Impact of the MDM2 splice-variants MDM2-A, MDM2-B and MDM2-C on cytotoxic stress response in breast cancer cells

Abstract

Background: The murine double minute 2 (MDM2) is an oncogene and a negative regulator of the tumor suppressor protein p53. MDM2 is known to be amplified in numerous human cancers, and upregulation of MDM2 is considered to be an alternative mechanism of p53 inactivation. The presence of many splice variants of MDM2 has been observed in both normal tissues and malignant cells; however their impact and functional properties in response to chemotherapy treatment are not fully understood. Here, we investigate the biological effects of three widely expressed alternatively spliced variants of MDM2; MDM2-A, MDM2-B and MDM2-C, both in unstressed MCF-7 breast cancer cells and in cells subjected to chemotherapy. We assessed protein stability, subcellular localization and induction of downstream genes known to be regulated by the MDM2-network, as well as impact on cellular endpoints, such as apoptosis, cell cycle arrest and senescence.

Results: We found both the splice variants MDM2-B and -C, to have a much longer half-life than MDM2 full-length (FL) protein after chemotherapy treatment indicating that, under stressed conditions, the regulation of degradation of these two variants differs from that of MDM2-FL. Interestingly, we observed all three splice variants to deviate from MDM2-FL protein with respect to subcellular distribution. Furthermore, while MDM2-A and -B induced the expression of the pro-apoptotic gene PUMA, this effect did not manifest in an increased level of apoptosis.

Conclusion: Although MDM2-B induced slight changes in the cell cycle profile, overall, we found the impact of the three MDM2 splice variants on potential cellular endpoints upon doxorubicin treatment to be limited.

Keywords: Breast cancer; Doxorubicin; MDM2; MDM2 splice variants; MDM2-A; MDM2-B; MDM2-C.

Fig. 1

(a) Schematic representation of MDM2 and the splice variants MDM2-A, -B and -C. MDM2 consists of 491 amino acids. Localization of the p53, pRb and MDM2/4 binding sites, NLS, NES, NoLS, the acidic domain, the Zn-finger domain and the RING-finger domain are indicated, as well as the exon distribution. (b) Validation of protein expression. MCF-7 breast cancer cells transfected with pCMV-GFP control (GFP-control), MDM2-A (75 kDa), -B (48 kDa) and -C (85 kDa) analyzed 24 h post transfection. GAPDH was used as loading control. Primary antibodies were Anti-MDM2 (N-20) Sc-813 (Santa Cruz) and GAPDH (SantaCruz). (c) Protein stability of the expressed splice variants. MCF-7 cells transfected with MDM2-FL, MDM2-A, -B and -C at 0, 1, 2 and 4 h post cycloheximide treatment, respectively. In addition to cycloheximide treatment, left panel shows cells without doxorubicin treatment, right panel shows cells treated with 1 μM doxorubicin for 24 h. Primary antibodies Anti-MDM2 (N-20) Sc-813 (Santa Cruz) and GAPDH (SantaCruz). Histograms under immunoblots represent averages of triplicate experiments and show levels of the MDM2-variants relative to GAPDH-levels for each sample

Fig. 2

Sub-cellular localization of the MDM2-splice variants. (a) MCF-7 cells transfected with pCMV-GFP (GFP-control), MDM2-FL, -A, -B and -C evaluated by indirect immunofluorescence for determination of cellular localization of the proteins. Top lane shows the nucleus by Hoechst staining (blue), second lane shows the cells expressing GFP (green), third lane shows the AlexaFlour 647 MDM2 antibody (Life Technologies, red), and lastly, an overlay of the tree previous pictures. (b) Percentage of cells with exclusively nuclear (dark blue bars), nuclear and cytoplasmic (blue bars) or exclusively cytoplasmic (light blue bars) localization of the MDM2-A, -B and -C. For each of the differently transfected cell samples the sub cellular localization was determined for 50 transfected cells with each of the constructs. The experiment was repeated in triplicate, with three independent transfections. Cells were counted by three independent investigators blinded to sample identity and each other’s results

Fig. 3

Expression of endogenous MDM2 after transfection with the MDM2-splice variants. Relative mRNA levels of MDM2 3UTR / MDM2 EXON3 (i.e. corrected for transfection efficacy) measured after MCF-7 cells were transfected with MDM2-A, -B and -C and after cell sorting by GFP expression, untreated cells (purple bars) or 1 μM doxorubicin treated (green bars) for 24 h. The experiment was repeated in triplicate, with three independent transfections. * = p ≤ 0.05

Fig. 4

Expression of PUMA upon overexpression of the MDM2 splice variants. Activation of PUMA after transfection with the splice variants by qPCR assay revealing relative mRNA levels of PUMA. MCF-7 cells transfected with pCMV-GFP (GFP-control), MDM2-A, -B and -C, sorted after 12 h based on GFP expression. Graphs show the untreated cells (purple bars) and the 24 h 1 μM doxorubicin (green bars). The experiment was repeated in triplicate, with three independent transfections. * = p ≤ 0.05, ** = p ≤ 0.01, *** = p ≤ 0.001

Fig. 5

Induction of apoptosis. Graphs show the percentage of apoptotic cells after transfection with pCMV (TOPO-Control) and the splice variants. Untreated cells (purple bars) and cells treated with 1 μM doxorubicin (green bars) were analyzed by AnnexinV assay 48 h post transfection. Each pillar represents the total of apoptotic and early apoptotic cells. The experiment was repeated in triplicate with three independent transfections

Fig. 6

Cell cycle analysis. (a) MCF-7 breast cancer cells were transfected with pCMV-GFP (GFP-control), MDM2-A, -B and -C. DMSO control cells (left graph) or cells treated with 1 μM doxorubicin (right graph) for 24 h. The cells were analyzed by NucleoCounter-3000 for status of cell cycle progression. Bars represent cell debris (purple bars), cells in G1/G0-phase (green bars), cells in S-phase (pink bars) and cells in G2/M-phase (blue bars) respectively. (b) Cell proliferation analyzed by cell count. MCF-7 breast cancer cells transfected with pCMV-GFP (GFP-control), MDM2-A, -B and -C were counted to examine the cell proliferation 24, 48 and 72 h after transfection (data points at 72 h have been slightly shifted on the x-axis of the graph, for clarity). Both the experiments were repeated in triplicate, with three independent transfections

Fig. 7

Induction of senescence. Graphs show the percentage of senescent cells after transfection with pCMV (TOPO-Control) and the splice variants, untreated (purple bars) or treated with 0.25 μM doxorubicin (green bars). Cells were analyzed by β-galactosidase assay 8 days post transfection. The experiment was repeated in triplicate with three independent transfections