Animals
Experimental protocols followed ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines. Animal protocols were reviewed and approved by the Institute Animal Care and Use Committee at the University of Texas Health Science Center in San Antonio (UTHSCSA). Bmp2 floxed allele (Bmp2fx/fx) mice was described previously 42,43, and cross-bred with the Sp7-Cre mice (Jackson Laboratory, Farmington, CT, USA) to generate Bmp2 conditional KO (cKO) mice. Bmp2 cKO (Bmp2ko/ko) and Bmp2 control mice (Bmp2fx/fx, Sp7-Cre, wild type, and Bmp fx/−) were employed, and four animals for each group were used for this study.
Cell culture
Mouse dental immortalized papilla mesenchymal (iMDP3)44, Bmp2fx/fx, and Bmp2ko/ko mesenchymal cells 45 were cultured in αMEM (Invitrogen, San Diego, CA, USA) with 10% bovine serum and penicillin (100 unit/ml) and streptomycin (100 µg/ml), and grown at 37°C. Human fetal kidney (HEK293T) cells were obtained from ATCC (Manassas, VA, USA) and grown in DMEM (Invitrogen) containing 10% bovine serum plus penicillin (100 unit/ml) and streptomycin (100 µg/ml).
Microscopic analysis of teeth
Animals were anesthetized with Ketamine (Sigma-Aldrich). The morphology of teeth was observed using a stereomicroscope. X-ray radiography was performed to examine bone and tooth changes recorded by a Faxitron radiograph inspection unit (Field X-ray Corporation, Lincolnshire, IL, USA). Digitized images of the incisors and molars were measured and assayed by the AnalySIS software to calculate the size and width of selected components. Morphology of mandibular incisors from 1-, 3-month-old control and Bmp2 cKO mice was observed by Scanning electron microscopy (SEM). The incisor surface was washed in 0.1M sodium cacodylate buffer and fixed in 2.5% (w/v) glutaraldehyde (Sigma-Aldrich) in 0.1M cacodylate buffer for 20 min and then washed in sodium cacodylate buffer and dehydrated in alcohol. The fractured incisor surface was fixed in hexamethyldisilane and sputter-coated with gold. The specimen was observed by SEM at 20 kV (JEOLJSM 6610 LV; JEOL, Inc., Peabody, MA, USA). For Micro-CT, mouse mandibles from the control and Bmp2 cKO mice were scanned using a high-resolution scanner (Scanco Medical AG, Basserdorf, Switzerland) and The Micro-CT-generated DICOM data were analyzed using Micro-view (GE Healthcare, Milwaukee, WI, USA).
Mechanical property measurements
Mandibular incisors of 1-month-old control and Bmp2 cKO mice were collected and embedded in Acrymount embedding resin (Electron Microscopy Sciences, Hatfield, PA, USA). Mechanical properties of mandibular incisors were measured using a testing machine and shear and compression tests were performed (ReNew Model 1125 Upgrade Package, MTS Systems Corporation, Eden Prairie, MN, USA). Data were analyzed using Image-Pro Plus Software (Media Cybernetics, Inc., Rockville, MD, USA).
Immunohistochemistry
The postnatal mouse mandibles were dissected, fixed, embedded, and then frontally sectioned at 4 µm. All experimental methods were followed in accordance with relevant guidelines. The slides were incubated with either anti-Dlx3 or anti-Dmp1, anti-Dsp, and anti-Sp7 antibodies, respectively at 4ºC overnight. The sections were treated with IgG as negative control (Dako Carpinteria, CA, USA). Subsequently, the slides were incubated with biotinylated secondary antibody (Vector Laboratory Inc., Burlingame, CA, USA), followed by Vectastain Elite ABC reagent (Vector Laboratory Inc.) and stained with DAB and counterstained with Mayer’s hematoxylin. The slides were observed by light microscopy. For double cell fluorescent immunostaining, iMDP3 cells were treated with or without 100 ng/ml of BMP2 (R&D System Inc., Minneapolis, MN, USA), and fixed with 70% ethanol at room temperature. The cells were blocked with 10% donkey serum. After washing, the cells were added with rabbit anti-Dlx3, goat anti-Sp7, and with rabbit anti-GCN5 and goat anti-Sp7 antibodies overnight at 4°C. After washing, the samples were incubated with the secondary antibodies of donkey anti-rabbit Alexa Fluo® 488 green (1:300) and donkey anti-goat Alex Fluo® 568 red (1:300) for 1h. Hoechst was used for nuclear staining. The slides were observed under fluorescent microscopy (Nikon, TE2000-5, JAN).
Alkaline phosphatase assay and alizarin red S staining
Cells were grown in differentiation medium (DM) with DMEM containing 10% fetal bovine serum, 1% antibiotics, 10 mM sodium β-glycerophosphate, 50 µg/ml ascorbic acid, and 100 nM dexamethasone and control with DMEM supplemented with 10% fetal bovine serum, 1% antibiotics for 0, 7 and 14 days. Then, the cells were fixed and washed in 1xPBS. In situ alkaline phosphatase (ALP) activity was carried out in accordance with the instructions. For cell mineralization, the cells were cultured in DM for 0, 7, and 14 days and fixed in 10% formalin as well as treated with 1% Alizarin red S dye (pH 4.2).
RNA extraction and quantitative real-time PCR (RT-qPCR)
The postnatal day 2 mandibles and maxillae of the control and Bmp2 cKO mice were homogenized (Bertin Technology, Rockville, MD, USA). Total RNA was isolated using TRIZOL reagent (Qiagen Inc., Valencia, CA, USA), treated with DNase I (Promega), and purified with the RNeasy Mini kit (Qiagen Inc.). RNA concentration was measured with a Bio-analyzer. Standard protocols were used to generate complementary DNA (cDNA). RT-qPCR was performed to quantitate levels of Dlx3, Dmp1, Dspp, Sp7, and cyclophilin A as an internal control using an ABI 7500 (Applied Biosystems, Foster City, CA, USA) and threshold values were calculated using SDS2 software (Applied Biosystems). Primers for RT-qPCR were represented in Table S1. Gene expression levels normalized to cyclophilin A value was calculated by ∆∆Ct method. The results from 3 separate experiments in triplicate were performed.
RNA-seq and gene expression analysis
The Bmp2fx/fx and Bmp2Ko/Ko dental mesenchymal cells were harvested, and RNA was extracted using TRIZOL reagent (Qiagen Inc.). The RNA-seq libraries were prepared from total RNAs in accordance with Illumina’s RNA specimen preparation protocol (Illumina Inc., San Diego, CA, USA). RNAs were barcoded, pooled, and sequenced with a HiSeq 2000 system with the 50 bp single read sequencing protocol and with targeted read counts of around 30 million reads per sample. Paired reads to the UCSC mm9 genome build were mapped by a TopHat2 aligner. To quantify gene expression, HTSeq was used to obtain raw read counts per gene and then converted to RPKM in accordance with gene length and total mapped read count per sample. Gene expression levels were measured by Log2-transformed RPKM. Analysis of differential expression and classification of functional annotation were assessed as described earlier 46.
Gene Ontology (GO) and Kyoto Encyclopedia of genes and genomes (KEGG) analysis
GO analysis was used to study the roles of all differentially expressed mRNAs (https://david.ncifcrf.gov/). DAVID-based KEGG analysis was used to determine the significant pathways related to the differentially expressed mRNAs. Fisher’s exact test and the x2 test were used to select the significant GO categories and pathways. The threshold of significance was *p < 0.05 and **p < 0.01.
Generation of the protein-protein interaction (PPI) network
The differentially expressed genes (DEGs) were imported into a STRING database, and the species was limited to “Mus musculus” to obtain PPI information. The network relationships with high confidence (≥ 0.75) were screened and imported into Cytoscape 3.6.2 to draw a PPI network diagram.
Western blot analysis
Tissues and cells were lysed with RIPA lysis buffer (Sigma-Aldrich). Protein levels were calculated by the BCA Protein Assay kit (Pierce Biotechnology Inc., Rockford, IL, USA). The same amounts of proteins were subjected to SDS-PAGE gels and Western blotting using specific antibodies as described above.
Protein phosphorylation
iMDP3 cells were induced either with or without BMP2 (100 ng/ml) for the given times and lysed with RIPA buffer (Sigma-Aldrich). The cell lysate was separated with SDS-PAGE gel electrophoresis and blotted with anti-Akt, anti-pAkt, anti-Erk42/44, anti-pErk42/44, anti-p38, anti-p-p38, and β-actin antibodies. To assess the phosphorylation of Dlx3, Sp7, and GCN5 induced by BMP2, cells were treated or untreated with BMP2 (100 ng/ml) and lysed. Dlx3, Sp7 and GCN5 proteins were immunoprecipitated with anti-Dlx3, or anti-Sp7, anti-GCN5 antibody, respectively. Immunoprecipitated Dlx3, Sp7, and GCN5 proteins were immunoblotted by anti-pTyr and anti-pThr antibodies. To ascertain if BMP2 induces phosphorylation of Dlx3, Sp7, and GCN5 at Tyr and Thr residues by Akt and Erk42/44 kinases, the cells were stimulated or un-stimulated with BMP2 as well as with or without Akt kinase inhibitor, GSK630693 (10–40 µM) or Erk kinase inhibitor, U0126 (10–40 µM, Sigma-Aldrich), respectively. The cells were lysed, and Dlx3, Sp7, and GCN5 were immunoprecipitated by anti-Dlx3, anti-Sp7, and anti-GCN5 antibodies. BMP2-mediated phosphorylation of Akt and Erk42/44 kinases inhibited by U0126 and GSK630693 were analyzed by Western blotting using anti-pAkt and anti-pErk42/44 antibodies.
Chromatin immunoprecipitation (ChIP) and Re-ChIP assays
ChIP was preceded in accordance with the ChIP-IT TM kit (Active Motif). Immunoprecipitation was processed with protein G agarose beads and 10 µg of anti-Dlx3 or anti-Sp7 antibody with end-over-end mixing overnight at 4°C. IgG was a negative control. Immunoprecipitated samples were washed, and cross-links reversed. Recovered materials were incubated with proteinase K and the DNA fragments were purified using DNA purification columns (Qiagen Inc.). To assess the dynamic BMP2 effect of binding of Dlx3 and Sp7 to Dspp and Dmp1 promoters, iMDP3 cells were treated or untreated with BMP2. The cell lysate was precipitated with anti-Dlx3 or anti-Sp7 antibody and DNA fragments were purified and amplified by qPCR. Primers for qPCR were shown in Table S2. To determine the effect of Dlx3 on mouse Sp7 promoter, cells were transfected with Dlx3 expression plasmid. After 48h, the ChIP assay was carried out as described above. DNA fragments were purified and amplified by the primers present in Table S3. For Re-ChIP, chromatin complexes from the first ChIP (anti-GCN5 or anti-pol II) were collected and centrifuged. The supernatant was diluted in Re-ChIP buffer. The second IP antibody (either anti-pol II or anti-GCN5) was added, followed by crosslink reversal and DNA purification. The first ChIP and Re-ChIP products were analyzed by PCR that would amplify the 379-bp and 231-bp fragments of Dspp and Dmp1 genes encompassing the GCN5 and pol II binding sites. The primers for PCR were present in Table S4.
Electrophoretic mobility shift assay (EMSA)
Nuclear proteins from iMDP3 cells were prepared as previously described 4. All manipulations were performed at 4°C. Protein levels were measured by the Bradford assay (Bio-Rad Laboratories, Inc., Hercules, CA, USA). For EMSA, the double-stranded DNA probes were labeled with [γ-32P] ATP and T4 polynucleotide kinase. EMSA was carried out described earlier 4. All oligonucleotides used in EMSA were shown in Table S5.
Transfection assay
Plasmids of the mouse Dspp promoters were created and described earlier 4. In brief, the Dspp elements between nucleotides (nt) -97 and + 54, -318 and + 54, -791 and + 54, -1,243 and + 54, and − 2,644 and + 54 were inserted into pGL-3 basic luciferase plasmid (Promega), to generate plasmids, p97, p318, p791, p1243, and p2644. These reporter plasmids were co-transfected with either pcDNA-Dlx3 or/and pcDNA-Sp7 plasmids, empty pcDNA3.1 plasmid as control with a pRL-TK plasmid (Promega) into iMDP3 cells by Lipofectamine2000TM. For mutant DNA constructs, mutant plasmids were created by a mutagenesis kit using p318 DNA construct as a template (Invitrogen). DNA sequencing was employed to verify corrective constructs. After 48h transfection, the cells were collected, and Dspp promoter activity was assayed with the dual luciferase reporter assay system. The Dspp promoter activity was measured by the ratio of firefly/Renilla luciferase for each construct. The experimental data are present to the mean ± S.E. from 3 separate experiments in triplicate.
Immunoprecipitation
For in vitro protein-protein interaction, full length, NH2- and COOH-fragments of mouse Dlx3, Sp7, and GCN5 cDNAs were subcloned into pGEX tagged with glutathione fusion protein gene, respectively. The pGST fusion protein expression and purification were preceded in accordance with the manufacturer’s instruction (Amersham Pharmacia Biotech, Piscataway, NJ, USA). GST-Sp7 proteins were incubated with GST-Dlx3 proteins and with GST-GCN5 proteins, respectively, and then added anti-Sp7 or anti-Dlx3, anti-GCN5 antibodies in the reaction mixtures with end-over-end mixing. IgG was the negative control. After incubation, protein G agarose beads (Invitrogen) were added to the reaction and further incubation. After centrifugation, the supernatant was removed, and an equal volume of 2 x SDS gel-loading buffer was added to the beads and then heated for 5 min. The samples were subjected to Western blotting. For in vivo immunoprecipitation, the full-length, NH2- and COOH-domains of Sp7, Dlx3, and GCN5 genes were inserted into mammalian expression vector pcDNA tagged with Flag or c-Myc (Sigma-Aldrich), respectively. iMDP3 and HEK293 cells were transfected with pcDNA-Flag-Sp7, pcDNA-Myc-Dlx3 and pcDNA-Flag-Sp7, pcDNA-Myc-GCN5 genes. After 48h transfection, the cells were washed and scraped. The scarped samples were centrifuged. The supernatant was collected and rotated after adding anti-Flag or anti-Myc antibody (Thermo Fisher Scientific), further incubated with protein G agarose beads (Invitrogen). After the reaction, the samples were washed with lysis buffer and added SDS gel-loading buffer. After centrifugation, the supernatant was collected, electrophoresed to SDS-PAGE gels and a Western blotting assay was performed.
Sp7 and histone H3 acetylation by GCN5
In vitro Sp7 acetylation was performed by a HAT assay kit (Active Motif). In brief, GST fusion Sp7 (5µg) or 50 µM of histone H3 peptide (Active motif) was incubated with 500 ng of recombinant p300 catalytic domain (Active motif) or recombinant GCN5 protein (Active motif). For acetylation inhibition, anacardic acid (150 µM) was added to the reaction. After incubation, the reaction mixture was added by stop solution, followed by 20 µl of the peptide substrate. 100 µl of the final developing solution was then added to each well and incubated for 15 min in dark at room temperature. The samples were read at fluorescence with excitation at 360–390 nm and emission at 450 to 470 nm on a SpectraMAX Gemini XS plate reader (Molecular Devices, San Jose, CA, USA). To determine the GCN5 effect on Sp7 acetylation in vivo, pcDNA-Sp7, and pcDNA-GCN5 or pcDNA3.1 as control were transfected into iMDP3 cells respectively. After 48h transfection, the cells were washed, lysed, and centrifuged. The supernatant was collected and rotated after adding an anti-Sp7 antibody, then incubated with protein G agarose beads (Invitrogen). After the reaction, the beads were washed with lysis buffer, added SDS gel-loading buffer, and boiled. After centrifugation, the supernatant was collected and electrophoresed SDS-PAGE gel and blotted with anti-Sp7 and anti-acetyl-lysine antibodies (Santa Cruz Biotechnology Inc.). For in vivo assay of Sp7 acetylation induced by BMP2, the pcDNA-Myc-Sp7 vector was transfected into iMDP3 cells and then treated or untreated with BMP2 (100 ng/ml), and with or without 60 nm of Trichostatin A for 2h, 12h, and 24h and then lysed. The co-immunoprecipitation was carried out as described above using an anti-Myc antibody to immuno-precipitate the complexes and subjected to Western blot using anti-Sp7 and anti-acetyl lysine antibodies.
Statistical analysis
Experimental data are shown as means ± S.D. from 3 independent experiments in triplicate. The difference in statistical significance was analyzed with a one-way t-test using GraphPad Prism8 software (GraphPad Software, Inc. La Jolla, CA, USA). The statistical differences among various groups were significant at *p < 0.05 and **p < 0.01.