Regulation of E2F1 activity via PKA-mediated phosphorylations

Abstract

E2F1 becomes activated during the G1 phase of the cell cycle, and posttranslational modifications modulate its activity. Activation of G-protein coupled receptors (GPCR) by many ligands induces the activation of adenylate cyclases and the production of cAMP, which activates the PKA enzyme. Activated PKA elicits its biological effect by phosphorylating the target proteins containing serine or threonine amino acids in the RxxS/T motif. Since PKA activation negatively regulates cell proliferation, we thought that activated PKA would negatively affect the activity of E2F1. In line with this, when we analyzed the amino acid sequence of E2F1, we found 3 hypothetical consensus PKA phosphorylation sites located at 127-130, 232-235, and 361-364 positions and RYET, RLLS, and RMGS sequences. After showing the binding and phosphorylation of E2F1 by PKA, we converted the codons of Threonine-130, Serine-235, and Serine-364 to Alanine and Glutamic acid codons on the eukaryotic E2F1 expression vector we had previously created. We confirmed the phosphorylation of T130, S235, and S364 by developing monoclonal antibodies against phospho-specific forms of these sites and showed that their phosphorylation is cell cycle-dependent. According to our results, PKA-mediated phosphorylation of E2F1 by PKA inhibits proliferation and glucose uptake and induces caspase-3 activation and senescence.

Keywords: E2F1; PKA; cell cycle; forskolin; proliferation; senescence.

Figure 1

PKA activation destabilizes E2F1 and mutants of E2F1 can be ectopically expressed. (A) HEK293 cells were treated with forskolin (20 mM) and cellular lysates were collected for indicated times. Approximately 100-mg total protein lysates were fractionated on 10% SDS-PAGE and blot was labeled with an anti-E2F1 antibody. The blot was stripped off and relabeled with an anti-GAPDH antibody. Band intensities were determined using ImageJ, and the relative abundance of the E2F1 level was determined by dividing E2F1 band intensities to that of GAPDH. (B) HEK293 cells were pretreated with PKA inhibitor H9 for 1 h, then forskolin was added for indicated times and levels of E2F1; GAPDH relative change in expression of E2F1 was determined as mentioned. Band intensities were determined using ImageJ and relative abundance of E2F1 level was determined by dividing E2F1 band intensities to that of GAPDH. (C,D) Subconfluent HEK293 cells cultured DMEM in 100 mm plates were transfected with 30-ug expression vectors of Alanine mutants (C) and Glutamic acid mutants (D) of E2F1 using calcium phosphate protocol for 48 h. Cellular lysates were collected in lysis buffer and 100-ug total protein lysates were fractionated on 10% SDS-PAGE, and the blot was labeled with an anti-E2F1 antibody. The blot was stripped off and relabeled with an anti-GAPDH antibody (WT-E2F1 was used as a control in statistical analysis. “a”: statistically significant decrease, P < 0.0001).

Figure 2

PKA and E2F1 bind to each other, and activation of PKA causes a decline inthe protein level of E2F1. (A,B) Physical interaction between E2F1 and PKA was explored with reciprocal immunoprecipitations. E2F1 and PKA were immunoprecipitated from untreated and 1 h- forskolin-treated HEK293 cells. E2F1 and PKA were immunoprecipitated from 1mg of total cellular lysate and blotted with E2F1 and PKA antibodies. (C) In vitro kinase reaction was performed with a PKA catalytic subunit. Wild type E2F1 was immunoprecipitated from 3-mg total cellular lysate prepared from HEK293 cells overexpressing the wild type E2F1. After immunoprecipitation, the pellet was washed 3 times in 1X PKA reaction buffer and resuspended in 50 mL of reaction buffer. It then underwent in vitro kinase reaction with the addition of the PKA catalytic subunit PKAca (purchased from NEB-P600S) and 10 mM ATP. The reaction mix was incubated for 30 min at 37oC. We terminated the reaction by incubating the samples at 65 °C for 5 min, then washed the reaction mix according to the manufacturer’s protocol; we then resuspended the pellet in SDS-loading dye, incubated it at 95 °C for 5 min, and fractionated the supernatant and lysate in duplicates using 10% SDS-PAGE. Immunoblotting was performed with E2F1 and an phospho-PKA substrate antibody.

Figure 3

PKA phosphorylation site mutants differentially affect proliferation of HEK293 cells. To determine the effect of Alanine and Glutamic acid mutants of E2F1 on proliferation, HEK293 cells were reverse transfected with 100 ng of mock and vectors in 96-well plates and incubated for 72 h. Cell viability was determined using MTT assay (WT-E2F1 was used as a control in statistical analysis. “a”: statistically significant increase, P < 0.0001; “b”: statistically significant increase, P < 0.0001).

Figure 4

PKA phosphorylation site mutants differentially affect expression of Ki67 in HEK293 cells. HEK293 cells were reverse transfected with 100 ng of mock and vectors in 96-well plates and incubated for 72 h. Then, Ki67 immunofluorescence staining was performed (Ki67: red; DAPI: blue; Merge: red and blue).

Figure 5

PKA phosphorylation site mutants differentially affect Caspase activation and Glucose uptake. (A) HEK293 cells were reverse transfected with mock-, wild-type, or mutants of E2F1 for 48 h, and caspase-3 activity was determined by colorimetric Caspase-3 activation assay (Bio vision K106-200). (B) HEK293 cells were reverse transfected with mock-, wildtype, or mutants of E2F1 for 24 h in 96-well plates and incubated for 24 h. A Cayman Glucose Uptake Cell Based Assay Kit was used to determine the level of glucose uptake (CTCF: corrected total cell fluorescence). (WT-E2F1 was used as a control in statistical analysis. “a”: statistically significant decrease, P < 0.0001; “b”: statistically significant increase, P < 0.0001; “c”: statistically significant decrease, P < 0.005).

Figure 6

Synchronization of 293T cells using double thymidine block. HEK293 cells were synchronized by double-thymidine block for 48 h; then, the cells were washed with PBS and released by adding complete DMEM (without thymidine). The cells were then collected at 0-2-4-6-8-12-16-20-24 h postrelease for propidium-iodide staining; cell cycle analysis was done using flow cytometry.

Figure 7

E2F1 is phosphorylated by the PKA at G0/G1 stages of the cell cycle. (A) To determine the specificity of antibodies in our antibodies, we treated serum-starved HEK293 cells with forskolin (20 uM) for 1h, and 100 ug of cell lysates was fractionated on 10%SDS-PAGE; blots were labeled with 1/200 dilutions of the supernatant of our monoclonal hybridomas.(B) Coomasie-blue staining of purified supernatants of our monoclonal hybridomas by column. (C,D,E) We ectopically expressed Alanine mutants of E2F1 for 48h; then, 100 ug of cell lysates were fractionated on 10% SDS-PAGE and blots were labeled with corresponding antibodies. (F) HEK293 cells were synchronized by double-thymidine block for 48 h, released, and the cellular lysates of the released cells were prepared between at 2–24 h. postrelease. 100 mg of total cell lysates were fractionated on 10% SDS-PAGE. Immunoblot was first labeled with anti-E2F1 antibody (1:1000), stripped off, and relabeled with anti-pT130, anti-pS235, and-pS364 antibodies (1:1000) and with antimouse HRP (1:5000). The bottom portion of the blot was also labeled with anti-GAPDH as a loading control. Analysis of SDS-PAGE results: the fold changes in phosphorylation were determined by dividing ImageJ values of band intensities of phospho-bands to that of the E2F1 (zero hour (0 h) protein level was used as a control in statistical analysis. “a”: statistically significant increase, P < 0.0001; “b”: statistically significant decrease, P < 0.0001; “c”: statistically significant decrease, P = 0.0002).

Figure 8

PKA-mediated phosphorylation of E2F1 induces senescence. HEK293 cells were reverse transfected with mock and vectors in 96-well plates and incubated for 48 h. Senescence-associated β-galactosidase staining was then performed. Staining was determined using ImageJ and analyzed with a GraphPad Prism (WT-E2F1 was used as a control in the statistical analysis. “a”: statistically significant increase, P < 0.0001).