INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PROGRAMMED CELL DEATH is a highly conserved evolutionary event crucial for normal development and homeostasis. Deregulated cell death has been associated with disease entities such as cancer (36, 42), human immunodeficiency virus infection (29), and more recently cardiovascular disease (22, 38). Notably, apoptosis has been detected in cardiac tissue after ischemia followed by reperfusion (17), oxidative stress injury (15), postinfarction (20), and in patients with end-stage heart failure (37, 41). Because ventricular myocytes retain a limited regenerative potential after birth, the loss of potentially viable cells through an apoptotic process may profoundly influence cardiac structure/function. Although the molecular mechanisms that govern apoptosis in the heart are poorly defined, there is increasing awareness that certain cellular factors can either promote or suppress the cell death process.

Nuclear factor-B (NF-B) is a ubiquitously expressed transcription factor that is regulated by the inhibitor protein B (IB). IB binds to and sequesters NF-B in the cytoplasm, preventing NF-B from translocating to the nucleus. Signal-induced activation of NF-B involves the phosphorylation of IB at serine residues 32 and 36. This leads to the ubiquitination and degradation of IB by the proteasome, allowing NF-B to translocate to the nucleus and affect gene transcription (14). Recently, an anti-apoptotic function for NF-B has been described (5, 50, 52). This is supported by studies in which cells defective for NF-B signaling were found to be more sensitive to proapoptotic signals than normal wild-type cells (49, 50). Furthermore, transgenic mice incapable of NF-B activation die at embryonic day 14.5 from excessive apoptosis and severe liver degeneration (46). Together, these observations support a critical role for NF-B in abrogating death-promoting signals and apoptosis.

Tumor necrosis factor- (TNF-) is a pleiotropic cytokine with diverse biological functions that include cell proliferation, inflammation, and apoptosis. Although TNF- is known to strongly activate NF-B, there is emerging evidence suggesting that TNF- predominately triggers apoptosis in cells that are either deficient or defective for NF-B activation (5, 49, 50). The relative spatial and temporal expression of TNF- in the heart, particularly during heart hypertrophy and end-stage heart failure (cf. Refs. 21, 33), raises the possibility that TNF- directly modulates NF-B activity and the apoptotic process. However, whether NF-B suppresses death-promoting signals mediated by TNF- in ventricular myocytes is unknown and has not been formally tested. Therefore, in the present study, we examined the significance of the NF-B signaling pathway in ventricular myocytes by determining whether a block to NF-B activation would unmask the cytotoxic actions of TNF- and render ventricular myocytes susceptible to apoptosis. In this report, we provide the first direct evidence to support a role for NF-B as an anti-apoptotic factor in ventricular myocytes. Furthermore, our data show that a functional NF-B signaling pathway is crucial for preventing death-promoting signals and apoptosis in ventricular myocytes mediated by TNF-.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell culture and transfection. Neonatal ventricular myocytes were isolated from 2-day-old Sprague-Dawley rat hearts and were submitted to primary culture as described previously (9). After an overnight incubation in DMEM-Ham's nutrient mixture F-12 (1:1), 17 mM HEPES, 3 mM NaHCO₃, 2 mM L-glutamine, 50 µg/ml gentamicin, and 10% fetal bovine serum (FBS), cells were transferred to serum-free medium as previously described (9, 24). Myocytes were infected with recombinant adenoviruses and transfected after removal of viral stocks with NF-B luciferase reporter plasmid in DMEM containing DEAE-dextran as previously described (9). Myocytes were stimulated with 10 ng/ml of human recombinant TNF- (R&D Systems) or 10 µM C₂ ceramide (Sigma Chemical, St. Louis, MO) in serum-free media for 24-72 h. This concentration of ceramide was previously shown to trigger apoptosis of ventricular myocytes (26). Luciferase activity was normalized to -galactosidase activity to control for differences in transfection efficiency and was expressed as relative light units. Data were obtained from at least n = 3 independent myocyte cultures with replicates of three for each condition. Results were compared by Student's t-test, using a significance level of P  0.05.

Recombinant adenoviruses. Adenoviruses were propagated, harvested, purified, and titered from 293 cells as previously reported (23). The cDNA epitope-FLAG-tagged derivative of the IB mutant containing serine-to-alanine substitutions at amino acid positions 32 and 36 was generously provided by D. Ballard (6) and was subcloned into the Hind III/Xba I sites of an adenovirus shuttle plasmid. Recombinant adenovirus was generated by homologous recombination in 293 cells as previously reported (10, 23). Viral infection was controlled for by using the adenovirus designated AdCMV, which contains the cytomegalovirus (CMV) enhancer-promotor without a cDNA insert. Myocyte cultures were infected with 20 plaque-forming units per cell of recombinant adenovirus for 4 h. This titer of virus achieves gene delivery to 95% of neonatal ventricular cells under these conditions (23).

Western blot analysis. For immunodetection of IB and A20 proteins, cardiac myocytes were harvested in buffer containing 0.5% SDS, 150 mM NaCl, and 50 mM Tris · HCl, pH 7.4 (RIPA buffer). Cell lysates (100 µg) were resolved on a 10% SDS-polyacrylamide gel at 140 V for 4 h and were electrophoretically transferred to a polyvinylidine difluoride (PVDF) membrane (Roche Diagnostics). For detection of IB-FLAG-tagged proteins, the PVDF filter was incubated with a rabbit antibody directed toward IB clone C2 (1 µg/ml; Santa Cruz Biotechnology) or a polyclonal antibody directed toward human A20 protein (25). Bound proteins were visualized by chemiluminescence reaction with horseradish peroxidase-conjugated antibodies against mouse or rabbit IgG using enhanced chemiluminescence reagents (Amersham).

Electromobility gel shift assay. Nuclear extracts of cardiac myocytes were prepared as previously described by de Moissac et al. (9) with modifications. Briefly, 3 × 10⁶ cells were pelleted and resuspended in 200 µl of 10 mM HEPES, pH 7.9, 60 mM KCl, 1.0 mM EDTA, 1.0 mM dithiothreitol, protease inhibitors, and 0.3% Nonidet P-40. Cells were allowed to swell on ice for 15 min and then were centrifuged at 1,000 g at 4°C. The supernatant was extracted, and the remaining cell pellet was resuspended in 50 µl of 200 mM HEPES, pH 7.9, 0.4 M NaCl, 1.0 mM EDTA, 1.0 mM EGTA, 1 mM dithiothreitol, and 1 mM phenylmethylsulfonyl fluoride at 4°C for 15 min. The nuclear extract was centrifuged for 5 min at 10,000 g and was stored at 80°C. Analysis of DNA-binding activities by electromobility shift analysis was carried out as previously described using a ³²P-radiolabeled duplex oligonucleotide probe containing NF-B consensus binding sites 5'-AGTTGAGGGGACTTTCGCAGGC-3'. DNA binding reactions (20 µl) were carried out on ice and contained 5 µg nuclear extract, 2 µg double-stranded probe poly(dI-dC) (Pharmacia), 10 µg BSA in 20 mM HEPES, pH 7.9, 5% glycerol, 1 mM EDTA, and 5 mM dithiothreitol. Nuclear-protein complexes were resolved on a 5% polyacrylamide gel in 1× Tris-borate-EDTA (pH 8.0) and were detected by autoradiography.

Viability analysis. Total cell number was determined before and after stimulation with TNF- to ensure that cells did not aberrantly detach from plates and that equivalent cell numbers were available for viability analysis. Myocytes stimulated with TNF- were assessed for viability by staining cells with the vital dyes calcein-acetoxymethyl ester (AM) (2 µM) and ethidium homodimer-1 (2 µM) for 30 min (Molecular Probes, Eugene OR; see Refs. 22 and 24). Cells were washed and mounted on glass slides and visualized using an Olympus AX70 Research microscope equipped with an excitation and emission filter set to simultaneously detect the number of live (green) and dead (red) cells, respectively. The relative number of green vs. red cells was determined from at least 200 cells/condition.

Detection of apoptosis. Nuclear morphology and nucleosomal DNA fragmentation of cardiac nuclei were determined by staining myocytes with Hoechst 33258 dye for nuclear DNA as previously described (22, 24). Replicate cultures using 200 cells for each condition were utilized. Genomic DNA was isolated from ventricular myocytes for nucleosomal DNA fragmentation by gel electrophoresis as previously described (22, 24).

Statistical analysis. Data were obtained from at least n = 3 independent cell cultures with replicates of three for each condition. Results were compared by Student's t-test, using a significance level of P  0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

To establish whether ventricular myocytes are functionally coupled to biological signals that lead to the downstream activation of NF-B by TNF-, ventricular myocytes were transfected with a luciferase reporter gene containing putative binding sites for NF-B and were stimulated with TNF- (9). A 12-fold increase (P < 0.01) in NF-B-dependent gene transcription was observed in the presence of TNF- compared with vehicle-treated control cells (Fig. 1). Moreover, stimulation of myocytes with TNF- resulted in a threefold induction of the endogenous A20 protein, a protein known to be regulated by NF-B (25, 43; Fig. 2). Furthermore, gel shift experiments indicated that NF-B binding activity was increased in myocytes stimulated with TNF- compared with vehicle-treated control cells (Fig. 3, lane 2 vs. lane 3). Moreover, competition binding assays with 100-fold excess probe (lane 7) as well as supershift experiments with antibodies directed toward the p65 subunit of NF-B (lane 6) confirmed that the higher migrating complex contained the p65 subunit of NF-B. Together these findings confirm that ventricular myocytes are functionally coupled to biological signals that link TNF- to NF-B DNA binding and NF-B-dependent gene transcription.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Tumor necrosis factor- (TNF-)-mediated nuclear factor-B (NF-B)-dependent gene activation in postnatal ventricular myocytes. Myocytes were transfected with a luciferase reporter plasmid containing tandem NF-B binding elements and were stimulated with TNF- (10 ng/ml) in the presence and absence of inhibitor protein B (IB) mutant (MT). After 24 h, a 12-fold induction (P < 0.01) of NF-B transcription was observed in the presence of TNF- compared with vehicle-treated cells. In the presence of the IB mutant, TNF--mediated NF-B gene activation was repressed to basal levels. Control (CNTL) myocytes were transfected with the eukaryotic expression vector pcDNA3 lacking the "cis-acting" NF-B response elements. Data are expressed as mean ± SE folds of increase (P < 0.05). Experiments were repeated at least three times with independent culture conditions with 3 replicates for each condition.

View larger version (77K): [in this window] [in a new window]   Fig. 2.   Expression of the endogenous A20 protein in ventricular myocytes. Top: Western blot analysis of A20 in cardiac cell lysate. TNF- results in a threefold increase in the endogenous NF-B target protein A20 compared with vehicle control cells. Induction of A20 by TNF- is prevented by the IB mutant. Bottom: Ponceau-S stained filter to demonstrate equivalent protein loading.

View larger version (65K): [in this window] [in a new window]   Fig. 3.   Electromobility gel shift analysis for NF-B in ventricular myocytes. Equivalent amounts of nuclear extract from ventricular myocytes were prepared after interventions and analyzed for NF-B binding activity with a 32P-labeled oligonucleotide probe containing NF-B binding sites. Lane 1, free probe; lane 2, vehicle-stimulated myocyte cells (CNTL); lane 3, TNF- (10 ng/ml)-stimulated myocytes; lane 4, myocytes expressing the IB mutant (IBMT) and stimulated with TNF-; lane 5, myocytes expressing the IBMT alone; lane 6, supershift analysis of nuclear extract from cells incubated with rabbit antibody directed toward the p65 subunit of NF-B (see METHODS for details); arrow indicates p65 subunit of NF-B; lane 7, competition binding analysis of nuclear extract with 100-fold excess cold probe. Arrow indicates the higher migrating band to be NF-B containing the p65 subunit. NS, not significant.

To formally test whether TNF- is cytotoxic and provokes apoptosis of ventricular myocytes, ventricular myocytes were stimulated with TNF- and stained with the vital dyes calcein-AM and ethidium homodimer-1 to identify the live and dead cells, respectively. As shown in Fig. 4A, myocytes stimulated with TNF- for up to 72 h were indistinguishable from vehicle-treated control cells with respect to cell viability (P = 0.31), indicating that TNF- alone was not cytotoxic to myocytes and did not provoke cell death. Importantly, no significant difference in cell number was observed after stimulating cells with TNF- at any of the time points tested, verifying that cells were not preferentially lost by TNF- stimulation (Fig. 4B). In contrast, ventricular myocytes stimulated with the cell-permeable sphingolipid C₂ ceramide (10 µM), a by-product of TNF- stimulation in ventricular myocytes (39) known to provoke apoptosis at the concentration utilized (26), resulted in widespread cell death compared with vehicle-treated cells or those stimulated with TNF- (Fig. 4A). This substantiates that ceramide, but not TNF-, was toxic to myocytes.

View larger version (33K): [in this window] [in a new window]   Fig. 4.   TNF- provokes widespread cell death in the absence of NF-B activation. A: immunofluorescent staining of ventricular myocytes stimulated for 24-72 h with TNF- (10 ng/ml). Cell viability was determined using calcein-AM (green) and ethidium homodimer-1 (red) to distinguish the number of live vs. dead cells, respectively. TNF- alone had no effect on cell viability compared with vehicle-treated control cells or those infected with a control adenovirus (AdCMV) at any of the time points tested. Ceramide (10 µM) triggers widespread cell death, indicated by the presence of red-staining myocytes. B: cell counts before and after treatment with TNF-. No significant difference in cell number was observed before or after treatment with TNF- at any of the time points tested: 24 h, before vs. after, P = 0.363; 48 h, before vs. after, P = 0.309; 72 h, before vs. after, P = 0.294. C: histogram of cell viability from ventricular myocytes shown in A. Stimulation of myocytes with TNF- alone does not provoke apoptosis of ventricular myocytes compared with vehicle-treated controls, P = 0.31. However, TNF- provoked a 2.2-fold increase in myocyte death (P < 0.001) in myocytes expressing the IB mutant (IBMT) and those defective for NF-B activation.

However, it has recently been shown that the cytotoxic responses to TNF- can be enhanced by agents such as actinomycin D or cycloheximide, which inhibit transcription and translation, respectively (30). The fact that these agents unmask the cytotoxic actions of TNF- suggests that the de novo activation of downstream genes that are cytoprotective are important for preventing TNF--mediated cell death.

In this regard, the transcription factor NF-B has been suggested to be important in preventing death-promoting signals and apoptosis mediated by TNF-. Because NF-B activity is governed by IB, which binds to and sequesters NF-B in the cytoplasm, we determined whether NF-B is necessary for suppressing apoptosis in ventricular myocytes by testing whether a block to NF-B activation with a mutant form of IB would render ventricular myocytes susceptible to TNF--induced cell death. For these experiments, we generated a replication defective adenovirus encoding an IB molecule containing serine-to-alanine point substitutions at amino acid positions 32 and 36, respectively. This renders IB defective for phosphorylation and degradation, thereby preventing signal-induced nuclear activation of NF-B (3, 4). As shown by electromobility gel shift analysis, TNF--mediated NF-B nuclear DNA binding was inhibited to basal levels in cells expressing the IB mutant (Fig. 3, lane 4 vs. lane 3). Moreover, in the presence of the IB mutant, TNF--mediated NF-B-dependent gene transcription was inhibited to levels comparable to vehicle-treated control cells (Fig. 1). Furthermore, TNF--mediated activation of the endogenous A20 protein was also impaired in the presence of the IB mutant (Fig. 2), indicative of impaired NF-B activation.

Together these findings verify that the IB mutant was functionally active in ventricular myocytes in suppressing signal-induced activation of NF-B by TNF-. Importantly, myocytes infected with the control adenovirus were not different from uninfected control cells with respect to viability (Fig. 4A), confirming that adenoviral infection was not toxic to myocytes (24). In contrast, cells expressing the IB mutant and stimulated with TNF- displayed a significant 2.2-fold increase in the incidence of cell death compared with vehicle-treated control cells or those stimulated with TNF- (Fig. 4C; P < 0.001). Moreover, genomic DNA isolated from myocytes expressing the mutant IB and stimulated with TNF- displayed evidence of apoptosis, as demonstrated by an increase in nucleosomal DNA laddering (Fig. 5). Similarly, myocytes defective for NF-B activation and stimulated with TNF- displayed characteristic features of apoptosis by Hoechst 3325 dye compared with control cells (de Moissac and Kirshenbaum, unpublished observation). Together, these findings support our contention that activation of the NF-B signaling pathway is crucial for suppressing death-promoting signals and apoptosis in ventricular myocytes that would otherwise be provoked by TNF-.

View larger version (49K): [in this window] [in a new window]   Fig. 5.   TNF- provokes apoptosis in ventricular myocytes defective for NF-B activation. Stimulation of ventricular myocytes with TNF- alone does not provoke apoptosis of ventricular myocytes. Nucleosomal DNA laddering is observed in myocytes expressing IB mutant (IBMT), in the presence of TNF-.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

To our knowledge, the data provide the first direct evidence for the operation of the NF-B signaling pathway for the suppression of apoptosis induced by TNF- in ventricular myocytes. Furthermore, our data indicate that TNF- alone does not provoke apoptosis in ventricular myocytes that are functionally coupled to the downstream activation of NF-B. A role for TNF- as a modulator of cardiac function has been proposed and substantiated by the relative spatial and temporal expression of this cytokine in the myocardium but notably by its upregulation during mechanical load (47) in surviving myocytes after infarction (35) and end-stage heart failure (31, 48). The unexpected and counterintuitive lack of apoptosis in ventricular myocytes with TNF-, despite evidence elsewhere (11, 26), suggests that TNF- likely activates dual signaling cascades in a cell- and context-specific manner, with one pathway leading to apoptosis while the other pathway, mediated through NF-B, dominates to suppress prodeath signals and apoptosis (5, 34, 51).

The physiological/pathophysiological role of TNF- in the heart is unknown. However, the fact that TNF- does not provoke apoptosis in the presence of a functional NF-B signaling pathway seen here suggests that TNF- may have an alternative role as a stress response factor (19, 32). In this regard, recent in vitro and in vivo studies have shown that TNF- can modulate contractile function (21) and gene expression (12) characteristic of the dilated failing heart (27). Moreover, because elevated TNF- levels have been detected in cardiac pathologies such as allograft rejection (45), postinfarction, end-stage heart failure (47), and viral myocarditis (28), it is tempting to speculate that NF-B contributes to the inflammatory response of these conditions by suppressing or blunting the apoptotic response of cells to TNF- (2, 13). This notion is supported by recent studies in which endothelial cells, key mediators of the inflammatory response, were found to be resistant to TNF-, whereas endothelial cells defective for NF-B activation readily underwent apoptosis provoked by TNF- (1, 40). These observations are consistent with the findings of the present study which demonstrate that interference with signal-induced activation of NF-B unmasks the cytotoxic effects of TNF-, resulting in apoptosis of ventricular myocytes.

The mode by which NF-B suppresses apoptosis is unknown but may be related to the activation of downstream genes that regulate the apoptotic process. This is supported by the fact that several anti-apoptotic factors, including cellular inhibitors of apoptosis (c-IAP), c-IAP1, c-IAP2, (7), A20 (8), and IEX-1L (53), are known transcriptional targets of NF-B. These factors can reportedly block the activation of caspase 8, the proximal caspase in the TNF-/CD95/Fas signaling pathway that propagates apoptotic signals through activation of the death-inducing signaling complex (44).

Whether these factors are present and functionally active in ventricular myocytes is unknown. However, the fact that cell-permeable C₂ ceramide but not TNF- was cytotoxic to myocytes affirms the notion that activation of signaling molecules downstream of TNF- receptor are crucial for suppressing prodeath signals and apoptosis (16). Although our data substantiate a cytoprotective role for NF-B in ventricular myocytes, it must be stated that protection from apoptosis may not be a universal feature of NF-B, since under certain instances such as the case with Sindbis virus, NF-B may trigger rather than prevent apoptosis (18). Furthermore, the role played by NF-B in cardiac disease conditions is unknown and awaits further investigation. Thus whether NF-B operates as a pro- or anti-apoptotic factor may rely on the context of cell type and the ensuing stimulus. Nevertheless, under the conditions tested, our data provide the first direct evidence to support a role for the suppression of TNF--mediated apoptosis in ventricular myocytes by NF-B and highlights the importance of coordinated regulation of NF-B by TNF- to modulate the apoptotic response during disease states. Our current investigations are directed toward elucidating the impact of the NF-B signaling pathway and the downstream effector proteins that regulate apoptosis in cardiac disease conditions.