Fig 2: Analysis of transcript variants of MDM4 in retinoblastoma. A, B) Genomic organization of MDM4 showing two previously identified splice variants for this gene and the corresponding mRNA. MDM4-S results from skipping of exon 6 and MDM4-A results from skipping of exon 9 (dark gray shading). (C) Domain structure of the full length MDM4 protein with known protein modification sites (phosphorylation (P) and sumoylation (S)) ([18]). The predicted size is 55 kDa with an observed size on SDS-PAGE of ~75 kDa. (D) The loss of exon 6 in MDM4-S results in a frameshift after residue 114 and a translational termination after residue 140 (diagonal line). The predicted size of MDM4-S is 15 kDa and the observed size on SDS-PAGE is 27 kDa [18], [27]. (E) The loss of exon 9 in MDM4-A results in an in frame fusion that is missing 50 residues spanning a portion of the acidic domain. (F) Representative reads from transcriptome analysis showing the loss of exon 6 and exon 9 in the retinoblastoma orthotopic xenografts. These alternative spliced transcripts were verified by PCR (H) and Sanger sequencing (I). (G) Representative reads from transcriptome analysis of MDM2 showing very little detection for alternative spliced transcripts.

Fig 3: MDMX overexpression partially reverses the inhibitory effect of TRPM7 inhibition on MCF7 cell migration.A, MDMX overexpression partially rescues cell migration repression mediated by NS8593 in MCF-7 cells with stably expressed MDMX. MCF-7-pcDNA3 and MCF-7-Myc-MDMX cells were plated on 6-well plate to reach ~90% confluence. The cells were starved with DMEM plus 0.5% FBS for 24 h. After that, the cells were scratched and incubated in DMEM plus 0.5% FBS with DMSO or NS8593 (30 µM). Images were taken at 0 or 24 h after scratch (Scale bar = 500 µm). Percentage of healed wound area at 24 h related to 0 h time point was graphed as described in Experimental procedures. The asterisk indicates statistical significance (p value <0.05). B, MDMX overexpression partially rescues cell migration repression mediated by TRPM7 ablation in MCF-7 cells with stably expressed MDMX. MCF-7-pcDNA3 and MCF-7-Myc-MDMX cells were plated on 60 mm dishes for 24 h and then transfected with control siRNA or siRNA targeting TRPM7 (siTRPM7_2). Six hours after transfection, 2 × 106 cells were plated in each well of a 6-well plate to reach ~90% confluence. The cells were starved with DMEM plus 0.5% FBS for 24 h and then scratched. Images were taken at 0 or 24 h after scratch (Scale bar = 500 µm) and analyzed as described in Experimental procedures. The asterisk indicates statistical significance (p value <0.05). C, MDMX overexpression partially reverses cell migration repression mediated by NS8593 in MCF7 cells with tetracycline-inducible MDMX overexpression. MCF7 cells with tetracycline-inducible MDMX overexpression were plated on 6-well plate with or without 2.25 µg/ml of tetracycline to reach ~90% confluence. The cells were starved with DMEM plus 0.5% FBS for 24 h. After that, they were scratched and incubated in DMEM plus 0.5% FBS with DMSO or NS8593 (30 µM). Images were taken at 0 or 24 h after scratch (Scale bar = 500 µm) and analyzed as described in Experimental procedures. The asterisk indicates statistical significance (p value <0.05). D, MDMX overexpression partially reverses cell migration repression mediated by TRPM7 knockdown in MCF7 cells with tetracycline-inducible MDMX overexpression. MCF7 cells with tetracycline-inducible MDMX overexpression were plated on 60 mm dishes for 24 h and then transfected with control siRNA or siRNA targeting TRPM7 (siTRPM7_2). Six hours after transfection, both cells were trypsinized and counted. 2 × 106 cells were plated in each well of a 6-well plate to reach ~90% confluence. The cells were starved with DMEM plus 0.5% FBS for 24 h. After that, they were scratched and images were taken at 0 or 24 h after scratch (Scale bar = 500 µm) and analyzed as described in Experimental procedures. The asterisk indicates statistical significance (p value <0.05).

Fig 4: KD of Mdmx impeded MCF-7 cells in a p53-dependent manner. MCF-7 cells expressing Mdmx shRNA or control shRNA (shMdmx wobble control) were treated and analyzed as described in Figure 2, with the exception that half the samples were also treated with the p53 inhibitor pifithrin-a (2.5 µM PFTa). For the analysis of cell numbers (a) and GFP positivity (b), cells were treated for 5 days. PFTa offered statistically significant protection against Mdmx shRNA, both for cell numbers (a, P=0.0027) and also for numbers of GFP positive cells (b, P=0.0018). For the measurement of Mdmx and p21 expression, cells were treated for 3 days and analyzed by RT-PCR (c). All the experiments represent triplicates. Errors bars represent S.D. of mean. P-values were calculated by Student's t-test

Fig 5: KD of Mdmx growth inhibited wt p53 basal-like SKBR7. Wt p53 basal-like SKBR7 cells transduced with either shMdmx constructs or shRNA controls were either not treated or treated with DOX (2.5 and 5 ng/ml) for 8 days and analyzed by western blotting (a) with protein detection of Mdmx, p53 and USP7. A colony assay of 2.5 ng/ml DOX-treated samples and controls demonstrated significant growth inhibition with Mdmx KD as visualized in b and as quantified in c