Endogenous human MDM2-C is highly expressed in human cancers and functions as a p53-independent growth activator

Abstract

Human cancers over-expressing mdm2, through a T to G variation at a single nucleotide polymorphism at position 309 (mdm2 SNP309), have functionally inactivated p53 that is not effectively degraded. They also have high expression of the alternatively spliced transcript, mdm2-C. Alternatively spliced mdm2 transcripts are expressed in many forms of human cancer and when they are exogenously expressed they transform human cells. However no study to date has detected endogenous MDM2 protein isoforms. Studies with exogenous expression of splice variants have been carried out with mdm2-A and mdm2-B, but the mdm2-C isoform has remained virtually unexplored. We addressed the cellular influence of exogenously expressed MDM2-C, and asked if endogenous MDM2-C protein was present in human cancers. To detect endogenous MDM2-C protein, we created a human MDM2-C antibody to the splice junction epitope of exons four and ten (MDM2 C410) and validated the antibody with in vitro translated full length MDM2 compared to MDM2-C. Interestingly, we discovered that MDM2-C co-migrates with MDM2-FL at approximately 98 kDa. Using the validated C410 antibody, we detected high expression of endogenous MDM2-C in human cancer cell lines and human cancer tissues. In the estrogen receptor positive (ER+) mdm2 G/G SNP309 breast cancer cell line, T47D, we observed an increase in endogenous MDM2-C protein with estrogen treatment. MDM2-C localized to the nucleus and the cytoplasm. We examined the biological activity of MDM2-C by exogenously expressing the protein and observed that MDM2-C did not efficiently target p53 for degradation or reduce p53 transcriptional activity. Exogenous expression of MDM2-C in p53-null human cancer cells increased colony formation, indicating p53-independent tumorigenic properties. Our data indicate a role for MDM2-C that does not require the inhibition of p53 for increasing cancer cell proliferation and survival.

Figure 1. An mdm2 splice variant transcript, mdm2-C is highly expressed in MDM2 over-expressing cells.

A. Quantitation using Image J after northern blot analysis of RNA from untreated MANCA, SJSA-1, ML-1 and K562 cell samples. An mdm2 DNA probe to exon 12 was PCR generated and radiolabelled with α ³²P dCTP. Transcript levels were compared to K562 for basal expression and normalized for RNA levels to gapdh. An average of four independent experiments is shown. Error bars indicate standard error. B. Schematic of mdm2 messages detected using a Taqman probe for exons 6 and 7 (6-7 probe) or forward primer for 4:10 and reverse primer for 12 (4:10-12 primer) for mdm2-C. C. qRT-PCR with 4:10 forward and exon 12 reverse vs. Taqman probe to exon 6 and 7 of mdm2 were performed to detect mdm2-C and mdm2 (with exons 6 and 7) transcripts. The mdm2-C transcripts were detected via Syber Green and mdm2 (with exons 6 and 7) transcripts were detected via Taqman technology. Transcript levels were compared to K562 for basal expression and normalized for RNA levels to gapdh. An average of three independent experiments is shown. Error bars indicate standard error. D. qRT-PCR of RNA using Taqman technology from p53 target genes, p21 and puma after DNA damage treatment with 8μM etoposide for 3 hours. Data from each cell line is presented as normalized to its own untreated control sample for fold activation and normalized for RNA levels to gapdh. An average of three independent experiments is shown. Error bars indicate standard error.

Figure 2. An MDM2-C specific antibody named C410, detects the MDM2-C protein.

A. Schematic of full length MDM2 (MDM2-FL) and MDM2-C. The retained proteins’ biochemical functional domains are shown as color codes. The peptide sequence used as an immunogen containing the splice junction of human MDM2-C is shown. Glycine (G) and Cysteine (C) residues were added to the N-terminus of the peptide to facilitate conjugation to an immunoreactive protein, keyhole limpet hemocyanin (KLH). B. ³⁵S methionine was used as a radioactivity source to label in vitro translated proteins. In vitro translations of pcDNA3-mdm2-FL and pcDNA3-P2mdm2-C using the TNT coupled wheat germ extract system. Resulting MDM2-FL and MDM2-C protein were electrophoresed on a10% SDS-PAGE in a 5:1:1 ratio. The gel was transferred to a nitrocellulose membrane and exposed to film for significant MDM2-C protein product detection. Wheat germ lysate without DNA was used as a negative control. C. Immunoprecipitation of ³⁵S methionine radioactive-labeled in vitro translated MDM2-FL and MDM2-C proteins using MDM2 C410 and pre-immune polyclonal serum antibodies. Protein ratios as shown in B were used in the pull down assay. Samples were electrophoresed on a 10% SDS-PAGE gel transferred to a nitrocellulose membrane and exposed to film for protein detection. This is representative of three independent experiments. D. In vitro translated protein made in rabbit reticulocyte lysate (RRL lanes 1 and 2) were compared to protein translated in wheat germ extract (WGE lanes 3 and 4). These proteins were detected with either antibody 4B11 (top panel) or 2A9 (bottom panel). HRP-conjugated anti-mouse and anti-rabbit were used as secondary antibodies.

Figure 3. Human MDM2-C migrates at 98 kDa.

Immunoprecipitation using MDM2 C410 and pre-immune polyclonal serum antibodies with HeLa in vitro translated MDM2-C extracts using pre immune sera, MDM2 C410 and N-20 polyclonal antibodies. Samples were electrophoresed onto a 10% SDS-PAGE gel in duplicate. A. One half was stained with coomassie blue for protein detection. B. Samples from half of gel were transferred to nitrocellulose membrane and MDM2 was detected with 4B11 monoclonal antibody. HRP-conjugated anti-mouse was used as secondary antibody. Arrows depict MDM2-C protein.

Figure 4. High expression of endogenous MDM2-C protein in G/G mdm2 SNP309 MDM2 over-expressing cells.

A. Western blot analysis of whole cell extracts from: MANCA, SJSA-1, ML-1 and K562 cells. MDM2-C protein levels were analyzed via MDM2 C410 polyclonal serum antibodies (C410, lanes 1-4) and total MDM2 was detected with the monoclonal 4B11 (lanes 9-12 and long exposure for lane 12). Actin was used as a loading control. Pre immune polyclonal serum was used as a negative control. HRP-conjugated anti-mouse and anti-rabbit were used as secondary antibodies. This is representative of three independent experiments. Bi. MANCA cells were lysed either as whole cell extracts (WCE) or into cellular compartments- cytosolic (CYTO) and Chromatin (CHR). Proteins were detected as in A. MDM2-C protein levels were analyzed via MDM2 C410 polyclonal serum antibodies (C410, lanes 1-3) and total MDM2 was detected with the monoclonal 4B11 (lanes 1-9 and long exposure for lane 9). Bii. Tubulin and Fibrillarin were used to show efficient cellular fractionation of extract. C. Spinning disk confocal microscopy of MANCA cells. Cells were fixed, permeabilized and incubated with p53, MDM2 C410 and pre-immune polyclonal serum antibodies. Slides were incubated with secondary Alexa-conjugated goat anti-rabbit and FITC-conjugated goat anti-mouse. DAPI was used to stain the cell nuclei. Pictures were taken at 60X magnification. Arrows indicate regions of Mdm2-C protein nuclear localization.

Figure 5. MDM2-C increases with estrogen treatment in ER+ Breast cancer cells and is located in the cytoplasm and nucleus.

A. MCF-7 and T47D cells were grown and treated with 10nM estrogen (E2) for five days. Cells were lysed and proteins were analyzed via western blot. MDM2 C410 polyclonal serum anti-rabbit antibody was used for protein detection. Pre-immune polyclonal serum was used to detect background signal. HRP-conjugated anti-mouse and anti-rabbit were used as secondary antibodies. Actin was used as a loading control. This is representative of three independent experiments. Bi. MCF-7 and T47D cells were lysed for whole cell or chromatin fractionation extracts. 50μg of samples were resolved using 10% SDS-PAGE and MDM2 C410 polyclonal sera (lanes 1 - 6) and MDM2 monoclonal antibody, 4B11 (lanes 13 - 18) were used for protein detection. Pre immune was used to detect background signal (lanes 7 - 12). Secondary antibodies used were same as in A. Bii. Tubulin and Fibrillarin were used to determine fractionation purity. C. Spinning disk microscopy of MCF-7 and T47D cells. Cells were grown on coverslips and treated with 10nM E2 for five days. Cells were fixed and incubated with p53 antibody and Mdm2 C410 polyclonal antibody for protein detection. Pre-immune serum was used as a negative control for staining. Slides were incubated with secondary Alexa-conjugated goat anti-rabbit and FITC-conjugated goat anti-mouse. DAPI was used to stain the nuclei. Cells were visualized at 60X magnification. Arrows depict MDM2-C localization regions.

Figure 6. MDM2-C is highly expressed in liposarcoma and breast carcinoma tissues.

A. Immunohistochemistry of lipoma and liposarcoma tissues using MDM2 C410 and pre-immune polyclonal serum antibodies. Biotinylated secondary antibody, ABC Reagent and DAB from Vector Labs were used. Pictures were obtained at 20X and 40x magnification. H&E refers to Hematoxylin and Eosin counterstaining. B. Breast tissue arrays (TMA-1008) were purchased from protein biotechnologies and automated histology was performed at the Molecular Cytology Core Facility at Memorial Sloan Kettering Cancer Center. Antibodies MDM2 C410 and MDM2 4B11 were used for staining. Mouse IGG was used for negative staining. H&E refers to Hematoxylin and Eosin counterstaining.

Figure 7. Human MDM2-C interacts with a variety of proteins in vivo.

Immunoprecipitation from MANCA whole cell extracts using MDM2 C410 and pre-immune polyclonal serum antibodies antibodies. Samples were electrophoresed onto a 10% SDS-PAGE gel in duplicate. A. One half was stained with coomassie blue for protein detection and protein bands from A were excised for analysis via LC/MS/MS. B. Samples from half of gel A were transferred to nitrocellulose membrane and MDM2 protein was detected using the MDM2 monoclonal antibody mix (4B2, 2A9, 4B11). A. Bracket represents region where bands were cut out for LC/MS/MS analysis. B. Arrows depict MDM2-C protein.

Figure 8. MDM2-C has p53-independent transformation activity.

A. Analysis of exogenously expressed MDM2-FL or MDM2-C in H1299 cells with simultaneous expression of p53. Increasing amount of pcDNA3-mdm2-FL and pcDNA-P2mdm2-C plasmid DNAs were transiently transfected with a constant amount of SN3 plasmid (p53) into H1299 p53-null cells. PGL2 Basic plasmid was used as a DNA normalizer. Cells extracts were prepared 48 hours after transfection and analyzed for protein expression by western blot analysis using the Mdm2 monoclonal antibody mix (4B11) and p53 monoclonal antibody mix (240, 1801, 421). Actin was used as a loading control. HRP-conjugated anti-mouse and anti-rabbit were used as secondary antibodies. This is a representative of three independent experiments. B. 48 hours after transfection, cells were lysed and protein extracts were utilized in a luciferase assay reaction. Samples were compared to PGL2 Basic plasmid and normalized for amount of protein. An average of three independent experiments is shown. Error bars represent standard error. C. Colony formation assay in H1299. Cells were transiently transfected with the plasmid pBaBe-puro-mdm2-FL or pBaBe-puro-mdm2-C and after 24 hours, 2000 cells were plated into media (RPMI with 2ug/ml puromycin). Cells were allowed to grow for 3 weeks. Picture represents one experiment. Two experiments were carried out in duplicates. D. Colony counts after colony formation assay. An average of two experiments carried in duplicates is shown above. Error bars represent standard error. * Asterisks represent p-value compared to vector control.