Cell isolation and culture

C57BL/6 mice aged 8 weeks were sacrificed by cervical dislocation to extract bone marrow macrophages (BMMs). The mouse corpse was disinfected with 75% alcohol, and the femur and tibia were separated, along with the surface muscle tissue. The bone marrow was then flushed into a 10-cm dish. At 37 °C and under 5% CO2 incubation, the cells were cultivated in an α-MEM medium (Gibco-BRL, Gaithersburg, MD, USA) supplemented with 10% FBS (Gibco-BRL), 1% penicillin/streptomycin (1%, Gibco-BRL), and 30 ng/mL M-CSF (R&D, Minneapolis, MN, USA). After 48 h, the medium was changed to eliminate the unattached cells, and the attached cells were used for subsequent experiments. Mouse embryonic fibroblasts (C3H10) were purchased from ATCC (Manassas, VA, USA) and cultured in an incubator (37 °C, 5% CO2). The growth medium was DMEM supplemented with 10% FBS and 1% penicillin/streptomycin.

Cell Counting Kit-8 assay

The CCK-8 test was used to evaluate Tetrandrine’s cytotoxicity on BMMs and C3H10. Briefly, 1 × 105 cells were added to each well of a 96-well plate and cultivated for 48 and 96 h with Tetrandrine (Selleck, Houston, Texas, United States). The CCK8 reagent (10 uL) (Dojindo Molecular Technology, Kumamoto, Japan) was then added to each well and cultured for an hour. The absorbance at 450 nm was measured by a microplate reader (Bio-Tek Instruments, Winooski, VT, USA).

In vitro osteoclastogenesis assay

BMMs (5 × 103) were added to each well of a 96-well plate and incubated overnight. BMMs were then incubated with 30 ng/mL M-CSF and 50 ng/mL RANKL (R&D), with or without Tetrandrine at a certain concentration (0, 0.25, 0.5, and 1 μM). The medium was changed every 2 days for 7 days, and the cells were then rinsed twice with PBS before fixing with 4% paraformaldehyde (PFA) at room temperature for 15 min, and finally stained with tartrate-resistant acid phosphatase (TRAP) (Sigma-Aldrich, St. Louis, MO, USA) for 30 min.

Bone resorption assay

The bottom of a 96-well plate was filled with bovine bone slices (Corning Inc., Corning, NY, USA), and 1 × 104 BMMs were seeded in each well. After 24 h, BMMs were differentiated with 30 ng/mL of M-CSF, 50 ng/mL RANKL, and different concentrations of Tetrandrine (0, 0.5, and 1 μM). The medium was changed every 2 days until the 7th day. After osteoclast differentiation, the bovine bone slices were removed and rinsed thrice with PBS. After drying the bone slices, we photographed them with a TM1000 desktop scanning electron microscope (Hitachi High-Technologies, Tokyo, Japan) and the bone absorption area was analyzed with ImageJ software (NIH, Bethesda, MD, USA).

Osteoblast differentiation

C3H10 cells were osteogenically induced in a DMEM medium comprising 50 μg/mL ascorbic acid and 10 μM β-glycerophosphate (Sigma-Aldrich). After osteogenic induction for 7 days, alkaline phosphatase (ALP) staining was performed with BCIP/NBT kit (CWBIO, Beijing, China), and the ALP activity was quantified by an alkaline phosphatase detection kit (Beyotime, Nanjing, China). After 14 days of osteogenic induction, alizarin red S (ARS) staining was performed with 1% alizarin red staining solution (Sigma-Aldrich) to evaluate the calcium deposition in the extracellular matrix. To quantify the mineralization, staining was eluted with 10% glacial acetic acid and the absorbance was measured at 405 nm by a microplate reader (Bio-Tek Instruments).

Cell transfection

BMMs were harvested into 6-well or 12-well plates, and the medium was replaced every 2 days. When the cell density reached 50–60%, they were stimulated with RANKL for 4 h before being transfected with TRAIL siRNA, TRAIL vector, or negative controls (control siRNA and control vector) using the TSnanofect transfection reagent (Tsingke Biotechnology Co, Beijing, China) according to the manufacturer’s instructions. Tsingke Biotechnology Co. supplied TRAIL siRNA, TRAIL vector, and negative controls (control siRNA and control vector). The TRAIL siRNA sequences are listed in Table 1.

Table 1 Sequences of primers and siRNARNA extraction and quantitative real-time polymerase chain reaction (qPCR)

TRIzol Reagent (CWBIO) was used to extract total RNA from BMMs and C3H10 cells, as well as from the bones of mice used in the OVX experiment. The HiFiScipt cDNA Synthesis kit (CWBIO) was used to reverse transcribed total RNA into cDNA, and the cDNA was amplified via quantitative real-time PCR. qPCR was performed using the UltraSYBR Mixture (Takara Bio, Otsu, Japan) with the ABI Prism 7500 Fast System (Applied Biosystems, Foster City, CA, USA) under the following PCR cycling conditions: 95 °C for 5 min, 40 cycles at 95 °C for 10 s, 60 °C for 30 s, followed by a final step at 4 °C for 10 min. GAPDH was used as endogenous control, and the expression level of each gene was calculated using the 2 − ΔΔCt method. The primer sequences are listed in Table 1.

Protein extraction and Western blotting

Total protein was extracted from BMMs and C3H10 cells and from the bones of mice used in the OVX experiment using a radioimmunoprecipitation assay lysis buffer (RIPA) (Sigma-Aldrich), and the protein concentration was measured using a bicinchoninic acid protein (BCA) assay kit (EMD Millipore, Bedford, MA, USA). Protein equivalents were separated using 10% SDS PAGE and transferred onto PVDF membranes (EMD Millipore). The membranes were blocked for an hour at room temperature with 5% skim milk in Tris-buffered saline with Tween 20. After incubating primary antibodies overnight at 4 °C, the membranes were nurtured with corresponding HRP conjugated secondary antibodies (Proteintech Group, Chicago, IL, USA.) The bands were visualized employing ECL Luminous Liquid (EMD Millipore). ImageJ software was used to compute the relative grey level of proteins (NIH). Anti-TRAIL was procured from Santa Cruz Bio (Santa Cruz, CA, USA). Anti-CTSK, TRAP, NFATc1, OPG, RUNX2, OCN, RANKL, COL1a, p-P65, P65, p-P38, P38, p-JNK, JNK, p-ERK, ERK, p-IKBα, IKBα, IKKα, IKKβ, and p-IKKα/β were all purchased from Cell Signaling Technology (Danvers, MA, USA).

Fluorescent staining and actin ring formation of osteoclasts

The BMMs were cultivated in a 96-well plate and stimulated for osteoclastogenesis. The cells were then fixed in 4% PFA for 15 min before being broken into 0.1% Triton X-100 for 30 min. Anti-TRAIL antibody (1:200) was added to cells overnight at 4 °C after blocking in 10% goat serum for an hour. After rinsing with PBS, the fluorescent secondary antibody was added and incubated for an hour at room temperature (Alexa Fluor 488, green). The F-actin ring was stained with rhodamine phalloidin at room temperature for an hour in a dark room. The cells were then rinsed with PBS and stained with DAPI. Finally, the cells were observed using a fluorescence microscope (NIKON TE2000, Nikon Corporation, Minato, Tokyo, Japan).

Co-immunoprecipitation (IP) assay

The Co-IP experiment was conducted according to the protocol of the Co-IP kit (Thermofisher Scientific, MA, USA). Briefly, the cell lysates were immunoprecipitated with 2 μg anti-TRAIL antibody at 4 °C overnight and then incubated for an hour at room temperature with 0.25 mg protein A/G-agarose. The magnetic beads were then eluted using Elution Buffer, magnetically separated, and the supernatant containing the anti-TRAIL antibody was saved. The bound proteins were resolved using 8% SDS-PAGE and then incubated with anti-TRAIL and anti-ubiquitin.

The OVX-induced bone loss model

Eight-week-old C57BL/6 female mice (n = 15) weighing 20–25 g were anesthetized with 3 mg/mL pentobarbital for bilateral ovariectomy (OVX). The mice were randomly assigned to 3 groups of 5 mice each. The sham and vehicle groups received corn oil (MedChemExpress, NJ, USA) containing 10% DMSO (MedChemExpress), whereas the Tetrandrine group received tetrandrine (30 mg/kg) every 3 days for 42 days. The isolated femur and tibia were then fixed in 4% PFA for 48 h before micro-CT and histological experiments.

Enzyme-linked immunosorbent assay

The serum of mice was collected, and the enzyme-linked immunosorbent assay of IL-6, TNFα, OPG, and RANKL in the serum was performed with the ELISA kit (ProteinTech Group) according to the manufacturer’s instruction.

Micro-CT scanning

A previously published approach was employed to analyze the fixed tibia using a micro-CT scanner (1072 μCT System; Sky Scan, Bruker microCT, Kontich, Belgium) (Boerckel et al. 2014). The following parameters are set: X-ray energy and current were 70 kV and 300 μA, respectively, the aluminum filter was 0.5 mm, the rotation step was 0.4°, and the isometric resolution was 9 μm. The NRecon program (Bruker microCT, kontich, Belgium) was used for reconstruction. For rebuilding, the settings used were as follows: (1) smoothing; (2) ring artifact correction; (3) 40% beam hardening correction. Each sample was evaluated utilizing the software CTAn (Bruker microCT, Kontich, Belgium) for BV/TV (trabecular bone volume per total volume), Tb.N (the trabecular number), Tb.Th (the trabecular thickness), and Tb.Sp (trabecular separation).

Histological analysis

The fixed femur and tibia were immersed in 10% ethylenediaminetetraacetic acid (EDTA), decalcified for 2 weeks at 37 °C, and then embedded in paraffin to prepare the histological sections. These sections were stained with Masson trichrome, TRAP, and hematoxylin–eosin (HE) staining solutions and then photographed using the Aperio Scanscope high-quality microscope (MT Waverley, Vic, Australia). Image Pro Plus software was used to calculate the number of trap-positive cells (Media Cybernetics, Bethesda, MD, USA).

Calcein-alizarin red S labeling and von Kossa staining

On days 0 and 4, 20 mg/kg calcein (Sigma, C0875-5 g, 1 mg/mL in 2% NaHCO3 solution) and 40 mg/kg alizarin red S (Sigma, a5533-25 g, 2 mg/mL in H2O) were injected intraperitoneally. The mice were euthanized on the seventh day, and their bones were fixed, dehydrated, and embedded with embedded-812 (Electron Microscope Science). The samples were cut into 5-μm slices with a hard tissue cutter (RM2265, Leica, Wetzlar, Germany). Fluorescent labeling images were obtained with a microscope (BX51, Olympus).

For von Kossa staining, the femur samples were sliced into 6-μm dense pieces. Subsequently, the sections were irradiated with ultraviolet light in 2% silver nitrate solution for 1 h before being incubated with 5% sodium thiosulfate for 2 min. Finally, using a Diplan light microscope, the staining results were examined (Leitz).

Statistical analysis

The data from triplicate trials were evaluated utilizing Graphpad Prism 9 windows (Graphpad Software Inc., San Diego, CA, USA) and expressed as the mean ± SEM. For group comparison, one-way or two-way ANOVA and the Student’s t-test were used, with P < 0.05 indicating a statistically significant difference.