Wild-type and cancer-related p53 proteins are preferentially degraded by MDM2 as dimers rather than tetramers

Abstract

The p53 tumor suppressor protein is the most well studied as a regulator of transcription in the nucleus, where it exists primarily as a tetramer. However, there are other oligomeric states of p53 that are relevant to its regulation and activities. In unstressed cells, p53 is normally held in check by MDM2 that targets p53 for transcriptional repression, proteasomal degradation, and cytoplasmic localization. Here we discovered a hydrophobic region within the MDM2 N-terminal domain that binds exclusively to the dimeric form of the p53 C-terminal domain in vitro. In cell-based assays, MDM2 exhibits superior binding to, hyperdegradation of, and increased nuclear exclusion of dimeric p53 when compared with tetrameric wild-type p53. Correspondingly, impairing the hydrophobicity of the newly identified N-terminal MDM2 region leads to p53 stabilization. Interestingly, we found that dimeric mutant p53 is partially unfolded and is a target for ubiquitin-independent degradation by the 20S proteasome. Finally, forcing certain tumor-derived mutant forms of p53 into dimer configuration results in hyperdegradation of mutant p53 and inhibition of p53-mediated cancer cell migration. Gaining insight into different oligomeric forms of p53 may provide novel approaches to cancer therapy.

Keywords: 20S proteasome; MDM2; mutant p53; nuclear export; p53 dimer; p53 tetramer.

Figure 1.

MDM2 preferentially binds and degrades dimer-forming mutant versions of p53. (A) Polypeptides from wild-type or L344A p53 CTDs (residues 293–393; 0.5 mM) were titrated into a solution containing 0.02 mM an MDM2 NTD polypeptide (residues 10–139), and their interactions were measured by ITC as described in the Materials and Methods. The MDM2 NTD bound the mutant p53 CTD (L344A) with an affinity of 1.7 µM. (B) The top panel shows a binding curve of p53 to MDM2 from one representative experiment. Purified full-length wild-type or E343K mutant p53 (0.03 mM) proteins were used to coat ELISA plates followed by addition of full-length purified wild-type MDM2 protein at the indicated concentrations. Complexes were detected using anti-MDM2 antibody (N-20) and enzyme-conjugated secondary antibody as described in the Materials and Methods. The bottom panel graphically depicts the cumulative binding data from six replicative experiments. The data represent the mean ± SEM for six biological replicates with two technical replicates each. (***) P-value = 0.0002, calculated by nonparametric one-way ANOVA as described in the Materials and Methods. (C, top panel) A diagram of a p53 protein showing the TAD (residues 20–60), proline domain (Pro; residues 63–97), DBD (residues 94–312), OD (residues 324–355), and C-terminal REG (residues 363–393). The positions of the dimer-forming (D) and monomer-forming (M) mutations are indicated. The two panels below the diagram are typical experiments showing results of transfecting U2OS cells with increasing amounts of Flag-MDM2 (0, 375, and 750 ng) and a constant amount of HA-tagged wild-type or OD mutant versions of p53 (150 ng) as indicated. Twenty-four hours later, cells were harvested, and lysates were used for immunoblotting with the indicated antibodies. (D) U2OS cells were transfected with 1.2 µg of Flag-MDM2 and 150 ng of HA-tagged wild-type or OD mutant p53, as indicated. Twenty-four hours after transfection, 100 µg/mL cycloheximide (CHX) was added, and cells were harvested at the indicated times. (Top panels) Cell lysates were subjected to immunoblotting with the indicated antibodies. (Bottom panels) Quantification of the immunoblotting data was carried out using ImageJ software.

Figure 2.

MDM2-dependent cytoplasmic localization of dimer-forming p53 variants. (A, top panels) U2OS cells were transfected with 150 ng of HA-p53 (wild type or mutants) in the absence or presence of 900 ng of Flag-MDM2. Twenty-four hours after transfection, cells were treated with 5 ng/mL LMB or DMSO for 5 h. Cell lysates were subjected to immunoblotting with the indicated antibodies. (Bottom panels) Quantification of the immunoblotting data was carried out using ImageJ software. In each case, values for p53 protein variants were plotted relative to their untreated counterparts (DMSO), which were taken as 1. The bar graphs represent the average of three independent experiments. P-values were calculated by comparing the LMB-treated samples with the DMSO-treated samples within each pair (B, left) Cellular localization of 150 ng of wild-type or mutant versions of p53 was detected in U2OS cells transfected with the p53 variants indicated at the top in the absence or presence of 750 ng of Flag-MDM2. Twenty-four hours after transfection, cells were fixed with 4% paraformaldehyde and analyzed by immunofluorescence microscopy after staining with anti-HA antibodies (green). Arrows indicate the presence of p53 in the cytoplasm. For the full images, including MDM2, see Supplemental Figure S5A. (Right) Quantification of cell distribution was done for three independent experiments, as described in the Materials and Methods. (C, top) U2OS cells were transfected with 150 ng of HA-p53 (wild type or mutants) in the absence or presence of 1050 ng of Flag-MDM2. Cell extracts were fractionated into cytoplasmic (Cyto) and nuclear (Nuc) portions and then analyzed by immunoblotting with the indicated antibodies. PARP and GFP were used as cytosolic and nuclear controls, respectively. (Bottom) Quantification of the immunoblotting data was carried out using ImageJ software. The graph represents the average of three independent experiments. P-values were analyzed by comparing the cytoplasmic fraction versus the nuclear fraction within each pair.

Figure 3.

Dimeric p53 is underubiquitinated by MDM2 and structurally distinct from wild-type p53. (A,B) Wild-type U2OS cells were transfected with 160 ng of HA-p53 (wild type or the indicated mutant forms) in the absence or presence of 1.2 µg of Flag-MDM2 as indicated along with 0.8 µg of Flag-ubiquitin (A) or 0.7 µg of His-UbKO monoubiquitin (B). Cells were treated with 20 µM MG132 for 4 h before harvesting, and cell lysates were immunoprecipitated with anti-HA antibody followed by immunoblotting with anti-p53 antibody (FL393-G) to detect ubiquitinated p53. (C) U2OS cells were transfected with 200 ng of constructs expressing wild type or the indicated dimer-forming mutant versions of HA-p53 or the tumor-derived core domain mutant p53 (R175H). Twenty-four hours later, cell lysates were prepared and subjected to immunoprecipitation (IP) with anti-p53 PAb 240 or Mab 1620 antibodies followed by immunoblotting with anti-HA antibody. (D) Forty nanograms of p53 proteins (wild-type p53, dimeric p53_E343A, and monomeric p53_L330A) that was purified from HCT116 cells grown in medium containing photo-L-methionine was premixed without or with increasing amounts of recombinant purified MDM2 (190–760 ng). After irradiation with UV as described in the Materials and Methods, samples were resolved by SDS-PAGE, and cross-linked p53 species were detected by immunoblotting using anti-p53 antibody (Fl-393).

Figure 4.

Tetrameric but not dimeric p53 is degraded by the 26S proteasome. (A) U2OS parental (par) or U2OS CRISPR cell lines expressing mutated dimeric p53 (E343K) (het 10, hom 8, and hom 84) were harvested, and lysates were treated with increasing amounts of glutaraldehyde (0, 0.01% and 0.1%) for 20 min at room temperature and then used for immunoblotting with p53 antibodies (1801/DO1). (B) U2OS parental (par) and U2OS CRISPR cell lines expressing mutated dimeric p53 (E343K) (hom 8 and hom 84) were transfected with 30 nM nonspecific siRNAs (siC) and two different Rpn2 siRNAs. Forty-eight hours later, cells were harvested, and lysates were used for immunoblotting with antibodies versus the indicated proteins. (C) U2OS cells were either untreated (NT) or transfected with 30 nM three different Rpn2 siRNAs and 30 nM two different nonspecific siRNAs (siC and siLuc). Forty-eight hours later, cells were harvested, and lysates were subjected to glutaraldehyde cross-linking as in A (right panel) or used for immunoblotting with the antibodies versus the indicated proteins (left panel). (D) Model depicting possible degradation mechanisms of wild-type and dimeric versions of p53. As shown in many previous studies, wild-type p53 is degraded mostly by the well-characterized ubiquitin-dependent 26S proteasomal degradation pathway, while dimeric p53 with a partially unstructured conformation is preferentially degraded by the 20S proteasome in a ubiquitin-independent manner.

Figure 5.

Mapping and characterizing the MDM2 N-terminal region required for binding to the dimer form of the p53 C terminus. (A) Fluorescence anisotropy binding studies were performed to measure the binding of an MDM2 peptide (residues 33–43) to wild-type (left panel) or the L344A dimer-forming mutant (right panel) versions of the p53 CTD_293–393 polypeptide. L344A p53 CTD_293–393 bound the MDM2 peptide with an affinity of 121 µM ± 1 µM. (B) Fluorescence anisotropy binding was performed to measure the interaction between the MDM2 NTD (amino acids 10–139) and wild-type or mutant p53 OD (residues 326–355) peptides. The wild-type p53 OD peptide was provided at concentrations of 50 nM (red +), 100 nM (red •), or 2 µM (green Δ) and were compared with dimeric mutant peptide p53 (L344A) at 100 nM (blue □). The binding curve shape is not suitable for accurate quantification because the peptide had a tendency to aggregate shortly after the analysis. (C) U2OS cells were transfected with increasing amounts of Flag-tagged version of wild-type (WT) or the indicated mutant forms of MDM2 (0, 0.5, 1, and 2 µg) along with 142 ng of HA-p53. Twenty-four hours later, cell lysates were prepared and used for immunoblotting with the indicated antibodies. (D) U2OS cells were transfected with 300 ng of wild-type or the indicated mutant forms of Flag-MDM2 along with wild-type HA-p53 (150 ng was used with wild-type MDM2, and 75 ng was used with MDM2 mutants). Cell lysates were subjected to immunoprecipitation with anti-Flag antibody followed by immunoblotting with anti-HA antibody. (E) U2OS cells were transfected with 300 ng of wild-type or mutant forms of HA-p53 as indicated in the presence of 1500 ng of wild-type or mutant (L38P) Flag-MDM2. Twenty-four hours later, cells were fixed with 4% paraformaldehyde and visualized by immunofluorescence microscopy after staining with anti-HA antibodies (green). Arrows indicate localization of p53 in the cytoplasm. For the full images, including MDM2, see Supplemental Figure S11A.

Figure 6.

MDM2 preferentially degrades and regulates cancer-derived core domain mutant forms of p53 when dimeric. (A) U2OS cells were transfected with increasing amounts of Flag-MDM2 (0, 500, and 1000 ng) and constant amounts of wild-type or the indicated core domain mutant forms of HA-p53 (200 ng), each of which contained either a wild-type OD or dimer-forming mutation (E343K) in the OD, as indicated. Twenty-four hours after transfection, cell lysates were prepared and used for immunoblotting with the indicated antibodies. Quantification of the immunoblotting data was carried out using ImageJ software. The graph represents the average of five independent experiments. P-values were calculated by comparing p53 with MDM2 with the initial levels of p53 without MDM2. (B) U2OS cells were transfected with mutant p53 (R175H) containing either wild-type OD or E343K dimer-forming mutant OD (200 ng) in the absence or presence of 1200 ng of Flag-MDM2. Twenty-four hours after transfection, cultures were scratch-wounded with a 200-µL pipette tip and treated with 2.5 µg/mL mitomycin C to prevent cell proliferation. Images of wound closure were captured by phase-contrast microscopy at 0 h and after 24 h. The graphed quantification of migration is presented as the distance between the edges of each scratch relative to 0 h. The graph represents the average of three independent experiments. Student's t-test was used to compare different mutant p53 (175) variants with or without the presence of MDM2.