DEC1 regulates breast cancer cell proliferation by stabilizing cyclin E protein and delays the progression of cell cycle S phase

Abstract

Breast cancer that is accompanied by a high level of cyclin E expression usually exhibits poor prognosis and clinical outcome. Several factors are known to regulate the level of cyclin E during the cell cycle progression. The transcription factor DEC1 (also known as STRA13 and SHARP2) plays an important role in cell proliferation and apoptosis. Nevertheless, the mechanism of its role in cell proliferation is poorly understood. In this study, using the breast cancer cell lines MCF-7 and T47D, we showed that DEC1 could inhibit the cell cycle progression of breast cancer cells independently of its transcriptional activity. The cell cycle-dependent timing of DEC1 overexpression could affect the progression of the cell cycle through regulating the level of cyclin E protein. DEC1 stabilized cyclin E at the protein level by interacting with cyclin E. Overexpression of DEC1 repressed the interaction between cyclin E and its E3 ligase Fbw7α, consequently reducing the level of polyunbiquitinated cyclin E and increased the accumulation of non-ubiquitinated cyclin E. Furthermore, DEC1 also promoted the nuclear accumulation of Cdk2 and the formation of cyclin E/Cdk2 complex, as well as upregulating the activity of the cyclin E/Cdk2 complex, which inhibited the subsequent association of cyclin A with Cdk2. This had the effect of prolonging the S phase and suppressing the growth of breast cancers in a mouse xenograft model. These events probably constitute the essential steps in DEC1-regulated cell proliferation, thus opening up the possibility of a protein-based molecular strategy for eliminating cancer cells that manifest a high-level expression of cyclin E.

Figure 1

The expression of DEC1 in human breast tumor tissue samples and the effect of DEC1 on the cell proliferation. (a and b) Immunohistochemistry (IHC) analysis of DEC1 expression in adjacent normal breast tissues (n=18) and breast carcinoma (n=30). Each sample was incubated with antibody against DEC1. Positive staining and negative staining are indicated by brown and blue staining, respectively. (c and d) MCF-7 cells and T47D cells were transfected with Flag-DEC1 or control vectors (Ctrl). Cells were then cultured in selective medium (500 μg/ml G418) and subjected to colony formation and MTT assays (days, time after transfection). Each bar represents the mean±S.D. from five independent experiments. *P<0.05. (e) Western blot analysis of endogenous DEC1 expression. MCF-7 cells were transfected with DEC1 siRNAs (siRNA-1 and siRNA-2) for 24 h and probed with anti-DEC1 and anti-β-actin. Each bar represents the mean±S.D. from three independent experiments. (**P<0.01). (f) Colony formation assay, MCF-7 cells were transfected with the shDEC1 vector, followed by 2-week selection with hygromycin B. Hygromycin-resistant clones are shown in the left panel. The right panel shows the corresponding quantitative analyses. Only representative data from three independent experiments are shown. Scale bar, 500 μm

Figure 2

The expression dynamics of DEC1 in cell cycle and the effects of DEC1 expression on cyclin E stability. (a) Endogenous DEC1 levels during the cell cycle. MCF-7 cells were synchronized at G2/M by treating with 50 ng/ml nocodazole for 16 h. The cells were collected at the indicated time points following the removal of nocodazole. The cells were analyzed by FACS, and DEC1 level in the total cell lysate was determined by western blotting using anti-DEC1 antibody. (b and c) DEC1 affects the level of cyclin E protein but not in its transcription. The transcript and protein levels of cyclin E, p53 and p21 were measured in MCF-7 cells that overexpressed cyclin E only or cyclin E plus DEC1 by using reverse transcription PCR or by western blotting using specific antibodies against cyclin E, p53 and p21. (d) Western blot of cyclin E in DEC1-overexpressing T47D and MCF-7 cells. (e) T47D and MCF-7 cells transfected with DEC1 shRNA and probed with anti-DEC1 and anti-cyclin E antibodies. (f) DEC1 regulates cyclin E in a dose-dependent way. MCF-7 cells were transfected with Myc-tagged cyclin E and either empty vector or 0.5, 1 and 4 μg Flag-DEC1. Lysates were analyzed by western blot with anti-myc, anti-Flag or anti-β-actin. (g) DEC1 regulates the level of cyclin E protein in both the presence and absence of serum. MCF-7 cells were transfected with the indicated plasmids for 12 h and were then starved for 24 h in serum-free medium. The histogram in the right plot shows the quantitative analysis of the bands. (h) Western blot analysis of the effect of DEC1 on the stability of cycin E under serum starvation condition. MCF-7 cells were transfected with or without DEC1 to detect the change in cyclin E protein level. The graph shows the relative intensity of the cycin E bands at different time points. In all experiments (a–h), β-actin expression or GAPDH mRNA level was used as a reference

Figure 3

DEC1 regulates the stability of cyclin E in MCF-7 via Fbw7-mediated cyclin E ubiquitin-proteasome pathway. (a and b) Western blot analysis of the effect of DEC1 on the half-life of cyclin E. MCF-7 cells were transfected with Myc-cyclin E and Flag-DEC1 or empty vector. The cells were then treated with CHX only (a) or both CHX and MG132 (b), and harvested at the indicated time periods. The cell extract was subjected to western blot with anti-Myc, anti-Flag and anti-β-actin antibody. The histogram in the right panel (a and b) shows the quantitative analysis of cyclin E bands. (c) MCF-7 cells were transfected with HA-tagged cyclin E and Myc-Ub or empty vector, followed by treatment with MG132 for 8 h before harvest. The cell extract was immunoprecipitated with anti-HA antibody and then probed with anti-Myc antibody. (d) MCF-7 cells were transfected with plasmids expressing the indicated Flag-tagged Fbw7 isoforms together with Myc-tagged cyclin E or empty vector or DEC1, followed by western blot analysis with anti-Myc antibody. (e) MCF-7 cells were transfected with Myc-tagged cyclin E, Flag-Fbw7α and Flag-DEC1 or empty vector, followed by treatment with MG132. Clear cell extracts were probed with anti-Myc and anti-Flag antibodies. (f) MCF-7 cells were transfected with Myc-cyclin E only or together with Flag-DEC1 in the presence of sh-Fbw7 or sh-c. The histogram shows the quantitative analysis of cyclin E protein levels after normalization to β-actin (bottom). Data are the means±S.D. (g) MCF-7 cells were transfected with Myc-tagged cyclin E, Flag-Fbw7α and GFP-DEC1 or empty vector, treated with MG132 for 8 h before harvest. The clear cell extracts were immunoprecipitated with anti-Myc antibody, and then probed with anti-Flag, anti-GFP and anti-Myc antibody as indicated. β-Actin was used as a negative control. (h) Effect of DEC1 on the interaction between Fbw7α and cyclin E under serum starvation stress. MCF-7 cells were transfected as in (g), and cultured in serum-free medium, and then treated with 10 μM MG132 for 8 h before harvest. The cells were collected and subjected to immunoprecipitation assay as in (g). (i and j) Effect of DEC1 on the interaction between endogenous Fbw7α and cyclin E. MCF-7 cells were transfected with GFP-DEC1 or empty vector, and cultured with or without serum, and then treated with MG132 for 8 h before harvest. The clear cell extracts were immunoprecipitated with anti-IgG or anti-cyclin E antibody, probed with anti-Fbw7, anti-GFP and anti-cyclin E antibody as indicated. β-Actin was used as a negative control. (k) Ubiquitination status of endogenous cyclin E in DEC1- and Fbw7-silenced cells. MCF-7 cells were transfected with either shDEC1 or shFbw7α, or both. Cell extract was immunoprecipitated with anti-cyclin E antibodies, followed by western blotting using anti-Ub or anti-cyclin E antibody. Total cell lysate was also analyzed by western blotting using anti-β-actin antibody. To test the expression and immunoprecipitation of cyclin E, mouse monoclonal (SB62a) secondary antibody against rabbit IgG light chain (HRP, ab99697) was used at a 1 : 5000 dilution in (g–k)

Figure 4

Interaction between DEC1 and cyclin E. (a) MCF-7 cells were transfected with Flag-tagged DEC1 and Myc-tagged cyclin E. After 24 h of transfection, the cell extract was subjected to immunoprecipitation with anti-Flag antibody, followed by western blot analysis with anti-Myc antibody. (b) MCF-7 cells transfected as in (a) were subjected to serum starvation for 24 h. The cell extract was subjected to immunoprecipitation with anti-Flag antibody followed by western blot analysis with anti-Myc antibody. (c and d) CoIP experiment showing the interaction between endogenous DEC1 and cyclin E in serum-plus and serum-free conditions. MCF-7 cells were subjected to immunoprecipitation with anti-cyclin E antibody followed by western blot analysis with anti-DEC1 antibody. Immunoprecipitation carried out with anti-IgG antibody was used as control. Mouse monoclonal (SB62a) secondary antibody against rabbit IgG light chain (HRP) was used at a 1 : 5000 dilution. (e) Western blot analysis of GST pull-down assay showing the interaction between cyclin E and DEC1. The cell extract was incubated with glutathione-agarose beads coated with purified GST or GST-tagged cyclin E as indicated at the top. After extensive washing, bound proteins were eluted, resolved by SDS-PAGE, and probed with anti-DEC1 and anti-GST antibodies. (f and g) Interaction between DEC1 and cyclin E as detected by a mammalian two-hybrid system for cells without and with serum starvation. Cyclin E and DEC1 were expressed from pBIND-cyclin E and pACT-DEC1, respectively. The MCF-7 cells were transfected with pG5-luc and the empty vectors (pACT and pBIND) as indicated. Positive control cells were transfected with pBIND-ID and pACT-MyoD. Each bar represents the mean±S.D. from three independent experiments. **P<0.01 compared with cells transfected with pACT and pBIND. (h and i) Co-localization of DEC1 and cyclin E in MCF-7 cell nuclei. MCF-7 cells were transfected with Myc-cyclin E and Flag-DEC1. Flag-antibody complex and Myc-antibody complex were visualized with FITC and TRITC, respectively. Nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI). (j) CoIP experiments showing the dynamics of the interaction between DEC1 and cyclin E. MCF-7 cells were transfected with Flag-DEC1, and then subjected to immunoprecipitation with anti-Flag antibody followed by western blot analysis with anti-cyclin E antibody. HRP (ab99697) was used as secondary antibody

Figure 5

DEC1 promotes the complex formation and activity of cyclin E/Cdk2 and inhibits cell cycle progression through the S phase. (a) Effect of DEC1 on the formation of cyclin E/Cdk2 complex. MCF-7 cells were transfected with HA-Cdk2 and Myc-cyclin E only or HA-Cdk2, Myc-cyclin E and Flag-DEC1, and the clear cell extract was immunoprecipitated with anti-Myc antibodies and then probed with anti-HA antibody. (b) Effect of DEC1 on the sequent formation of the cyclin A/Cdk2 complex as detected by CoIP. MCF-7 cells were transfected with empty vector only, HA-Cdk2 and Myc-cyclin A only or HA-Cdk2, Myc-cyclin A and Flag-DEC1, and the clear cell extract was immunoprecipitated with anti-Myc antibody and then probed with anti-HA antibody. (c) Mammalian two-hybrid assay analysis of the effect of DEC1 on the interaction between cyclin E and Cdk2. Cyclin E, Cdk2 and DEC1 were expressed from pBIND-cyclin E, pACT-Cdk2 and Flag-DEC1, respectively. Each bar represents the mean±S.D. from three independent experiments. **P<0.01 compared with cells transfected with pACT and pBIND. (d) Effect of DEC1 on the activity of Cdk2/cyclin E kinase. Columns, mean (n=3)±S.D.; *P<0.05 compared with controls. (e) Immunofluorescence assay showing the effect of DEC1 on the sublocation of cyclin E. MCF-7 cells were transfected with Myc-cyclin E only or Myc-cyclin E and Flag-DEC1, and were fixed and stained with rabbit anti-Myc antibody (red) and then counterstained with DAPI (blue) for nucleus detection. (f) Effect of DEC1 on the sublocation of Cdk2. MCF-7 cells were transfected with HA-Cdk2 only or HA-Cdk2 and Flag-DEC1 and stained with mouse anti-HA antibody (green), and then counterstained with DAPI (blue) for nucleus detection. (g and h) CoIP assay showing the interaction between DEC1 and Cdk2. Control cells or cells overexpressing HA-Cdk2 only, Flag-DEC1 only or Flag-DEC1 and HA-Cdk2 were subjected to immunoprecipitation with anti-HA antibody (g) and with anti-Flag antibody (h). (i) Mammalian two-hybrid assay analysis of the interaction between DEC1 and Cdk2. DEC1 and Cdk2 were expressed from pACT-DEC1 and pBIND-Cdk2, respectively. Columns, mean (n=3)±S.D.; **P<0.01 compared with controls

Figure 6

DEC1 inhibits the progression of S phase in the cell cycle. (a and b) Western blot of cyclin E and cyclin B in control and DEC1-overexpressing cells. The plots show the relative intensities of the bands in the blots. MCF-7 cells were transfected with empty vector or Flag-DEC1, and then synchronized at G2/M stage by treating with 50 ng/ml nocodazole for 20 h. The cells were analyzed at different time points after release from nocodazole treatment. The clear cell extract was subjected to western blot analysis with anti-cyclin E, anti-cyclin B or anti-β-actin antibody. (c) MCF-7 cells were treated as in (a and b) and then subjected to immunoprecipitation using anti-cyclin E antibody followed by western blot with anti-Cdk2, anti-DEC1 and anti-β-actin antibody. (d) Effect of DEC1 on the dynamics of cyclin E and Cdk2 as demonstrated by mammalian two-hybrid assay. Cyclin E, Cdk2 and DEC1 were expressed from pACT-cyclin E, pBIND-Cdk2 and Flag-DEC1, respectively. MCF-7 cells were transfected with the indicated plasmids, synchronized and harvested at the indicated time points. Each bar represents the mean±S.D. from three independent experiments. **P<0.01 compared with cells transfected with pACT and pBIND. (e) Effect of DEC1 on the cell cycle progression (on the left panel) as demonstrated by FACS assay. Quantified analysis is shown by histogram in the right panel. MCF-7 cells were transfected with DEC1 or control vector, synchronized and harvested at the indicated time points

Figure 7

DEC1 inhibits the proliferation of cells overexpressing cyclin E and inhibits tumor xenograft growth. (a and b) MCF-7 cells were transfected with Myc-cyclin E only, Myc-cyclin E and Flag-DEC1 or control vector. The cells were then cultured in selective medium (500 μg/ml G418) and subjected to colony formation and MTT assays. Each bar represents the mean±S.D. from five independent experiments. *P-value was determined by ANOVA with Bonferroni test (*P<0.05). (c) Representative colonies of each experimental group are shown. MCF-7 cells transfected with control vector, Myc-cyclin E or Myc-cyclin E and Flag-DEC1 were selected with 800 μg/ml G418 for 15 days. The cells were then collected and suspended in a soft agar. Photographs of the colonies were taken 30 days after seeding. Scale bar, 100 μm. All experiments were repeated at least three times. (d) Tumor growth by subcutaneously implanted MCF-7 cells (6 × 10⁵ cells) transfected with pCMV10-3 × Flag or pCMV10-3 × Flag-DEC1 and screened with G418. P-value was determined by unpaired t-test. (e) Tumor formation 36 days after the mice were injected with the tumor cells. Left: mice injected with control cells. Right: mice injected with DEC1 overexpressed cells. (f and g) Tumor images (f) and tumor weight (g) 48 days after mice were injected subcutaneously with MCF-7 cells overexpressed Flag-DEC1 or empty vector. n=5 mice per group in (d, f and g). (h) Representative immunohistochemical data for H&E and DEC1 on paraffin-embedded section of subcutaneous tumors in (f) and (g), generated by MCF-7 cells overexpressed DEC1 or empty vector. (i) Western blot analysis of the expression of cyclin E and DEC1 in control (Ctrl) and Flag-DEC1-overexpressing tumors