Api5 contributes to E2F1 control of the G1/S cell cycle phase transition

Abstract

Background: The E2f transcription factor family has a pivotal role in controlling the cell fate in general, and in particular cancer development, by regulating the expression of several genes required for S phase entry and progression through the cell cycle. It has become clear that the transcriptional activation of at least one member of the family, E2F1, can also induce apoptosis. An appropriate balance of positive and negative regulators appears to be necessary to modulate E2F1 transcriptional activity, and thus cell fate.

Methodology/principal findings: In this report, we show that Api5, already known as a regulator of E2F1 induced-apoptosis, is required for the E2F1 transcriptional activation of G1/S transition genes, and consequently, for cell cycle progression and cell proliferation. Api5 appears to be a cell cycle regulated protein. Removal of Api5 reduces cyclin E, cyclin A, cyclin D1 and Cdk2 levels, causing G1 cell cycle arrest and cell cycle delay. Luciferase assays established that Api5 directly regulates the expression of several G1/S genes under E2F1 control. Using protein/protein and protein/DNA immunoprecipitation studies, we demonstrate that Api5, even if not physically interacting with E2F1, contributes positively to E2F1 transcriptional activity by increasing E2F1 binding to its target promoters, through an indirect mechanism.

Conclusion/significance: The results described here support the pivotal role of cell cycle related proteins, that like E2F1, may act as tumor suppressors or as proto-oncogenes during cancer development, depending on the behavior of their positive and negative regulators. According to our findings, Api5 contributes to E2F1 transcriptional activation of cell cycle-associated genes by facilitating E2F1 recruitment onto its target promoters and thus E2F1 target gene transcription.

Figure 1. Api5 depletion induces cell accumulation in G1 phase and decreases cell proliferation.

A. H1299 cells were synchronized by a double thymidine block, released and collected (Figure S1). Api5, E2F1 and OPA1 expression through the cell cycle was analyzed by Western blot (WB). B. Endogenous Api5 (green) and E2F1 (red) immunostaining were performed on H1299 and HeLa cells; nuclei were stained with PI (blue). White arrows show cells undergoing mitosis. [Scale bar 20 µm]. C. Cytometric quantification of experiments described in (A). Cell cycle analysis was performed using FlowJo. D. Cell growth was analyzed by cell counting. H1299 cells were transfected by the indicated siRNAs.

Figure 2. Api5 depletion induces a decrease in E2F1-target gene mRNA levels without affecting E2F1 expression.

A. Api5 and E2F1 reciprocal regulation. After depletion of Api5 and E2F1 with specific siRNAs, H1299 cells were made quiescent by serum starvation for 48 h (0% FBS) and then allowed to progress through the cell cycle over 10 hours by FBS addition (10 h refeeding). Control cells were grown in 10% fetal bovine serum media (10% FBS). Api5, E2F1 and β-actin expression was analyzed by Western blot (WB). B. E2F1 and β-actin expression was analyzed by Western blot (WB) after Api5 overexpression. HeLa cells were transfected with different amounts of a plasmid encoding Api5. C. Api5 and E2F1 immunocytochemistry. HeLa cells were transfected with the indicated siRNAs or treated with etoposide for 16 hours. Endogenous immunodetection of Api5 (green) and E2F1 (red). Nuclei were stained with PI (blue). [Scale bar 20 µm]. D. Retinoblastoma expression under Api5 overexpression or depletion. Hela cells were transfected or not with a plasmid encoding Api5 or an siRNA directed against Api5. Western blot analysis was performed with an anti-retinoblastoma antibody. E. siRNA efficiency of quantitative PCR samples. HeLa cells were treated with Api5 or E2F1 siRNAs. Api5 and E2F1 protein content was assessed by Western blot (left panel) to evaluate siRNA efficiencies by densitometric analysis (relative to scrambled treated cells) (right panel). F. E2F1 target gene expression. HeLa cells were treated with Api5 or E2F1 siRNA, or with both siRNAs. Api5, E2F1, cyclin E, cyclin A, cyclin D1, p16 INK4a, Cdk2 and Cdc25A relative mRNA levels were measured by quantitative PCR, using HPRT1 and GUSB mRNA levels as references. Asterisks indicate a significant decrease compared to control conditions (scrambled siRNA treated cells). Statistical significance was determined by two-tailed Student’s t test (*, p<0.05). G. Cyclin A expression. Western blot analysis of cellular levels of cyclin A protein after transfection of H1299 cells with Api5 or E2F1 siRNAs.

Figure 3. Api5 stimulates the transcriptional function of E2F1.

A. Luciferase experiments were performed in HeLa cells co-transfected with a plasmid encoding the firefly luciferase under the control of the wild type cyclin E promoter (WT E2F1) or under the control of the mutant cyclin E promoter in which the E2F1 response element had been mutated (mut E2F1); pRL-CMV encoding the Renilla luciferase for normalization and the indicated siRNAs. B. Luciferase experiments were performed in HeLa cells co-transfected with a plasmid encoding the firefly luciferase under the control of the SV40 viral promoter; pRL-CMV encoding the Renilla luciferase for normalization and the indicated siRNAs. C. Luciferase experiments were carried out in HeLa cells co-transfected with a plasmid encoding the firefly luciferase under the control of the SKP2 promoter; pRL-CMV encoding the Renilla luciferase for normalization and the indicated siRNAs. D. Luciferase experiments were carried out in stable HeLa cells expressing the indicated shRNAs (scrambled siRNA: shSCR or directed against Api5: shApi5) transfected with a plasmid encoding the firefly luciferase under the control of the wild type cyclin E promoter (WT E2F1) or under the control of the mutant cyclin E promoter in which the E2F1 response element has been mutated (mut E2F1); pRL-CMV encoding the Renilla luciferase for normalization and, when indicated, a plasmid (R-2HA2Flag-Api5) encoding Api5 whose RNA is resistant to the shApi5. E. Western blot analysis with the shSCR and shApi5 HeLa cell lines: detection of recombinant HA-Api5 resistant to shApi5 (HA), endogenous Api5 (Api5) and β-actin (Actin).

Figure 4. Api5 is a chromatin associated factor facilitating E2F1 association to specific target promoters.

A. Saline chromatin extracts. Western blot analysis of chromatin extracts with Api5 antibody and TBP and histone H3 antibodies as control. Chromatin fractions were prepared by digesting nuclei with DNase and extracting with 0.4/0.8/1 and 2 M NaCl. Residual pellets correspond to proteins not extracted at 2 M NaCl. B. Api5 co-immunoprecipitation. HeLa cells were transfected with HA-tagged Api5 expression vector. Putative Api5 associated proteins were analyzed by Western Blot (WB) from co-immunoprecipitated protein complexes using anti-HA antibody. Western blot against high molecular weight forms of FGF2 was used as a positive control. C. HA-E2F1 and HA-Api5 ChIP experiments. HeLa cells were transfected with HA-tagged Api5 or HA-tagged E2F1 expression vectors (Mock = empty vector). The percentage of input reflects the enrichment of SKP2, cyclin E and E2F1 promoters after immunoprecipitation with an HA antiboby or with a normal rabbit serum (NRS). PCR amplified regions of the albumin promoter were used as a negative control. The ChIP results were obtained from 3 independent replicates; the error bars indicate the standard error. Statistical significance was determined by two-tailed Student’s t test (*, p<0.05). D. E2F1 ChIP experiment. Enrichment of SKP2, cyclin E and E2F1 promoters (% of input) after immunoprecipitation with E2F1 antiboby. Cells were transfected with an Api5 expression vector (Api5) or an siRNA directed against Api (siApi5). Chromatin immunoprecipitations were done using an anti-E2F1 polyclonal rabbit antibody (ChIP E2F1) or with normal rabbit serum (CHIP NRS). PCR amplified regions of the albumin promoter were used as a negative control. The ChIP results were obtained from 3 independent experiments; the error bars indicate the standard error. Statistical significance was determined by two-tailed Student’s t test (*, p<0.05).