted in accelerated tooth movement while providing protection to the roots.
Conclusions
The SOD3 signaling cross-talked with IL-33 and differentially regulated cementoblast-like cells, osteoblast progenitors, and osteoclast precursor cells. Combined treatment of sST2 and sod3 gene transfer reorchestrated the IL-33-SOD3 signaling and led to accelerated bone remodeling coupled with reinforced root repair, achieving the co-occurrence of tooth movement acceleration and root protection in mice.
Highlights
-
•
SOD3 signaling interacted with IL-33 to regulate orthodontic tooth movement.
-
•
Soluble suppression of tumorigenicity 2(sSt2) promoted the rate of tooth movement.
-
•
The combination of sSt2 and SOD3 accelerated tooth movement and protected the roots.
Orthodontic tooth movement (OTM) occurs when mechanical loading evokes cellular responses that cause tissue remodeling and spatial reconstruction of the tooth-alveolus complex. ,, In response to mechanical stimuli, the alveolar bone and cementum undergo constant remodeling, the cellular basis of which comprises functional and phenotypic changes of osteoblasts, cementoblasts, osteoclasts and cementoclasts, and their progenitors or precursor cells. ,, Increasing efforts have been devoted to accelerating tooth movement by enhancing osteoclastogenesis, a key rate-limiting factor of this process. ,, However, enhanced osteoclast activity would inevitably incur aggravation of root resorption, which was contrary to doctors’ desires. , Neither local nor targeted drug delivery can easily resolve this contradiction because of the high proximity of alveolar bone and root surfaces, as well as the high homology between osteoclasts and cementoclasts. There is a dilemma that factors accelerating tooth movement aggravate root resorption whereas factors facilitating root protection impair bone resorption and retard tooth movement, posing a fascinating challenge to realize the co-achievement of tooth movement acceleration and root protection.
Because it’s difficult to find a single factor playing contradictory roles between osteogenesis and cementogenesis, the application of combined reagents may provide a solution to address this challenge. Our previous studies had demonstrated that orthodontic load–induced periodontal ligament (PDL) cells abundantly release interleukin-33 (IL-33), which negatively impacts cementum and bone turnover by suppressing cementogenesis and mineralization, as well as inhibiting osteoclastogenesis. , Soluble suppression of tumorigenicity 2 (sST2), a decoy receptor for IL-33, acts as a bait receptor that inhibits the IL-33/ST2L signaling pathway by competing with ST2L for IL-33. , Antagonizing IL-33 signaling might fasten the remodeling processes of cementum and alveolar bone. Nevertheless, only when it is supplemented by another factor to reinforce root protection in the pro-osteoclastogenic context can the IL-33 inhibition therapy become a promising approach to accelerate tooth movement. Cementum undergoes active remodeling, in which root resorption is subsequently repaired by cementoblast-mediated remineralization and cementogenesis. ,, Signals orchestrating the balance between resorption and repair determine the extent of loss of root tissue. Augmenting the cementoblastogenic signaling might counteract the procementoclastic effect and skew the balance towards cementum repair.
To achieve the co-occurrence of tooth movement acceleration and root protection, an ideal supplemental factor should effectively promote cementum repair without excessively enhancing bone remineralization. This underscores the importance of a promineralization factor that differentially regulates osteogenesis and cementogenesis. One of the characteristics that distinguishes cementum from bone is its higher resistance to the compressive stress-induced local hypoxia and oxidative damage. , The endogenous antioxidant system might provide a clue. A proteomic study highlighted the uniqueness of cementum by high expressions of superoxide dismutase (SOD3), an endogenous antioxidant in human and murine tissues.
Considering the IL-33-antagonizing effect of sST2, which might fasten the remodeling processes of cementum and alveolar bone, as well as the promineralizing effect of SOD3, we hypothesized that combined treatment of sST2 and adenovirus-mediated SOD3 overexpression would lead to accelerated bone remodeling coupled with reinforced root repair, achieving the co-occurrence of tooth movement acceleration and root protection. To elucidate the mechanism by which these combined reagents work, the downstream signaling and IL-33-SOD3 crosstalk were to be examined in OTM. SOD3 has been shown to inhibit IL-33/ST2-mediated type II innate lymphoid cell activation in allergic asthma, thus mitigating allergic airway inflammation. However, the effects of IL-33 on SOD3 signaling and the role of IL-33-SOD3 signaling in OTM remain unexplored. Thus, this study aimed to investigate the effect of combined administration of sST2 and SOD3 overexpression on alveolar bone and cementum during OTM and elucidate the mechanism by which IL-33-SOD3 signaling regulates cementoblast-like cells, osteoblast progenitors, and osteoclast precursor cells, respectively.
Material and methods
This study was approved by the ethical committees of West China School of Stomatology, Sichuan University (WCHSIRB-D-2022-368).
Immortalized murine cementoblast cell line 30 ( OCCM-30) cells were cultured in α-MEM medium (Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin-streptomycin (Gibco). For IL-33 treatment, the recombinant mouse IL-33 (rmIL-33; R&D, Minneapolis, Minn) was added to the medium at a concentration of 20 ng/mL. For chemical inhibitor treatment, SCH772984 (10 and 20 μM; Selleck, Houston, Tex) or SP600125 (10 and 20 μM; Selleck) was added to the medium.
Bone marrow stromal cells (BMSCs) were extracted from the femur of 6-week-old C57BL/6J male mice under sterile conditions. The bone marrow was flushed with α-MEM medium (Gibco) containing 10% FBS (Gibco) and 1% penicillin-streptomycin (Gibco) using a 1 mL syringe. The culture medium was renewed halfway through 24 hours of incubation in α-MEM medium (Gibco), and a complete medium change was performed after an additional 48 hours of culture. The BMSCs were passaged once they reached approximately 80% confluence. For IL-33 treatment, rmIL-33 (R&D) was added to the medium at a 20 ng/mL concentration.
The Raw264.7 cells were cultivated in Dulbecco’s modified Eagle medium (DMEM; Gibco) containing 10% FBS (Gibco) and 1% penicillin-streptomycin (Gibco). For IL-33 treatment, rmIL-33 (R&D) was added to the culture medium at a 20 ng/mL concentration.
The OCCM-30 cells and BMSCs were cultured on the BioFlex plates coated with type I collagen (Flexcell, Burlington, NC). Once the cells reached a confluence of 80%, they were loaded with force generated by the FX-5000 system (Flexcell) for 24 hours, with a frequency of 0.5 Hz and a deformation magnitude of 10%. In the IL-33 treatment group, the rmIL-33 was added to the medium at a 20 ng/mL concentration.
The passage 2 BMSCs, OCCM-30 cells, and Raw264.7 cells were cultivated in 6-well plates. Upon reaching 80% confluence, the cells were transfected with siRNA or plasmids. Specifically, 200 μL Opti-MEM, 6 μL EndoFectin, and 6 μL siRNA (GenePharma, Shanghai, China) or 1 μg plasmids (GeneCopoeia, Rockville, Md) were each mixed in 6-well plates for the transfection process. The sequences used were si- ctrl (forward: UUCUCCGAACGUGUCACGUTT; reverse: ACGUGACACGUUCGGAGA) and si- sod3 (forward: CCGGUUGAGAAGAUAGACATT; reverse: ATTUGUCUAUCUUCUCAACCGGTT). After being incubated for 10 minutes, 800 μL medium containing serum was added to each well of the 6-well plates. After an 8-hour cultivation period, the culture medium for BMSCs or OCCM-30 cells was replaced with an osteogenic induction solution containing 1×10 -8 mmol/L dexamethasone (Solarbio, Beijing, China), 0.2 mmol/L ascorbic acid (Solarbio), and 10 mmol/L β-sodium glycerate (Sigma, St. Louis, Mo). For Raw264.7 cells, the culture medium was renewed with DMEM (Gibco) supplemented with 50 ng/mL RANKL (R&D).
The total RNA was isolated using the Trizol reagent (Invitrogen, Carlsbad, Calif) and subsequently tested by Nanodrop (Thermo, Waltham, Mass). The cDNA synthesis was performed using a reverse transcription kit (Takara, Shiga, Japan) according to the instructions. The polymerase chain reaction amplification was conducted on an ABI Q3 thermocycler (Thermo) with a polymerase chain reaction kit (Takara). The data were calculated as 2 -△△t. The primer sequences (Sangon, Shanghai, China) used in this study are listed in the Table .
Table
Primers used in quantitative real-time polymerase chain reaction
| Genes | Forward (5’-3’) | Reverse (5’-3’) |
|---|---|---|
| sod3 | CCTTCTTGTTCTACGGCTTGCAT | TCGCCTATCTTCTCAACCAGG |
| Alkaline phosphatase (alp) | CCAACTCTTTTGTGCCAGAGA | GGCTACATTGGTGTTGAGCTTTT |
| Runt-related transcription factor 2 (runx2) | GCTTCATTCGCCTCACAAA | GCACTCACTGACTCGGTTGG |
| Osteopontin (opn) | CAGGGAGGCAGTGACTCTTC | AGTGTGGAAAGTGTGGCGTT |
| Osteocalcin (ocn) | CTGACCTCACAGATCCCAAGC | TGGTCTGATAGCTCGTCACAAG |
| Osterix (osx) | ATGGCGTCCTCTCTGCTTG | TGAAAGGTCAGCGTATGGCTT |
| Bone sialoprotein (bsp) | ATGGAGACGGCGATAGTTCC | CTAGCTGTTACACCCGAGAGT |
| Nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 1 (nfatc1) | GACCCGGAGTTCGACTTCG | TGACACTAGGGGACACATAACTG |
| Matrix metallopeptidase 9 (mmp9) | CTGGACAGCCAGACACTAAAG | CTCGCGGCAAGTCTTCAGAG |
| c-fos | CGGGTTTCAACGCCGACTA | TTGGCACTAGAGACGGACAGA |
| Cathepsin K (ctsk) | GAAGAAGACTCACCAGAAGCAG | TCCAGGTTATGGGCAGAGATT |
| Tartrate-resistant acid phosphatase (trap) | CACTCCCACCCTGAGATTTGT | CATCGTCTGCACGGTTCTG |
| Glyceraldehyde-3-phosphate dehydrogenase (gapdh) | CAAGTCCCACACAGCAGCTT | AAAGCCGAGCTGCCAGAGTT |
The cells were lysed with protein extraction lysates containing protease inhibitors and phosphatase inhibitors. The protein concentration in the supernatant was measured by a BCA assay kit (Beyotime, Shanghai, China). The proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The voltage of sodium dodecyl sulfate-polyacrylamide gel electrophoresis was initially at 80 V for 30 minutes, followed by 100 V to the end. The membrane transfer was conducted at a current of 200 mA, with the duration adjusted because of the molecular weight. After the transfer, the membranes were blocked with 5% skim milk (Biofroxx, Einhausen, Germany) at room temperature for 1 hour and incubated with specific primary antibodies, GAPDH (SAB, Miami, Fla), SOD3 (Affinity, Wilmington, Del), Tubulin (HuaBio, Zhejiang, China), OPN (1:1000; Proteintech, Wuhan, China), BSP (1:300; Santa Cruz Biotechnology, Dallas, Tex), OSX (1:1000; Abcam, Waltham, Mass), OCN (1:1000; Affinity), alkaline phosphatase (ALP) (1:1000; HuaBio), RUNX2 (1:1000; HuaBio), P-PI3K (1:1000, CST), PI3K (1:1000; CST, Danvers, Mass), P-protein kinase B (AKT) (1:1000; CST), AKT (1:1000; CST), P-c-Jun N-terminal kinase (JNK) (1:1000; Huabio), JNK (1:1000; Huabio), P-extracellular signal-regulated kinase (ERK) (1:1000, Huabio), ERK (1:1000; Huabio), P-p38 mitogen-activated protein kinase (P38) (1:1000; CST), P38 (1:1000; CST), P-glycogen synthase kinase 3 (GSK3) (1:1000; CST), β-catenin (1:1000; Huabio), CTSK (1:1000; Huabio), NFATC1 (1:1000; Thermo), MMP-9 (1:1000; Huabio), and c-Fos (1:1000; Huabio) at 4°C overnight, respectively. The membrane was incubated with horse radish peroxidase conjugated anti-rabbit or anti-mouse IgG (SAB) and subsequently visualized using chemiluminescent substrate (US Everbright Inc, Jiangsu, China). Protein levels were quantified using Image J software (National Institutes of Health, Bethesda, Md).
The cells were cultured in an osteogenic induction solution for 7 days after transfection with either siRNA or plasmids. The cells were fixed with 4% paraformaldehyde at room temperature for 15 minutes and stained with an ALP kit (Beyotime) in the dark at room temperature for 15 minutes according to the instructions. The images were captured by a camera or a microscope.
For the determination of ALP activity, the cells were lysed in protein lysis buffer on ice for 10 minutes. The supernatant was collected after centrifugation at 12000 g at 4°C for 15 minutes. Part of the supernatant was applied for total protein detection by a BCA assay kit (Beyotime), and the other part was used for ALP detection by an ALP assay kit (Beyotime). The results were calculated as the amount of ALP activity per unit of total protein.
After being transfected by either siRNA or plasmids, the cells were cultured in an osteogenic induction medium for 14 days, with the medium refreshed every other day. The cells were fixed with 4% paraformaldehyde at room temperature for 15 minutes and then incubated in a 0.1% alizarin red staining (ARS) solution (pH = 4.3) at room temperature for 10 minutes. The images were acquired by a camera or microscope. For ARS semiquantification, 1 mL of 10% cetylpyridinium chloride was added to each well of the 24-well plates and then incubated at 37°C for 30 minutes. The solution above was transferred to 96-well plates for OD value measurements at a wavelength of 562 nm.
The Raw264.7 cells were cultured in DMEM (Gibco) supplemented with 50 ng/mL RANKL (R&D) for 6 days after transfection with either siRNA or plasmids. The cells were fixed with 4% paraformaldehyde on ice for 15 minutes. For configuring the solution, acetone and absolute ethyl alcohol were added to the wells at a 1:1 ratio and then incubated at–20°C for 1 minute. For configuring the staining working solution (Wako, Osaka, Japan), acid phosphatase substrate solution A, acid phosphatase substrate solution B, and 10x tartaric acid solution were applied to the wells at a 90:1:10 ratio and incubated at 37°C in the dark for 30 minutes. The images were captured by a microscope.
For the tartarate-resistant acid phosphatase (TRAP) activity assay, the cells were lysed in a protein lysis buffer on ice for 10 minutes. The supernatant was collected after centrifugation at 12000 g at 4°C for 15 minutes. A portion of the supernatant was employed for total protein detection by a BCA assay kit (Beyotime), whereas another portion of the supernatant was applied for TRAP detection by a TRAP assay kit (Beyotime). The results were expressed as the amount of TRAP activity per unit of total protein.
The C57BL/6J male mice aged 8 weeks were housed under standard laboratory conditions (22°C, 50% humidity, and 12-hour light/dark cycle) in groups of 5 per cage with ad libitum access to chow and water. No additional environmental enrichment was provided beyond standard nesting materials.
A total of 30 C57BL/6J male mice aged 8 weeks were used for OTM models. The murine OTM models were generated as described previously. After intraperitoneal injection of 0.5% ketamine (Sigma) at a dosage of 20 mL/kg for anesthesia, a nickel-titanium coil spring (Mingxin, Shandong, China) was ligated with a 0.1 mm diameter stainless steel wire (Hongxiang, Zhejiang, China) between the mandibular incisor and maxillary first molar on the right side. The spring was activated to exert a tension force of 30 g. The maxillary left first molar, which was not subjected to tension force, served as a control. The orthodontic force loading was applied for a duration of 14 days. Along with the OTM models, the mice were subjected to additional reagent administration, according to which the mice were divided into 6 experimental groups: PBS (6 μL; Gibco), recombinant sST2 (6 μL, 30 μg/mL; R&D), adenovirus-GFP (Ad-GFP, 6 μL, 1×10 10 pfu; Genechem, Shanghai, China), Ad-SOD3 (6 μL, 1×10 10 pfu; Genechem), Ad-GFP (1×10 10 pfu) plus sST2 (30 μg/mL)(6 μL), and Ad-SOD3 (1×10 10 pfu) plus sST2 (30 μg/mL)(6 μL). Randomization was applied to allocate mice to the 6 experimental groups (5 animals per group) using a computer-generated random number sequence in Microsoft Excel (RAND function; Microsoft, Redmond, Wash). The adenoviruses were synthesized by Genechem (Genechem). The reagents were injected into the PDL using a microsyringe with a 30-gauge needle (Hamilton, Bonaduz, Switzerland) at the palatal gingiva of the maxillary right first molar every other day within the 14-day time frame. At the 14th day, the mice were sacrificed, and the maxillae were harvested and fixed with 4% paraformaldehyde (Biosharp, Beijing, China) at 4°C for 24 hours. The samples were used for micro-computed tomography scanning and TRAP staining.
The maxillae were subjected to micro-computed tomography scanning (Scano Medical AG, Bassersdorf, Switzerland) with the following parameters: a voltage of 70 kV, a current of 200 μA, duration of 300 ms, and a resolution of 10 μm. The scanning data were used for 3-dimensional reconstruction, measurements of tooth movement distance, and quantification of root resorption by MIMICS software (version 20.0; Materialise, Leuven, Belgium). Tooth movement distance was measured as the distance between the distal contact point of the crown of the first molar and the mesial contact point of the crown of the second molar. Root resorption was quantified by the volume of pits in the distal root of the first molar. The initial assumption was that the root surface exhibited a smooth convex curvature. The volume of root resorption was determined by calculating the difference between this assumed smooth surface and the actual depressions observed. , The area of study was specifically focused on the distal root of the maxillary first molar.
The maxillae were immersed in 10% ethylene diamine tetra-acetic acid (Biofroxx) for 28 days to facilitate decalcification. The decalcifying solution was refreshed every other day. Subsequently, the maxillae were embedded in paraffin and sectioned at a thickness of 4 μm. The paraffin sections were then heated at 65°C for 2 hours, dewaxed in xylene (Solarbio), and hydrated with a series of gradient alcohols. The sections were stained with a trap kit (Wako). A 200 μL mixture, composed of acid phosphatase substrate solution A, acid phosphatase substrate solution B, and 10x tartaric acid solution, in a ratio of 90:1:10, was added to the sections, which were then incubated at 37°C for 1 hour. After being washed with PBS (Gibco) for 3 times, the cell nuclei were stained by 5% methyl green (Solarbio) at room temperature for 10 minutes. Subsequently, the images were acquired by a microscope (Leica, Wetzlar, Germany).
Statistical analysis
The data were expressed as mean ± standard deviation. Data analysis was performed using GraphPad Prism (version 9.0; GraphPad Software Inc, La Jolla, Calif). An independent-samples t test was used for comparison between 2 groups. One-way analysis of variance was employed for comparisons among groups. Multiple comparisons between pairs were conducted using the Tukey honest significant difference test. P <0.05 was considered statistically significant.
Results
Endogenous SOD3 enhanced osteogenic differentiation and mineralization of cementoblast-like cells. Cementoblast-like (OCCM-30) cells had increased expressions of SOD3 in osteogenic induction cultures, accompanied by upregulated levels of osteopontin (OPN), bone sialoprotein (BSP), osterix (OSX), and osteocalcin (OCN) proteins ( Fig 1 , A ). siRNA-mediated knockdown of SOD3 remarkably suppressed the osteogenic phenotypes and mineralization of cultured OCCM-30 cells by reducing the expressions of ALP, runt-related transcription factor 2 (RUNX2), OPN, OCN, BSP, and OSX ( Fig 1 , B-D ). Conversely, SOD3 overexpression enhanced osteogenic differentiation and mineralization in OCCM-30 cell cultures ( Fig 1 , E-G ).
SOD3 promoted cementoblastogenic differentiation and mineralization of OCCM-30 cells: A, Representative images of western blot showing SOD3 and cementoblastogenic protein expressions by OCCM-30 cells cultured in an osteogenic induction medium for 1, 3, 5, 7, and 14 days; B, The protein and mRNA levels (n = 4) of SOD3 in OCCM-30 cells at 48 hours after transfection with si-sod3; C, The protein and mRNA levels (n = 3) of cementoblastogenesis-related markers in OCCM-30 cells treated with control ( si-ctrl ) or si-sod3 ; D, Representative images of ALP staining and ALP activity analysis (n = 4) and ARS staining and semiquantitative analysis (n = 6) for OCCM-30 cell cultures treated with si-ctrl or si-sod3 . Scale bar: 200 μm; E, The protein and mRNA levels (n = 4) of SOD3 in OCCM-30 cells at 48 hours after transfection with control ( plasmid-ctrl ) or sod3-overexpression plasmids ( plasmid-sod3 ); F, The protein and mRNA levels (n = 6) of cementoblastogenesis-related markers in OCCM-30 cells transfected with plasmid-ctrl or plasmid-sod3 ; G, Representative images of ALP staining and ALP activity analysis (n = 4) and ARS staining and semiquantitative analysis (n = 3) in OCCM-30 cell cultures treated with plasmid-ctrl or plasmid-sod3 . Scale bar: 200 μm.
SOD3 promoted cementogenesis in OCCM-30 cells via ERK-dependent and SOD3-ERK positive feedback mechanisms.
The PI3K, AKT, JNK, ERK, and P38 signalings were suppressed by SOD3 knockdown and activated by SOD3 overexpression ( Fig 2 , A and Supplementary Fig 1 ). Among these signalings, JNK and ERK were mostly affected. Administration of the ERK inhibitor SCH attenuated the osteogenic phenotypes of OCCM-30 cells, exhibiting a more pronounced inhibitory effect compared with that induced by SOD3 siRNA. In contrast, the osteogenesis-inhibiting effects were comparatively weaker in the OCCM-30 cell cultures treated with JNK inhibitor SP, with significant suppression of RUNX2 expression but minimal impact on other osteogenic phenotypes ( Fig 2 , B ). It is of interest to note that SOD3 expression was inhibited by ERK inhibitor SCH772984 (SCH), an ERK1/2 inhibitor but not by JNK inhibitor SP600125 (SP) ( Fig 2 , C ). The SOD3-induced ERK activation and ERK-mediated SOD3 expression suggested a SOD3-ERK positive feedback loop, augmenting the pro-osteogenic signaling.
Promotion of cementoblastogenic differentiation from OCCM-30 cells via ERK-dependent mechanism and a SOD3-ERK positive feedback loop: A, The activation of PI3K-AKT and mitogen-activated protein kinase (MAPK) signaling pathways was detected in OCCM-30 cells with knockdown or overexpression of sod3 ; B, The protein levels of cementoblastogenesis-related markers in OCCM-30 cells at 48 hours after administration of SCH772984 (at concentrations of 10 or 20 μM) or SP600125 (at concentrations of 10 or 20 μM); C, The protein levels of cementoblastogenesis-related markers in OCCM-30 cells at 48 hours after transfection with plasmid-sod3 in the presence or absence of SCH772984 (10 μM) or SP600125 (10 μM). Data were shown as representative images or mean ± standard deviation. ∗ P <0.05, ∗∗ P <0.01.
IL-33 inhibited SOD3 expression, as well as SOD3-mediated osteogenic differentiation and mineralization of cementoblast-like cells by inhibiting ERK signaling. IL-33 administration inhibited SOD3 expression in OCCM-30 cells ( Fig 3 , A ). In vitro, tensile force loading upregulated SOD3 expressions, accompanied by increased levels of osteogenesis-related genes ( Fig 3 , B ). IL-33 administration inhibited SOD3-mediated expressions of osteogenic markers in OCCM-30 cells, including ALP, RUNX2, OPN, and OSX OCCM-30 ( Fig 3 , C ), and abrogated the promineralizing effect of SOD3 on OCCM-30 cell cultures ( Fig 3 , D ). Examinations on pathways showed that IL-33 activated P38 but inhibited ERK signaling ( Fig 3 , E ), with no significant impact on AKT, PI3K, and JNK signalings ( Supplementary Fig 2 ), suggesting that the IL-33-mediated inhibition of SOD3 is at least partly via inhibiting the ERK signaling ( Fig 3 , F ).
