Bone morphogenetic protein 2 mediates dentin sialophosphoprotein expression and odontoblast differentiation via NF-Y signaling

Abstract

Dentin sialophosphoprotein (DSPP), an important odontoblast differentiation marker, is necessary for tooth development and mineralization. Bone morphogenetic protein 2 (BMP2) plays a vital role in odontoblast function via diverse signal transduction systems. We hypothesize that BMP2 regulates DSPP gene transcription and thus odontoblast differentiation. Here we report that expression of BMP2 and DSPP is detected during mouse odontogenesis by in situ hybridization assay, and BMP2 up-regulates DSPP mRNA and protein expression as well as DSPP-luciferase promoter activity in mouse preodontoblasts. By sequentially deleting fragments of the mouse DSPP promoter, we show that a BMP2-response element is located between nucleotides -97 and -72. By using antibody and oligonucleotide competition assays in electrophoretic mobility shift analysis and chromatin immunoprecipitation experiments, we show that the heterotrimeric transcription factor Y (NF-Y) complex physically interacts with the inverted CCAAT box within the BMP2-response element. BMP2 induces NF-Y accumulation into the nucleus increasing its recruitment to the mouse DSPP promoter in vivo. Furthermore, forced overexpression of NF-Y enhances promoter activity and increases endogenous DSPP protein levels. In contrast, mutations in the NF-Y-binding motif reduce BMP2-induced DSPP transcription. Moreover, inhibiting BMP2 signaling by Noggin, a BMP2 antagonist, results in significant inhibition of DSPP gene expression in preodontoblasts. Taken together, these results indicate that BMP2 mediates DSPP gene expression and odontoblast differentiation via NF-Y signaling during tooth development.

FIGURE 1.

Effect of BMP2 on DSPP expression in mouse preodontoblasts. A, quantitative RT-PCR analysis of DSPP mRNA expression from MD10-F2 cells treated with or without BMP2 (100 ng/ml) at 12, 24, 48, and 72 h. Values were expressed as fold increase versus cells without BMP2 as 1.0-fold. * indicates significant differences between the BMP2-treated and -untreated cells (*, p < 0.05). B, qRT-PCR products from one of three experiments were run onto 1.5% agarose gels and stained with ethidium bromide. M, DNA marker; Neg, negative control; Cyclo, cyclophilin serves as an internal control. Con and bmp2 lanes show MD10-F2 cells without or with BMP2 stimulation, respectively. C, DSPP protein expression in MD10-F2 cells in the presence or absence of BMP2. MD10-F2 cells treated with or without BMP2 were incubated with anti-DSP polyclonal antibody. Mouse IgG1 was used as a negative control. The samples were then incubated with the secondary antibody conjugated to Alexa Fluo® 488 (Molecular Probes). Alexa Fluo® 488 staining images were obtained under the same parameters in an Olympus wide field microscope and quantified by means of MetaMorph software. Bri and Fluo indicate bright and fluorescent fields, respectively.

FIGURE 2.

The DSPP promoter contains a BMP2-response element. A, schematic representation of the constructs used in the luciferase (Luc) assay. B, BMP2 stimulates DSPP promoter activity. Transient MD10-F2 transfectants in the presence or absence of BMP2 (100 ng/ml) for 12 h were used to determine transcriptional activity of those chimeric constructs. The value obtained from the control group (pGL3 basic only) was taken as 1-fold, and fold increases were calculated by dividing the individual value by the control group value and plotted as a graph showing the mean ± S.E. from three independent experiments in triplicate. Significant differences comparing BMP2-treated groups with BMP2-untreated groups are shown with the following probability levels: *, p < 0.05; **, p < 0.001.

FIGURE 3.

NF-Y binding to the BMP2-response element in the mouse DSPP promoter. A, ³²P-labeled double strand probe between nt –97 and –72 was incubated with MD10-F2 cell nuclear extracts in the presence or absence of 50- and 100-fold molar excesses of unlabeled competitor DNA described in Table 1. WT, wild type. B, mouse DSPP proximal promoter nucleotide sequences are shown from nt –100 to –66. Several potential transcription factor-binding sites are present in this element. C, comparison of mouse, rat, and human DNA sequences from –100 to –66 positions. Dots represent nucleotide identity. D, NF-Y antibody supershift assay. MD10-F2 cell nuclear extracts were preincubated with antibody to NF-Y subunits or serum, respectively, and then with ³²P-labeled wild-type (lanes 1–9) or ³²P-labeled NF-Y probes (lanes 10–18). E, NF-Y-binding site schematic in the mouse DSPP promoter. The black box shows the NF-Y-binding box. The arrows indicate specific primers for ChIP assay. F, ChIP assay. ChIP analysis was performed as described under “Experimental Procedures.” Lane 2 shows a negative control. Lane 3 shows the TFIIB antibody was bound to TFIIB-binding site in the GAPDH promoter region. The immunoprecipitated and purified DNA fragments as a template were amplified by GAPDH PCR primers. Lane 4 shows input DNA amplified by mouse DSPP primers. Lane 5 shows the DSPP target was effectively immunoprecipitated by the anti-NF-YB antibody.

FIGURE 4.

DSPP, BMP2, and NF-Y gene expression in tooth organs during development. A–L, in situ hybridization of mouse tooth developmental stages from E15 to PN1 with BMP2 (B, D, and F) and DSPP (H, J, and L) antisense probes. Hematoxylin was shown in A, C, E, G, I, and K. od, odontoblasts; am, ameloblasts; p, dental pulp cells; c, cartilage; eo, enamel organ; dp, dental papilla; e, enamel; d, dentin. Bar scale is 200 μm. M, effect of BMP2 on ALP activity in MD10-F2 cells. After the cells were treated with BMP2 (100 ng/ml) or without BMP2 (con) for 3 and 6 days, ALP activity was determined as described under “Experimental Procedures.” Data were means ± S.E. of three wells, significantly different from the control without BMP2. *, p < 0.001. N, RT-PCR amplification of the three NF-Y subunits were carried out. Ten μl of PCR products were analyzed on 1.5% agarose gels. Amplification products of mouse β-actin served as an internal control. M, 100-bp DNA ladder. Arrows indicate expected PCR products of the three NF-Y subunits, NF-YA, NF-YB, and NF-YC, and β-actin genes. Con, negative control; 10-F2, MD10-F2; 2wk, 10 wk, and 13 wk indicate mouse molars at aged 2, 10, and 13 weeks after birth, respectively.

FIGURE 5.

Effect of BMP2 on NF-Y expression in MD10-F2 cells. A, qRT-PCR analysis of NF-Y subunit mRNA expression in MD10-F2 cells. Cells were grown in the presence or absence of BMP2 at the given time points. NF-Y mRNA expression in MD10-F2 cells in the absence of BMP2 served as a 1.0-fold increase. * indicates significant differences between the BMP2-treated and -untreated cells (p < 0.05). B, qRT-PCR products from one of three experiments were run on 1.5% agarose gels and stained with ethidium bromide. M, DNA marker; Neg, negative control; Cyclo, cyclophilin served as an internal control. Con and bmp2 show MD10-F2 cells without or with BMP2 induction, respectively. C, MD10-F2 cells stimulated with or without BMP2 for the indicated times were subjected to ChIP with anti-NF-YB antibody. Immunoprecipitated and purified DNA fragments were subjected to PCR with primers covering the NF-Y-binding site of the mouse DSPP promoter.

FIGURE 6.

Inhibition of BMP 2 effect of NF-Y expression in MD10-F2 cells by cycloheximide. MD10-F2 cells were treated with BMP2 (100 ng/ml) in the presence or absence of CHX (10 μg/ml) at the indicated time points. Protein expression of NF-YA, NF-YB, NF-YC, DSP, and β-actin was detected by Western blot analysis as described under “Experimental Procedures.”

FIGURE 7.

NF-Y-binding site is important in BMP2-mediated DSPP promoter activity. A, illustration of wild-type and mutant DSPP promoter-luciferase reporter gene constructs. The diagram shows the relative positions and sequences of the NF-Y-binding site with mutated sequences shown in lowercase. B, effect of BMP2 on DSPP promoter activity. MD10-F2 cells were transfected with p97wt-luc, p97mut-luc, or empty vector (pGL3-basic only). pRL-TK (Renilla luciferase) co-transfection served as an internal control. These transfectants were treated or untreated with recombinant BMP2 (100 ng/ml) for 12 h before cell harvest. The value (ratio between firefly and Renilla luciferase) was compared with the control group (pGL3-basic only). The fold luciferase activity was determined by dividing the individual values by the control group value. The data are expressed as the mean ± S.E. from at least three separate experiments performed in triplicate. Significant differences comparing BMP2-treated cells with untreated cells are shown with following probability levels: **, p < 0.01. C, effect of BMP antagonist Noggin on BMP2-mediated DSPP transcription. Transient transfection p97 analysis in the presence or absence of BMP2 (100 ng/ml) for 12 h in MD10-F2 cells is shown with the effects of Noggin at different doses. The results obtained from three separate transfections show the mean ± S.E. Significant differences compared with groups untreated with Noggin are shown with the following probability levels: *, p < 0.05; **, p < 0.001.

FIGURE 8.

Effect of NF-Y on DSPP expression in MD10-F2 cells. A, cells were transiently co-transfected with either p97wt-luc, p97mut-luc, or an empty vector (pGL3-basic only) and an NF-Y (NF-YA, NF-YB, and NF-YC) or pcDNA 3.1 (mock) expression vector. All cells were co-transfected with pRL-TK (Renilla luciferase) and served as an internal control. After a 48-h transfection, transfectants were harvested, and luciferase activity was measured. The value (ratio between firefly and Renilla luciferase) was obtained from the control group (pGL3-basic empty vector only). The fold luciferase activity was determined by dividing the individual value by the control group value. Bars represent the mean ± S.E. from at least three separate experiments performed in triplicate. B, endogenous DSP protein expression by NF-Y. MD10-F2 cells were transfected with 5 μg of either pcDNA 3.1 (mock) or NF-Y (NF-YA, NF-YB, and NF-YC) expression vector. Total cellular proteins were isolated after a 72-h post-transfection. Forty μg of total cellular lysates were run on 10% SDS-polyacrylamide gels and subsequently electroblotted. Membranes were probed with anti-mouse DSP antibody. Lanes 1 and 2, pcDNA 3.1; lanes 3 and 4, NF-Y. Arrowhead shows detected DSP. Arrow indicates the internal control, β-actin.