Figure 1.
Coordinated expression of FOXM1 and CDC25A during the cell cycle progression.
(A) Representative flow cytometry analysis shows cell cycle progression in U2OS cells. U2OS were synchronized with nocodazole for 16 h and then stimulated to re-enter the cell cycle by addition of medium containing 10% fetal bovine serum. Asynchronous cells (Asynch) were included as controls. Cell cycle was analyzed by flow cytometry at release (16 h Noc) and at 1, 3, 6, 12, and 18 h after release. (B) Mean percentage (± SD) of cells in each phase of the cell cycle following 16 h nocodazole treatment and after 1, 3, 6, 12, 18 h release (N = 3). Asynchronous (Asynch) cells were included as controls. The analysis indicates that the cells at early, middle and late G1 phase are at 1, 3 and 6 h respectively, S phase is maximal at 12 h while G2/M phase occurs at 16 h Noc and 18 h after release. (C) Cells at the continuous cell cycle phases were collected and processed for western blotting analysis with antibodies against FOXM1, CDC25A, CDK1 and CDK2. FOXM1 expression increased as cells progressed from G1 through S and into G2/M, and degraded when cells exited to G2/M (1 h). CDC25A and CDK1 exhibited a similar expression profile. CDK2 expression levels were relatively constant throughout the cell cycle. β-Actin expression was used as a loading control.
Figure 2.
The FOXM1 transcription factor regulates CDC25A gene expression.
(A) FOXM1 increased the transactivation of CDC25A promoter. U2OS cells were co-transfected with pGL3-CDC25A promoter reporter along with increasing doses of pACT-FOXM1 for 48 h. pACT vector was used as the empty vector, and pRL-SV40 expressing Renilla luciferase was used as an internal control. Data represent the mean ± SD (N = 3). *, p<0.05; ***, p<0.001. (B) siRNA-mediated silencing of FOXM1 decreased CDC25A expression. U2OS cells were transiently transfected with siRNA pools targeting FOXM1 or non-target control for 48 h. A U2OS non-transfected control was also used. The protein expression of FOXM1 and CDC25A were detected by Western blot. β-Actin expression was used as a loading control. The western blot was repeated three times and the protein expression was analyzed by the Adobe Photoshop CS4 software (Adobe Systems, San Jose, CA). The difference of band density between the FOXM1-siRNA and control siRNA was analyzed by the paired t test. ***, p<0.001.
Figure 3.
FOXM1 regulates the gene transcription of CDC25A by direct DNA binding as well as the E2F pathway.
(A) Schematic diagram of CDC25A promoter showing the 3′ distal end containing two E2F binding sites, the 5′ proximal end including three putative FOXM1 binding sites and excluding E2F binding sites, and a 224 bp fragment containing the three putative FOXM1 binding sites. (B) FOXM1 increased the transactivation of full-length and truncated CDC25A promoter constructs, indicating both direct and indirect FOXM1 induction. U2OS cells were co-transfected with pACT-FOXM1 or pACT empty plasmid and the reporter plasmids containing full length and truncated CDC25A promoters. pRL-SV40, a Renilla luciferase report vector, was co-transfected as the control. Data were normalized to Renilla luciferase activities and are presented as the mean ± SD (N = 5). ***, p<0.001. (C) FOXM1 recruits p300, but not PCAF or CBP, to activate CDC25A promoter. U2OS cells were co-transfected with pACT-FOXM1 (or pACT control) and pCMV-p300, or pCMV-pCAF, or pCMV-CBP (or empty vector control), along with pGL3-CDC25A promoter for 48 h. pRL-SV40 was co-transfected as the normalizing control. Data are normalized to Renilla activity, and represented as the mean ± SD. The difference was analyzed by paired t test. ***p<0.001, Δ>0.05. (D) Representation of the wild-type and mutated E2F consensus binding sequences at CDC25A promoter. (E) Two E2F binding sites are required for FOXM1-activated CDC25A promoter transactivation. U2OS cells were co-transfected with pACT-FOXM1 or empty vector control and the reporter plasmids containing the indicated CDC25A promoters (wild-type or mutated E2F binding sites). Data are presented as the mean fold induction of FOXM1 activity over control ± SD. *** p<0.001.
Figure 4.
FOXM1 directly binds to the putative FOXM1 consensus binding sequences located on the CDC25A promoter.
(A) FOXM1 binding to CDC25A promoter was confirmed by ChIP-qPCR. Primer set A was designed to span three FOXM1 binding sites, and primer set B, targeting a region of the promoter that did not include the FOXM1 binding sites, was used as the control. Chromatin immunoprecipitations of U2OS cells were prepared using anti-FOXM1 antibody. Semi-quantitative PCR was performed using the primer sets A and B, and the PCR products were detected by electrophoresis (Top). Quantitative PCR was performed using the primer sets A and B, and the binding activity of FOXM1 to CDC25A promoter was evaluated by the fold enrichment method. (B–C) The three FOXM1 binding sites on the CDC25A promoter are functionally redundant (B) Schematic diagram showing the mutations of the three FOXM1 binding sites on the CDC25A promoter. (C) U2OS cells were transfected with pACT-FOXM1 or empty vector control, along with the indicated CDC25A promoter-luciferase reporter construct. Targeted mutation of single FOXM1 binding sites did not significantly affect CDC25A transactivation. Mutations of sites 1 and 2, sites 2 and 3, and sites 1, 2, and 3 significantly reduced the transactivation of the CDC25A-luciferase reporter. Data were normalized to Renilla luciferase activities and are presented as the mean wild-typefold induction over empty vector control ± SD (N = 3). Data were subjected to one-way ANOVA (significance level α = 0.05) and Dunnett's multi-comparison post-hoc tests (mutations vs. wild-type). * P<0.05.
Figure 5.
CDC25A regulated FOXM1 transcriptional activity through CDK1-mediated phosphorylation sites.
(A) CDC25A activated the transcriptional activity of FOXM1 in U2OS cells. U2OS cells were co-transfected with pACT-FOXM1 or/and pACT-CDC25A, along with the pGL3-6×FOXM1-Luc plasmids containing 6 FOXM1 DNA binding sequences, and pRL-SV40, a Renilla luciferase reporter vector was used as the control. Data were normalized to Renilla luciferase activities and are presented as the mean ± SD (N = 3). *** p<0.001. (B) siRNA-mediated inhibition of CDK1 blocked CDC25A-mediated FOXM1 transcriptional activity. U2OS cells were co-transfected with pACT-CDC25A, pACT-FOXM1 and pGL3-6×FOXM1-Luc plasmids, or pACT-FOXM1 and pACT-CDC25A together with siRNAs against CDK1, CDK2, CDK4 or CDK6 or control siRNA, and pRL-SV40 was used as the normalizing control. Forty-eight hours after transfection, the cells were lysed, and firefly and Renilla luciferase activities were measured. Data were normalized to Renilla luciferase activities and are presented as the mean ± SD (N = 3). Data were subjected to one-way ANOVA (significance level α = 0.05) and Dunnett's multi-comparison post-hoc tests (CDK siRNA vs. CTR siRNA). *** p<0.001. (C) CDC25A activated FOXM1 transcriptional activity via the CDK phosphorylation sites T600, T611, and T620 and a LXL docking motif at L656. U2OS cells were co-transfected with pACT-CDC25A and pACT-FOXM1 (wild-type or the mutations of phosphorylation sites/LXL motif), along with pGL3-6×FOXM1-Luc plasmids. pRL-SV40 was used as the normalizing control. Data were normalized to Renilla luciferase activities and are presented as the mean ± SD (N = 3). Data were subjected to one-way ANOVA (significance level α = 0.05) and Dunnett's multi-comparison post-hoc tests (mutations vs. wild-type). *** p<0.001.
Figure 6.
CDC25A phosphatase activates FOXM1 transcriptional activity by direct protein-protein interaction.
(A) CDC25A and FOXM1 proteins physically interact. The protein-protein interaction of FOXM1 and CDC25A was confirmed by co-immunoprecipitation. HEK293T cells were co-transfected with p3×FLAG-CMV-14-FOXM1 (FOXM1-3×FLAG) or the empty vector control p3×FLAG-CMV-14, along with pCMV3Tag9-CDC25A (CDC25A-3×MYC) or the empty vector control (pCMV3Tag9). Lysates were immunoprecipitated with anti-FLAG agarose resin, separated by PAGE, and electroblotted to PVDF. Western blot analysis showed that CDC25A-3×MYC co-immunoprecipitates with FOXM1-3×FLAG, supporting the hypothesis that these two expressed proteins interact. (B) The native protein interaction between FOXM1 and CDC25A was detected by co-immunoprecipitation. 6×106 U2OS cells were lysed in MPER buffer containing 1× protease/phosphatase inhibitors. Whole cell lysates were pre-cleared with protein G sepharose and normal rabbit IgG overnight at 4°C with end-over-end mixing. The cell lysates were incubated with anti-CDC25A antibody overnight at 4°C and then with protein G sepharose for 1 h at 4°C. The resin was washed three times, and the eluted protein complex was separated by PAGE and analyzed using the anti-FOXM1 antibody by Western blotting. (C) Schematic diagram showing FOXM1 deletion constructs used to identify FOXM1 domains critical to the interaction with CDC25A. Sequences encoding the specific FOXM1 deletion constructs were subcloned in pBIND. (D) A mammalian two-hybrid assay reveals a critical role for the FOXM1 C-terminus in interactions with CDC25A. pBIND construct (FOXM1, deletion, or empty) was co-transfected with pACT construct (CDC25A or empty) and pG5luc reporter. Data represent the mean ± SD, normalized to Renilla luciferase activities (N = 3). Co-expression of the C-terminal FOXM1 and CDC25A revealed a robust interaction. In a construct lacking the C-terminus, the interaction between FOXM1 and CDC25A was significantly diminished. *** p<0.001. (E) Schematic diagram showing the mutations of the CDK/cyclin phosphorylation sites and LXL motif at the C-terminal of FOXM1 protein. pBIND construct (FOXM1 wild-type and FOXM1 with mutations of phosphorylation sites/LXL motifs) was co-transfected with pACT construct (CDC25A or empty) and pGL5-luc reporter. Data represent the mean ± SD, normalized to Renilla luciferase activities (N = 3). Data were subjected to one-way ANOVA (significance level α = 0.05) and Dunnett's multi-comparison post-hoc tests (mutations vs. wild-type). *** p<0.001. (F) Mutation of T600 and T611 diminishes FOXM1 protein association with CDC25A by co-immunoprecipitation. U2OS cells were transiently transfected with either FOXM1-3×FLAG construct encoding wild-type FOXM1 (lane 1) or FOXM1 with the following mutations: T600A (lane 2), T611A (lane 3), T620A (lane 4) and L656A (lane 5). U2OS cell lysates were prepared 48 hours after transfection and FOXM1 expression in these cell lysates was detected by Western blot. The same amount of protein was co-immunoprecipitated with anti-CDC25A antibody, and the co-immunoprecipitated proteins were subjected to Western blot with an anti-FLAG antibody. The CDC25A expression in the co-immunoprecipitated proteins was used to test the efficiency of co-immunoprecipitation. The phosphorylation of T600 and T611 enhanced the protein interaction of FOXM1 and CDC25A.
Figure 7.
CDC25A phosphatase enzyme activity is required for the CDC25A-activated FOXM1 transcriptional activity and protein-protein interaction between CDC25A and FOXM1.
(A) CDC25A phosphatase enzyme activity is required for the CDC25A-activated FOXM1 transcriptional activity. The C431 site in CDC25A protein was changed via site-directed mutagenesis to a serine to create a phosphatase-dead mutant. U2OS cells were co-transfected with pACT-FOXM1 and/or pACT-CDC25A (pACT-CDC25A C431S), along with the pGL3-6×FOXM1-Luc plasmids containing 6 FOXM1 DNA binding sequences. pRL-SV40, a Renilla luciferase reporter vector, was used as the control. Data were normalized to Renilla luciferase activities and are presented as the mean ± SD (N = 5). *** p<0.001. (B) A mammalian two-hybrid assay confirmed a critical role for CDC25A phosphatase activity in mediating the FOXM1-CDC25A protein-protein interaction. U2OS cells were co-transfected with pBIND-FOXM1 (or pBIND control) and one of following: wild-type pACT-CDC25A, or pACT-CDC25A C431S phosphatase-dead mutant, or pACT control. pG5-luc was used as a normalizing control. Data were normalized to Renilla luciferase activities and are presented as the mean ± SD, (N = 5). Loss of phosphatase activity by CDC25A led to a significant reduction in the FOXM1-CDC25A interaction. ** p<0.01; *** p<0.001.