Ti particle preparation

Pure titanium (Ti) particles were obtained from Johnson Matthey (catalog #00681, Ward Hill). Manufacturer claims 90% of the Ti particles were less than 10.0 mm in diameter, with the average being 5.34 mm. In the earlier studies, the efficacy of these particles was demonstrated [15].

The particles were developed following the reported procedure of Chen et al. Furthermore, a QCL-1000 kit was employed to make sure that no endotoxin was present (BioWhittaker) [8]. Endotoxin-free Ti particles (100 mg/mL) were prepared and stored in PBS at 4 °C. For experiments on cells and animals, the Ti particle suspension was diluted to the desired concentration.

SOST adeno-associated virus vector for in vivo experiments and development of a cranial osteolysis model

The approval for animal utility in this study was provided by the Second Affiliated Hospital of Soochow University’s ethics committee. Herein, 40 C57BL/6 female mice aged 8 weeks and weighing 20 to 25 g were selected. Four groups of mice (i.e., (a) Sham group, (b) Ti (Ti) group, (c) SOST-RNAi (SOST-L) group, and (d) SOST-RNAi + Ti (SOST-L + Ti) group) were randomly assigned to each group. Adeno-associated virus vector (50 μL) was subcutaneously injected in the center of the skull of each mouse. The viral vectors injected into the SOST-L and SOST-L + Ti groups carried 1012 replicas of the siRNA SOST gene. One week later, mice were anesthetized with 50 mg/kg pentobarbital (1%) intraperitoneally, and a 10-mm sagittal incision was made in the skull’s center. The sham and SOST-L groups received an equal volume of PBS injection, while the mice in the Ti group and SOST-L + Ti group received 50 μL of Ti particles in PBS suspension (100 mg/mL) under the periosteum of the sagittal suture of the skull. After 2 weeks, the skull samples of the mice were taken for radiological, histological, and immunohistochemical examinations.

Micro-CT Analysis

The cranial specimens were examined via micro-CT (Scanoco) at 10 μm/layer after being preserved in 4% paraformaldehyde (n = 5 per group). The X-ray settings were 70 kV and 114 μA. The following data was obtained using micro-CT on a circular region of interest (ROI) with a 6-mm diameter in the middle of each skull: bone volume (BV), bone mineral density (BMD), bone volume/tissue volume (BV/TV), and the number of perforations.

Histological and immunohistochemical analysis

Skull samples (n = 5 per group) were decalcified in 10% ethylenediaminetetraacetic acid (EDTA, Biosharp) for a month after being soaked in formalin for 2 days. The specimens were then trimmed, mainly to preserve the Ti particle-covered parietal and anterior bones of the mice models. After dehydration and paraffin embedding, the cranial specimens were proceeded for hematoxylin and eosin (H&E) staining and cut into 5-μm sections. Images of the H&E staining results were obtained with the skull midline suture located in its center at 20 × magnification. The surface area of bone tissue (BS mm2) and eroded bone surface area (ES mm2) were calculated using Image Pro-Plus 6.0 software.

Immunohistochemical staining technique was occurred to detect the appearance of the following proteins: sclerostin (Abcam, ab86465, 1:200 dilution), β-catenin (Proteintech, ab66379, 1:200 dilution), OPG (Abcam, ab183910, 1:500 dilution), and RANKL (Abcam, ab45039, 1:200 dilution). Furthermore, osteocytes were labeled with sclerostin, OPG, and RANKL, while osteoclasts were labeled with TRAP staining (Cosmo Bio Co. Ltd). First, skull specimens were dewaxed, gradient hydrated and antigen extracted. Primary antibodies were overnight incubated at 4 °C. Afterward, the tissue slices were first rinsed and incubated with a buffer containing secondary antibody for 35 min at room temperature. Each tissue sample was photographed about the median sagittal suture of the skull under a 20 × and 40 × microscope, respectively. Cells that stained brown were immunoreactive positive cells and positively stained cells were counted in three sections from every group (n = 5) by two independent observers under a 20 × microscope.

Overexpression and silencing of SOST

The osteocytes MLO-Y4 cell lines were taken from the Chinese Academy of Sciences (CAS) cell bank and seeded in α-MEM containing 10% FBS (Gibco). Next, this culture mixture was incubated at 37 °C with a continuous supply of 5% CO2. Short-hairpin shRNA lentiviral particles taken for the Sigma-Aldrich Chemical Co. were used according to the provided protocol of the manufacturer to silence a portion of MLO-Y4 osteocytes. Briefly, following a GenBank search for the SOST gene sequence, the SOST shRNA interference sequence was designed using Invitrogen’s RNA interference and synthesized manually by Sangon Biotech. MLO-Y4 osteocytes were transfected via lentivirus units carrying disordered or SOST-specific shRNAs. Using Western blotting, the expression of the sclerostin protein was measured separately to see how well shRNA worked. The most efficient SOST-shRNA and SOST overexpressed sequences in MLO-Y4 osteocytes were screened and the detail sequences were in supplement Table 1. MLO-Y4 cells that expressed different levels of SOST were used in the next in vitro experiments.

Cell culture and treatments

After 24 h after seeding, MLO-Y4 cells were treated with Ti particles at doses of 0, 0.1, and 1.0 mg/ml (control). The medium was replaced by adding fresh medium every 3 days. The day the Ti particles were introduced to the cells is known as Day 0. The MLO-Y4 osteocytes were then divided into four groups and inoculated in 6-well Corning plates: the control group, the Ti group, the SOST-L + Ti group, and the SOST-H group. For 48 h, Ti particles at 1.0 mg/ml were added to MLO-Y4 cells in the Ti and SOST-L + Ti groups. All groups had their initial culture medium for osteoclasts collected, centrifuged to remove any Ti particles or other impurities suspended therein, and then replaced every 2 days with fresh medium.

Primary bone marrow-derived monocyte macrophage (BMM) extraction then induction of differentiation had BMMs remained extracted from the bone marrow of tibias and femurs of 4-week-old female mice C57BL/6. The cells were seeded in α-MEM containing 10% FBS and were incubated overnight at 37 °C with a continuous supply of 5% CO2. Next, the supernatant was then centrifuged to obtain suspension cells, and these cell suspensions were resuspended in 10% FBS containing α-MEM and M-CSF solution (30 ng/ml) (R&D systems). After 3 days, the adherent cells in the above culture dishes were collected, transferred to 6-well plates at a density of 4 × 105/well, and resuspended in α-MEM containing 50 ng/ml of RANKL (R&D systems), 30 ng/ml of M-CSF, and 10% FBS in α-MEM to stimulate their differentiation to osteoclasts. Cells after culturing for 3 days with TRAP staining and a nuclear number ≥ 3 were considered mature osteoclasts. Additionally, the characteristic markers of osteoclasts were identified (Supplement 1B-C).

In the indirect cell co-culture model, BMMs were induced to develop into osteoclasts using the method described above. Differently, when BMMs were transferred to 6-well plates, the medium that was collected from each group of MLO-Y4 osteocytes cultured, as described above, was added, with the addition of RANKL (50 ng/ml) and M-CSF (30 ng/ml). After the development of multinucleated cells, subsequent experiments were conducted.

Immunofluorescence staining

After being rinsed three times in PBS, MLO-Y4 osteocytes were fixed in 4% paraformaldehyde for 10 min, permeabilized with 0.1% Triton X-100 for 5 min, and incubated in 0.1% PBS-Tween (PBST) containing 5% bovine serum albumin (BSA) for 1 h. Secondary, OPG (1:500 dilution) and RANKL (1:500 dilution) primary antibodies were incubated with MLO-Y4 osteocytes at 4 °C overnight. The cells were washed thricely using PBS, followed by exposure to goat anti-rat (Alexa Fluor® 594, Abcam, ab15016, 1:500 dilution), goat anti-mouse (Alexa Fluor® 647, Abcam, ab150115, 1:500 dilution), and goat anti-rabbit (Alexa Fluor® 488, Abcam, ab150077, 1:500 dilution) secondary antibodies for an hour in the absence of light. Cell nuclei were stained for 5 min with DAPI (1:1000 dilution) after being thricely washed with PBST.

BMMs were inoculated into confocal culture dishes and were induced to differentiate osteoclasts, fix, and break membranes, as described previously. Osteoclasts were incubated with ActinGreenTM488 fluorescent ghost pen cyclic peptide (Invitrogen, R37110, two drops into 1.5 ml PBS) for 30 min in dark at room temperature. After PBS washing, nuclei were stained for 5 min with DAPI (1:1000 dilution). F-actin rings were observed using an immunofluorescence microscope (ZEISS).

ELISA

The levels of OPG, RANKL, IL-6, and TNF-α were evaluated in the culture medium supernatant of each group after 48 h of intervention in MLO-Y4 cells under varied circumstances using ELISA kits (R&D Systems). The O.D. was recorded at 450 nm using a fully automatic micro-enzymatic standard.

Protein expression and western blotting analysis

After being treated with lysis buffer, osteocytes and osteoclasts were rinsed twice with PBS, centrifuged at 15,000 rpm for 15 min, and kept on ice for 20 min. After collecting the supernatant, the protein content was determined using the BCA Protein Assay Kit (Beyotime, P0010). Then, separated 50 μg of each sample by SDS-PAGE (10%) and electroblotted on PVDF membranes (membrane soaking in methanol for 3 min before protein transfer). The membranes were incubated with primary antibodies for sclerostin (Abcam, ab86465, 1:1000 dilution), β-catenin (dephosphorylated form, CST, 8480 T, 1:1000 dilution), OPG (Abcam, ab183910, 1:2000 dilution), RANKL (Abcam, ab45039, 1:1000 dilution), TNF-α (Proteintech, 17590–1-AP, 1:1000 dilution), IL-6 (Abcam, ab9324, 1:2000 dilution), NFATc1 (CST, 5861S, 1:2000 dilution), RANK (Abcam, ab200369, 1:1000 dilution), CTSK (Abcam, ab19027, 1:2000 dilution), and TRAP (Abcam, ab19146, 1:1000 dilution) overnight at 4 °C after being blocked with 5% BSA for an hour at room temperature. The membrane was washed with TBST (tris-buffered saline containing tween), followed by treatment with horseradish peroxide (HRP), goat anti-rabbit IgG, and goat anti-Rat IgG (Multisciences, GRT007, 1:2000 dilution) for 60 min at room temperature. The membranes underwent three additional TBST washes before being exposed to electrochemiluminescence (ECL) and subjected to GIS analysis. In addition, to evaluate the expression of the target protein, we chose β-actin as the internal reference protein.

TRAP staining

Sections of skull specimens were prepared, as described above, and TRAP staining was performed on these sections, using the TRAP staining kit. In vitro, BMMs were inoculated at a concentration of 3 × 105/well on 6-well plates, and then groups of BMMs were induced to differentiate toward osteoclasts, as described previously. TRAP staining and microscopy were performed to visualize the differentiation of these cells. Using Image Pro-Plus 6.0, TRAP-positive cells with 3 + nuclei were identified as osteoclasts, followed by counting them.

Osteoclastic resorption assessment

The function of osteoclasts was assessed by the resorption assay. Briefly, BMMs were inoculated in osteo assay surface plates containing 24-wells (Corning, USA) at a concentration of 10 × 104/well. Then groups of BMMs were induced to differentiate toward osteoclasts as before. After 6 days, trypsin digestion and three PBS washes were performed on the cells. Following that, images were captured with a simple light microscope (Zeiss), and Image Pro-Plus (Version-6.0) was employed to analyze the region of the resorption depression.

Statistical evaluation

For each assay, three separate replicate experiments were conducted. The ANOVA and post hoc multiple comparisons (Tukey Kramer’s post hoc test) were used to identify differences between groups. The data have been presented as mean standard deviation (SD). SPSS version 27.0 was employed to analyze all the data, differences with P < 0.05 were considered significant.