E2F1-mediated upregulation of p19INK4d determines its periodic expression during cell cycle and regulates cellular proliferation

Abstract

Background: A central aspect of development and disease is the control of cell proliferation through regulation of the mitotic cycle. Cell cycle progression and directionality requires an appropriate balance of positive and negative regulators whose expression must fluctuate in a coordinated manner. p19INK4d, a member of the INK4 family of CDK inhibitors, has a unique feature that distinguishes it from the remaining INK4 and makes it a likely candidate for contributing to the directionality of the cell cycle. p19INK4d mRNA and protein levels accumulate periodically during the cell cycle under normal conditions, a feature reminiscent of cyclins.

Methodology/principal findings: In this paper, we demonstrate that p19INK4d is transcriptionally regulated by E2F1 through two response elements present in the p19INK4d promoter. Ablation of this regulation reduced p19 levels and restricted its expression during the cell cycle, reflecting the contribution of a transcriptional effect of E2F1 on p19 periodicity. The induction of p19INK4d is delayed during the cell cycle compared to that of cyclin E, temporally separating the induction of these proliferative and antiproliferative target genes. Specific inhibition of the E2F1-p19INK4d pathway using triplex-forming oligonucleotides that block E2F1 binding on p19 promoter, stimulated cell proliferation and increased the fraction of cells in S phase.

Conclusions/significance: The results described here support a model of normal cell cycle progression in which, following phosphorylation of pRb, free E2F induces cyclin E, among other target genes. Once cyclinE/CDK2 takes over as the cell cycle driving kinase activity, the induction of p19 mediated by E2F1 leads to inhibition of the CDK4,6-containing complexes, bringing the G1 phase to an end. This regulatory mechanism constitutes a new negative feedback loop that terminates the G1 phase proliferative signal, contributing to the proper coordination of the cell cycle and provides an additional mechanism to limit E2F activity.

Figure 1. p19INK4d is induced by E2F1, 2 and 3.

A. BHK-21 cells were cotransfected with pCMV, E2Fs and DP1 expression vectors (0.5–2.5 µg) and pBabe-Puro (0.5 µg), with total DNA amounts normalized with pCMV. Total RNA from puromycin-resistant cells was extracted and subjected to northern blot analysis. B. HEK-293 cells were cotransfected with E2F1 and DP1 expression vectors (2 µg) and cell lysates (100 µg) were immunoblotted. C. BHK-ER-E2F1 cells were transfected as indicated with pCMV, wild-type or mutant pRB expression vectors (2.5 µg) or wild-type or mutant E2F DO (100 nM), and pBabe-Puro (0.5 µg). After 24 h, cells were treated with 4-OHT for another 24 h. Total RNA was extracted from puromycin-resistant cells and subjected to northern blot analysis. Results showed in Figures are representative of at least two independent experiments.

Figure 2. E2F1 increases the transcriptional activity of p19INK4d.

A. BHK-ER-E2F1 cells were treated with 4-OHT for 8 h and subjected to nuclear run-on assay. Transcription rates of the indicated genes were normalized to that of β-tubulin. Results are representative of two independent experiments. B–D. Indicated cells were cotransfected with p19CAT (4.4 µg) and pCEFL-β-galactosidase (5 µg) reporter plasmids. CAT activity was determined 48 h after transfection and normalized to β-galactosidase activity. BHK-21 cells were transfected with reporter plasmids and 1 or 3 µg of E2F1 expression vector and grown in medium containing 10% or 1% FBS (B). BHK-ER-E2F1 cells were transfected with reporter plasmids, grown in medium containing 10% or 1% FBS, and treated with 4-OHT as indicated (C). BHK-ER-E2F1 were transfected with wild-type or mutant pRB expression vectors (6 µg) or wild-type or mutant E2F DO (100 nM) for 24 h. Cells were transfected with reporter plasmids for another 24 h before 4-OHT treatment (D). In panels B, C and D values are the average ± SD of three independent experiments, each performed in triplicates.

Figure 3. An E2F responsive element from the human p19 promoter is conserved across a wide range of mammalian species.

A. Sequence of the human p19 promoter (−961/−1 from p19 translation initiation site) with putative E2F binding sites indicated by boxes. Deviations from the E2F consensus are shown by underlined letters. B. Sequence alignment of p19 promoter regions from different mammalian species using the ClustalW software. Conserved residues are marked by asterisks and the box indicates the E2F-C site.

Figure 4. The p19INK4d promoter contains functional E2F binding sites.

A. EMSA was performed using oligonucleotides corresponding to the E2F-C (top panel) or E2F-D (middle panel) sites or the E2F consensus sequence (E2F CS) (bottom panel) as radiolabeled probes. BHK-21 nuclear extracts (N.E.) were incubated with probes alone or in the presence of 50-, 500-, or 1000-fold molar excess of each of the indicated unlabeled competitors. Results are representative of at least two independent experiments. B. BHK-ER-E2F1 cells were cotransfected with 4.4 µg of the indicated CAT reporters and 5 µg of pCEFL-β-galactosidase for 24 h. Cells were treated with 4-OHT as indicated. CAT activity was determined 48 h after transfection and normalized to β-galactosidase activity. Values are the average ± SD of two independent experiments, each performed in quadruplicate.

Figure 5. p19 periodic expression during cell cycle is dependent on E2F1.

A. BHK-21 cells were synchronized by serum deprivation. Total RNA was extracted at the indicated times following serum restoration and subjected to northern blot analysis (top panel). Corresponding dishes were subjected to immunoblotting (middle panel) and thymidine incorporation analysis (bottom panel). B. Synchronized BHK-21 cells were subjected to nuclear run-on assay at the indicated times after cell cycle re-entry (top panel). Levels of p19 protein, mRNA and de novo mRNA synthesis are shown (bottom panel). C. BHK-21 cells were treated to induce cell cycle arrest in different phases (for G0 arrest: 1% serum, for G1/S: 200 µM mimosine (mim), for G2: 2.5 µM etoposide (eto) or 100 µM adriamycin (adr), for M: 0.1 µg/ml nocodazole (noc)) and cotransfected with expression vectors for E2F1 (2.75 µg), DP1 (2.5 µg) and pBabe-Puro (0.5 µg). Total RNA from puromycin-resistant cells was extracted and subjected to northern blot analysis (top panel). Corresponding dishes were subjected to flow cytometry (bottom panel). D. BHK-21 cells were transfected with 100 nM wild-type or mutant E2F DO as indicated for 18 h. Cells were then arrested in G1 phase by treatment with 200 µM mimosine for 36 h. Total RNA was extracted and subjected to northern blot analysis (top panel). Corresponding dishes were subjected to thymidine incorporation assay (bottom panel). Values are the average ± SD of three independent experiments. Results showed in panels A, B, C, and D are representative of at least two independent experiments.

Figure 6. E2F1 sequentially induces cyclin E and p19 during the cell cycle.

A. WI-38 cells were synchronized by serum deprivation. Total RNA was extracted at the indicated times following serum restoration and subjected to northern blot analysis. B. EMSA was performed using oligonucleotides corresponding to the E2F-C site or the E2F consensus sequence (CS) as radiolabeled probes. WI-38 nuclear extracts (N.E.) were incubated with probes alone or in the presence of 20-, 50-, 100-, 200-, or 500-fold molar excess of the indicated unlabeled competitors. Relative quantification of DNA-protein complexes is shown (bottom panel). C and D. Synchronized BHK-ER-E2F1 cells were untreated (C) or treated with 4-OHT (D) for 3 h before cells were stimulated to re-enter the cycle. Total RNA was extracted at indicated time points and subjected to northern blot analysis. Results are representative of at least two independent experiments.

Figure 7. Periodic expression of p19 contributes to proper cell cycle regulation.

A. HEK-293 cells were transfected with E2F1 and DP1 expression plasmids and the indicated oligonucleotides (100 nM for DO and 500 nM for TFO) for 24 h. Total RNA was extracted and subjected to northern blot analysis. B. WI-38 cells were transfected with control or E2F TFO β and synchronized by serum deprivation 24 h after transfection. Total RNA was extracted at the indicated times following serum restoration and subjected to northern blot analysis. C. WI-38 cells were transfected with the indicated TFO for the indicated days and thymidine incorporation was assayed at the indicated time points. Data was compared using Mann-Whitney test (SPSS 11.5.1 LEADTOOLS). Asterisk indicates p<0.05. D. Corresponding dishes from day 4 were subjected to flow cytometry (top panel). Results are quantitated in bar graph (bottom panel). In A, B, C and D panels results are representative of at least three independent experiments.

Figure 8. p19 mediates a negative feedback mechanism that regulates E2F activity.

A simplified model of normal cell cycle progression. Following phosphorylation of pRb, free E2F induces p19 through binding to its promoter. This induction is delayed, compared to that of cyclin E (indicated by numbers 1 and 2), due to differences in the affinity of the E2F binding sites in both promoters. The E2F-dependent accumulation of p19 inhibits the CDK4 containing complexes, bringing the G1 phase to an end.