Human bone tissue collection

All experiments were approved by the Ethics Committee of the First Affiliated Hospital of Jinan University (Ethical approval number: KY-2021-065). All patients aged 49–81 years old were subjected to an examination of serum bone metabolic biomarkers: osteocalcin (OST), procollagen type I N-terminal propeptide (PINP), C-terminal cross-linking telopeptide of type I collagen (β-CTx), and alkaline phosphatase (ALP). Bone mineral density (BMD) was assessed using dual-energy X-ray absorptiometry (DXA) and was categorized into three groups: normal (femoral neck (FN)-BMD T score > −1.0), osteopenia ( − 2.5 < FN-BMD T score ≤ −1.0), and osteoporosis (FN-BMD T score ≤−2.5). Participants with other metabolic diseases and abnormalities in any of the screening laboratory tests (complete blood count, serum calcium, phosphorus, albumin, etc.) were excluded from our study. The detailed characteristics of each patient are listed in Table S1. Human tibial bone specimens were obtained from patients who received total knee arthroplasty at the Department of Bone and Joint Surgery of the First Affiliated Hospital of Jinan University and were immediately placed in a liquid nitrogen jar for preservation. All participants were aware of the purpose of the study and signed informed consent before participating in the study.

Establishment of global and conditional Trim21 knockout mice

Heterozygous recombinant Trim21 mice (Trim21−/+) with a C57BL/6 genetic background were purchased from Cyagen Biosciences (Guangzhou, China). Heterozygous recombinant Trim21 mice (Trim21flox/+) with a C57BL/6 genetic background were purchased from GemPharmatech (Nanjing, China). Animals were bred and maintained under specific pathogen-free (SPF) conditions. All animal procedures were approved by the Institutional Animal Care and Use Committee of Jinan University (Approval number: IACUC-20211216-15 and IACUC-202111213-02, respectively) and conformed to the “Guide for the Care and Use of Laboratory Animals” of the National Institute of Health in China. Briefly, Trim21 knockout (KO) (Trim21−/−) mice and WT (Trim2+/+) mice were generated by intercrossing the heterozygous targeted mice, followed by genotyping with a Quick Genotyping Assay Kit for Mouse Tail (Beyotime, Cat# D7283S, China). For the generation of Trim21 conditional knockout mice (Ctsk-cre; Trim21f/f) in OCs, Trim21f/f mice were crossed with Ctsk-cre mice. All mice were on a C57BL/6 background. Polymerase chain reaction (PCR) genotyping primers are listed in Table S2.

Cell culture and OB treatments

Bone marrow mesenchymal stem cells (BMSCs) were derived from 4- to 6-week-old Trim21+/+ and Trim21−/− mice. Primary murine calvarial OBs were isolated from 3-day-old mice. The preosteoblast cell line MC3T3-E1 was gifted by Prof. Chengzuo Qiu (Jinan University, Guangzhou). HEK293T cells were purchased from the American Type Culture Collection (ATCC, USA) and were maintained in our laboratory and regularly tested for mycoplasma contamination.44 BMSCs, OBs, and MC3T3-E1 cells were cultured in complete α-MEM containing 10% FBS and 1% penicillin/streptomycin.

For protein degradation evaluation, HEK293T cells were seeded into (2 × 105 cells per well) 6-well plates for 24 h. Cells were then transiently transfected with 1 μg of plasmid encoding HA-Vector or HA-Trim21 using Lipofectamine 2000 (Thermo Fisher Scientific, Cat# 11668019, USA) for the indicated time, followed by treatment with 5 μmol·L−1 MG132 (Selleck, Cat# S2619, USA) for another 24 h and immunoblotting analysis.

For osteogenic differentiation, BMSCs and OBs were reseeded at a density of 5 × 105 cells per well into 6-well plates or 2.5 × 105 cells into 12-well plates. BMSCs were then stimulated with a MesenCult Osteogenic Stimulatory Kit (Mouse) (StemCell, Cat# 0504, StemCell, Canada). MC3T3-E1 cells were seeded at a density of 2 × 105 cells per well into 6-well plates or 1 × 105 cells into 12-well plates. OBs and MC3T3-E1 cells were stimulated with osteogenic medium containing 10% FBS, 50 μg·mL−1 ascorbate, 10 μmol·L−1 β-glycerophosphate, 0.1 μmol·L−1 dexamethasone, and 10 μmol·L−1 glutamine. The osteogenic medium was replaced every 3 days for all cells.

To mimic inflammation-induced bone loss in vitro, we seeded MC3T3-E1 cells into 6-well plates at a density of 2.5 × 105 for 24 h. Then, the cells were treated with different concentrations of LPS (Beyotime, Cat# S1732, China) or TNF-α (CST, Cat# 89025 C, USA) for another 24 h, followed by the analysis of Trim21, IL-6, and osteogenic markers (Runx2 and Osterix) by quantitative RT‒PCR or immunoblotting assays.

Osteoclastogenesis in BMMs and the coculture system

BMMs were derived from 4- to 8-week-old Trim21+/+ and Trim21−/− mice or Trim21f/f mice, as well as Ctsk-cre; Trim21f/f mice, and then incubated in α-MEM containing 30 ng·mL−1 M-CSF (MCE, Cat# HY-P70553, USA) for 4 days to generate BMMs. For generation of mature OCs, BMMs were cultured with 30 ng·mL−1 M-CSF and 100 ng·mL−1 RANKL for at least 5 days, followed by a series of experiments. The same volume of phosphate-buffered saline (PBS) containing 30 ng·mL−1 M-CSF was applied as a control. For specific knockout of Trim21, BMMs from Trim21f/f mice were infected with lentivirus expressing EGFP or Cre recombinase (defined as LV-Con or LV-Cre, Obio Technology, China).

RAW264.7 monocytes/macrophages (ATCC, USA) were cultured and maintained as we described previously.45 For induction of the differentiation of RAW264.7 cells into OCs, 100 ng·mL−1 RANKL (R&D Systems, Cat# 462-TEC-010) was added to the cells for three more days, and the medium was changed every 2 days.

In the coculture model of BMSCs-BMMs, BMSCs from Trim21+/+ and Trim21−/− mice were placed in the upper chamber of Transwell inserts (Costar, Cat#3450, USA), while BMMs from Trim21+/+ mice were seeded in the lower chamber at a density of 4 × 105 cells per well with 30 ng·mL−1 M-CSF and 100 ng·mL−1 RANKL induced for 5 days, allowing cell‒cell communication without direct physical contact. This Transwell configuration allows a specific ratio of 2:1 for BMSCs to BMMs.

Plasmid construction

The pCMV-FLAG vector was gifted by Prof. Chang-Deng Hu (Purdue University). For cloning of mouse Trim21 genes, the cDNA of RAW264.7 cells was amplified by PCR using the primers 5′-CCGAATTCTCATGTCTCTGGAAAAGATGTGGG-3′ and 5′- TAAAGCGGCCGCTCACATCTTTAGTGGACAGAGCTT-3′. The purified PCR fragment was cloned and inserted into the abovementioned backbone using EcoRI and NotI sites for the construction of pCMV-FLAG-mus-Trim21. For construction of Myc-VN155-YAP1, the YAP1 gene was amplified from HA-YAP1 using the primers 5′-GCAGATCTGGATGGATCCCGGGCAGCAGCCG-3′ and 5′-TAAAGCGGCCGCCTATAACCATGTAAGAAAGCTTTCT-3′ and then subcloned and inserted into the BglII-NotI sites of pBiFC-VN155.44,46 The above constructs were verified by sequencing and immunoblotting analysis. The pCMV-HA-homo-Trim21 was previously constructed.47

Small interfering RNA assays

siRNA targeting Trim21 was purchased from GenePharma (Shanghai, China). The sequences are listed in Table S3. Transfection of siRNA was performed using Lipofectamine 2000 (Thermo Fisher, Cat# 11668019, USA) according to our previous study.47 Briefly, MC3T3-E1 or RAW264.7 cell lines were seeded in 6-well plates at a density of 2 × 105 cells per well for 24 h and then transiently transfected with 5 μL of Lipofectamine 2000 and 5 μL of siCon, siTrim21#2 or siTrim21#3 for 4 h. Then, the medium was changed to complete α-MEM, and the knockdown effect of Trim21 was verified by quantitative RT‒PCR and immunoblotting after 48 h.

Alizarin red S, ALP, and oil red O staining

Cells were incubated with osteogenic medium for 7–14 days to allow the formation of opaque calcified nodules. The cell samples were then washed once with phosphate-buffered saline (PBS), followed by fixation with 4% paraformaldehyde (PFA) for 20 min. Washed nodules were then stained with 0.1% Alizarin Red S (Beyotime, Cat# C0148S, China) solution for 30 min. Alkaline phosphatase (ALP) staining was carried out using an ALP staining kit (KeyGen, Cat# KGA353, China). After 1 week of adipogenic induction in BMSCs, the cells were subjected to Oil Red O staining according to the manufacturer’s protocols (Beyotime, Cat# C0158S, China). The stained images were observed and taken using a phase-contrast microscope (Nikon, Tokyo, Japan). The area stained by Alizarin Red S, ALP, or Oil Red O was quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA) from over 5 random fields.

Tartrate-resistant acid phosphatase (TRAP) staining

Cells and mouse knee sections were fixed and stained with TRAP solution using a kit according to the manufacturer’s instructions (Sigma-Aldrich, Cat# 387 A, Germany). Cells with more than 3 nuclei were considered OCs. Five high-power fields (200 ×) were randomly selected for OC counting. Mouse knee sections were selected at 4 high-power fields (200 ×) in the tibial metaphysis for OC counts. TRAP-positive cells were visualized, and the number of OCs per field and the number of nuclei per TRAP+ cell were quantified by ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Quantitative real-time PCR (RT‒PCR)

For mRNA expression analysis, total RNA was purified using TRIzol (TaKaRa, Cat# 9109, Japan) and was reverse-transcribed into cDNA using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher, Cat# 4368814, USA). In brief, the cancellous bone specimens were fragmented, and TRIzol was added to lyse cells to purify mRNA, followed by the reverse transcription of cDNA and quantitative RT‒PCR. Quantitative RT‒PCR was performed on a QuantStudio 5 using Fast SYBR Green PCR Master Mix (Thermo Fisher, Cat# 4385612, USA) with 500 nmol·L−1 primers. The primers used for quantitative RT‒PCR are listed in Table S4. The relative expression levels of the indicated genes were calculated using the 2−ΔΔCt method, with expression levels normalized to that of GAPDH.46

Coimmunoprecipitation (Co-IP)

Co-IP assays were performed to explore the endogenous interaction among BCL9, YAP1, and Trim21 in MC3T3-E1 cells following the same procedure as described previously.19,47 Briefly, cells with or without induction were harvested and then lysed by sonication in Western and IP Lysis Buffer (Beyotime, Cat# P0013, China) containing NaF, PMSF, Na3VO4, and protease inhibitors. After incubation on ice for 30 min, 40 μL of 50% protein A agarose bead slurry (CST, Cat# 9863, USA) was used to preclear cell extracts for 30 min at 4 °C, followed by incubation with antibodies against YAP1 (CST, Cat# 14074, USA, 1:50), BCL9 (Abcam, Cat# ab37305, UK, 1:50), Trim21 (Novus, Cat# NBP1-33548, China, 1:100) or control IgG (CST, Cat# 2729, USA) overnight at 4 °C. Antibody complexes were then incubated with 40 μL of 50% protein A agarose bead slurry for 4 h at 4 °C on a rotating platform. Immunoprecipitated proteins were washed five times with cold lysis buffer and then subjected to immunoblotting analysis. HRP AffiniPure Mouse Anti-Rabbit IgG Light Chain (Abbkine, Cat# A25022, USA) was used as the secondary HRP-conjugated antibody to avoid interference from the IgG heavy chain.

Immunoblotting assay

Protein concentration was determined by the BCA method, and 20–50 µg of protein was first separated by 8%–15% sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE) gels (Beyotime, Cat# P0012A, China), followed by visualization with primary and secondary antibodies using a Tanon 5200 Luminescent Imaging Workstation (Tanon, China) as described previously47. Briefly, the lumbar spine of the mouse was dissected by removing the surrounding muscles and the transverse and spinous processes with a rongeur. Then, the cone and lamina of the lumbar vertebra were disrupted using a tissue disrupter with the aid of liquid nitrogen, followed by the addition of Western and IP Lysis Buffer (Beyotime Biotechnology, P0013). The protein supernatant was collected after centrifugation for subsequent experiments. Samples for other cell experiments were prepared routinely. The antibodies used were as follows: YAP1 (CST, Cat# 14074, USA, 1:1 000), β-catenin (CST, Cat# 8480 S, USA, 1:1 000), GAPDH (CST, Cat# 2118, USA, 1:1 000), FLAG (CST, Cat# 8146, USA, 1:1 000), Trim21 (Novus, Cat# NBP1-33548, China, 1:1 000), AXIN1 (CST, Cat# 2087 S, USA, 1:1 000), Osterix (Abcam, Cat# ab209484, UK, 1:1 000), BCL9 (Abcam, Cat# ab37305, UK, 1:1 000), CTSK (Abcam, Cat# ab19027, UK, 1:1 000), NFATC1 (SANTA, Cat# sc-7294, USA, 1:1 000), and Runx2 (Abcam, Cat# ab236639, UK, 1:1 000).

Immunofluorescence (IF) staining of cells

OCs or OBs induced as described above were then fixed with 4% PFA for 20 min, permeabilized with 0.1% (v/v) Triton X-100 for 10 min, and blocked with 5% skim milk for 1 h. Cells were then incubated with primary antibodies (YAP1, CST, Cat# 14074, USA, 1:50) at 4 °C overnight, washed with phosphate-buffered saline with Tween 20 (PBST) 3 times for 5 min, and then incubated with a fluorescent secondary anti-mouse antibody (Alexa Fluor 488, green, CST, Cat# 8878, USA, 1:100). F-actin (only for OCs) was stained with rhodamine phalloidin (Invitrogen, Cat# R415, USA) in the dark for 1 h followed by 4′, 6-diamidino-2-phenylindole (DAPI) staining for 10 min. The images were captured using a laser scan confocal microscope (Zeiss LSM 880, Germany). F-actin rings were visualized, and the number of F-actin rings per field (4 ×) was quantified by ImageJ software.

Biomolecular fluorescence complementation (BiFC) assay

HA-VC155-Trim21 was previously constructed, and the BiFC assay was performed essentially as previously described to analyze the interaction between YAP1 and Trim21 in living cells.47 In brief, HEK293T cells were seeded on coverslips in a 15 mm confocal dish for 24 h, and the plasmids encoding Myc-VN155-YAP1 and HA-VC155-Trim21, along with HA-cerulean, were cotransfected into cells for 4 h by Lipofectamine 2000 (Thermo Fisher, Cat# 11668019, USA). After 48 h of culture, the cells were fixed with 4% paraformaldehyde and then stained with DAPI for 5 min, followed by observation using a confocal microscope.

Tandem mass tag-based quantitative proteomics

Tandem mass tag (TMT)-based quantitative proteomic analysis was performed by Shanghai Applied Protein Technology Company (Shanghai, China). In brief, proteins were extracted using SDT lysis buffer (4% sodium dodecyl sulfate (SDS), 100 mmol·L−1 Tris/HCl pH 7.6, 0.1 mol·L−1 dithiothreitol (DTT)), and the protein content was determined using the BCA method. Next, protein samples were digested using the filter-aided proteome preparation (FASP) method as previously described48 and were desalted on C18 cartridges (Empore™ SPE Cartridges C18, standard density), dried under a vacuum, and then resuspended in 0.1% (v/v) formic acid. Each set of eluted peptides was labeled with a unique TMT isobaric tag (TMT126–128 for Trim21+/+, TMT129-131 for Trim21−/−) and analyzed by a Q Exactive mass spectrometer (MS, Thermo Fisher Scientific) coupled with an Easy-nLC 1000 system (Thermo Fisher Scientific). MS raw data were analyzed by Proteome Discoverer (Thermo Fisher Scientific, version 1.4) and then subjected to a database search using the MASCOT search engine (Matrix Science, Boston, MA, USA, version 2.2) for peptide identification. In this study, a 1.2-fold change (upregulation or downregulation) was used as a cutoff for biological significance based on the standard deviation and normalized peptide ratios. The top BLAST hits were represented by volcano plots and hierarchical clustering analysis. Pathway analysis was performed using the Kyoto Encyclopedia of Genes and Genomes database (http://www.genome.jp/kegg/pathway.html).

Double staining of the mouse skeleton

Mice were eviscerated, the skin was removed, and the resulting samples were transferred into acetone for 48 h after overnight fixation in 95% ethanol. Skeletons were then stained in Alcian blue and Alizarin Red S solution (Beyotime, Cat# C0148S, China) for 3 days at 37 °C and sequentially cleared in 1% KOH. Skeletons were then replaced in 1% KOH/20% glycerol for 3 days and passed through increasing concentrations of 1:1 glycerol/ethanol solution (20%, 50%, and 100%) for 1 day. Finally, the stained skeletons were dehydrated in glycerol for imaging and storage.

Mouse cranial bone defect model

Twelve-week-old Trim21+/+ or Trim21−/− mice (n = 5) were anesthetized by intraperitoneal injection of pentobarbital sodium (80 mg·kg−1), and then, the region of interest was disinfected twice. A 1.5 cm skin incision was made longitudinally along the midline of the skull to expose the parietal bone. After separation of the periosteum, a 2.5 mm hole was created in the left parietal bone far from the cranial suture using a Mini electric drill with a 2.5 mm diameter diamond bit (Shenzhen, China) at a low speed ( < 1 500 r·min−1). Next, the operation area was washed with PBS and then sutured. Four weeks after defect creation, mice were euthanized, and the cranium of the mice was fixed with 4% PFA solution for 48 h. Finally, the samples were transferred to 70% ethanol until processing for micro-CT and histology.

Ovariectomy (OVX)- and lipopolysaccharide (LPS)-induced osteoporosis mouse model

For the OVX model, Trim21+/+ or Trim21−/− mice (females; 12 weeks old) were randomly divided into two groups: the sham-operated group (n = 5) and the OVX group (n = 5). Bilateral ovariectomy was performed to induce osteoporotic bone loss under sodium pentobarbital-induced anesthesia in the OVX group, while in the sham-operated group, the ovaries were just exteriorized and then placed back. The incisions of the muscle and the skin were closed using 6-0 silk sutures. All of the mice were sacrificed 8 weeks after OVX modeling, and the mouse tibias were collected for micro-CT analysis, followed by the assessment of histological changes.

For the LPS model, Trim21+/+ or Trim21−/− mice (males; ~12 weeks old) and Trim21f/f or Ctsk-cre; Trim21f/f mice (~12 weeks old) were randomly divided into two groups: the PBS group (n ≥ 3) and the LPS group (n ≥ 5). The mice in the latter group were treated with LPS (Beyotime, Cat# ST1470, China; 5 mg·kg−1) by intraperitoneal injection on Day 0 and Day 4, and then, the bone specimens were collected on Day 8 for micro-CT analysis as previously described with minor modifications.49,50

Micro-CT analysis

After the mice were sacrificed, the right tibia or craniums were isolated and fixed with 4% PFA for 48 h and then maintained in 70% ethanol. High-resolution ex vivo micro-CT (SkyScan1176, Bruker micro-CT, Kontich, Belgium) was used for image acquisition, and the scans were then integrated into 2D images. The trabecular region from 0.1 mm to 1.0 mm below the tibia growth plate was reconstructed for 3D visualization, and cortical bones were scanned at mid-diaphysis of the tibia. Quantitative analyses of the morphometric parameters were conducted using the appropriate software package, and the following morphometric indices were analyzed: bone volume per tissue volume (BV/TV), trabecular thickness (Tb.Th), trabecular number (Tb.N), trabecular separation (Tb.Sp), cortical BV/TV (Ct.BV/TV) and cortical bone thickness (Ct.th).

Von kossa and masson staining

Tibias were fixed with 4% PFA at room temperature for 48 h. Bones were dehydrated using a gradual series of ethanol (70%, 95%, and 100%), infiltrated, and embedded without decalcification in methyl methacrylate. von Kossa (Servicebio, Cat# G1043, China) and Masson (Servicebio, Cat# GP1032, China) staining were performed on tibial sections according to the manufacturer’s instructions. The proportion of positively stained areas in the total area was captured.

Histological staining and histomorphometric analysis

For histological analysis, tissues were fixed with 4% PFA for 48 h and incubated in DEPC-EDTA solution for decalcification. Then, the specimens were embedded in paraffin and sectioned at 5 μm. Sections were then used for H&E (Leagene, DH0006) and Safranin O (S/O) staining according to the manufacturer’s instructions (Servicebio, Cat# GP1051, China). Fat cell density was quantified using ImageJ software (National Institutes of Health, Bethesda, USA) from 5 random fields (200 ×) of metaphysis for each sample. For dynamic histomorphometric analysis, mice were injected with calcein (Sigma, Cat# C0875, Germany) at a dose of 20 mg·kg−1 of body weight on Day 7 and Day 1 before being sacrificed. Tissues were fixed with 4% PFA for 48 h, and 9 μm-thick sections were prepared for double-labeling fluorescent analysis. The mineral apposition rate (MAR) was analyzed using a laser scanning confocal microscope (Leica Microsystems, Mannheim, Germany).

Immunohistochemistry (IHC) and IF for tissues

The bone section was prepared as we previously described.51 Briefly, tibial bone sections were boiled in Tris-EDTA (pH 9) buffer solution to retrieve antigens, quenched with 3% hydrogen peroxide, and then treated with 0.5% Triton X-100. Next, 1% goat serum was used to block tissues at 37 °C for 30 min. For IHC staining, sections were treated with primary antibody against YAP1 (CST, Cat# 14074, USA, 1:200) overnight at 4 °C. Tissue sections were then washed with PBST, incubated with secondary antibody and visualized using a diaminobenzidine (DAB) kit (CST, Cat# 8059, USA). Cell nuclei were counterstained with hematoxylin. Images were taken under a light microscope (Nikon, Tokyo, Japan). For IF staining, the bone sections were treated with primary antibodies, including those against Sox9 (CST, Cat# 82630, USA, 1:100) and β-catenin (Abcam, Cat# AB32572, USA, 1:100). After PBST washes, tissue sections were incubated with a peroxidase-conjugated anti-fluorescein antibody (CST, Cat#4413, USA, 1:500). Antifade mounting medium with DAPI was used to mount the slides. The images were captured using a laser scan confocal microscope (Zeiss LSM 880, Germany) and analyzed by Image-Pro Plus software.

Statistical analysis

All data are expressed as the mean ± standard deviation (SD) with sample sizes indicated in either the figures and/or legends. For comparisons between the two groups, statistical analyses were performed by Student’s t test. One-way ANOVA was used to compare the effects of more than two groups. A P value of <0.05 was considered statistically significant.