Study approval

The human study was conducted in accordance with the principles of the Declaration of Helsinki, and was approved by the Ethics Committee of the Shanghai Jiao Tong University of Medicine Affiliated Sixth People’s Hospital. Written informed consent was received from participants or parents of minors prior to inclusion in the study. The zebrafish experiments were approved by the Institutional Animal Care and Use Committee of Shanghai Research Center for Model Organisms. And the zebrafish facility is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International. The mice were bred and maintained under specific pathogen free conditions in the institutional animal facility of the Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences. All mice experiments were conducted in accordance with a protocol approved by the Animal Care and Use Committee of Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences.

Clinical and genetic analysisHuman subjects and clinical analysis

We investigated probands with SEMD and their available family members. All individuals underwent a thorough clinical inquiry, physical examination, biochemical and bone-densitometry analyses, and skeletal radiography. The diagnosis of SEMD was based on clinical manifestations and radiological findings. In addition, 750 healthy unrelated Chinese subjects were enrolled as controls.

Genome-wide linkage analysis

Genomic DNA was extracted and purified from peripheral blood leukocytes with traditional methods. Genotyping of the samples was performed using Illumina’s GenomeStudio Genotyping Module, with data collected according to Infinium HD Assay Super protocol. Multipoint parametric linkage analysis and haplotyping for a dominant inheritance model with complete penetrance were conducted by the MERLIN program, version 1.1.2. Putative protein-coding genes within candidate region were examined using UCSC Genome Browser on human Feb. 2009 (GRCh37/hg19) assembly (http://genome.ucsc.edu/).

Whole-exome sequencing

Qualified DNA sample was fragmented, ligated to adapters, amplified by ligation-mediated PCR, and hybridized to exome array for enrichment. Agilent 2100 Bioanalyzer and quantitative PCR were used to estimate the magnitude of enrichment of captured ligation-mediated PCR products. High-throughput sequencing were conducted for each qualified capture library loaded on Illumina Hiseq platforms. The derived data were then processed by Illumina base-calling Software for base-calling with default parameters, and were generated as paired-end reads. High quality reads were mapped to the human reference genome GRCh37 with Burrows-Wheeler Aligner (BWA V0.7.15), and duplicate reads were removed by Picard-tools (v2.5.0). Genome Analysis Toolkit (GATK) was used for local realignment around InDels and base quality score recalibration following best practices. HaplotypeCaller of GATK (v3.3.0) was applied for detection of genomic variation including SNPs and InDels, and high-confident variant calls were obtained using hard-filtering methods. SnpEff tool was conducted for a series of annotations for variants. Variants with an allele frequency >0.1% in the 1 000 Genomes Project database (http://www.1000genomes.org/), the ESP6500 (http://evs.gs.washington.edu/EVS/), and the ExAC database (http://exac.broadinstitute.org/) were excluded.

Sanger sequencing

The variants in candidate pathogenic genes, identified by combination of linkage analysis and WES, was confirmed by Sanger sequencing following PCR amplification in all available family members. Amplificated PCR products were purified with shrimp alkaline phosphatase (Promega) and exonuclease I (Epicentre). Sequencing reaction was performed using BigDye Terminator cycle sequencing ready reaction kit, version 3.1 (Applied Biosystems), and the products were detected and analyzed using an ABI Prism 3130 automated sequencer (Applied Biosystems) and Polyphred.

Cell culture

Mice BMSCs were isolated from tibia and femur bone marrow cavity of 4-week-old C57BL/6 mice: bone marrow was flushed out with pre-cooled PBS, followed by centrifugation and lysis of erythrocytes. The cell precipitates were gently resuspended and cultured in α-MEM supplemented with 10% FBS and 1% penicillin/streptomycin. HEK293T cells and C3H10T1/2 cells were purchased from the American Cell Type Culture Collection (ATCC), and HEK293T cells were cultured in DMEM containing 10% FBS and 1% penicillin/streptomycin, C3H10T1/2 cells were cultured in α-MEM containing 10% FBS and 1% penicillin/streptomycin. All the cells were cultured in a 5% CO2 humidified incubator at 37 °C.

Transfection and infection

Phage vector overexpressing Flag-tagged full-length human mutant CCN2 (p.Arg22Pro) or wild-type CCN2 were constructed. HEK293T cells were transfected with the wild-type or mutant vector together with pLenti-EGFP according to the Effectene Transfection protocol (QIAGEN, 301425), with pLenti-EGFP as a control. For lentivirus vector construction, we overexpressed ZsGreen, Flag-tagged full-length human mutant CCN2 or wild-type CCN2 using a pLenti vector (pLenti-hCCN2-3xFlag-ZsGreen-Puro). Lentivirus packaging was conducted according to the VSVG-delta 8.9 system. BMSCs or C3H10T1/2 were infected with lentivirus for 24 h, and treated with puromycin for 48 h. The ZsGreen lentivirus was used as a control.

In vitro osteoblast differentiation

For in vitro osteoblast differentiation, BMSCs treated with lentivirus were induced by α-MEM with 10% FBS, 1% penicillin/streptomycin, 50 μg/mL L-ascorbic acid (Sigma, A5960), and 1.08 mg/mL β-Glycerophosphate disodium salt hydrate (Sigma, G9422). On day 7, BMSCs were fixed with 4% paraformaldehyde (PFA) for 10 min and subjected to ALP staining according to the manual of BCIP/NBT Alkaline Phosphatase Color Development Kit (Beyotime, C3206). For ALP activity quantitative analysis, osteoblasts were incubated with Alamar Blue (Invitrogen) for 1 h and read with a luminometer (Envision) for cell number quantification. Then the osteoblasts were incubated with phosphatase substrate (Sigma, S0942) dissolved in the solution (6.5 mmol/L Na2CO3, 18.5 mmol/L NaHCO3, 2 mmol/L MgCl2) for 20 min. The ALP activity was read with a luminometer (Envision) at A405, and normalized as A405/Alamar Blue.

Protein secretion analysis

For CCN2 secretion analysis, HEK293T cells were transfected with indicated plasmids. Cells were harvested after 48 h and cells were lysed with 500 μL EBC buffer (50 mmol/L Tris, pH 7.5, 120 mmol/L NaCl, and 0.5% NP-40) supplemented with a protease inhibitor cocktail (MCE, HY-K0010) and centrifuged at 12 000 r/min. Culture medium was collected into 15 mL centrifuge tube and centrifuged at 2 000 r/min. Cell lysates and medium were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotted with primary antibody against Flag (Sigma, F3165-5MG, 1:5 000) or GFP (Proteintech, 66002-1-Ig, 1:5 000) overnight at 4°C and secondary antibody of anti-mouse-immunoglobulins/HRP (Dako, P0260). After incubation with specific antibodies, we used enhanced chemiluminescence kit (Millipore) to detect the protein signals. Our method of quantification is to do a ratio between the destination bands and their matching internal reference bands after quantification. The ratio of each destination band is then normalized and presented below each destination band.

Immunofluorescence

For cell immunofluorescence analysis, C3H10T1/2 cells infected with indicated plasmids after 48 h were fixed in 4% PFA for 15 min and blocked in PBS with 3% horse serum and 0.3% Triton X-100 (Sangon Biotech, A110694-0500) for 30 min, then incubated with mouse-anti-Flag (Sigma, F3165-5MG, 1:200) overnight at 4 °C. Donkey anti-mouse-Cy3 (Jackson ImmunoResearch, 715-165-150, 1:1 000) was used as secondary antibody. DAPI (Sigma, D8417, 1:1 000) was used for counterstaining. Cells were mounted with anti-fluorescence mounting medium (Dako, S3023) and images were captured with a Leica SP8 confocal microscope.

RT-PCR analysis

Total RNA of cells was isolated in Trizol (Invitrogen) and reverse transcribed into cDNA using Transcriptor First strand cDNA synthesis kit (Roche). RT-PCR analysis was conducted with FastStart Universal SYBR Green Master (Roche) in a LightCycler480 PCR system (Roche). Relative expression of target genes was normalized to Gapdh or Hprt, and was calculated with delta-delta CT method. The primer used are listed as follows: hCCN2-qF, 5’-AAGGGCAAAAAGTGCATCCG-3’; hCCN2-qR, 5’-TCTTCTTCATGACCTCGCCG-3’; mGAPDH-qF: 5’-TTCCTACCCCCAATGTGTCC-3’; mGAPDH-qR, 5’-GGTCCTCAGTGTAGCCCAAG-3’; mAlp-qF, 5’-CGGGACTG GTACTCGGATAA-3’;mAlp-qR, 5’-ATTCCACGTCGGTTCTGTTC-3’; mCcn2-qF, 5’-CTGCAGGCTAGAGAAGCAGAG-3’; mCcn2-qR, 5’-GATGCACTTTTTGCCCTTCT-3’; mRunx2-qF, 5’-CCAACCGAGTCATTTAAGGCT-3’; mRunx2-qR, 5’-GCTCACGTCGCTCATCTTG-3’; mBglap-qF, 5’-GCAGCACAGGTCCTAAATAG-3’; mBglap-qR, 5’-GGGCAATAAGGTAGTGAACAG-3’; mHprt-qF, 5’-GTTAAGCAGTACAGCCCCAAA-3’; mHprt-qR, 5’-AGGGCATATCCAACAACAAACTT-3’.

Human serum CCN2 levels measurement

CCN2 protein levels in human serum were determined using Elisa kits (DRG Instruments GmbH, EIA-5195) following the manufacturer’s instructions. The intra-assay coefficients of variation (CVs) for CCN2 were 1.3% at a level of 110.1 ng/mL, 2.6% at a level of 49.9 ng/mL, 3.9% at a level of 129.7 ng/mL. The inter-assay CVs for CCN2 were 9.7% at a level of 27.2 ng/mL, 9.8% at a level of 15.0 ng/mL, 10.6% at a level of 31.0 ng/mL, 6.2% at a level of 32.2 ng/mL.

CCN2 knock-down studies in ZebrafishZebrafish care, maintenance, and microinjections

Adult wild-type zebrafish (AB strain) were maintained at 28.5 °C on a 14 h light/10 h dark cycle,66 and 5 to 6 pairs of them were set up for nature mating every time. The generated embryos were maintained at 28.5 °C in fish water (0.2% Instant Ocean Salt in deionized water), and were washed and staged according to previous studies.67 For ccn2a gene knock-down experiment, the ccn2a-e2i2-MO (5’-AACAGCCAAGATCCTTACCTGTGCA-3’) and the standard control MO (5’-CCTCTTACCTCAGTTACAATTTATA-3’), was designed (Gene Tools, http://www.gene-tools.com/) and microinjected into fertilized one-cell stage embryos.68 RT-PCR analysis and Sanger sequencing were performed to confirm the efficacy of the ccn2a-e2i2-MO. Total RNA of 30 to 50 embryos per group was extracted and reverse transcribed into cDNA. The primers of ccn2a spanning exon 1 and exon 3 were 5’-CTCTGCTGTTCCTGACTTTCTT-3’ (forward) and 5’-ACTTCCCAGGCACTTTCAC-3’ (reverse). The primers of ef1α, used as the internal control, were 5’-GGAAATTCGAGACCAGCAAATAC-3’ (forward) and 5’-GATACCAGCCTCAAACTCACC-3’ (reverse). For rescue experiments, 4 ng ccn2a-e2i2-MO was co-injected with 100 pg pcDNA3.1 containing nonmutant zebrafish ccn2a cDNA per embryo. The coding region of the wild-type zebrafish ccn2a was synthed by Sangon Biotech and subcloned into pcDNA3.1 vector (Invitrogen).

Calcein staining and image acquisition

Zebrafish at 5-dpf were firstly washed with fish water and immersed in 0.2% calcein solution for 10 min. And the zebrafish were then rinsed thoroughly in fish water again and anaesthetized with 0.016% MS-222 (tricaine methanesulfonate, Sigma-Aldrich). Finally, the zebrafish were oriented on ventral side and mounted with 3% methylcellulose (Sigma-Aldrich) in a depression slide for observation by fluorescence microscopy (Nikon SMZ 1500) and subsequently photographed with digital cameras. Qualification of the RFI of head skeleton bone mass was performed using morphometric analysis (NIS-Elements D3.1). To optimally visualize the expression patterns, a subset of images was adjusted for levels, brightness, contrast, hue and saturation (Adobe Photoshop 7.0 software). A total of 10 zebrafish for each group were quantified and averaged.

Ccn2fl/fl;Prx1Cre mice model analysis Mice model generation

Ccn2fl/fl mice bearing loxP sites flanking exons 1–5 of the Ccn2 gene were provided by GemPharmatech (Nanjing, China). The Prx1Cre mice strain was a gift from Andrew McMahon. Ccn2fl/fl mice were crossed with Prx1Cre mice to generate Ccn2fl/fl;Prx1Cre mice. All mice analyzed were maintained on the C57BL/6 background.

Micro-CT analysis

Left femurs isolated from age- and sex-matched 6-week-old Ccn2fl/fl;Prx1Cre mice and control littermates were fixed in 70% ethanol and scanned using SkyScan-1176 micro-computed tomography (μCT) (Bruker micro CT, Belgium) system. Scans were performed using 8.96 μm voxel size, 50 KV, 500 μA and 0.4 degrees rotation step (180 degrees angular range). The 1.6 version of NRecon software (Bruker) was used for three-dimensional (3D) reconstruction and viewing of images, and 3D photos of bones were performed using 3.3.0 version of CTvox software (Bruker). After 3D reconstruction, the 1.13 version of CTan software (Bruker) was used for bone analysis, and the segmentation were range from 85/120 to 255/255. For trabecular bone, micro-CT evaluation was performed on a 2 mm region of metaphyseal spongiosa in the distal femur, located 0.5 mm above the growth plate. For cortical bone, measurements were performed on a 0.5 mm region of the mid diaphysis of the femur. Trabecular bone measurements included BMD, BV/TV, Tb.Th, Tb.N and Tb.Sp. Cortical bone measurements included Ct.Th.

Histological staining, immunofluorescence, and immunohistochemistry

Femur and tibiae bones from mice were fixed with 4% PFA for 48 h at 4 °C, decalcified in 10% EDTA for 2 weeks, dehydrated in alcohol, cleared with xylene, and embedded in paraffin. Each sample was sectioned sagittally using microtome (Lexica Microsystems Nussle GmbH) at a thickness of 8 μm for staining. H/E staining were performed using a standard protocol. For TRAP staining, sections were stained with the Leukocyte Acid Phosphatase Kit (Sigma, 387A-1KT) for 1 h at 37 °C following the manufacturers’ instructions. For immunofluorescence, sections were blocked in PBS with 10% horse serum and 0.3% Triton X-100 (Sangon Biotech, A110694-0500) for 1 h and then stained overnight at 4 °C with rabbit-anti-Collagen II (Abcam, ab34712) or rabbit-anti-Collagen I (Abcam, ab21286). Secondary antibodies were used according to the species of the primary antibody, and DAPI (Sigma, D8417) was used for counter staining. Slides were mounted with anti-fade fluorescence mounting medium (Dako, S3023). Images were acquired with microscope (Olympus BX51, Tokyo, Japan).

RNA sequencing and analysis

We collected femurs and tibias of wild-type and Ccn2-deficient mice, removed the surrounding soft tissues, and isolated BMSCs: bone marrow was flushed out with pre-cooled PBS, followed by centrifugation and lysis of erythrocytes. The cell precipitates were gently resuspended in culture medium (α-MEM supplemented with 10% FBS and 1% penicillin/streptomycin), seeded (2 × 106 cells per dish) in 100-mm culture dishes, and incubated at 37 °C under 5% CO2 conditions. After 48 h, nonadherent cells were washed away with PBS, adherent cells were cultured for another 5 days. We collected the cells and obtained RNA for transcriptome sequencing (fold change greater than 2.0 and P value less than 0.05). The data were analyzed on the free online platform of Majorbio Cloud Platform (www.majorbio.com).

Statistics

Statistical analysis was performed using the GraphPad Prism 9 software (GraphPad Software). Cell-based experiments were performed at least twice. Animals were randomized into different groups and at least 3 mice were used for each group, unless otherwise stated. All quantitative data are presented as mean ± SD. Student’s t test, one-way ANOVA, and Mann–Whitney test was used, as appropriate, for statistical evaluations of group comparisons. Statistical significance is indicated by *, where P < 0.05, **, where P < 0.01, ***, where P < 0.001, ****, where P < 0.000 1, and all tests were two-sided.