The adenovirus E1A oncogene induces innate immune rejection of tumors by sensitizing tumor cells to apoptosis in response to injuries, such as those inflicted by macrophage-produced TNF α and NO. E1A sensitizes cells to TNF by repressing its activation of NF-κB-dependent, antiapoptotic defenses. This suggested the hypothesis that E1A blockade of the NF-κB activation response might be the central mechanism of E1A induced cellular sensitivity to other proapoptotic injuries, such as macrophage-produced NO. However, creation of E1A-positive NIH-3T3 mouse cell variants with high-level, NF-κB-dependent resistance to TNF did not coselect for resistance to apoptosis induced by either macrophage-NO or chemical-NO, as the hypothesis would predict. E1A expression did block cellular recovery from NO-induced mitochondrial injury and converted the reversible, NO-induced cytostasis response of cells to an apoptotic response. This viral oncogene-induced phenotypic conversion of the cellular injury response of mouse and human cells was mediated by an E1A-related increase in NO-induced activation of caspase-2, an apical initiator of intrinsic apoptosis. Blocking caspase-2 activation or expression eliminated the NO-induced apoptotic response of E1A-positive cells. These results define an NF-κB-independent pathway through which the E1A gene of human adenovirus sensitizes mouse and human cells to apoptosis by enhancement of caspase-2-mediated mitochondrial injury.

Adenoviruses are common respiratory pathogens, which like many viruses commandeer cell cycle machinery to enhance the efficiency of viral replication. The central adenoviral gene in this activity is the early region 1A gene, E1A,3 which is a promiscuous modulator of viral and cellular gene transcription and the primary mediator of the cell cycle takeover (1). When expressed in abortive infection or ectopically during stable cellular transfection, E1A functions as an oncogene that immortalizes mammalian cells, similarly to functionally homologous oncogenes of other DNA tumor viruses and cellular oncogenes (2, 3). In addition to its cell cycle control and related cellular immortalizing activities, E1A expression has the side effect of sensitizing cells to apoptosis in response to a variety of immunological, chemical, and physical stimuli (reviewed in Ref. 4). It is partly for this reason that targeted E1A gene delivery has been used in various forms of cancer therapy, including those using oncolytic adenoviruses and liposomal delivery of E1A genes to tumor cells (reviewed in Ref. 5). Despite the well-described E1A-induced sensitivity of tumor cells to apoptosis in response to injury by immune killer cells (NK cells, activated macrophages, and CTL), TNF α, TRAIL, Fas ligand, chemotherapeutic drugs, and irradiation, there is an incomplete understanding of the cellular pathways and molecular mechanisms through which E1A mediates the conversion of cells from the apoptosis-resistant to the apoptosis-sensitive phenotype (reviewed in Ref. 4). One hypothesis is that E1A targets and modulates the activity of one or more apoptotic pathways in expresser cells, as a collateral effect of E1A cell cycle takeover. One candidate for such a pathway is the antiapoptotic, NF-κB activation response that is known to be repressed by E1A, resulting in increased cellular sensitivity to TNF α (reviewed in Ref. 4).

Activated macrophages, one component of the innate immune antitumor defense, repress cellular mitochondrial function, an effect that can result in either tumor cell cytostasis or apoptosis, depending on the cell being targeted (6, 7). NO is the main macrophage cytotoxin that kills E1A-positive tumor cells (8). One reported NO effect on cells is NF-κB activation (9, 10, 11, 12, 13), which can defend cells against NO-induced apoptosis (14, 15, 16). These observations suggested the hypothesis that the NF-κB activation response is the key pathway that is targeted by E1A to sensitize tumor cells to macrophage-NO-induced apoptosis. We report evidence in this study that this hypothesis is incorrect and that E1A sensitizes cells to NO-induced apoptosis through a mechanism that is independent of E1A-induced NF-κB repression. The data indicate that the alternative mechanism involves E1A-enhanced, NO-induced activation of caspase-2 that may directly injure mitochondria, resulting in triggering of the intrinsic apoptosis pathway (17, 18). These results, combined with those reported previously, indicate that there are two complementary effects of E1A that sensitize tumor cells to macrophage killing and that are involved in the antitumor cell effects of E1A in the context of the host innate immune response. One is E1A repression of death receptor-triggered NF-κB activation that could be involved in E1A-induced tumor cell sensitivity to TNF α (19, 20), TRAIL (21, 22, 23, 24), and Fas ligand (25). The other, reported in this study, involves NF-κB-independent enhancement of caspase-induced mitochondrial injury inflicted by killer cell cytolytic mechanisms such as NO.

NIH-3T3 cells expressing Ad5 E1A 12S proteins (MT12–1) and human H4 cells (a subclone of the well characterized human fibrosarcoma cell line, HT-1080) expressing genomic E1A (H4-E1A, P2AHT2A) (26, 27) and their derivatives were maintained in DMEM plus antibiotics and 5% calf serum (complete medium). TNF-resistant cell variants (20 ng/ml TNF; Biosource) were selected for at least 2 wk, and their incidence was estimated by colony counting. E1A-positive NIH-3T3 cells that stably overexpress NF-κB, p65/RelA were created by retroviral transduction as follows. A murine stem cell virus retroviral vector was used to express GFP and p65/RelA from a bicistronic message. pMSCVRelA-ires-GFP (28), provided by William Sha (University of California, Berkeley, CA), was used (29) to generate the retroviral vector by cotransfection of 293 cells with pCLeco (provided by Inder Verma, The Salk Institute, La Jolla, CA), a packaging plasmid expressing structural proteins of Moloney murine leukemia virus (30). E1A, p65/RelA, Bax, cIAP2, Bcl-2 (Santa Cruz Biotechnology) and actin (Sigma-Aldrich) were detected by immunoblotting using ECL. Comparisons of cell protein expression were made by analysis of scanned images of immunoblots using Molecular Dynamics ImageQuant v. 5.1 software. Band intensities were normalized to the levels of actin expression, and these actin-normalized intensities were used for comparisons of fold-changes in protein expression. Assays of the NF-κB-dependent transcription response to TNF stimulation were done using the dual-luciferase transfection method described previously (20), with cells tested after 4 h of TNF (20 ng/ml) stimulation. In brief, cells were transfected with two reporter plasmids, pNF-κB-firefly luciferase and TK-Renilla luciferase, using GenePORTER 2 (Genlantis) lipid-based transfection. Cell lysates were tested for firefly and Renilla luciferase expression by the dual-luciferase assay (Promega). Renilla luciferase levels were used to normalize transfection efficiencies between cell lines as follows. Firefly luciferase activity was divided by Renilla luciferase specific activity (Renilla luciferase activity divided by the cellular protein concentration). These results are presented as normalized κB-LUC.

TNF-induced cell killing (20 ng/ml) was measured by radiolabel release from the indicated cell types as described (20) without or with cell pretreatment with cycloheximide (10 ug/ml; Sigma-Aldrich) to block protein synthesis, or with the I-κB kinase inhibitor, Bay-117085 (4 μM; Calbiochem), to specifically block NF-κB-dependent transcription activation. Macrophage-induced killing was measured by radiolabel release from target cells as described (8, 31). Macrophage NO production was blocked with NG-monomethyl-l-arginine methyl ester (l-NAME) (1 mM; Calbiochem). For chemical NO killing, cells were treated with DETA-NONOate (Cayman Chemical). The quality of the cell death response was assessed by flow cytometry (LSR, BD Biosciences) of cells stained with Hoechst and propidium iodide (PI, Vybrant Assay 5, Invitrogen).

Mitochondrial membrane potential (MMP) was estimated by flow cytometry of cells stained with tetramethylrhodamine ethyl ester perchlorate (TMRE; 100 nM; Invitrogen). ATP levels were estimated using the somatic cell ATP assay (Sigma-Aldrich), with cell lysates tested at baseline and the indicated times after NO injury under three culture conditions: 1) high (25 mM) glucose, 2) low (5.5 mM) glucose plus 2-deoxyglucose (20 mM; Sigma-Aldrich), or 3) oligomycin (20 μM; Sigma-Aldrich). Glycolysis was measured by monitoring the conversion of 5-[3H]-glucose to [3H]2O, essentially as described by Ashcroft et al. (32). One × 106 cells were suspended in 0.5 ml of complete medium in the absence or presence of 20 μM oligomycin for 30 min at 37°C. [5-[3H]-glucose (PerkinElmer; d-(5-[3H] (N)), 10–20 Ci (370–740 GBq)/mmol) was added at 20 μCi/ml to complete medium without or with DETA-NONOate (250 μM). 50 μl samples taken at 30, 60, 120, and 180 min were added to 50 μl of 0.2 N HCl in an uncapped PCR tube that was put in a scintillation tube containing 250 μl of H2O, so that the contents of the PCR tube did not mix with the H2O in the scintillation tube. The scintillation tube was capped, and [3H]2O was allowed to diffuse for 24 h. Glucose use, as measured by [3H]2O production, was linear at these time points. 1 mole of [5-[3H]-glucose fully consumed will produce 1 mol of [3H]2O. The amount of [3H]2O was determined in pmol for each time point and plotted vs time to determine the rate of glucose consumption. Final values of glucose use were expressed as pmol/h/1 × 106 cells.

Caspase activity was inhibited with the pancaspase inhibitor, zVAD-fmk (100 μM; Calbiochem), or the caspase-2-selective inhibitor, zVDVAD-fmk (200 μM; Calbiochem). Caspase-2 activity was tested in lysates of 2 × 106 cells using the caspase-2 activity assay (R&D Systems). Caspase-2 expression was compared by immunoblotting (Santa Cruz Biotechnology).

E1A-positive cells were infected with lentiviral particles containing shRNAs against mouse caspase-2 obtained from Sigma-Aldrich. Transduced cells were selected in puromycin. Single cell clones were obtained by FACS and tested for reduced expression of caspase-2 and for continued expression of the parental-cell-level of E1A protein by immunoblotting. The clonal cell line, E1AiC2, that was selected for further testing was generated using the following siRNA target sequence: CCGGCCTTAAAGGTAATGCTGCCATCTC GAGATGGCAGCATTACCTTTAAGGTTTTT.

Statistical analysis was done using Student’s t test or two-way ANOVA, using SigmaStat 3.11 software (Systat Software).

A few E1A-positive 3T3 cells survived apoptotic killing by recombinant TNF (4.5 ± 0.4 × 10−3 cells; mean ± SEM; n = 3). These TNF-resistant (TR) clones expressed parental-cell levels of E1A (TR1 and TRB1, Fig. 1,A) but were as resistant to TNF-induced cytotoxicity as E1A-negative cells (Fig. 1 C, untreated group).

FIGURE 1.

Characterization of E1A-positive, NIH-3T3 cell variants selected for TNF resistance (TR1, TRB1) or forced to overexpress NF-κB p65/RelA (A1, B5), compared with E1A-positive (E1A) and E1A-negative (3T3) cell controls. A, Comparative immunoblots of cell lysates for the indicated proteins. B, TNF-induced, NF-κB-dependent transcription responses. Cells were transiently transfected with a κB-luciferase reporter and tested for response to TNF stimulation (20 ng/ml, 4 h). The numbers above bars represent the fold induction of NF-κB-dependent transcription that was significant for the TNF-resistant clones, TR1 and A1 (n = 3; p < 0.05), but not for the TNF-sensitive parental cells, E1A. C, NF-κB-dependent cellular sensitivity to TNF-induced cytotoxicity. TNF-resistant clones, TR1 and A1, were compared with E1A-positive, TNF-sensitive (E1A) and E1A-negative, TNF-resistant (3T3) control cells for sensitivity to TNF (20 ng/ml)-induced radiolabel release, before (“untreated”) and after blockade of cellular protein synthesis by treatment with cycloheximide (CHX) or of TNF-induced NF-κB activation by treatment with the I-κB kinase inhibitor, Bay-117085 (Bay). The TNF-resistant, E1A-positive clones, TR1 and A1, were significantly more sensitive to TNF-induced cytotoxicity after repression of the NF-κB-dependent activation response, using both treatments, like E1A-negative 3T3 cells (n = 3, p < 0.05).

FIGURE 1.

Characterization of E1A-positive, NIH-3T3 cell variants selected for TNF resistance (TR1, TRB1) or forced to overexpress NF-κB p65/RelA (A1, B5), compared with E1A-positive (E1A) and E1A-negative (3T3) cell controls. A, Comparative immunoblots of cell lysates for the indicated proteins. B, TNF-induced, NF-κB-dependent transcription responses. Cells were transiently transfected with a κB-luciferase reporter and tested for response to TNF stimulation (20 ng/ml, 4 h). The numbers above bars represent the fold induction of NF-κB-dependent transcription that was significant for the TNF-resistant clones, TR1 and A1 (n = 3; p < 0.05), but not for the TNF-sensitive parental cells, E1A. C, NF-κB-dependent cellular sensitivity to TNF-induced cytotoxicity. TNF-resistant clones, TR1 and A1, were compared with E1A-positive, TNF-sensitive (E1A) and E1A-negative, TNF-resistant (3T3) control cells for sensitivity to TNF (20 ng/ml)-induced radiolabel release, before (“untreated”) and after blockade of cellular protein synthesis by treatment with cycloheximide (CHX) or of TNF-induced NF-κB activation by treatment with the I-κB kinase inhibitor, Bay-117085 (Bay). The TNF-resistant, E1A-positive clones, TR1 and A1, were significantly more sensitive to TNF-induced cytotoxicity after repression of the NF-κB-dependent activation response, using both treatments, like E1A-negative 3T3 cells (n = 3, p < 0.05).

Close modal

TR cell variant expression of selected proapoptotic and antiapoptotic proteins was investigated to understand their loss of TNF sensitivity. Loss of both Bax and Bak impairs TNF-induced apoptosis of E1A-positive cells, but expression of either protein is sufficient for apoptosis (33). Bax expression in TR variants at levels comparable to those of parental cells (Fig. 1,A, upper panel) excluded Bax/Bak loss as the explanation for their TNF resistance. TRB1 overexpressed NF-κB p65/RelA (Fig. 1,A, upper panel; 4-fold increase compared with parental, E1A-positive cells), suggesting that selection for TNF-resistant, E1A-positive cells could coselect for overexpression of antiapoptotic, NF-κB-related proteins. Several NF-κB-dependent cellular proteins, such as Bcl-2, c-IAPs, and c-FLIP, can block apoptosis (34). Consistent with this, one TR variant (TR1) overexpressed c-IAP2 (Fig. 1,A, middle panel; 4-fold increase), and another (TRB1) overexpressed Bcl-2 (Fig. 1 A, middle panel; 2-fold increase), when compared with parental, E1A-positive cells. None of these antiapoptotic proteins were overexpressed by three other TR variants, indicating that TNF selection for other resistance mechanisms can also occur in E1A-positive cells.

We have reported that transient overexpression of NF-κB p65/RelA can block TNF-induced apoptosis of E1A-positive cells (20). We therefore tested the phenotypes of E1A-positive cells forced to stably overexpress p65/RelA (Fig. 1,A, bottom panel). Two such clonal lines, A1 and B5, retained parental cell levels of E1A but expressed much higher levels of p65/RelA (Fig. 1,A, bottom panel). These clones, along with TNF-selected, E1A-positive variants were compared with parental, E1A-positive cells for TNF-induced, NF-κB-dependent transcription responses and for sensitivity to TNF-induced cytotoxicity. Data for a representative clone from each group of TNF-resistant clones (TNF selected, TR1; forced RelA overexpression, A1) are shown in Fig. 1, B and C. Other cell variants described in Fig. 1,A were tested and showed similar results (data not shown). These TR variants and p65/RelA overexpressers exhibited relief of E1A repression of TNF-induced, NF-κB-dependent transcription (Fig. 1,B) and were resistant to TNF cytotoxicity (Fig. 1,C, untreated group). Furthermore, their TNF cytotoxic resistance depended on NF-κB activation, because the variants, like E1A-negative NIH-3T3 cells, became sensitive to TNF following pretreatment with cycloheximide (an inhibitor that blocks NF-κB-dependent protein expression) and when NF-κB activation was specifically blocked by the IκB kinase inhibitor, Bay 11–7085 (Fig. 1 C, CHX and Bay, respectively).

The hypothesis that NF-κB repression is the central mechanism through which E1A sensitizes cells to both TNF- and NO-induced apoptosis predicted that selection of E1A-positive cells with NF-κB-dependent TNF resistance should coselect for resistance to killing by macrophage-produced NO. However, the TNF-resistant, E1A-positive variants remained susceptible to NO-dependent macrophage killing (Fig. 2,A). E1A-specific killing of these cells by macrophages was blocked with the NO-synthetase inhibitor, l-NAME (gray bars). Retention of NO sensitivity of these TNF-resistant, E1A-positive cells was confirmed in experiments using the chemical NO generator, DETA-NONOate (Fig. 2 B). These data indicated that the mechanism(s) through which E1A sensitizes cells to NO are independent of E1A repression of NF-κB activation to the level that is sufficient to induce TNF sensitivity.

FIGURE 2.

Sensitivity of TNF-resistant, E1A-positive cell variants to cytotoxicity by macrophage-NO and chemical-NO, compared with macrophage-sensitive, E1A-positive (E1A) and macrophage-resistant, E1A-negative (3T3) control cells as evidenced by radiolabel release. A, Macrophage-NO cytotoxicity. Radiolabeled target cells were incubated on monolayers of activated macrophages at a 50:1 macrophage:target ratio, in the absence or presence of 1 mM l-NAME to block macrophage NO-synthase activity. TNF-resistant TR1, A1, and B5 cells were highly sensitive to NO-dependent macrophage killing (n = 3, p < 0.05). B, Chemical-NO cytotoxicity. Radiolabeled cells were incubated with the NO generator, DETA-NONOate, at a concentration of 250 uM. All TNF-resistant, E1A-positive variants remained significantly more sensitive to NO cytotoxicity than E1A-negative control cells (3T3) (n = 3, p < 0.05).

FIGURE 2.

Sensitivity of TNF-resistant, E1A-positive cell variants to cytotoxicity by macrophage-NO and chemical-NO, compared with macrophage-sensitive, E1A-positive (E1A) and macrophage-resistant, E1A-negative (3T3) control cells as evidenced by radiolabel release. A, Macrophage-NO cytotoxicity. Radiolabeled target cells were incubated on monolayers of activated macrophages at a 50:1 macrophage:target ratio, in the absence or presence of 1 mM l-NAME to block macrophage NO-synthase activity. TNF-resistant TR1, A1, and B5 cells were highly sensitive to NO-dependent macrophage killing (n = 3, p < 0.05). B, Chemical-NO cytotoxicity. Radiolabeled cells were incubated with the NO generator, DETA-NONOate, at a concentration of 250 uM. All TNF-resistant, E1A-positive variants remained significantly more sensitive to NO cytotoxicity than E1A-negative control cells (3T3) (n = 3, p < 0.05).

Close modal

NO can induce either apoptosis or necrosis, depending on the cell type tested (35). NF-κB activation is an anti-apoptotic defense, but does not defend against necrotic cell death (36, 37, 38). Therefore, it was possible that the NF-κB-independence of the NO sensitivity of E1A-positive cells was because the cells were dying as a result of NO-induced necrosis, not apoptosis. However, flow cytometry analysis showed that the NO-treated, E1A-positive cells died by apoptosis (Fig. 3).

FIGURE 3.

Flow cytometry analysis of the quality of NO-induced cell death and changes in MMP of E1A-positive vs E1A-negative cells 18 h after injury. A, Changes in forward light scatter/side scatter and Hoechst/PI staining patterns of E1A-positve and E1A-negative cells following exposure to 250 μM DETA-NONOate. Percentages denote the percentage of cells in each quadrant out of the total cell population. B, Changes in MMP. E1A-positive and E1A-negative cells were treated with 250 μM DETA-NONOate for 18 h and then stained with TMRE to estimate MMP. Loss of MMP by NO-injured (left panel, shaded histogram) compared with uninjured (solid line) E1A-positive cells was caspase-dependent (zVAD-blocked, NO-injured; dashed line). E1A-related changes in MMP in response to NO, expressed as a percentage of the MMP of untreated control cells (right panel; n = 3, p < 0.05). C, Conversion of the NO injury response of E1A-positive cells, by zVAD blockade of caspase activity, to a response that resembled that of E1A-negative cells (i.e., identical to forward light scatter/side scatter and Hoechst/PI patterns of E1A-negative cells in Fig. 3 A).

FIGURE 3.

Flow cytometry analysis of the quality of NO-induced cell death and changes in MMP of E1A-positive vs E1A-negative cells 18 h after injury. A, Changes in forward light scatter/side scatter and Hoechst/PI staining patterns of E1A-positve and E1A-negative cells following exposure to 250 μM DETA-NONOate. Percentages denote the percentage of cells in each quadrant out of the total cell population. B, Changes in MMP. E1A-positive and E1A-negative cells were treated with 250 μM DETA-NONOate for 18 h and then stained with TMRE to estimate MMP. Loss of MMP by NO-injured (left panel, shaded histogram) compared with uninjured (solid line) E1A-positive cells was caspase-dependent (zVAD-blocked, NO-injured; dashed line). E1A-related changes in MMP in response to NO, expressed as a percentage of the MMP of untreated control cells (right panel; n = 3, p < 0.05). C, Conversion of the NO injury response of E1A-positive cells, by zVAD blockade of caspase activity, to a response that resembled that of E1A-negative cells (i.e., identical to forward light scatter/side scatter and Hoechst/PI patterns of E1A-negative cells in Fig. 3 A).

Close modal

Classical flow cytometry traits of cells undergoing apoptosis—reduced cellular size, increased cellular granularity, and increased chromatin condensation before loss of cell membrane integrity (39)—were used to contrast the quality of the NO-induced, cell death responses of E1A-positive vs E1A-negative cells. Cells were stained with the DNA dye, Hoechst 33342, to measure the increased intensity of nuclear staining associated with chromatin condensation and with the vital dye, PI, to measure loss of cell membrane integrity. Increased NO-induced apoptosis of E1A-positive cells was evidenced by the typical flow cytometry traits: 1) decreased cell size and increased granularity of the majority of the cell population, as revealed by an NO-induced leftward shift in forward light scatter and upward shift up of side scatter, respectively (Fig. 3,A, left panel) and 2) markedly increased chromatin condensation before loss of cell membrane integrity (Fig. 3,A, right panel; increased percentage of Hoechst-positive/PI-negative cells among E1A-positive cells, upper right panel = 85% vs E1A-negative cells, lower right panel = 5%). In contrast, NO treatment of E1A-negative cells caused reduced size of a smaller subpopulation of cells (forward light scatter), little change in granularity (side scatter), and a simultaneous increase in Hoechst and PI staining of a minority (25%) of cells (Fig. 3 A, lower right panel; double-positive cells = 25%), characteristics of necrotic death.

The E1A mechanism(s) of sensitization to NO were considered in the context of NO effects on mitochondrial function because of its central role in the cell death response (Fig. 3,B). MMP is the main determinant of mitochondrial respiration and ATP synthesis. NO-induced loss of MMP causes progression to apoptosis (40). The net mitochondrial injury and recovery response of NO-treated, E1A-positive vs E1A-negative cells was contrasted using TMRE fluorescence as an indicator of MMP at 18 h after injury (Fig. 3,B). The MMP of NO-injured, E1A-positive cells was reduced in a manner that was caspase dependent, because loss of MMP was blocked by the pancaspase inhibitor, zVAD (Fig. 3,B, left panel; compare shaded histogram (NO-treated) vs dashed-line histogram (zVAD-blocked, NO-treated); solid line histogram = control). In contrast, there was no reduction of MMP of most E1A-negative cells at 18 h after injury (Fig. 3,B, lower histogram). This observation of the caspase dependence of the E1A-related blockade of cellular recovery of MMP was observed in repeated experiments (Fig. 3,B, right panel; n = 3). A small fraction of NO-injured, E1A-positive cells exhibited increased TMRE fluorescence (Fig. 3,B, right-shoulders on histograms), consistent with caspase-independent mitochondrial swelling or hyperpolarization (41, 42). zVAD blockade of a net loss of MMP by E1A-positive cells increased survival and changed the quality of the NO response to one that resembled E1A-negative cells, as assessed by flow cytometry (compare the patterns of NO plus zVAD-treated, E1A-positive cells in Fig. 3,C with those of NO-treated, E1A-negative cells in Fig. 3 A). This suggested that the E1A-related sensitivity to NO-induced apoptosis resulted from caspase-dependent, irreversible mitochondrial dysfunction (loss of MMP), in contrast to NO-injured, E1A-negative cells that sustained their MMP.

Macrophage-NO represses mitochondrial respiration, resulting in a transient cytostatic effect on tumor cells (6, 43). This same NO-induced, reversible cytostasis was observed with E1A-negative mouse cells in our studies (data not shown). NO reversibly inhibits mitochondrial respiration by competing with oxygen at cytochrome oxidase, complex IV in the electron transport chain (44, 45). Cellular recovery from this respiration blockade depends on NO-induced compensatory glycolysis and reversal of F1F0 ATP synthase activity that extrudes protons from the mitochondrial matrix, thereby sustaining the MMP (40, 46). We compared NO effects on cellular ATP levels to investigate the reasons for E1A-related progression to apoptosis. NO caused identical reductions in intracellular ATP levels in E1A-negative and E1A-positive cells, to ∼20% of pretreatment levels (Fig. 4, • vs ♦, respectively). This is the reduction predicted for complete blockade of respiration in cells targeted by activated macrophages or treated with chemical NO (6, 43). ATP levels in E1A-negative cells returned toward pretreatment levels within 4 hours (Fig. 4, •). This ATP level recovery was blocked by 2-deoxyglucose (Fig. 4, ○) and oligomycin (Fig. 4, □) and therefore required glycolysis and F1F0 ATP synthase activity, respectively. E1A-positive cells did not recover from NO injury despite unlimited glucose (Fig. 4, ♦), but their ATP levels did not fall to the levels observed with glycolysis blockade (Fig. 4, 2DG-treated cells, ○). Comparative assays revealed that there was no blockade of compensatory glycolysis in E1A-positive vs E1A-negative cells in response to stimulation with NO or treatment with oligomycin (Fig. 4,B). Therefore, the E1A effect on the postinjury cellular ATP level resulted from an inability of the mitochondria to restore MMP, despite compensatory glycolysis. Residual ATP levels in E1A-positive cells were sufficient for progression to apoptosis (Fig. 3 A), in contrast to the necrotic death of E1A-negative cells that followed ATP exhaustion in glucose depleted conditions (data not shown). This pattern is consistent with reports that some residual level of cellular ATP is required for apoptosis and that necrotic death follows ATP exhaustion (47, 48).

FIGURE 4.

Effects of E1A expression on bioenergetic responses of NO-injured cells. A, ATP levels over time after chemical-NO treatment of E1A-positive (E1A) and E1A-negative (3T3) cells as the percentage of levels in untreated control cells (n = 3, all conditions). ATP levels decreased in both cell types to ∼20% control within 30 min, with a further reduction to ∼10% control in 2-deoxyglucose-treated E1A-negative cells (p < 0.05). There was a time-related recovery of ATP levels in E1A-negative cells (p < 0.05), but not in E1A-positive cells. ATP levels were repressed at 4 h in E1A-negative cells treated with either 2-deoxyglucose or oligomycin (p < 0.05). B, Comparison of the stimulus-induced glycolytic responses of E1A-positive vs E1A-negative cells at 3 h after treatments. Responses to NO and oligomycin were not significantly different for these two cell types (n = 3).

FIGURE 4.

Effects of E1A expression on bioenergetic responses of NO-injured cells. A, ATP levels over time after chemical-NO treatment of E1A-positive (E1A) and E1A-negative (3T3) cells as the percentage of levels in untreated control cells (n = 3, all conditions). ATP levels decreased in both cell types to ∼20% control within 30 min, with a further reduction to ∼10% control in 2-deoxyglucose-treated E1A-negative cells (p < 0.05). There was a time-related recovery of ATP levels in E1A-negative cells (p < 0.05), but not in E1A-positive cells. ATP levels were repressed at 4 h in E1A-negative cells treated with either 2-deoxyglucose or oligomycin (p < 0.05). B, Comparison of the stimulus-induced glycolytic responses of E1A-positive vs E1A-negative cells at 3 h after treatments. Responses to NO and oligomycin were not significantly different for these two cell types (n = 3).

Close modal

The caspase dependence of MMP loss in E1A-positive cells (Fig. 3,B) suggested that the E1A effect on the cellular bioenergetic response might involve enhancement of caspase-induced mitochondrial injury in response to NO. NO induces activation of caspase-2 (49, 50), which is unique among initiator caspases, because it directly injures the mitochondria and does not process other caspase zymogens (17, 18, 51). E1A induces increased basal expression of selected caspases, as a result of E1A activation of E2F transcription factors (52). Because the effects of E1A on caspase-2 function and activation were unknown, we tested the requirement for caspase-2 activity in NO-induced apoptosis of E1A-positive mouse and human fibroblast cells (Fig. 5). Initial experiments were used to define optimal concentrations and times of incubation with the NO generator, DETA-NONOate, to detect induction of caspase-2 activity in lysates of the E1A-positive mouse (250 μM, 4 h) and human (750 uM, 6 h) cells. Under these conditions, E1A-positive cells from both species exhibited increased NO-induced caspase-2 activity when contrasted with E1A-negative cells, which showed little or no change in activity compared with that detected in untreated controls (Fig. 5 A).

FIGURE 5.

E1A enhancement of NO-induced caspase-2 activity, expression, and caspase-2-dependent apoptosis. A, NO (DETA-NONOate)-induced increase in caspase-2 activity in E1A-positive mouse (3T3; 250 μM/4 h) and human (H4; 750 μM/6 h) cells. Caspase-2 activities are represented for each cell type as the percentage of the activities in lysates of the respective, E1A-negative or E1A-positive cells, without NO treatment (n = 3, p < 0.05). B, Caspase-2 overexpression and autoprocessing in E1A-positive NIH-3T3 cells under basal (untreated; NO−) and NO-induced (DETA-NONOate-treated; NO+; 250 μM/4 h) conditions. Upper bands = full length (FL); p31 bands = autoprocessed, intermediate form; p12 bands = fully processed, active form. C, zVDVAD blockade of chemical NO-induced apoptosis of E1A-positive mouse (3T3) and human (H4) cells (n = 3, p < 0.05). D, shRNA-induced knockdown of caspase-2 expression in a clone of E1A-positive 3T3 cells (E1AiC2; left panel) reduced their sensitivity to killing by both NO-producing RAW264.7 macrophages (MP-NO) and DETA-NONOate (250 μM; chem-NO) (right panel; n = 3, p < 0.05).

FIGURE 5.

E1A enhancement of NO-induced caspase-2 activity, expression, and caspase-2-dependent apoptosis. A, NO (DETA-NONOate)-induced increase in caspase-2 activity in E1A-positive mouse (3T3; 250 μM/4 h) and human (H4; 750 μM/6 h) cells. Caspase-2 activities are represented for each cell type as the percentage of the activities in lysates of the respective, E1A-negative or E1A-positive cells, without NO treatment (n = 3, p < 0.05). B, Caspase-2 overexpression and autoprocessing in E1A-positive NIH-3T3 cells under basal (untreated; NO−) and NO-induced (DETA-NONOate-treated; NO+; 250 μM/4 h) conditions. Upper bands = full length (FL); p31 bands = autoprocessed, intermediate form; p12 bands = fully processed, active form. C, zVDVAD blockade of chemical NO-induced apoptosis of E1A-positive mouse (3T3) and human (H4) cells (n = 3, p < 0.05). D, shRNA-induced knockdown of caspase-2 expression in a clone of E1A-positive 3T3 cells (E1AiC2; left panel) reduced their sensitivity to killing by both NO-producing RAW264.7 macrophages (MP-NO) and DETA-NONOate (250 μM; chem-NO) (right panel; n = 3, p < 0.05).

Close modal

In addition to these E1A-related increases in NO-induced caspase-2 activity in cells from both species, immunoblotting analysis revealed increases in caspase-2 protein expression and processing in mouse cells, under both control and NO-induced conditions (Fig. 5,B). There was a 5- to 9-fold increase in the expression of full length and intermediate (p31) forms of caspase-2, respectively, in E1A-positive vs E1A-negative NIH-3T3 cells, in the absence of NO treatment (Fig. 5,B, lane 3 vs lane 1, respectively). In addition, there was a barely detectable level of complete processing of caspase-2 to the p12 form in untreated, E1A-positive cells (Fig. 5,B, lane 3 vs lane 1; 9-fold increase in intensity of the p12 band). NO treatment of E1A-negative NIH-3T3 cells resulted in no detectable change in the pattern of caspase-2 expression or processing (Fig. 5,B, lane 2 vs lane 1, respectively). In contrast, there was increased NO-induced processing of caspase-2 in E1A-positive cells, as evidenced by a 7-fold increase in the fully processed, p12 form, accompanied by a 2-fold decrease in both the full-length and intermediate (p31) forms of caspase-2 in NO-treated (NO+) compared with untreated (NO−), E1A-positive cells (Fig. 5,B, lane 4 vs lane 3, respectively). These patterns of caspase-2 expression and processing are similar to those reported previously for NIH-3T3 cells transiently overexpressing caspase-2 (53). There was no consistent change in basal caspase-2 protein expression detected in E1A-positive human cells, when compared with E1A-negative control cells (data not shown). The observations that E1A expression in human fibroblasts induces increased caspase-2 activity in response to NO stimulation (Fig. 5 A) in the absence of an E1A-induced increase in caspase-2 protein expression suggests that these may be two independent E1A activities.

To investigate the requirement for E1A-enhanced caspase-2 activation in NO-induced apoptosis, E1A-positive mouse and human cells were treated, before injury, with the caspase-2-selective, cell permeable pentapeptide inhibitor, zVDVAD-fmk (54) (Fig. 5,C). This treatment protected E1A-positive cells from NO-induced apoptosis. Complementary experiments were done using a caspase-2 knockdown, E1A-positive cell line, E1AiC2, created using shRNA technology (Fig. 5,D). These cells retained normal E1A expression but had barely detectable caspase-2 expression (Fig. 5,D, left panel). The specificity of caspase-2 repression in these cells was confirmed by their continued expression of caspase-3 and caspase-9 (Fig. 5,D, left panel). Caspase-2 knockdown significantly reduced E1AiC2 sensitivity to both macrophage killing and chemical-NO-induced apoptosis (Fig. 5 D, right panel). These complementary studies with zVDVAD inhibition of caspase-2 and shRNA knockdown of caspase-2 expression indicate that caspase-2 activity is required for E1A-induced enhancement of cellular sensitivity to NO-induced apoptosis, whether mediated by activated macrophages or by injury from chemically generated NO that is independent of any other macrophage cytotoxic mechanisms.

These data indicate that E1A enhancement of NO-triggered, caspase-2-mediated mitochondrial injury is a key initiating event in E1A-induced tumor cell sensitization to killing by macrophage-NO. These results define a new pathway through which E1A sensitizes cells to apoptotic injuries. E1A induces an NF-κB-independent, gain-of-function effect, resulting in increased caspase-2 activation, caspase-dependent blockade of mitochondrial respiration recovery after NO injury and increased mitochondria-triggered “intrinsic” apoptosis in response to NO. This E1A activity complements the previously reported E1A-induced loss-of-function effect—repression of the NF-κB activation response—that sensitizes cells to death-receptor-triggered “extrinsic” apoptosis (20).

In our experiments, NO-induced comparable, initial inhibition of respiration of E1A-negative and E1A-positive cells (Fig. 4,A). The E1A-related difference in the response to NO involved the cellular response that followed this initial repression of respiration. E1A-negative cells exhibited a glycolysis-dependent recovery of mitochondrial function after NO injury (Fig. 4) and resumed normal growth after a period of cytostasis (data not shown). This is comparable to what has been reported for macrophage-NO and chemical-NO injury of various types of tumor cells (6, 43). In contrast, E1A-positive cells never recovered from NO-induced mitochondrial injury but proceeded to apoptosis, despite unlimited glucose and an intact, compensatory glycolytic response (Fig. 3 and 4 B). These results are consistent with the reported relationship between the cellular bioenergetic response to NO and survival and suggest the existence of a caspase-dependent mechanism through which E1A represses the ability of mitochondria to benefit from the MMP-restoring effect of NO-induced compensatory glycolysis (40, 46).

These data suggest that E1A-enhancement of NO-induced caspase-2 activation may be the direct cause of irreversible mitochondrial injury, which then triggers apoptosis. Caspase-2 is unique among initiator caspases (17, 18, 55, 56). It can permeabilize mitochondria, uncouple mitochondrial respiration, and directly trigger release of mitochondrial apoptotic factors. These caspase-2 effects can be independent of Bid cleavage, do not require caspase-2 cleavage of other caspase zymogens, do not require expression of either Bax or Bak, and are unaffected by Bcl-2 overexpression. The E1A-related caspase-2 activation in our assays, detected within the few first hours after NO injury (Fig. 5,A), was temporally associated with the failure of recovery of mitochondrial respiration of E1A-positive cells (Fig. 4). In contrast, respiration recovered quickly in E1A-negative cells, most of which survived NO injury and resumed normal growth, after a period of cytostasis.

E1A expression was observed to induce increased caspase-2 activation in both mouse and human cells following NO stimulation (Fig. 5,A). However, we could only detect an E1A-related increase in caspase-2 protein expression in mouse fibroblasts (Fig. 5 B) and not in human H4 fibroblasts. It has been shown previously that E1A induces overexpression of other caspase proteins (i.e., caspases 3, 7, 8, and 9) in primary mouse fibroblasts and a different human fibroblastic cell line (IMR-90) through the Rb-binding and E2F-enhancing activities of E1A (52). Therefore, whereas one mechanism by which E1A might enhance caspase-2-dependent cellular activity is induction of increased caspase zymogen expression, our data suggest the existence of another mechanism by which E1A enhances the caspase-2 activation response, independently of effects on caspase protein expression. The question of whether these two, possibly different activities, map to the same E1A domain and cell protein binding activity is the focus of ongoing study.

In addition to its cell sensitizing effect for macrophage-NO injury, E1A enhancement of caspase-2-mediated triggering of the intrinsic, mitochondria-mediated, apoptosis pathway might also be mechanistically involved in E1A-induced sensitization of tumor cells to killing by other proapoptotic injuries that trigger intrinsic apoptosis, such as chemotherapeutic drugs and irradiation (reviewed in Refs. 57 and 58). This E1A effect on caspase-2 activation might also partly explain the increased sensitivity of E1A-expressing tumor cells to perforin/granzyme B-mediated killing by cytolytic lymphocytes (25, 59), because granzyme B can activate caspase-2 (60, 61) and kill tumor cells in a caspase-2-dependent manner (62). Further studies of the molecular mechanisms through which E1A enhances caspase-2 activation in response to diverse proapoptotic stimuli might provide a basis for development of strategies to optimize the use of E1A-expressing, oncolytic adenoviruses for adjuvant therapy of cancer (63). Because most E1A effects are mediated by modulation of cellular gene transcription, studies of the cellular mechanisms through which E1A enhances caspase-2 activation in response to proapoptotic stimuli might also provide a better understanding of the regulatory control of the activation of this initiator caspase.

We thank Sharon Ryan for technical support.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by the James A. and Marion C. Grant Fund and by Public Health Service Grants CA86727 (to J.L.C.) and CA76491 (to J.M.R.). J.R.R. is the James A. and Marion C. Grant Fellow in Immunology and Infectious Diseases.

3

Abbreviations used in this paper: E1A, early region 1A oncogene of adenovirus serotype 5; PI, propidium iodide; MMP, mitochondrial membrane potential; TMRE, tetramethylrhodamine ethyl ester perchlorate; shRNA, short hairpin RNA; TR, TNF-resistant, E1A-positive cell variants.

1
Ben-Israel, H., T. Kleinberger.
2002
. Adenovirus and cell cycle control.
Front. Biosci.
7
:
d1369
-d1395.
2
Ralston, R., M. Bishop.
1983
. The protein products of the myc and myb oncogenes and adenovirus E1a are structurally related.
Nature
306
:
803
-806.
3
Felsani, A., A. M. Mileo, M. G. Paggi.
2006
. Retinoblastoma family proteins as key targets of the small DNA virus oncoproteins.
Oncogene
25
:
5277
-5285.
4
Cook, J. L., J. M. Routes.
2005
. Adenovirus E1A gene-induced tumor cell rejection through cellular sensitization to immune and nonimmune apoptotic injuries.
Front. Biosci.
10
:
1396
-1414.
5
Chu, R. L., D. E. Post, F. R. Khuri, E. G. Van Meir.
2004
. Use of replicating oncolytic adenoviruses in combination therapy for cancer.
Clin. Cancer Res.
10
:
5299
-5312.
6
Granger, D. L., R. R. Taintor, J. L. Cook, J. B. Hibbs, Jr.
1980
. Injury of neoplastic cells by murine macrophages leads to inhibition of mitochondrial respiration.
J. Clin. Invest.
65
:
357
-370.
7
Cook, J. L., A. M. Lewis, Jr.
1987
. Immunological surveillance against DNA virus-transformed cells: correlations between natural killer cell cytolytic competence and tumor susceptibility of athymic rodents.
J. Virol.
61
:
2155
-2161.
8
Miura, T. A., K. Morris, S. Ryan, J. L. Cook, J. M. Routes.
2003
. Adenovirus E1A, not human papillomavirus E7, sensitizes tumor cells to lysis by macrophages through nitric oxide- and TNF-α-dependent mechanisms despite up-regulation of 70-kDa heat shock protein.
J. Immunol.
170
:
4119
-4126.
9
Umansky, V., S. P. Hehner, A. Dumont, T. G. Hofmann, V. Schirrmacher, W. Droge, M. L. Schmitz.
1998
. Co-stimulatory effect of nitric oxide on endothelial NF-κB implies a physiological self-amplifying mechanism.
Eur. J. Immunol.
28
:
2276
-2282.
10
Connelly, L., M. Palacios-Callender, C. Ameixa, S. Moncada, A. J. Hobbs.
2001
. Biphasic regulation of NF-κB activity underlies the pro- and anti-inflammatory actions of nitric oxide.
J. Immunol.
166
:
3873
-3881.
11
Kim, S. J., J. S. Chun.
2003
. Protein kinase C α and ζ regulate nitric oxide-induced NF-κB activation that mediates cyclooxygenase-2 expression and apoptosis but not dedifferentiation in articular chondrocytes.
Biochem. Biophys. Res. Commun.
303
:
206
-211.
12
Chun, K. S., H. H. Cha, J. W. Shin, H. K. Na, K. K. Park, W. Y. Chung, Y. J. Surh.
2004
. Nitric oxide induces expression of cyclooxygenase-2 in mouse skin through activation of NF-κB.
Carcinogenesis
25
:
445
-454.
13
Clancy, R. M., P. F. Gomez, S. B. Abramson.
2004
. Nitric oxide sustains nuclear factor κB activation in cytokine-stimulated chondrocytes.
Osteoarthritis Cartilage
12
:
552
-558.
14
D'Acquisto, F., F. de Cristofaro, M. C. Maiuri, G. Tajana, R. Carnuccio.
2001
. Protective role of nuclear factor κB against nitric oxide-induced apoptosis in J774 macrophages.
Cell Death Differ.
8
:
144
-151.
15
Relic, B., M. Bentires-Alj, C. Ribbens, N. Franchimont, P. A. Guerne, V. Benoit, M. P. Merville, V. Bours, M. G. Malaise.
2002
. TNF-α protects human primary articular chondrocytes from nitric oxide-induced apoptosis via nuclear factor-κB.
Lab. Invest.
82
:
1661
-1672.
16
Feng, Z. W., V. Tan, K. S. Khoo, K. J. Leck.
2003
. NF-κB plays a protective role in nitric oxide-induced neuronal apoptosis.
Ann. Acad. Med. Singapore
32
:
S30
-S31.
17
Guo, Y., S. M. Srinivasula, A. Druilhe, T. Fernandes-Alnemri, E. S. Alnemri.
2002
. Caspase-2 induces apoptosis by releasing proapoptotic proteins from mitochondria.
J. Biol. Chem.
277
:
13430
-13437.
18
Robertson, J. D., M. Enoksson, M. Suomela, B. Zhivotovsky, S. Orrenius.
2002
. Caspase-2 acts upstream of mitochondria to promote cytochrome c release during etoposide-induced apoptosis.
J. Biol. Chem.
277
:
29803
-29809.
19
Shao, R., M. C. Hu, B. P. Zhou, S. Y. Lin, P. J. Chiao, R. H. von Lindern, B. Spohn, M. C. Hung.
1999
. E1A sensitizes cells to tumor necrosis factor-induced apoptosis through inhibition of IκB kinases and nuclear factor κB activities.
J. Biol. Chem.
274
:
21495
-21498.
20
Cook, J. L., T. A. Walker, G. S. Worthen, J. R. Radke.
2002
. Role of the E1A Rb-binding domain in repression of the NF-κB-dependent defense against tumor necrosis factor-α.
Proc. Natl. Acad. Sci. USA
99
:
9966
-9971.
21
Routes, J. M., S. Ryan, A. Clase, T. Miura, A. Kuhl, T. A. Potter, J. L. Cook.
2000
. Adenovirus E1A oncogene expression in tumor cells enhances killing by TNF-related apoptosis-inducing ligand (TRAIL).
J. Immunol.
165
:
4522
-4527.
22
Hu, B., H. Zhu, S. Qiu, Y. Su, W. Ling, W. Xiao, Y. Qi.
2004
. Enhanced TRAIL sensitivity by E1A expression in human cancer and normal cell lines: inhibition by adenovirus E1B19K and E3 proteins.
Biochem. Biophys. Res. Commun.
325
:
1153
-1162.
23
Shao, R., D. F. Lee, Y. Wen, Y. Ding, W. Xia, B. Ping, H. Yagita, B. Spohn, M. C. Hung.
2005
. E1A sensitizes cancer cells to TRAIL-induced apoptosis through enhancement of caspase activation.
Mol. Cancer Res.
3
:
219
-226.
24
Ye, X., Q. Lu, Y. Zhao, Z. Ren, X. W. Ren, Q. H. Qiu, Y. Tong, M. Liang, F. Hu, H. Z. Chen.
2005
. Conditionally replicative adenovirus vector carrying TRAIL gene for enhanced oncolysis of human hepatocellular carcinoma.
Int. J. Mol. Med.
16
:
1179
-1184.
25
Cook, J. L., T. A. Potter, D. Bellgrau, B. A. Routes.
1996
. E1A oncogene expression in target cells induces cytolytic susceptibility at a post-recognition stage in the interaction with killer lymphocytes.
Oncogene
13
:
833
-842.
26
Cook, J. L., D. L. May, B. A. Wilson, B. Holskin, M. J. Chen, D. Shalloway, T. A. Walker.
1989
. Role of tumor necrosis factor-α in E1A oncogene-induced susceptibility of neoplastic cells to lysis by natural killer cells and activated macrophages.
J. Immunol.
142
:
4527
-4534.
27
Frisch, S. M..
1991
. Antioncogenic effect of adenovirus E1A in human tumor cells.
Proc. Natl. Acad. Sci. USA
88
:
9077
-9081.
28
Bhattacharya, D., D. U. Lee, W. C. Sha.
2002
. Regulation of Ig class switch recombination by NF-κB: retroviral expression of RelB in activated B cells inhibits switching to IgG1, but not to IgE.
Int. Immunol.
14
:
983
-991.
29
Mitchell, T. C., D. Hildeman, R. M. Kedl, T. K. Teague, B. C. Schaefer, J. White, Y. Zhu, J. Kappler, P. Marrack.
2001
. Immunological adjuvants promote activated T cell survival via induction of Bcl-3.
Nat. Immunol.
2
:
397
-402.
30
Naviaux, R. K., E. Costanzi, M. Haas, I. M. Verma.
1996
. The pCL vector system: rapid production of helper-free, high-titer, recombinant retroviruses.
J. Virol.
70
:
5701
-5705.
31
Cook, J. L., J. B. Hibbs, Jr, A. M. Lewis, Jr.
1980
. Resistance of simian virus 40-transformed hamster cells to the cytolytic effect of activated macrophages: a possible factor in species-specific viral oncogenicity.
Proc. Natl. Acad. Sci. USA
77
:
6773
-6777.
32
Ashcroft, S. J., L. C. Weerasinghe, J. M. Bassett, P. J. Randle.
1972
. The pentose cycle and insulin release in mouse pancreatic islets.
Biochem. J.
126
:
525
-532.
33
Degenhardt, K., R. Sundararajan, T. Lindsten, C. Thompson, E. White.
2002
. Bax and Bak independently promote cytochrome C release from mitochondria.
J. Biol. Chem.
277
:
14127
-14134.
34
Kucharczak, J., M. J. Simmons, Y. Fan, C. Gelinas.
2003
. To be, or not to be: NF-κB is the answer: role of Rel/NF-κB in the regulation of apoptosis.
Oncogene
22
:
8961
-8982.
35
Borutaite, V., G. C. Brown.
2003
. Nitric oxide induces apoptosis via hydrogen peroxide, but necrosis via energy and thiol depletion.
Free Radical Biol. Med.
35
:
1457
-1468.
36
Li, Y., W. Zhang, L. L. Mantell, J. A. Kazzaz, A. M. Fein, S. Horowitz.
1997
. Nuclear factor-κB is activated by hyperoxia but does not protect from cell death.
J. Biol. Chem.
272
:
20646
-20649.
37
Ma, Y., V. Temkin, H. Liu, R. M. Pope.
2005
. NF-κB protects macrophages from lipopolysaccharide-induced cell death: the role of caspase 8 and receptor-interacting protein.
J. Biol. Chem.
280
:
41827
-41834.
38
Song, J. M., R. H. Lee, J. S. Jung.
2005
. Roles of NF-κB and bcl-2 in two differential modes of cell death of mouse cortical collecting duct cells.
Kidney Blood Press. Res.
28
:
101
-110.
39
Darzynkiewicz, Z., S. Bruno, G. Del Bino, W. Gorczyca, M. A. Hotz, P. Lassota, F. Traganos.
1992
. Features of apoptotic cells measured by flow cytometry.
Cytometry
13
:
795
-808.
40
Moncada, S., J. D. Erusalimsky.
2002
. Does nitric oxide modulate mitochondrial energy generation and apoptosis?.
Nat. Rev. Mol. Cell Biol.
3
:
214
-220.
41
Vander Heiden, M. G., N. S. Chandel, E. K. Williamson, P. T. Schumacker, C. B. Thompson.
1997
. Bcl-xL regulates the membrane potential and volume homeostasis of mitochondria.
Cell
91
:
627
-637.
42
Gao, W., Y. Pu, K. Q. Luo, D. C. Chang.
2001
. Temporal relationship between cytochrome c release and mitochondrial swelling during UV-induced apoptosis in living HeLa cells.
J. Cell Sci.
114
:
2855
-2862.
43
Stuehr, D. J., C. F. Nathan.
1989
. Nitric oxide. A macrophage product responsible for cytostasis and respiratory inhibition in tumor target cells.
J. Exp. Med.
169
:
1543
-1555.
44
Cleeter, M. W., J. M. Cooper, V. M. Darley-Usmar, S. Moncada, A. H. Schapira.
1994
. Reversible inhibition of cytochrome c oxidase, the terminal enzyme of the mitochondrial respiratory chain, by nitric oxide. Implications for neurodegenerative diseases.
FEBS Lett.
345
:
50
-54.
45
Brown, G. C., C. E. Cooper.
1994
. Nanomolar concentrations of nitric oxide reversibly inhibit synaptosomal respiration by competing with oxygen at cytochrome oxidase.
FEBS Lett.
356
:
295
-298.
46
Beltran, B., A. Mathur, M. R. Duchen, J. D. Erusalimsky, S. Moncada.
2000
. The effect of nitric oxide on cell respiration: A key to understanding its role in cell survival or death.
Proc. Natl. Acad. Sci. USA
97
:
14602
-14607.
47
Leist, M., B. Single, A. F. Castoldi, S. Kuhnle, P. Nicotera.
1997
. Intracellular adenosine triphosphate (ATP) concentration: a switch in the decision between apoptosis and necrosis.
J. Exp. Med.
185
:
1481
-1486.
48
Lemasters, J. J., T. Qian, C. A. Bradham, D. A. Brenner, W. E. Cascio, L. C. Trost, Y. Nishimura, A. L. Nieminen, B. Herman.
1999
. Mitochondrial dysfunction in the pathogenesis of necrotic and apoptotic cell death.
J. Bioenerg. Biomembr.
31
:
305
-319.
49
Uehara, T., Y. Kikuchi, Y. Nomura.
1999
. Caspase activation accompanying cytochrome c release from mitochondria is possibly involved in nitric oxide-induced neuronal apoptosis in SH-SY5Y cells.
J. Neurochem.
72
:
196
-205.
50
Moriya, R., T. Uehara, Y. Nomura.
2000
. Mechanism of nitric oxide-induced apoptosis in human neuroblastoma SH-SY5Y cells.
FEBS Lett.
484
:
253
-260.
51
Enoksson, M., J. D. Robertson, V. Gogvadze, P. Bu, A. Kropotov, B. Zhivotovsky, S. Orrenius.
2004
. Caspase-2 permeabilizes the outer mitochondrial membrane and disrupts the binding of cytochrome c to anionic phospholipids.
J. Biol. Chem.
279
:
49575
-49578.
52
Nahle, Z., J. Polakoff, R. V. Davuluri, M. E. McCurrach, M. D. Jacobson, M. Narita, M. Q. Zhang, Y. Lazebnik, D. Bar-Sagi, S. W. Lowe.
2002
. Direct coupling of the cell cycle and cell death machinery by E2F.
Nat. Cell. Biol.
4
:
859
-864.
53
Butt, A. J., N. L. Harvey, G. Parasivam, S. Kumar.
1998
. Dimerization and autoprocessing of the Nedd2 (caspase-2) precursor requires both the prodomain and the carboxyl-terminal regions.
J. Biol. Chem.
273
:
6763
-6768.
54
Talanian, R. V., C. Quinlan, S. Trautz, M. C. Hackett, J. A. Mankovich, D. Banach, T. Ghayur, K. D. Brady, W. W. Wong.
1997
. Substrate specificities of caspase family proteases.
J. Biol. Chem.
272
:
9677
-9682.
55
Lassus, P., X. Opitz-Araya, Y. Lazebnik.
2002
. Requirement for caspase-2 in stress-induced apoptosis before mitochondrial permeabilization.
Science
297
:
1352
-1354.
56
Paroni, G., C. Henderson, C. Schneider, C. Brancolini.
2002
. Caspase-2 can trigger cytochrome C release and apoptosis from the nucleus.
J. Biol. Chem.
277
:
15147
-15161.
57
Aghi, M., R. L. Martuza.
2005
. Oncolytic viral therapies: the clinical experience.
Oncogene
24
:
7802
-7816.
58
Dilley, J., S. Reddy, D. Ko, N. Nguyen, G. Rojas, P. Working, D. C. Yu.
2005
. Oncolytic adenovirus CG7870 in combination with radiation demonstrates synergistic enhancements of antitumor efficacy without loss of specificity.
Cancer Gene Ther.
12
:
715
-722.
59
Klefstrom, J., P. E. Kovanen, K. Somersalo, A. O. Hueber, T. Littlewood, G. I. Evan, A. H. Greenberg, E. Saksela, T. Timonen, K. Alitalo.
1999
. c-Myc and E1A induced cellular sensitivity to activated NK cells involves cytotoxic granules as death effectors.
Oncogene
18
:
2181
-2188.
60
Harvey, N. L., J. A. Trapani, T. Fernandes-Alnemri, G. Litwack, E. S. Alnemri, S. Kumar.
1996
. Processing of the Nedd2 precursor by ICE-like proteases and granzyme B.
Genes Cells
1
:
673
-685.
61
Li, H., L. Bergeron, V. Cryns, M. S. Pasternack, H. Zhu, L. Shi, A. Greenberg, J. Yuan.
1997
. Activation of caspase-2 in apoptosis.
J. Biol. Chem.
272
:
21010
-21017.
62
Bergeron, L., G. I. Perez, G. Macdonald, L. Shi, Y. Sun, A. Jurisicova, S. Varmuza, K. E. Latham, J. A. Flaws, J. C. Salter, et al
1998
. Defects in regulation of apoptosis in caspase-2-deficient mice.
Genes Dev.
12
:
1304
-1314.
63
Dobbelstein, M..
2004
. Replicating adenoviruses in cancer therapy.
Curr. Top Microbiol. Immunol.
273
:
291
-334.