TNF is implicated in the attenuation of neutrophil constitutive apoptosis during sepsis. Antiapoptotic signaling is mediated principally through the TNF receptor-1 (TNFR-1). In adherent neutrophils, when β-integrin signaling is activated, TNF phosphorylates TNFR-1 and activates prosurvival and antiapoptotic signaling. Previously, we identified the δ-PKC isotype and phosphatidylinositol (PI) 3-kinase as critical regulators of TNF signaling in adherent neutrophils. Both kinases associate with TNFR-1 in response to TNF and are required for TNFR-1 serine phosphorylation, NF-κB activation, and inhibition of apoptosis. The purpose of this study was to examine the role of δ-PKC and PI 3-kinase in the assembly of TNFR-1 signaling complex that regulates NF-κB activation and antiapoptotic signaling. Coimmunoprecipitation studies established that PI 3-kinase, δ-PKC, and TNFR-1 formed a signal complex in response to TNF. δ-PKC recruitment required both δ-PKC and PI 3-kinase activity, whereas PI 3-kinase recruitment was δ-PKC independent, suggesting that PI 3-kinase acts upstream of δ-PKC. An important regulatory step in control of antiapoptotic signaling is the assembly of the TNFR-1-TNFR-1-associated death domain protein (TRADD)-TNFR-associated factor 2 (TRAF2)-receptor interacting protein (RIP) complex that controls NF-κB activation. Inhibition of either δ-PKC or PI 3-kinase decreased TNF-mediated recruitment of RIP and TRAF2 to TNFR-1. In contrast, TRADD recruitment was enhanced. Thus δ-PKC and PI 3-kinase are positive regulators of TNF-mediated association of TRAF2 and RIP with TNFR-1. Conversely, these kinases are negative regulators of TRADD association. These results suggest that δ-PKC and PI 3-kinase regulate TNF antiapoptotic signaling at the level of the TNFR-1 through control of assembly of a TNFR-1-TRADD-RIP-TRAF2 complex.
- tumor necrosis factor receptor-1-associated death domain protein
- receptor interacting protein
- tumor necrosis factor receptor-associated factor 2
- antiapoptotic signaling
human neutrophils are important in host defense against bacterial infections but can also contribute to the tissue damage of inflammation if not appropriately regulated. Mature neutrophils are end-stage cells and have a relatively short life span. Once released into the circulation, neutrophils undergo constitutive apoptosis or programmed cell death. During sepsis and other inflammatory diseases, neutrophil apoptosis is attenuated (21, 32, 34, 44, 46). Enhanced neutrophil survival at the site of inflammation promotes increased bactericidal activity but may also play a role in acute inflammatory damage.
Proinflammatory cytokines, such as TNF, have been implicated in the enhancement of neutrophil activity through activation of proinflammatory signaling and inhibition of apoptosis (2, 12, 19, 32, 34, 36, 40, 44, 46, 48, 51, 63). TNF is a unique cytokine whose signaling pathways are linked to both proapoptotic and antiapoptotic responses (for review, see Refs. 2, 3, 10, 28, 43, 64). A neutrophil's response to TNF is dependent on numerous environmental factors, including input from other signaling pathways. For example, TNF is an incomplete secretagogue and requires cell adherence to matrix proteins and engagement of β-integrins to trigger superoxide anion generation, degranulation, and activation of phosphatidylinositol 3-kinase (PI 3-kinase) (39, 49, 50). This binary signaling triggered by binding of TNF and engagement of β-integrins also results in activation of NF-κB and inhibition of neutrophil apoptosis. Our laboratory's previous studies demonstrated that TNF decreased DNA fragmentation by 60% in fibronectin (FN)-adherent neutrophils (36). Thus, in adherent neutrophils, TNF triggers prosurvival, antiapoptotic, and proinflammatory cellular events.
Neutrophils possess two TNF receptors, TNFR-1 (55–60 kDa) and TNFR-2 (75–80 kDa), and both proapoptotic and antiapoptotic signaling are regulated principally by TNFR-1 (3, 42, 54). Binding of TNF to TNFR-1 leads to the recruitment of several effector proteins, including TNFR-1-associated death domain protein (TRADD), receptor interacting protein (RIP), TNFR-associated factor 2 (TRAF2), and fas-associated death domain protein (FADD), which form a signaling complex (1, 29–31, 56, 57). In the classic signaling model, binding of TNF to TNFR-1 leads to the recruitment of TRADD, which then acts as a scaffold for recruitment of the effector proteins, RIP, TRAF2, and FADD (10). RIP and TRAF2 regulate both antiapoptotic and proinflammatory pathways, primarily through activation of transcription factors such as NF-κB (6, 17, 35, 58, 60). FADD is essential for TNF-induced apoptosis through its association with and activation of caspase 8 (8, 30). Recent studies suggest an alternative model for the assembly of TNFR-1 signaling complex (27, 47). It is now proposed that TNF triggers the rapid assembly of a TNFR-1-TRADD-TRAF2-RIP signaling complex at the level of the plasma membrane (complex I) (47). This complex controls the activation of NF-κB but not apoptosis. Complex I is a transient complex, and TRADD-TRAF2-RIP complex can disassociate from TNFR-1 to form a second cytosolic complex (complex II) with FADD and caspase 8/10. Apoptosis would be activated by complex II under conditions in which the signal from complex I fails to activate NF-κB.
How might the binary signaling activated by TNF in adherent neutrophils favor the activation of NF-κB and not apoptosis? Previously, our laboratory identified the PKC-δ isotype (δ-PKC) and PI 3-kinase as important components of this binary signaling pathway, which regulates TNF antiapoptotic signaling (36, 37, 39). PI 3-kinase is a known activator of NF-κB and is involved in antiapoptotic signaling triggered by other proinflammatory cytokines, such as IL-1 and granulocyte-macrophage colony-stimulating factor (7, 20, 24, 38, 45, 52, 65). In addition, PI 3-kinase can regulate δ-PKC activity (41, 59). Inhibition of either PI 3-kinase or δ-PKC significantly blocked TNF-mediated activation of NF-κB and inhibition of apoptosis (14, 19, 36, 62).
The mechanism(s) by which PI 3-kinase and δ-PKC regulate TNF-induced antiapoptotic signaling in adherent neutrophils has yet to be elucidated. A possible regulatory site is at the level of the TNFR-1 complex. Both δ-PKC and PI 3-kinase associate with TNFR-1 in response to TNF in adherent neutrophils (37, 39). TNF triggers phosphorylation of TNFR-1 on both serine and threonine residues (13, 15, 23, 37, 61). Receptor phosphorylation is an important mechanism for regulation of receptor function, and δ-PKC is required for TNF-mediated serine phosphorylation of TNFR-1 (37). Furthermore, in vitro, TNFR-1 is a direct substrate of δ-PKC, suggesting that these kinases may be acting at the level of the TNFR-1. The purpose of this study was to examine the role of δ-PKC and PI 3-kinase in the TNF-mediated assembly of discrete components into the TNFR-1 signaling complex, which regulates NF-κB activation and antiapoptotic signaling in adherent neutrophils.
MATERIALS AND METHODS
Recombinant human TNF-α was obtained from R&D Systems (Minneapolis, MN). Rottlerin and LY-294002 were obtained from Calbiochem (San Diego, CA). Human plasma FN was purchased from Life Technologies (Gaithersburg, MD). The mouse monoclonal anti-human CD120a (TNFR-1) was obtained from Cell Sciences (Norwood, MA). Polyclonal rabbit anti-human-δ-PKC, rabbit anti-human RIP, goat anti-human TNFR-1, protein A/G+ agarose, goat anti-mouse IgG-HRP, donkey anti-goat IgG-HRP, and goat anti-rabbit IgG-HRP were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The mouse monoclonal anti-human TRADD, rabbit polyclonal anti-PI 3-kinase p85, and rabbit polyclonal anti-human TRAF2 were purchased from Upstate Biotechnology (Lake Placid, NY). The mouse monoclonal anti-human FADD and positive control Jurkat cell lysates were obtained from BD Transduction Laboratories (San Diego, CA). δV1.1 PKC-Tat peptide and a Tat carrier control peptide were obtained from Dr. D. Mochly-Rosen (Stanford University). EGTA, goat anti-mouse IgG agarose, sodium-orthovanadate, 4-(2-aminoethyl)-benzenesulfonyl fluoride, leupeptin, protease inhibitor cocktail, and phosphatase inhibitor cocktail were obtained from Sigma (St. Louis, MO). SuperSignal ULTRA chemiluminescence substrate and dimethylpimelimidate (DMP) were obtained from Pierce (Rockford, IL).
Preparation of human neutrophils.
Neutrophils were isolated from heparinized venous blood (10 U/ml) obtained from healthy adult volunteers. Standard isolation techniques (9) were used employing Ficoll-Hypaque centrifugation, followed by dextran sedimentation and hypotonic lysis to remove residual erythrocytes. Cells were suspended in 10 mM HEPES buffer (pH 7.3) and incubated on FN-coated plates for 30 min at 37°C. FN-coated wells were prepared according to the method of Nathan (50) using a concentration of 3.4 μg/well.
Immunoprecipitation of TNFR-1 and FADD.
TNFR-1 coimmunoprecipitation (co-IP) experiments were carried out as described previously (37, 39). Neutrophils were incubated in suspension or in FN-coated six-well plates at a concentration of 50 × 106 cells/well at 37°C for 30 min. For experiments examining the effects of rottlerin (5 μM) or LY-294002 (10 μM), inhibitors were added 15 min before the addition of either TNF or buffer. For experiments examining the inhibitory effects of δV1.1 PKC-Tat peptides, neutrophils were pretreated with buffer, δV1.1 PKC-Tat peptide (1 μM), or Tat carrier peptide (1 μM) alone for 30 min at room temperature. The neutrophils were then plated onto FN-coated wells and incubated for 30 min at 37°C before the addition of either TNF or buffer, as described above. Samples were incubated with either TNF (25 ng/ml) or buffer for 5 min and placed on ice. The cells were lysed in immunoprecipitation (IP) buffer and vortexed for 20 min at 4°C to solubilize the membrane fraction. The IP buffer consisted of 10 mM HEPES, pH 7.4, 150 mM NaCl, 5 mM EDTA, 1 mM sodium-orthovanadate, 20 μM 4-(2-aminoethyl)-benzenesulfonyl fluoride, 0.2% Nonidet P-40, 5 μg/ml leupeptin, Sigma phosphatase inhibitor cocktail, and Sigma protease inhibitor cocktail. A low concentration of Nonidet P-40 was specifically selected to preserve protein associations to examine assembly of the TNFR-1 signaling complex (17, 18, 37, 39, 47).
For TNFR-1 IP experiments, cell lysates were incubated overnight at 4°C with a mouse monoclonal anti-TNFR-1 and then with anti-mouse IgG agarose for 1 h at 4°C. The IgG agarose pellet was washed four times with IP buffer, and the bound proteins were eluted by incubation for 5 min at 95°C in 2× SDS-PAGE sample buffer containing β-mercaptoethanol.
For co-IP experiments quantitating TNFR-1 and recruitment of TRAF2 and FADD, the mouse monoclonal anti-TNFR-1 was cross-linked to the IgG agarose with DMP, according to the method of Harlow and Lane (26). FADD was immunoprecipitated by using a mouse monoclonal anti-FADD cross-linked to IgG agarose with DMP.
Western blot analysis.
Immunoprecipitated proteins were run on a 4–12% gradient SDS-PAGE and transferred to nitrocellulose membranes. The membranes were blocked for 1 h at room temperature with Tris-buffered saline (TBS), pH 7.5, containing 0.1% Tween 20 and 1% BSA-3% casein, as described previously (37). Primary antibodies were diluted in TBS containing 0.1% Tween 20 and 1% BSA-3% casein, and membranes were incubated for 1 h at room temperature. The membranes were washed and incubated for 1 h with the corresponding horseradish peroxidase-conjugated secondary antibody diluted 1:20,000 in TBS containing 0.1% Tween 20 and 1% BSA-3% casein. The membranes were washed, and immunoreactive bands were visualized by using Pierce SuperSignal West Dura Extended Duration chemiluminescence substrate. Co-IP of proteins was quantitated by densitometric analysis of Western blots using Scan Pro, and the values were expressed in arbitrary densitometry units (ADU). In some experiments, 1% of the cell lysates were collected from the different experimental samples, as described by Devin et al. (17), and cellular content of the TNFR-1 signaling complex components was determined by Western blotting to ascertain whether incubation with TNF or the different inhibitors altered cellular content of any components of the TNFR-1 signaling complex.
Results are expressed as means ± SE. Data were analyzed using Student's t-test.
TNF-mediated recruitment of δ-PKC to TNFR-1 complex: effect of δ-PKC and PI 3-kinase inhibitors.
TNF triggers the association of δ-PKC and PI 3-kinase with the TNFR-1 complex in adherent neutrophils (37, 39). To further investigate the regulation of the formation of this complex, we examined the role of adherence in this process and determined whether kinase activity was a requirement for association of δ-PKC and PI 3-kinase with the TNFR-1 complex. In previous studies, time course experiments established that, within 5 min of exposure to TNF, both δ-PKC and PI 3-kinase were activated, δ-PKC and PI 3-kinase coimmunoprecipitated with the TNFR-1 complex, and there were significant functional alterations, as evidenced by phosphorylation of the TNFR-1, activation of NF-κB, and generation of superoxide anion (36, 37, 39). Moreover, at this time point, there was no loss of cell-associated TNFR-1 as a result of receptor shedding (37). Based on these experimental results, a 5-min TNF incubation period was selected to examine the assembly of the TNFR-1 signaling complex.
TNFR-1 was immunoprecipitated from neutrophils in suspension, FN-adherent neutrophils incubated with buffer alone for 5 min, and FN-adherent neutrophils incubated with TNF (25 ng/ml) for 5 min. For adherence studies, neutrophils were incubated for 30 min at 37°C in FN-coated wells to allow cell adherence. Co-IP of δ-PKC with TNFR-1 was determined by Western blot analysis. There was little δ-PKC associated with TNFR-1 in nonadherent neutrophils (Fig. 1). Interestingly, adherence alone resulted in a 46% increase in δ-PKC recruitment to TNFR-1 in the absence of TNF [4,468 ± 592 ADU (buffer-treated nonadherent cells) vs. 6,520 ± 412 ADU (buffer-treated adherent cells), means ± SE, P < 0.04, n = 4; Fig. 1]. The addition of TNF to FN-adherent cells greatly enhanced δ-PKC recruitment to the TNFR-1 by 67% [6,520 ± 412 ADU (buffer-treated FN adherent) vs. 10,864 ± 602 ADU (TNF-treated FN adherent), P < 0.01, n = 4]. Thus the activation of binary signaling through TNF binding and β-integrin engagement resulted in over a twofold increase in δ-PKC recruitment to the TNFR-1 (P < 0.01, n = 4; Fig. 1).
To determine the role for δ-PKC and PI 3-kinase activity in their recruitment, FN-adherent neutrophils were incubated in the absence or presence of the δ-PKC inhibitor rottlerin (5 μM) or PI 3-kinase inhibitor LY-294002 (10 μM) for 15 min before the addition of TNF or buffer. As shown in Fig. 2, preincubation with either rottlerin or LY-294002 before the addition of TNF effectively blocked TNF-mediated recruitment of δ-PKC to TNFR-1 receptor complex [10,308 ± 162 ADU (TNF) vs. 7,217 ± 488 ADU (TNF + rottlerin) or 6,405 ± 748 ADU (TNF + LY-294002), P < 0.001, n = 7]. In contrast, neither rottlerin nor LY-294002 alone had any significant effect on FN-adherence-mediated δ-PKC recruitment to the TNFR-1 complex in the absence of TNF [6,382 ± 209 ADU (buffer-treated FN adherent) vs. 6,512 ± 968 ADU (buffer + rottlerin) and 5,559 ± 846 ADU (buffer + LY-294002), P = not significant (NS), n = 4]. These results demonstrate that both δ-PKC and PI 3-kinase activity are required for TNF-mediated recruitment of δ-PKC to the TNFR-1 signaling complex. It should be noted that these kinase inhibitors were added after the neutrophils were plated onto FN-coated wells, and this may account for the lack of significant inhibitory effect on adherence-based recruitment of δ-PKC to TNFR-1. Neither adherence to FN nor the addition of TNF, rottlerin, or LY-294002 resulted in any significant differences in the concentration of TNFR-1 immunoprecipitated or in total cellular concentrations of δ-PKC (data not shown). Thus the changes in δ-PKC concentration documented by Western blotting reflect alterations in δ-PKC association with TNFR-1 complex and not alterations in receptor concentration or cellular kinase concentrations.
Recruitment of PI 3-kinase to TNFR-1 signaling complex: effect of δ-PKC and PI 3-kinase inhibitors.
We next determined whether PI 3-kinase recruitment to TNFR-1 complex required PI 3-kinase and/or δ-PKC activity. As described above, TNFR-1 was immunoprecipitated from adherent neutrophils pretreated with either rottlerin or LY-294002 before the addition of buffer or TNF. In response to TNF addition, there was a 57% increase in PI 3-kinase recruitment to TNFR-1 [6,418 ± 501 ADU (buffer) vs. 10,104 ± 238 ADU (TNF), P < 0.0001, n = 6; Fig. 3 ]. Preincubation with the PI 3-kinase inhibitor LY-294002 significantly inhibited TNF-mediated recruitment of PI 3-kinase to TNFR-1 complex [6,179 ± 903 ADU (TNF + LY-294002) vs. 10,104 ± 238 ADU (TNF), P < 0.01, n = 6]. In contrast, inhibition of δ-PKC had no significant effect on TNF-mediated PI 3-kinase recruitment [8,955 ± 1,079 ADU (TNF + rottlerin) vs. 10,104 ± 238 ADU (TNF), P = NS, n = 6], indicating that PI 3-kinase recruitment to TNFR-1 requires intrinsic PI 3-kinase activity but is independent of δ-PKC activity.
Effect of cell-permeant δ-PKC peptide antagonist on TNF-mediated recruitment of δ-PKC to TNFR-1 complex.
Recent reports have raised questions as to the specificity of the δ-PKC inhibitor rottlerin. Rottlerin has now been shown to be a potent inhibitor of other PKC isotypes as well as other kinases, including calmodulin-dependent protein kinase, MAPK-activating protein kinase-2, and p38-regulated/activated kinase (16). To establish a role specifically for δ-PKC in the assembly of the TNFR-1 signaling complex, a more specific inhibitor is required. A specific δ-PKC antagonist has recently been reported (11). The δ-PKC antagonist, δV1.1 PKC-Tat peptide, is derived from the first unique variable region (V1) of δ-PKC and coupled to a membrane-permeant peptide sequence in the human immunodeficiency virus Tat gene product. As shown in Fig. 4, preincubation with δV1.1 PKC-Tat peptide inhibited TNF-mediated δ-PKC association with TNFR-1 complex by 51% [5,290 ± 649 ADU (TNF + δ-PKC-Tat) vs. 10,839 ± 1,206 ADU (TNF + Tat carrier), P < 0.002, n = 6]. Conversely, pretreatment with the Tat carrier peptide alone had no significant effect on TNF-mediated recruitment of δ-PKC to TNFR-1 signaling complex [10,342 ± 357 ADU (TNF) vs. 10,839 ± 1,206 ADU (TNF + Tat carrier), P = NS, n = 6]. The δV1.1 PKC-Tat peptide, in contrast to rottlerin, also inhibited adherence-mediated δ-PKC recruitment to TNFR-1 in the absence of TNF [7,855 ± 218 ADU (buffer) vs. 5,540 ± 253 ADU (buffer + δ-PKC-Tat), P < 0.001, n = 4]. This is most likely the result of the different methods of cell treatment with the various δ-PKC inhibitors. Neutrophils were incubated with the δV1.1 PKC-Tat peptide inhibitor before adherence, whereas rottlerin was added following neutrophil addition to FN-coated wells. Furthermore, the δV1.1 PKC-Tat peptide inhibitor mechanism of action is through inhibition of translocation of δ-PKC (11). As with TNF-treated adherent neutrophils, the Tat carrier alone had no effect on δ-PKC recruitment to TNFR-1 in the absence of TNF [7,855 ± 218 ADU (buffer) vs. 6,832 ± 1,130 ADU (buffer + Tat carrier), P = NS, n = 4]. Thus recruitment of δ-PKC to the TNFR-1 complex was inhibited with two different types of δ-PKC inhibitors, rottlerin and the δ-PKC-specific δV1.1 PKC-Tat peptide.
Role of δ-PKC in TNF-mediated recruitment of RIP to TNFR-1 signaling complex.
In adherent neutrophils, TNF activation of NF-κB is controlled principally by TNFR-1. TNF binding to TNFR-1 triggers the assembly of a multicomponent signaling complex, which includes TRADD, RIP, and TRAF2, elements that are required for NF-κB activation and whose recruitment to TNFR-1 is regulated independently of each other (1, 6, 10, 17, 29, 30, 56). Our previous studies demonstrated that inhibition of either δ-PKC or PI 3-kinase significantly decreased TNF-mediated activation of NF-κB (36). We next examined the role of δ-PKC and PI 3-kinase in the assembly of TNFR-1 signaling complex. Neutrophils were pretreated with δV1.1 PKC-Tat peptide, Tat carrier peptide, or buffer alone before the addition of buffer or TNF, as described above. TNFR-1 was immunoprecipitated, and the co-IP of RIP with the TNFR-1 complex was determined by Western blotting. As shown in Fig. 5, the addition of TNF resulted in a 52% increase in RIP recruitment to the TNFR-1 complex in FN-adherent neutrophils [6,678 ± 609 ADU (buffer) vs. 10,171 ± 427 ADU (TNF), P < 0.001, n = 8]. Pretreatment with δV1.1 PKC-Tat peptide resulted in a 43% decrease in RIP association with the TNFR-1 signaling complex [5,616 ± 762 ADU (TNF + δ-PKC-Tat peptide) vs. 9,690 ± 972 ADU (TNF + Tat carrier), P < 0.01, n = 8]. Incubation with the Tat carrier peptide had no significant effect on TNF-mediated RIP recruitment to the TNFR-1 signaling complex [9,690 ± 972 ADU (TNF + Tat carrier) vs. 10,171 ± 56 ADU (TNF), P = NS, n = 8]. The effect of rottlerin pretreatment on TNF-mediated RIP association with TNFR-1 complex was also examined. Rottlerin pretreatment resulted in a 44% decrease in TNF-mediated recruitment of RIP to TNFR-1 [10,801 ± 748 ADU (TNF) vs. 6,057 ± 466 ADU (TNF + rottlerin), P < 0.001, n = 4]. These results demonstrate that, in FN-adherent neutrophils, TNF triggers recruitment of RIP to TNFR-1 complex, and δ-PKC activity is required for this association.
Role of PI 3-kinase in TNF-mediated recruitment of RIP to TNFR-1 signaling complex.
We next examined the role of PI 3-kinase in TNF-mediated association of RIP with TNFR-1 in FN-adherent neutrophils. Preincubation with LY-294002 before the addition of TNF decreased the association of RIP with the TNFR-1 signaling complex by 59% [10,801 ± 748 ADU (TNF) vs. 4,399 ± 887 ADU (TNF + LY-294002), P < 0.0001, n = 6, Fig. 6 ]. Thus, PI 3-kinase activity, as well as δ-PKC activity, is required for TNF-mediated recruitment of RIP to TNFR-1 in adherent neutrophils.
Role of δ-PKC in TNF-mediated association of TRAF2 with TNFR-1 signaling complex.
TRAF2 and RIP are recruited to the TNFR-1 complex independently and have different roles in the activation of NF-κB (17, 18). To determine whether TNF-mediated TRAF2 recruitment to TNFR-1 is also mediated by δ-PKC, the effect of pretreatment with δV1.1 PKC-Tat peptide on TNF-mediated TRAF2 recruitment was examined. Adherent neutrophils were incubated as described above, and TNFR-1 was immunoprecipitated. TRAF2 with a molecular mass of 56 kDa is approximately the same size as the TNFR-1 antibody heavy chain and cannot be visualized by Western blotting following IP of TNFR-1. To prevent release of the antibody heavy and light chain during IP of the TNFR-1, the TNFR-1 antibody was cross-linked to anti-mouse IgG agarose beads with DMP, as described in materials and methods. The addition of TNF to adherent neutrophils resulted in a 78% increase in TRAF2 recruitment to the TNFR-1 complex [Fig. 7; 7,066 ± 815 ADU (buffer) vs. 12,589 ± 678 ADU (TNF), P < 0.001, n = 5]. δV1.1 PKC-Tat peptide pretreatment significantly inhibited TNF-mediated TRAF2 recruitment by 43% [12,051 ± 1,037 ADU (TNF + Tat) vs. 6,911 ± 799 ADU (TNF + δ-PKC Tat peptide), P < 0.006, n = 5]. Pretreatment with the Tat carrier peptide alone had no significant effect on TNF-mediated TRAF2 recruitment [12,589 ± 678 ADU (TNF) vs. 12,051 ± 1,037 ADU (TNF + Tat), P = NS, n = 5]. Similar amounts of TNFR-1 were immunoprecipitated under each experimental condition (Fig. 7). Therefore, similar to RIP, TNF-mediated recruitment of TRAF2 to TNFR-1 signaling complex is δ-PKC dependent in adherent neutrophils.
Role for δ-PKC in TNF-mediated association of TRADD with TNFR-1 signaling complex.
Both RIP and TRAF2 associate with the scaffold protein TRADD (10, 29–31, 57). We next examined whether δ-PKC had a role in the recruitment of TRADD to TNFR-1. As shown in Fig. 8, the addition of TNF enhanced association of TRADD with TNFR-1 by 59% in FN-adherent neutrophils [6,462 ± 551 ADU (buffer) vs. 10,256 ± 653 ADU (TNF), P < 0.001, n = 7]. Surprisingly, preincubation with δV1.1 PKC-Tat peptide before the addition of TNF resulted in a 37% increase in TRADD association with TNFR-1 [16,102 ± 1,380 ADU (TNF + δ-PKC-Tat peptide) vs. 11,799 ± 467 ADU (TNF + Tat carrier), P < 0.01, n = 7]. The Tat carrier peptide alone had no significant effect on TNF-mediated TRADD association [11,799 ± 467 ADU (TNF + Tat carrier) vs. 10,256 ± 653 ADU (TNF), P = NS, n = 7]. These results suggest that δ-PKC acts as a negative modulator of TNF-triggered TRADD association with TNFR-1.
Role for δ-PKC in TNF-mediated association of FADD with TNFR-1 signaling complex.
TNF also triggers the association of TRADD with FADD, an important signaling component in TNF-mediated caspase activation and cell death (1, 30, 31). Our previous studies demonstrated that inhibition of either δ-PKC or PI 3-kinase inhibited TNF-mediated antiapoptotic signaling. We hypothesized that δ-PKC and PI 3-kinase might be negative regulators of TNF-mediated apoptosis and block the activation of apoptosis through interference with TRADD-FADD interactions. To ascertain the role of δ-PKC in recruitment of FADD to TNFR-1, we examined TNF-mediated recruitment of FADD with TNFR-1 signaling complex. As shown in Fig. 1, adherence alone can trigger some δ-PKC association with the TNFR-1. Therefore, we determined FADD association in both FN-adherent and nonadherent neutrophils. FADD did not coimmunoprecipitate with TNFR-1 in response to TNF under any of our experimental conditions (Fig. 9). FADD did not associate with TNFR-1 under experimental conditions in which TRADD, TRAF2, and RIP associated with TNFR-1. Furthermore, FADD association with the TNFR-1 was not detectable even under conditions of increased TRADD association with TNFR-1 in the presence of δV1.1 PKC-Tat peptide (data not shown). FADD is, however, present in human neutrophils. As shown in Fig. 9, equivalent amounts of FADD were immunoprecipitated from FN-adherent neutrophils incubated in the presence or absence of TNF (25 ng/ml for 5 min), indicating that short incubations with TNF had no effect on the cellular levels of FADD. Moreover, FADD did coimmunoprecipitate with TRADD (data not shown). Thus, whereas FADD is present in human neutrophils and associates with TRADD, FADD does not associate with TNFR-1 in response to TNF in either adherent or nonadherent neutrophils.
TNF is a unique proinflammatory cytokine whose signaling is linked to both antiapoptotic and proapoptotic signaling pathways. The cellular response to TNF is cell type dependent and influenced by input from other signaling pathways, such as engagement of integrins (25, 36, 37, 39, 49, 50). In the neutrophil, the binding of TNF and ligation of β-integrins activate signaling that favors antiapoptotic signaling. A principal pathway in this process is the activation of NF-κB (60). Our previous studies identified δ-PKC and PI 3-kinase as important regulators of this signaling pathway (36, 37, 39). Inhibition of either δ-PKC or PI 3-kinase resulted in the inhibition of TNF-mediated activation of NF-κB and TNF-mediated inhibition of apoptosis (36). The results of the present study suggest a mechanism for δ-PKC and PI 3-kinase-mediated regulation of TNF antiapoptotic signaling, specifically at the level of TNFR-1 through regulation of the assembly of the TNFR-1-TRADD-RIP-TRAF2 complex, a complex that controls NF-κB activation.
How might δ-PKC and PI 3-kinase regulate the assembly of the TNFR-1 signaling complex? A possible regulatory site is at the level of the TNFR-1 itself. TNF triggers the recruitment of δ-PKC and PI 3-kinase to the TNFR-1 signaling complex in adherent neutrophils, a process in which β-integrins play an important role (37, 39). The results of this study demonstrate that β-integrin-mediated neutrophil adherence, in the absence of TNF, can trigger δ-PKC recruitment to the TNFR-1 complex, whereas β-integrin activation and TNF cooperate to promote full activation of TNFR-1-based signaling. Indeed, TNF-mediated activation of c-Jun kinase is enhanced in adherent endothelial cells compared with those in suspension (55). The results of the present study suggest that integrins can augment proinflammatory cytokine signaling through enhanced recruitment of signaling components.
The association of δ-PKC with TNFR-1 requires both δ-PKC and PI 3-kinase activity, whereas PI 3-kinase association is PI 3-kinase dependent but δ-PKC independent. Our laboratory's previous studies determined that δ-PKC and PI 3-kinase are both components of the same TNF-mediated antiapoptotic signaling pathway (36). Co-IP of PI 3-kinase with a proinflammatory cytokine signaling complex is not unique to TNFR-1. IL-1 also triggers the association of PI 3-kinase with the IL-1 receptor I complex, and this association is thought to have an important role in IL-1-mediated activation of NF-κB (24, 45, 52). Granulocyte-macrophage colony-stimulating factor, another proinflammatory cytokine, also triggers the association of δ-PKC and PI 3-kinase with each other, indicating a role for these kinases in proinflammatory cytokine signaling (22). The present study suggests that PI 3-kinase acts upstream of δ-PKC signaling, a finding that is consistent with PI 3-kinase functioning as an activator of δ-PKC (41, 59).
A possible consequence of association of δ-PKC and PI 3-kinase with TNFR-1 may be alterations in protein-protein interactions required for signaling. Our results demonstrate that the formation of the TNFR-1-δ-PKC-PI 3-kinase complex has important regulatory consequences in the assembly of the TNFR-1-TRADD-TRAF2-RIP signaling complex and in TNF-mediated antiapoptotic signaling. Inhibition of either δ-PKC or PI 3-kinase resulted in significant alterations both in TNF-mediated assembly of the TNFR-1-TRADD-RIP-TRAF2 signaling complex and in the activation of NF-κB (this study and Ref. 36). RIP, TRAF2, and TRADD are all recruited to TNFR-1 in response to TNF, and each has an important role in NF-κB activation (17, 18, 31). The TNFR-1-TRADD-TRAF2-RIP complex regulates NF-κB activation through the recruitment and activation of IKK to the TNFR-1 signaling complex (1, 17, 18, 35, 58, 66). Once activated, IKK then phosphorylates the NF-κB inhibitor IκB, leading to its ubiquitination and rapid degradation (33). The degradation of IκB releases NF-κB for translocation to the nucleus and activation of transcription of target genes (4, 5).
RIP is a serine/threonine kinase composed of a COOH-terminal death domain, an intermediate domain, and an NH2-terminal kinase domain (29, 57). RIP is recruited independently to the TNFR-1 complex, where it interacts with TRADD via its death domain as well as with TRAF2 through the intermediate domain to form a complex (6). RIP can also interact directly with TNFR-1 by death domain interactions, but this interaction is thought to be relatively weak compared with interactions with TRADD (29). The importance of RIP in NF-κB activation has been demonstrated in RIP-deficient Jurkat cells and in RIP−/− cells, where TNF-mediated activation of NF-κB is inhibited (17, 18, 35, 58). Pretreatment of neutrophils with either the PI 3-kinase inhibitor LY-294002 or the δ-PKC inhibitors δ-PKC-Tat peptide or rottlerin significantly inhibited TNF-mediated recruitment of RIP to the TNFR-1 complex. Thus, in adherent neutrophils, δ-PKC and PI 3-kinase act as positive regulators of TNF-mediated association of RIP with TNFR-1.
Inhibition of δ-PKC also inhibited TNF-mediated recruitment of TRAF2 to the TNFR-1 complex in neutrophils. TRAF2 is also required for NF-κB activation and mediates IKK recruitment to the TNFR-1 (17, 18). TRAF2 is composed of two separate domains, a COOH-terminal TRAF domain and NH2-terminal ring and zinc finger domain (30, 53). Unlike RIP, FADD, and TNFR-1, TRAF2 does not bind to TRADD via its COOH-terminal death domain but rather with the NH2-terminal region. Although TRAF2 also interacts with RIP through its TRAF domain, the presence of RIP is not required for TNF-mediated TRAF2 recruitment to the TNFR-1 complex (17). In RIP−/− cells, TNF-mediated TRAF2 recruitment is not inhibited. The decreased recruitment of TRAF2 following inhibition of δ-PKC is most likely a direct effect and not merely the consequence of attenuated RIP recruitment to the TNFR-1 complex. These results suggest that δ-PKC and its upstream regulator PI 3-kinase are positive regulators of NF-κB activation through regulation of association of RIP and TRAF2 with TNFR-1 signaling complex.
An unexpected finding in our study was that, whereas δ-PKC inhibition attenuated RIP and TRAF2 recruitment to the TNFR-1 complex, δ-PKC inhibition significantly enhanced TNF-mediated association of TRADD with the TNFR-1 complex. TRADD contains a COOH-terminal death domain, which mediates its death domain/death domain interactions with TNFR-1, RIP, and FADD and a novel NH2-terminal region that binds to TRAF2 (29, 31). As TRADD recruitment was enhanced in response to δ-PKC inhibition, it is possible that RIP and/or TRAF2 interfere with TRADD interactions with TNFR-1. Enhanced recruitment of TRADD in response to TNF has been reported in RIP−/− fibroblasts compared with wild type, but not in TRAF2−/− fibroblasts, suggesting RIP interferes with TRADD association with the receptor (17). These findings are consistent with what is known about the structure of these effector proteins. Both TRADD and RIP associate with each other and with TNFR-1 via their death domains (29). RIP may be competing with TRADD for binding to TNFR-1, and decreased recruitment of RIP may allow for greater TRADD association. However, these studies do not rule out the possibility that δ-PKC has a direct inhibitory effect on TRADD association with TNFR-1 complex. In either case, δ-PKC is acting as a negative regulator of TRADD association with TNFR-1.
TRADD association with TNFR-1 may be an important control site or bifurcation point regulating antiapoptotic/proapoptotic signaling. δ-PKC-mediated negative regulation of TRADD association with TNFR-1 may prevent the activation of apoptotic signaling and promote NF-κB activation. Inhibition of δ-PKC leads to decreased association of RIP with TNFR-1 coupled with increased association of TRADD in response to TNF. The net result is increased availability of TRADD death domain binding sites. A potential outcome of increased TRADD availability would be enhanced TRADD-FADD interaction and a shift to FADD/caspase 8 signaling. In our studies, we demonstrated that FADD could be immunoprecipitated from neutrophils, but were unable to detect co-IP of FADD with the TNFR-1 receptor in response to TNF in either adherent or nonadherent cells. The inability to detect FADD co-IP with TNFR-1 is consistent with recent reports that FADD does not associate with the transient TNFR-1-TRADD-TRAF2-RIP complex at the plasma membrane but rather forms a secondary complex with a TRADD-TRAF2-RIP complex in the cytoplasm following disassociation of these effector proteins from TNFR-1 (27, 47). The results of the present study are consistent with this signaling model: δ-PKC may be regulating antiapoptotic/proapoptotic signaling at the level of TRADD. Whether δ-PKC has a regulatory role in the formation of this cytoplasmic secondary complex remains to be determined.
In summary, TNF signaling is regulated at multiple control points. An important regulatory step in control of antiapoptotic signaling is the assembly of the signaling complex composed of TNFR-1-TRADD-TRAF2-RIP, a complex that controls NF-κB activation. Our studies demonstrate that δ-PKC and PI 3-kinase are important components of this process and act as positive regulators of NF-κB activation and negative regulators of apoptosis (present study, Refs. 36, 37, 39). Both δ-PKC and PI 3-kinase coimmunoprecipitate with the TNFR-1 signaling complex and are positive regulators of RIP and TRAF2 recruitment to TNFR-1 complex and negative regulators of TRADD association. The present study does not identify specific sites of protein-protein interaction between TNFR-1 and δ-PKC or PI 3-kinase, and further studies are required to ascertain whether these kinases interact directly with the TNFR-1 or indirectly through associations with other components of the TNFR-1 signaling complex. Control of assembly of the TNFR-1-TRADD-TRAF2-RIP complex may be mediated through receptor phosphorylation, and δ-PKC is a candidate kinase. δ-PKC has a major role in TNF-triggered serine phosphorylation of TNFR-1 (37). Specifically, TNF-triggered serine phosphorylation of TNFR-1 is inhibited by rottlerin as well as staurosporine, and δ-PKC coimmunoprecipitates with the TNFR-1 signaling complex in the correct time frame for receptor phosphorylation. Moreover, in vitro, only δ-PKC, but not α-PKC, βI-PKC, βII-PKC, or ζ-PKC, is competent to phosphorylate the receptor in a cofactor-dependent manner, indicating that TNFR-1 is a direct substrate for δ-PKC. Thus serine phosphorylation of the receptor may produce conformational changes that favor RIP and TRAF2 association and NF-κB activation, leading to inhibition of apoptosis and activation of proinflammatory signaling.
This work was supported by National Institutes of Health Grants GM-64552 (L. E. Kilpatrick) and AI-24840 (H. M. Korchak).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society