Selective unresponsiveness to the inhibition of p38 MAPK activation by cAMP helps L929 fibroblastoma cells escape TNF-α-induced cell death

Forskolin, prostaglandin E2 (PGE₂), epinephrine, propidium iodide (PI), and actinomycin D were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Antibodies against phospho-JNK, JNK, phospho-p38, phospho-CREB, CREB, and c-FLIP_L were from Cell Signaling Technology (Beverly, MA, USA). Antibodies against p38, DLC, actin, and MKP-1 were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Mouse TNF-α was purchased from R&D Systems (Minneapolis, MN, USA). D-JNKi1 was purchased from BioMol (Plymouth Meeting, PA, USA). SB203580 was from Calbiochem (San Diego, CA, USA). 6-MB-cAMP was from Biolog (Hayward CA, USA). ECL chemiluminescence kit was obtained from Amersham (Arlington Heights, IL, USA).

L929 cells were grown in Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum, 2 mM glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin. Small interfering RNAs (siRNAs) that target murine CREB were designed based on nucleotides 1084 to 1102 (#1) and 749 to 767 (#2) relative to the translation start site, respectively, and purchased from Dharmacon (Lafayette, CO, USA). pcDNA3.1 Xpress-DLC has been described previously [14]. Transfection was done with Amaxa nucleofection kit V (VCA-1003, program T-20, Gaithersburg, MD, USA), according to the manufacturer's protocol.

Intracellular cAMP was determined in L929 cells using the cAMP enzyme immunoassay kit purchased from Cayman Chemical (Ann Arbor, MI, USA). Samples were prepared exactly as described by the manufacturer.

Immunoblotting analysis was done as previously described [26]. Briefly, adherent cells were washed with PBS and harvested with a cell scraper (Costars, Cambridge, MA, USA) in ice-cold lysis buffer (0.5% NP-40, 20 mM Tris-Cl, pH 7.6, 250 mM NaCl, 3 mM EDTA, 3 mM EGTA, 1 mM sodium orthovanadate, 1 mM DTT, 10 mM PNPP, 10 μg/ml aprotinin). Cell lysates were resolved by SDS-PAGE before transferring to nitrocellulose membranes. Nitrocellulose membranes were then incubated with 5% (w/v) nonfat dry milk in washing buffer (20 mM Tris-Cl, pH 7.6, 150 mM NaCl, and 0.1% Tween 20) for 1 h at 37°C to block nonspecific protein binding. Primary antibodies (1:1000) were diluted in washing buffer containing 3% BSA and applied to the membranes for overnight at 4°C. After extensive washing, the membranes were incubated with peroxidase-conjugated antibodies for 1 h at room temperature and washed again. Immunoreactive bands were visualized with the ECL chemiluminescence kit.

Cells were harvested by trypsin digestion. Dual staining with FITC-conjugated Annexin V and PI was carried out to detect the induction of apoptotic cell death. Cells were washed with PBS and resuspended in 200 μL of HEPES buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 1.8 mM CaCl₂) containing 1 μg/ml Annexin V-FITC and 5 μg/ml PI (Annexin V/PI staining kit, BD Biosciences Pharmingen, San Diego, CA, USA). Following incubation for 15 min at room temperature, cells were analyzed by flow cytometry (FACSCalibur; BD Biosciences, Franklin Lakes, NJ, USA). Annexin V-positive/PI-negative cells were apoptotic, whereas Annexin V/PI double positive cells were necrotic. A simple way to detect necrosis is PI staining. After washing with PBS, the pellet was stained with PI at a concentration of 5 μg/ml in PBS and incubated at room temperature in the dark for 5 min, which was followed by flow cytometry.

The data were shown as mean ± standard deviations (SD). The Student's t-test was used to compare the difference between the two groups. The difference was considered statistically significant when p < 0.05.

Our previous data have shown that cAMP promotes TNF-α-induced cell death in fibroblasts [14]. However, it remains unknown whether the same regulation is also applicable to fibroblastoma cells. For this purpose, L929 fibroblastoma cells were pretreated with the most widely used cAMP elevation agent forskolin for 30 min [27], followed by stimulation with or without 10 ng/ml TNF-α for 24 h. Surprisingly, cell death assay with Annexin-V/PI double staining revealed that forskolin significantly suppressed TNF-α-induced cell death in L929 fibroblastoma cells (Figure 1). TNF-α induced marginal cell death in fibroblasts [14] (data not shown). However, TNF-α induced massive cell death in L929 cells (Figure 1). The majority of L929 cells undergoing cell death were Annexin-V/PI double positive (Figure 1), consistent with the previous finding that TNF-α treatment of L929 cells leads to a caspase-independent cell death with necrotic phenotype [28, 29]. A Simple way to detect necrosis is PI staining [28, 29]. Therefore, TNF-α-induced cell death in L929 cells can be simply analyzed with PI staining.

To make sure that forskolin suppresses TNF-α-induced cell death in L929 cells because of the elevation of cAMP, the cells were pretreated with physiologically relevant cAMP inducers PGE₂ and epinephrine [3, 30]. PI staining revealed that under the conditions that pharmacological agent forskolin significantly suppressed TNF-α-induced cell death in L929 cells, physiologically relevant cAMP inducers PGE₂ and epinephrine exhibited similar effects (Figure 2A). Furthermore, 6-MB-cAMP, a site-selective activator of PKA [14], also suppressed TNF-α-induced cell death in L929 cells (Figure 2A). All these agents led to increased intracellular cAMP in a time-dependent manner (Figure 2B). Moreover, the different ability of these agents to increase the concentration of intracellular cAMP was correlated with the extent these cAMP elevators suppressed TNF-α-induced cell death in L929 cells (Figure 2A). Taken together, these data suggest that cAMP-PKA pathway suppresses TNF-α-induced cell death in L929 cells.

Our previous data suggest that cAMP regulates TNF-α-induced cell death in fibroblasts via CREB-mediated transcription [14]. Consistent with this notion, forskolin-, PGE₂-, epinephrine-, and 6-MB-cAMP-induced CREB phosphorylation at Ser133 (Figure 3A), which is required for CREB activation [31‐33], corresponded to the extent these cAMP elevators activated intracellular cAMP (Figure 2B) and suppressed TNF-α-induced cell death in L929 cells (Figure 2A). To further analyze the mechanism by which cAMP suppresses TNF-α-induced cell death in L929 cells, the cells were pretreated with or without forskolin, followed by treatment with TNF-α in the presence or absence of the RNA synthesis inhibitor actinomycin D. Because 10 ng/ml TNF-α rapidly induced more than 70% cell death in the presence of actinomycin D (data not shown), the concentration of TNF-α was titrated down. 2 ng/ml TNF-α treatment for 12 h led to only 7% cell death in the absence of actinomycin D, which was significantly suppressed by forskolin (Figure 3B). However, in the presence of actinomycin D the same dose of TNF-α resulted in about 35% cell death, though actinomycin D itself showed no detectable effect on the survival of L929 cells (Figure 3B). These data are consistent with the previous finding in the literature that blockade of de novo protein synthesis significantly enhances TNF-α-induced cell death. [34, 35] The suppression of TNF-α-induced cell death by forskolin was abolished by actinomycin D (Figure 3B). These data suggest that cAMP suppresses TNF-α-induced cell death in L929 cells in a transcription-dependent manner.

The cAMP pathway activates several transcription factors, including CREB, CREM, and ATF-1 [31‐33]. Among them, CREB is the major effector of the cAMP pathway [31‐33]. We used CREB siRNAs to test whether CREB mediates the suppression by cAMP of TNF-α-induced cell death in L929 cells. Immunoblotting analysis revealed that CREB siRNAs specifically inhibited CREB expression and abolished the basal and forskolin-stimulated CREB phosphorylation (Figure 3C). Transfection of L929 cells with CREB siRNAs but not the negative control siRNA reversed the suppression of TNF-α-induced cell death by forskolin (Figure 3D). Moreover, CREB siRNA #2, which was more efficient than CREB siRNA #1, led to increased sensitivity of L929 cells to TNF-α-induced cell death. Consistently, the suppression by forskolin of TNF-α-induced cell death was also abolished by ACREB, a specific CREB inhibitor that utilizes its acidic amphipathic extension to prevent the basic region of CREB from binding to DNA [36] (data not shown). Taken together, these data suggest that CREB plays a pro-survival role in TNF-α-induced cell death in L929 cells, and cAMP suppresses TNF-α-induced cell death in L929 cells via CREB-mediated transcription.

Our recent work has revealed that, at least in fibroblasts, the crosstalk between cAMP-PKA-CREB pathway and either JNK or p38 pathway plays a key role in the regulation of cell death by cAMP [14, 15]. Now that cAMP suppresses TNF-α-induced cell death in L929 fibroblastoma cells via CREB-mediated transcription, it is of importance to investigate the effects of cAMP on the activation of JNK and p38 in this cell context. For this purpose, L929 cells were pretreated with or without forskolin for various periods of times and then stimulated with TNF-α for 15 min or left untreated. Immunoblotting analysis revealed that TNF-α-induced phosphorylation of JNK at Thr183 and Tyr185, which is required for JNK activation [20], was inhibited by forskolin in a biphasic manner. The inhibition occurred from 30 to 90 min and decreased at 120 min after the pretreatment with forskolin (Figure 4A). The kinetics in the inhibition of JNK activation by forskolin was correlated with its effects on intracellular cAMP (Figure 2B). However, forskolin pretreatment showed no significant effect on TNF-α-induced phosphorylation of p38 at Thr180 and Tyr182, which is required for p38 activation [21, 22], under the same conditions (Figure 4A). Similar results were obtained when L929 cells were pretreated with PGE₂, epinephrine, and 6-MB-cAMP (Figure 4B). The different ability of these agents to inhibit the activation of JNK was correlated with the extent these cAMP elevators increased intracellular cAMP (Figure 2B), activated CREB (Figure 3A), and suppressed TNF-α-induced cell death in L929 cells (Figure 2A). cAMP uncoupled JNK activation and p38 activation not only in response to TNF-α, but also in response to UV (Figure 4C). Furthermore, cAMP inhibited the basal level of JNK phosphorylation, but not p38 phosphorylation, in L929 cells (Figure 4D). Taken together, these data suggest that cAMP uncouples JNK activation and p38 activation in L929 cells.

Blockade of total JNK activity has been shown to result in impaired cell death in response to TNF-α in various types of cells [23‐25]. Because cAMP significantly inhibited TNF-α-induced JNK activation in L929 cells, it is of interest to know the role of JNK in TNF-α-induced cell death in L929 cells. For this purpose, L929 cells were pretreated with or without the selective JNK inhibitor D-JNKi1 (10 μM) [37], and then stimulated with TNF-α or left untreated. Cell death assay revealed that D-JNKi1 significantly suppressed TNF-α-induced cell death in L929 cells (Figure 5), suggesting that JNK contributes to TNF-α-induced cell death in L929 cells. Thus, our data suggest cAMP suppresses TNF-α-induced cell death in L929 cells via, at least partially, inhibition of JNK activity.

Our previous study has shown that cAMP negatively regulates p38 activation, thereby contributing to TNF-α-induced apoptosis in fibroblasts [14]. Now that L929 cells exhibited selective unresponsiveness to p38 inhibition by cAMP and showed impaired cell death in response to TNF-α with cAMP pretreatment, it is possible that the lack of an inhibitory effect of cAMP on the pro-survival activation of p38 by TNF-α helps L929 cells escape TNF-α-induced cell death. To test this scenario, L929 cells were pretreated with or without forskolin, followed by treatment with TNF-α in the presence or absence of the selective p38 inhibitor SB203580 (1 μM) [38, 39]. Cell death assay revealed that 1 μM SB203580 significantly reversed the protection of L929 cells by cAMP from TNF-α-induced cell death (Figure 6). Thus, our data suggest that the inhibition of p38 activation over-rides the pro-survival effects of cAMP in TNF-α-induced cell death. Loss of an inhibitory effect of cAMP on p38 activation might help certain types of tumor cells escape from TNF-α-induced cell death.

Our previous studies have revealed that the major effectors of cAMP-mediated inhibition of JNK or p38 activation are different. The induction of DLC is required for cAMP-mediated inhibition of p38 activation [14], whereas the induction of c-FLIP_L and MKP-1 is required for cAMP-mediated inhibition of JNK activation [15]. Because cAMP uncoupled JNK activation and p38 activation in L929 cells and loss of a cAMP-dependent inhibition of p38 activation might be the key mechanism by which L929 cells escapes TNF-α-induced cell death, it is of great importance to investigate whether DLC induction by cAMP is impaired in L929 cells. For this purpose, L929 cells were treated with forskolin for various periods of times. Immunoblotting analysis revealed that up-regulation of c-FLIP_L and MKP-1 occurred from 30 to 90 min and decreased at 120 min after the pretreatment with forskolin (Figure 7A), which was correlated with the inhibition of JNK activation (Figure 4). However, there was no detectable up-regulation of DLC under the same conditions (Figure 7A). Thus, DLC induction by cAMP is indeed impaired in L929 cells. Ectopic expression of DLC significantly inhibited p38 activation induced by TNF-α either in the presence or absence of forskolin (data not shown). Furthermore, enforced expression of DLC reversed the protection of L929 cells by forskolin from TNF-α-induced cell death (Figure 7B). Taken together, these data suggest that selective unresponsiveness to the inhibition of p38 activation by cAMP might result from impaired induction of DLC.