E2F7 represses a network of oscillating cell cycle genes to control S-phase progression

Abstract

E2F transcription factors are known to be important for timely activation of G(1)/S and G(2)/M genes required for cell cycle progression, but transcriptional mechanisms for deactivation of cell cycle-regulated genes are unknown. Here, we show that E2F7 is highly expressed during mid to late S-phase, occupies promoters of G(1)/S-regulated genes and represses their transcription. ChIP-seq analysis revealed that E2F7 binds preferentially to genomic sites containing the TTCCCGCC motif, which closely resembles the E2F consensus site. We identified 89 target genes that carry E2F7 binding sites close to the transcriptional start site and that are directly repressed by short-term induction of E2F7. Most of these target genes are known to be activated by E2Fs and are involved in DNA replication, metabolism and DNA repair. Importantly, induction of E2F7 during G(0)-G(1)/S resulted in S-phase arrest and DNA damage, whereas expression of E2F7 during G(2)/M failed to disturb cell cycle progression. These findings provide strong evidence that E2F7 directly controls the downswing of oscillating G(1)/S genes during S-phase progression.

Figure 1.

Identification of E2F7 target genes by ChIP-seq analysis. (A) E2F7 and E2F1 immunoblots on HU-synchronized HeLa cells. Alpha-Tubulin immunoblots served as loading control. (B) Examples of E2F7 ChIP peaks near transcription start sites of three different genes. Chromosome coordinates are indicated above the peaks. The bottom tracks show UTRs, coding regions, and introns indicated by thin or thick boxes, and lines respectively. The direction of transcription is from left to right in all shown genes. (C) ChIP-qPCR confirmation of a panel of target genes in HU-synchronized HeLa cells. ChIPs were performed using an E2F7 antibody or non-immune IgGs as controls. ChIP qPCR on a non-specific region 700 bp upstream of E2F binding sites of the E2F1 promoter (E2F1 exon 1) served as negative control for unspecific binding. Graphs represent average ± SD (n = 2) (D) Position weight matrix of highest overrepresented motif within E2F7 peaks compared with random DNA regions. (E) Distribution of the E2F7 binding motif among ChIP-seq peaks, based on proximity to nearest transcription start site (TSS).

Figure 2.

Validation of E2F7 target genes by microarray analysis after doxycycline-induced E2F7 overexpression. (A) qPCR analysis for E2F7 gene expression in cell lines stably expressing doxycycline inducible E2F7-EGFP (E2F7), E2F7-DBDmut-EGFP (DBDmut), or EGFP alone, cultured either with doxycycline (DOXY) or vehicle for 24 h. Data represent average ± SEM (n = 3), asterisks indicate P < 0.05 versus vehicle (B) GFP and E2F1 immunoblots on HeLa cell lysates harvested at different time points after doxycycline-induced E2F7-EGFP expression. Immunoblotting for γ-tubulin served as loading control. (C) Venn diagram to show overlap between down- (red) and upregulated (green) transcripts identified by gene expression microarray after 8 h E2F7 overexpression and transcripts related to genes with a E2F7 ChIP peak (blue) within 5 kb from their transcription start site (TSS). (D) Percentages of genes containing a ChIP peak within 5 kb from the TSS, categorized according to regulation in gene expression microarray after E2F7 overexpression, i.e., downregulated (DOWN), no change (NC), or upregulated (UP). Error bars represent standard deviation of 1000 randomized datasets with resampled gene names. Triple asterisk indicates P < 0.001 calculated from permutation test. (E) Heat map of PANTHER gene ontology enrichments in microarray and ChIP-seq showing similar enrichments of gene ontology clusters. Only clusters with Benjamini scores <0.05 were considered significantly enriched.

Figure 3.

E2F7 represses transcription of G₁/S but not G₂/M genes. (A) Gene expression of G1/S- and G2/M-regulated genes involved in DNA replication, DNA repair and mitosis at different time point after doxycycline (DOXY) dependent induction E2F7-EGFP (E2F7) or E2F7-DBDmut-EGFP in HU arrested cells, analyzed by qPCR. Graphs represent average ± SEM (n = 3). Asterisks indicate P < 0.05 versus its corresponding 0 h time point. (B) FACS plots show that cells treated with doxycycline or vehicle were in G₂ 9 h after HU release; green overlay indicates percentage of E2F7-GFP positive cells versus all propidium iodide stained cells. Gene expression analysis similar as in A was performed on G₁/S- and G₂/M-regulated genes in cells harvested 9 h after HU release and doxycycline treatment. Data represent average ± SEM (n = 3); asterisks indicate P < 0.05 versus vehicle. (C) Gene expression analysis of G₁/S- and G₂/M-regulated genes in Myc/Ras transformed wild-type and E2f7^−/−E2f8^−/− mouse embryonic fibroblasts (MEFs) by qPCR. Data represent average ± SEM (n = 5); Asterisks indicate P < 0.05 versus wild-type. (D) FACS plots showing a dramatic increase in 4 n Myc/Ras transformed wild-type and E2f7^−/−E2f8^−/− MEFs after 16 h of nocodazole (250 ngml⁻¹) Gene expression analysis of G₁/S-regulated genes in the MEFs of indicated genotypes by qPCR. Data represent average ± SEM (n = 3); Asterisks indicate P < 0.05 versus wild-type.

Figure 4.

Timed E2F7 overexpression results in a severe DNA replication defect. (A) Flow cytometric analysis of DNA content of propidium iodide (PI)-stained cells, which were induced to express E2F7-EGFP (E2F7 DOXY) or E2F7-DBDmutant-EGFP (DBDmut DOXY) during the last 12 h of HU-synchronization (−12 h HU). Vehicle treated cells containing the E2F7-EGFP (E2F7 vehicle) served as an additional control. (B) Flow cytometric analysis on HU-synchronized cells, where E2F7-EGFP expression was induced at the time point of HU release (0 h HU). (C) Flow cytometric analysis on cells synchronized by 72-h serum starvation (0.1% fetal bovine serum), where E2F7-EGFP expression was induced 24 h before serum addition (10% fetal bovine serum; −24 h serum).

Figure 5.

E2F7 overexpression causes DNA damage. (A) Flow cytometric analysis of asynchronously growing cells, where E2F7 or E2F7 DBD-mutant expression was induced for 24 h by adding doxycycline. Vehicle treated cells were used as controls. Cells were labeled with DAPI and antibodies against γ-H2AX a marker for DNA damage. (B) Quantification of γ-H2AX staining, showing average percentage of gated cells with γ-H2AX signal above background level. Data represent average ± SEM (n = 3–4). Triple asterisk indicates P < 0.0005 versus E2F7 Wt vehicle. (C) RAD51 immunoblots on lysates of asynchronized cells, where E2F7 or E2F7 DBD-mutant expression was induced for 24 h by adding doxycycline. γ-Tubulin immunoblotting served as loading control. (D) Quantification of flow cytometric analysis of Annexin V and DAPI staining on asynchronized cells, where E2F7 expression was induced for 24 or 48 h by adding doxcycline. Apoptotic cells were counted as Annexin positive and DAPI negative cells. Data represent average ± SEM (n = 3); double asterisk indicates P < 0.005 versus vehicle.