The Oricell Strain C57BL/6 Mouse BM-MSCs (No: MUBMX-01001) were purchased from Cyagen Biosciences (Guangzhou, China) and cultured in Oricell C57BL/6 Mouse BM-MSCs Complete Medium (No: MUBMX-90011) supplemented with 440 ml basal medium, 50 ml qualified fetal bovine serum, 5 ml penicillin-streptomycin, and 5 ml glutamine. Cells were maintained in a water-saturated atmosphere at 37°C in 5% CO₂. Cells were identified by the supplier according to the presence of cell surface markers and multipotency. Additionally, previous studies by other investigators have confirmed the stem cell identity of C57BL/6 Mouse BM-MSCs (Liu et al., 2013; Chen et al., 2020). An Ints7 translation-blocking Vivo-morpholino (MO, Oligo Sequence: TTGACGCCATGACCCGGACAGTTAC) and a negative control MO (Oligo Sequence: CCTCTTACCTCAGTTACAATTTATA) were purchased from Gene Tools LLC (Philomath, OR, United States). The Ints7 MO oligos bind to complementary RNA and block translation initiation in the cytosol by targeting the 5′ untranslated region (UTR) through the first 25 bases of the coding sequence. Abcd3 short hairpin RNA (shRNA, Oligo Sequence: UUGAAAUCUUUGCUGCUGC), Hdlbp shRNA (Oligo Sequence: AUCCUUGUAGGUUGGAGGG), and a negative control shRNA (Oligo Sequence: ACGUGACACGUUCGGAGAA), were purchased from GenePharma (Shanghai, China) and transfected into BM-MSCs with Lipofectamine 2000 (Invitrogen, United States). Abcd3/Hdlbp shRNAs are plasmid vector-based shRNAs capable of specifically degrading target mRNAs via complementary binding sequences. Second to third passage BM-MSCs were used for experiments in this study.

The RNeasy Plus Micro Kit (Qiagen, Düsseldorf, Germany) was used to extract total RNA from BM-MSCs, according to the manufacturer’s instructions. RNA was reverse transcribed into cDNA using the PrimeScript Reverse Transcription Kit (Vazyme, Nanjing, China). SYBR green-based quantitative PCR was performed using an ABI 7500 machine (Applied Biosystems, Foster City, CA, United States). All results were normalized against 18S rRNA expression. The primers used in this study are summarized in Supplementary Table 1.

Western blot analysis was performed as previously described with a minor change (Wu et al., 2021). In short, 48 h after inhibition using Ints7 MO oligos, BM-MSC protein lysates were separated by 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to 0.22-μm polyvinylidene difluoride membranes (Millipore, Billerica, MA, United States) and incubated with specific anti-INTS7 (Proteintech, Chicago, IL, United States) or anti-ABCD3 (Abcam, Cambridge, MA, United States) antibodies. A beta-tubulin antibody (Beyotime, Nantong, China) was used as an internal control. After incubation with horseradish peroxidase (HRP)-conjugated secondary antibodies (Thermo Scientific, Waltham, MA, United States), signals were detected by enhanced chemiluminescence substrate (Thermo Scientific). The resulting bands were analyzed using Image-Pro Plus Software.

Bone marrow mesenchymal stem cells were fixed in 4% (w/v) paraformaldehyde (PFA), blocked with 1% bovine serum albumin (BSA, w/v), and then reacted at 4°C overnight with anti-INTS7 primary antibody (Proteintech), as described previously (Shen et al., 2019). After washing with phosphate-buffered saline (PBS), the samples were incubated with a secondary fluorescent antibody (Thermo Scientific) for 1 h at room temperature. Slides were mounted using VECTASHIELD mounting medium with 4′,6-diamidino-2-phenylindole (DAPI; VECTOR, Burlingame, CA, United States). Sections were analyzed under a confocal laser-scanning microscope (Zeiss LSM800, Carl Zeiss, Oberkochen, Germany). Finally, five randomly selected views in each well were imaged and calculated for quantification of fluorescence intensity.

After 48 h of INTS7/ABCD3/HDLBP inhibition, BM-MSCs were inoculated into 96-well plates (3000 cells/well). The cell proliferation capacity was evaluated every 24 h using Cell counting kit-8 (Beyotime). Ethynyl-2-deoxyuridine (EdU) experiments were performed using an EdU labeling and detection kit (RiboBio, Guangzhou, China), according to the manufacturer’s instructions. BM-MSCs were treated with 50 μM EdU labeling medium and incubated for another 2 h. Next, cells were fixed with 4% paraformaldehyde and permeated using 0.5% Triton X-100. Anti-EdU working solution and DAPI staining solution were added. EdU-positive cells were observed and counted under a confocal laser-scanning microscope (Zeiss LSM800, Carl Zeiss, Oberkochen, Germany).

Bone marrow mesenchymal stem cells were harvested for cell-cycle analysis and fixed with cold ethanol overnight at 4°C. The cell suspension was centrifuged, and the collected pellets were washed with PBS and resuspended in 500 μl PBS. Cells were stained with propidium iodide in the dark using the CycleTEST™ PLUS DNA Reagent Kit (BD Biosciences, Franklin Lakes, NJ, United States), following the manufacturer’s protocol, and analyzed as previously described (Gao et al., 2020). The percentages of cells in the G0/G1, S, and G2/M phases were counted.

Apoptotic cells were detected by a terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL; BrightRed Apoptosis Detection Kit, Vazyme), as previously described (Zhao et al., 2019). Briefly, BM-MSCs were fixed with 4% (w/v) PFA and treated with proteinase K (10 μg/ml) for 10 min at room temperature, followed by incubation with BrightRed Labeling Buffer for 45 min at 37°C. Cells were washed with PBS three times then stained with DAPI. Images were captured, and TUNEL-positive cells were counted under a confocal laser-scanning microscope (Zeiss LSM800, Carl Zeiss, Oberkochen, Germany).

An osteogenesis-inducing differentiation medium kit (No: MUBMX-90021) was purchased from Cyagen Biosciences to enhance osteoblastic differentiation. Differentiated osteoblasts were stained with Alizarin Red S dye solution. An adipogenesis-inducing differentiation medium kit (No: MUBMX-90031) was purchased from Cyagen Biosciences to enhance adipogenic differentiation. Differentiated adipocytes were stained with Oil Red O dye solution.

Briefly, 5 μg of anti-INTS7 antibody (Proteintech) was incubated with total BM-MSC lysates for 2 h, followed by the addition of 20 μl of washed protein A/G beads (Santa Cruz Biotechnology) and incubated for another 16 h at 4°C. Beads were then pelleted and washed three times in 20× bed volume of lysis buffer. The bound protein was eluted by heating the beads at 95°C for 5 min with 2× SDS buffer. Finally, anti-INTS7 (Proteintech), anti-ABCD3 (Abcam), or anti-HDLBP (Proteintech) antibodies were used to detect the presence of each protein in the immunoprecipitation (IP) product. Inputs represented approximately 1/10 of the extract volume used for the IP experiment.

The gel particles were cut into blocks approximately 1 mm³ in size for in-gel digestion. Gel particles were first washed with deionized water, 50% acetonitrile (can), and 100% ACN. Next, the proteins were reduced and alkylated for 45 min at room temperature in the dark. The gel particles were then thoroughly washed, dried naturally, rehydrated at 4°C for 30 min, and subjected to protein digestion at 37°C for 12 h. Trifluoroacetic acid (TFA), at a final concentration of 0.1%, was added to stop the digestion, and the supernatants were transferred to fresh tubes. All extracts and supernatants were combined and dried in a SpeedVac. The digested peptides were desalted with StageTip (Thermo Scientific) and analyzed using an LTQ Orbitrap Velos mass spectrometer (Thermo Scientific). We searched the tandem mass spectrometry (MS/MS) spectra using the UniProt protein database to identify proteins, with a false discovery rate of 1% for both peptides and proteins. Three replicates were performed. For label-free quantification, the protein expression levels were estimated using the iBAQ algorithm embedded in MaxQuant. The annotated proteins are listed in Supplementary Table 2. The data presented in the study are deposited in the ProteomeXchange Consortium via PRIDE partner repository, accession number PXD028817. The data can be accessed using the following details: Username: reviewer_pxd028817@ebi.ac.uk; Password: fyaTWuvX

InterPro tool (Blum et al., 2021) was used to intercept key or representative domain sequences during prediction, and 3D models were constructed accordingly. Robetta (Yang et al., 2020), a 3D protein structure prediction software, was used to model the 3D structure of the ligand (ABCD3: ABC transporter-like domain/HDLBP: KH domain) and receptor (INTS7: Armadillo-type fold superfamily) sequences. We used the HawkDock protein complex prediction tool (Weng et al., 2019) to analyze the interaction between the ligand and receptor and optimize the interface between the interacting residues using the built-in molecular mechanics with generalized Born and surface area solvation (MM/GBSA) algorithm. PyMOL (version: 2.1.0) software was used to visualize the results and annotate the interaction residues.

After INTS7/ABCD3 inhibition for 48 h, BM-MSCs were incubated with 25 μM 2′,7′–dichlorofluorescein diacetate (DCFDA, Invitrogen, Carlsbad, CA, United States) for 45 min at 37°C. Cells were then washed with PBS, and the fluorescence intensities were immediately measured on a fluorescent plate reader or a confocal laser-scanning microscope (Zeiss LSM800, Carl Zeiss), as described previously (Yu et al., 2021).

Experiments were independently repeated at least three times. Data are presented as the mean ± standard deviation (SD). Student’s t-test and one-way analysis of variance (ANOVA) were used to evaluate significant differences between groups, *p < 0.05; ^**p < 0.01; ^***p < 0.001.

To explore the biological behaviors and potential molecular mechanisms through which INTS7 exerts effects in BM-MSCs derived from C57BL/6 mice, immunofluorescent staining was performed to characterize the subcellular localization of INTS7. We found that INTS7 protein was primarily localized in the cytoplasm rather than the nucleus, indicating that INTS7 may exert functions at the post-transcriptional or translational level (Figure 1A). Next, INTS7 was inhibited using Inst7-MO oligos, which reduced INST7 protein levels to 25% that of the negative control MO (Ctr; Figures 1B,C). CCK-8 assays showed that BM-MSC viability was prominently impaired in the Inst7-MO group compared with the Ctr group (Figure 1D). Flow cytometric analysis suggested that INST7 depletion resulted in cell-cycle arrest at the G0/G1 phase (Figures 1E,F). To further understand whether cell proliferation or apoptosis regulation was involved in observed effects on BM-MSC growth following INST7 depletion, an EdU-based flow cytometry analysis was performed, which revealed that the proportion of EdU-positive BM-MSCs in the INTS7-depleted group was significantly reduced compared with the Ctr group (Figures 1G,H). In addition, TUNEL staining analysis further indicated that INTS7 suppression could induce BM-MSC apoptosis (Figures 1I,J). These data suggested that INTS7 significantly promoted C57BL/6 mouse BM-MSC proliferation in vitro, partially due to an accelerated cell-cycle course and the attenuation of apoptotic behavior.

To evaluate the effects of INTS7 on osteoblast and adipocyte differentiation among BM-MSCs, Alizarin Red S and Oil Red O dye solutions were, respectively, used to identify osteoblasts and adipocytes. As shown in Figure 2, the ability of BM-MSCs to differentiate into osteoblasts was significantly weakened by INTS7 inhibition (Figures 2A,B), whereas the number of differentiated adipocytes derived from BM-MSCs presented a five-fold increase (Figures 2C,D). Subsequently, the expression levels of several key transcription factors and signaling molecules involved in either osteoblast or adipocyte differentiation in BM-MSCs were detected in the Ctr and Inst7-MO groups, which revealed the downregulation of several facilitating factors involved in osteogenic differentiation following INST7 suppression, including Runx2, Osteocalcin (Ocn), alkaline phosphatase (Alp), and Osterix (Osx) (Figure 2E). By contrast, the effective molecules (PPARγ, C/EBPα, and adipsin) regulating adipocyte differentiation were significantly increased in the Inst7-MO group compared with the Ctr group (Figure 2F). Collectively, these results implied that INTS7 inhibition in BM-MSCs could decrease osteoblastic differentiation and accelerate adipocytic differentiation.

To identify proteins that interact with INTS7 in BM-MSCs, total endogenous proteins were collected, subjected to IP experiments using an anti-INTS7 antibody, and subjected to IP-PAGE, in-gel digestion, liquid chromatography separation, and mass spectrometry analysis (Figure 3A). Overall, three proteins, INTS7, ABCD3, and HDLBP, were successfully identified and quantified in three independently repeated experiments (Figures 3B,C). Co-IP assays confirmed the interaction between INTS7 and ABCD3/HDLBP proteins (Figure 3D), suggesting that INTS7 affects the proliferation and differentiation of BM-MSCs potentially through a cooperative mechanism involving ABCD3 or HDLBP proteins. Moreover, 3D model construction and bioinformatics analysis were performed to characterize the interactions between INTS7 and ABCD3/HDLBP proteins. The predicted 3D structural model for the protein interaction docking between INTS7 and ABCD3 is exhibited in Figure 3E, and the binding interface from both front and back perspectives is further visualized in Figure 3F. Similarly, Figures 3G,H separately displayed the 3D protein interaction docking and visual binding interface, respectively, between INTS7 and HDLBP.

The biological functions of ABCD3/HDLBP were next examined in BM-MSCs. First, Hdlbp and Abcd3 were significantly knocked down using targeted shRNAs (Figures 4A,B). CCK8 assays indicated that Abcd3 silencing could prominently impair BM-MSC viability, whereas Hdlbp inhibition had no impact on BM-MSC growth (Figure 4C). Next, EdU staining revealed that the proportion of EdU-positive BM-MSCs in the sh-Abcd3 group was significantly decreased compared with that in the negative control (Ctr) group. However, the number of EdU-positive cells was not altered after Hdlbp downregulation (Figures 4D,E). Analogously, TUNEL staining analysis demonstrated that apoptosis was induced in BM-MSCs in the sh-Abcd3 group but not in the sh-Hdlbp group compared with the Ctr group (Figures 4F,G). These data supported an active role for ABCD3 in the regulation of BM-MSC growth.

Alizarin Red S and Oil Red O staining were performed to evaluate the effects of ABCD3/HDLBP on osteoblast and adipocyte differentiation in BM-MSCs. The results showed that the proportion of differentiated osteoblasts decreased by 30%, whereas a five-fold increase in the number of differentiated adipocytes was observed following Abcd3 suppression (Figures 5A–D). As expected, Hdlbp suppression exerted no impacts on either osteoblast or adipocyte differentiation in BM-MSCs (Figures 5A–D). Molecule marker detection after Abcd3 knockdown revealed the downregulation of factors (Runx2, Ocn, Alp, and Osx) associated with osteoblast differentiation and the increased expression of factors (PPARγ, C/EBPα, and adipsin) associated with adipocyte differentiation (Figures 5E,F). However, no changes in the expression levels of any of the examined markers were observed following Hdlbp depletion in BM-MSCs (Figures 5E,F). These data supported the finding that Abcd3 inhibition activated BM-MSC differentiation into adipocytes and inhibited osteogenic differentiation, whereas Hdlbp had no involvement in BM-MSC differentiation.

Western blot analysis found that the quantity of ABCD3 protein was significantly reduced in response to Ints7 silencing (Figures 6A,B), which indicated that ABCD3 was a potential downstream target of INTS7, and the INTS7–ABCD3 pathway may exert co-regulatory effects in BM-MSCs. As previously reported, MSCs are highly sensitive to oxidative stress compared with differentiated cell types, and excessive ROS may impair the differentiation and proliferation capacity of MSCs (Ko et al., 2012; Denu and Hematti, 2016). To explore whether oxidative stress suppression was involved in the regulatory effects of INTS7–ABCD3 on BM-MSC proliferation and osteoblastic differentiation, we examined ROS and antioxidant quantities in INTS7- or ABCD3-depleted BM-MSCs. The results showed that the ROS levels in the treatment groups were significantly upregulated compared with those in the control group (Figures 6C,D). In addition, the expression levels of antioxidants (including Sod1, Gpx1, Gpx4, Txnrd1, and Cat) decreased significantly in response to both INTS7 and ABCD3 depletion (Figure 6E). DNA damage is associated with the upregulation of oxidative stress. We performed immunostaining to detect the histone γ-H2AX (a marker of DNA double-strand breaks) to confirm this phenomenon. Figures 6F,G show that the percentage of γ-H2AX–positive cells increased following the depletion of both INTS7 and ABCD3. Moreover, adding of the antioxidant scavenger N-acetylcysteine (NAC) after INTS7/ABCD3 knockdown could both partially rescue the cell viability of BM-MSCs (Figure 6H). These findings suggested that the low levels of oxidative stress are potentially responsible for the INTS7–ABCD3 co-regulatory effects observed for BM-MSC proliferation and differentiation.