p53 and MDM2 are involved in the regulation of osteocalcin gene expression

Abstract

Osteocalcin (OC) is a major noncollagenous bone matrix protein and an osteoblast marker whose expression is limited to mature osteoblasts during the late differentiation stage. In previous studies we have shown osteosarcomas to lose p53 function with a corresponding loss of osteocalcin gene expression. Introduction of wild type p53 resulted in re expression of the osteocalcin gene. Using gel shift and chromatin immunoprecipitation assays, we have identified a putative p53 binding site within the rat OC promoter region and observed an increase in OC promoter activity when p53 accumulates using a CAT assay. The p53 inducible gene Mdm2 is a well-known downstream regulator of p53 levels. Our results showed a synergistic increase in the OC promoter activity when both p53 and MDM2 were transiently overexpressed. We further demonstrate that p53 is not degraded during overexpression of MDM2 protein. Increased OC expression was observed with concomitantly increased p53, VDR, and MDM2 levels in ROS17/2.8 cells during treatment with differentiation promoting (DP) media, but was significantly decreased when co-treated with DP media and the small molecule inhibitor of MDM2-p53 interaction, Nutlin-3. We have also observed a dramatic increase of the OC promoter activity in the presence of p53 and Mdm2 with inclusion of Cbfa-1 and p300 factors. Our results suggest that under some physiological conditions the oncoprotein MDM2 may cooperate with p53 to regulate the osteocalcin gene during osteoblastic differentiation.

Figure 1. p53 binds to the rOC promoter region

(A) The rat OC promoter region contains a putative p53 response element at the position of −690 to −667 bp. The TATA box site is also indicated. Black boxes and Numbers, exons. (B) EMSA assay. NE, nuclear extracts from p53 over-expressed ROS17/2.8 cells; Labeled, Biotin-labeled probes; Unlabeled, competitive probes. (C) ChIP assay. Rat ROS17/2.8 cells were cross-linked, sonicated and then immunoprecipitated using an anti-p53 monclonal antibody. The known p53RE sequence in the rat p21 promoter was used as a p53RE positive control. Input, sonicated and precleared supernatant; IP-p53, p53 antiboby-immunoprecipitated samples; IgG, negative controls treated with normal mouse IgG antibody. Suggestion for this figure: use a TRANSFAC matrix motif in place of the old paper, as it is more current and gives a better representation of the possible nucleotides at each location. Also, for 1B, you cannot be certain that p53 is binding to the EMSA motif as there is no supershift by an antibody specific to p53 and a cell lysate was used.

Figure 2. Leptomycin B and MG132 both increase the hOC promoter activity

The 884-bp pOSCAT3 construct was transiently co-transfected into ROS17/2.8 cells with p53 expression vector, with or without 5 nM leptomycin B (A) or 20 μM MG132 (B). The CAT assays were done 48 hr after transfection, and the data are representative of 6–8 independent experiments. P values were determined by Student’s t-test. **, P < 0.05. (C) p53 and Mdm2 proteins were detected by Western blotting with β-Actin as a control. Relative expression levels of p53 and Mdm2 are indicated after normalizing to β-Actin. LPB, leptomycin B.

Figure 3. Mdm2 synergizes with p53 in the regulation of the hOC promoter activity

(A) CAT assay for the human OC promoter activity was conducted in ROS17/2.8 cells with transient transfections of pOSCAT3, p53 and/or Mdm2 expression vectors. All the data is representative of 6–8 independent experiments. (B) Reduction of endogenous p53 and Mdm2 levels reduces the osteocalcin promoter activity. p53 and Mdm2 expression was reduced using respective shRNA in transient transfections with the pOSCAT3 and the CAT activity was measured 48 hr later. The results represent average +SE of triplicates of an experiment that was repeated thrice. P values were determined by Student’s t-test. **, P < 0.05. (C) and (D), p53 and Mdm2 levels were detected by Western blotting in ROS17/2.8 cells co-transfected with p53 and Mdm2 expression vectors or shRNAs. Relative expression levels of p53 and Mdm2 are indicated after normalizing to β-Actin. (E) Immunoprecipitation (IP) and immunoblotting (IB) assays for detection of p53-Mdm2 binding complex. Equal amounts of lysates from the above experiments were immunoprecipitated with p53 (the upper panel) or Mdm2 (the bottom panel) and followed by IB with Mdm2 or p53 antibodies, respectively.

Figure 4. Disruption of p53-Mdm2 interaction with Nutlin-3 reduces the hOC promoter activity

(A)Transient transfections were carried out with pOSCAT3, p53 and/or Mdm2 in the presence and absence of Nutlin-3. The experiment was repeated thrice in triplicates and a representative result is shown. P values were determined by Student’s t-test. **, P < 0.05. (B) p53 and Mdm2 proteins were detected by Western blotting with β-Actin as a control. Relative expression levels of p53 and Mdm2 are indicated after normalizing to β-Actin. (C) Treatment with Nutlin-3 reduces p53 association with Mdm2. Equal amounts of lysates from the above experiment was immunoprecipitated with Mdm2 followed by Western blotting with p53 antibodies to determine the amount of p53 bound to Mdm2 under the conditions of the experiment.

Figure 5. DP media and Nutlin-3 treatments cause to changes in the expression levels of rOC, p53, and MDM2

(A) to (C), Rat ROS17/2.8 cells were exposed to DP media with or without 10 μM Nutlins, and the mRNA levels of p53, Mdm2 and osteocalcin were determined using a quantitative RT-PCR method. Relative quantification of gene expression was determined by using fold change relative to Day 0 and normalized against 18sRNA. DP experiment with Nutlins could not be extended longer than 6 days due to massive apoptosis of cells. P values were determined by Student’s t-test. **, P < 0.05. (D) The expression of endogenous p53, VDR and Mdm2 proteins were detected using Western blotting in ROS17/2.8 cells subjected to the treatment of DP media with or without Nutlins. Relative expression levels of p53, Mdm2 and VDR proteins are indicated after normalizing to β-Actin.

Figure 6. Inclusion of known Cbfa-1 and p300 optimizes the hOC promoter activity in the presence of p53 and Mdm2

CAT assay for the human OC promoter activity was conducted ROS17/2.8 cells with transient transfections of pOSCAT3, p53, Mdm2, Cbfa-1 and p300 expression vectors. All the data is representative of 3–4 independent experiments.