Inhibition of p53 by adenovirus type 12 E1B-55K deregulates cell cycle control and sensitizes tumor cells to genotoxic agents

Abstract

Adenovirus E1B-55K represses p53-mediated transcription. However, the phenotypic consequence of p53 inhibition by E1B-55K for cell cycle regulation and drug sensitivity in tumor cells has not been examined. In HCT116 cells with constitutive E1B-55K expression, the activation of p53 target genes such as the p21, Mdm2, and Puma genes was attenuated, despite markedly elevated p53 protein levels. HCT116 cells with E1B-55K expression displayed a cell cycle profile similar to that of the isogenic HCT116p53(-/-) cells, including unhindered S-phase entry despite DNA damage. Surprisingly, E1B-55K-expressing cells were more sensitive to drug treatment than parental cells. Compared to HCT116 cells, HCT116p53(-/-) cells were more susceptible to both doxorubicin and etoposide, and E1B-55K expression had no effects on drug treatment. E1B-55K expression increased the rate of cell proliferation in HCT116 but not in HCT116p53(-/-) cells. Thus, deregulation of p53-mediated cell cycle control by E1B-55K probably underlies sensitization of HCT116 cells to anticancer drugs. Consistently, E1B-55K expression in A549, A172, and HepG2 cells, all containing wild-type (wt) p53, also enhanced etoposide-induced cytotoxicity, whereas in p53-null H1299 cells, E1B-55K had no effects. We generated several E1B-55K mutants with mutations at positions occupied by the conserved Phe/Trp/His residues. Most of these mutants showed no or reduced binding to p53, although some of them could still stabilize p53, suggesting that binding might not be essential for E1B-55K-induced p53 stabilization. Despite heightened p53 protein levels in cells expressing certain E1B-55K mutants, p53 activity was largely suppressed. Furthermore, most of these E1B-55K mutants could sensitize HCT116 cells to etoposide and doxorubicin. These results indicate that E1B-55K might have utility for enhancing chemotherapy.

Fig. 1.

Repression of p53 target genes by Ad12 E1B-55K protein. Colon carcinoma cell line HCT116 was stably transduced with lentiviral vector expressing GFP (HCT116 vector), a GFP-Ad12 E1B-55K fusion (HCT116 GFP-E1B-55K), or wt Ad12 E1B-55K in addition to GFP, in which the coding sequences of E1B-55K and GFP are separated with an internal ribosomal entry site (IRES) (HCT116 E1B-55K). The parental line along with the derived stable cell lines were untreated (NT) or treated with doxorubicin (Doxo) (0.1 μM) or etoposide (Etop) (1 μM) for 24 h. The cells were harvested for Western blotting using the indicated antibodies. In each lane, 30 μg of total cellular proteins was loaded. The blots were probed with extracellular signal-regulated kinase 1 and 2 (Erk1/2) and mitochondrial heat shock protein 60 (Hsp60) as loading controls. For the p21 blot, two different exposures are shown. For the E1B-55K blot, the samples were unheated to avoid the formation of E1B-55K aggregates at high temperature.

Fig. 2.

Metabolic stabilization of p53 by Ad12 E1B-55K protein. HCT116 parental cells and the HCT116-E1B-55K cell line stably expressing Ad12 E1B-55K were exposed to 25 μg/ml of cycloheximide, and at the indicated time points after its addition, cells were harvested for Western blotting with the monoclonal anti-p53 antibody DO-1. E1B-55K was detected using unheated samples with an anti-Ad12 E1B-55K antiserum. Equal amounts (30 μg) of total cellular proteins were loaded in each lane. Two different exposures of the p53 blot are shown for HCT116 cell extracts. The membranes were reprobed with anti-Erk1/2 antibody as a loading control.

Fig. 3.

Deregulation of cell cycle control by Ad12 E1B-55K protein. Subconfluent cultures of the HCT116 parental cell line and the isogenic line HCT116p53^−/− or their derivatives with stable E1B-55K expression were untreated or exposed to 0.1 μM doxorubicin. After 24 h, cells were subjected to flow cytometry analysis. (A) Representative cell cycle profiles of these cell lines with or without drug treatment are shown. (B) Quantitative distributions in different phases of the cell cycle of the indicated cell lines. Shown are average values from two independent experiments along with standard deviations.

Fig. 4.

Ad12 E1B-55K protein sensitizes HCT116 cells to genotoxic agents. HCT116 parental cell line and the isogenic line HCT116p53^−/− (HCT116p53KO) or their derivatives that were stably transduced with vector (HCT116-vector) or that expressed Ad12 E1B-55K (HCT116-55K) or a GFP-E1B-55K fusion (HCT116-GFP-55K) were untreated or exposed to various doses of etoposide, doxorubicin, or KU-55933. The fractions of viable cells after exposure to drug for 96 h were determined and are plotted against drug concentrations. Shown are average values from three independent experiments along with standard deviations. (A to C) Effects of E1B-55K expression in HCT116 cells on sensitivity to etoposide (A), doxorubicin (B), or KU-55933 (C). (D and E) Effects of E1B-55K expression in HCT116p53^−/− (HCT116p53KO) cells on sensitivity to etoposide (D) or doxorubicin (E). (F) Expression levels of E1B-55K in HCT116 and HCT116p53^−/− (HCT116p53KO) cells.

Fig. 5.

Effects of E1B-55K expression on cell proliferation. The parental HCT116 cell line and the indicated derivatives were seeded in equal number in triplicate in a 96-well plate. At the indicated time points, relative cell numbers were determined using a CellTiter-Glo kit. The average relative cell numbers along with standard deviations are plotted against time on a semilogarithmic scale.

Fig. 6.

Ad12 E1B-55K protein sensitizes diverse tumor cells to genotoxic agents. Non-small-cell lung cancer cell lines H1299 and A549, hepatocellular carcinoma cell line HepG2, and glioblastoma cell line A172 were stably transduced with empty vector or that expressing Ad12 E1B-55K. These stable cell lines were exposed to various doses of etoposide, and their drug sensitivity was determined as for Fig. 4. Shown are average values from three independent experiments along with standard deviations. (A to D) Effects of E1B-55K expression in H1299 (A), A549 (B), HepG2 (C), or A172 (D) cells on sensitivity to etoposide. (E and F) Normal human fetal lung fibroblast cell line IMR-90 was transfected with an empty vector (vector) or a vector expressing E1B-55K. The transfected cells were treated with dosed etoposide (E) or doxorubicin (F). (G) The expression of E1B-55K in these cell lines was assessed by Western blotting.

Fig. 7.

Coprecipitation of wt Ad12 E1B-55K and various mutants with p53. (A) Predicted secondary structural elements based on Chou-Fasman and Robson-Garnier methods as implemented in the MacVector software. White and black boxes denote α helices and β strands, respectively. Nonstructured or loop-turn regions are shown as thick lines. Corresponding positions of linker insertions within Ad2 E1B-55K are indicated with arrows (51). Insertions that suppressed the interaction of Ad2 E1B-55K with p53 are denoted with arrows with open heads and those that did not affect this interaction with black arrows. The Sin3A-binding site near the N terminus, the C-terminal repression domain, and the p53-binding core region are indicated. Bioinformatics prediction using the PONDR program (Molecular Kinetics, Inc., Indianapolis, IN) indicates that the central part (amino acids [aa] 100 to 450, thick underline) of Ad12 E1B-55K is possibly ordered, whereas the N-terminal region (aa 1 to 100, thin underline) and the C-terminal tail (aa 450 to 482, thin underline) may be disordered. (B and C) Vector-transduced HCT116 cells (vector) or those transduced with a vector expressing wt E1B-55K or a specific mutant as indicated were subjected to IP with a rabbit anti-E1B-55K antiserum. The immunoprecipitates were extensively washed and then resolved by SDS-PAGE, and the coprecipitated p53 was examined using Western blotting. Cell extracts (input, about 5% of materials used for IP) were also loaded in the gels. The E1B-55K blots were done using unheated samples as for the other figures. Two exposures of the IP blots are shown. The results shown are representative of three independent IP experiments.

Fig. 8.

Effects of E1B-55K mutations on p53 stabilization and activity. HCT116 cells were stably transduced with vector expressing wt Ad12 E1B-55K or the indicated point mutant. As a positive control for p53 activation, the vector-transduced HCT116 cells were exposed to etoposide at 1 μM as indicated for 24 h. The cell extracts of these stable cell lines as well as the parental cells were subjected to SDS-PAGE and Western blotting with the indicated antibodies. Equal amounts of total extracts were loaded in each lane. PCNA was detected as a loading control.

Fig. 9.

Chemosensitization by Ad12 E1B-55K mutants. The parental HCT116 cell line and a panel of derived cell lines expressing wt Ad12 E1B-55K or the indicated mutant as in Fig. 8 were untreated (NT) or exposed to the indicated dose of doxorubicin or etoposide. Cell viability was assessed as for Fig. 4. Shown are average values from three independent experiments along with standard deviations. Statistical significance was calculated using Student's t test.