| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 281, Issue 23, 16128-16138, June 9, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Signaling through Protein Kinase C
Induces NADPH Oxidase-mediated Oxidant Generation and NF-
B Activation in Endothelial Cells*
1
From the
Department of Pharmacology and Center for Lung and Vascular Biology, the University of Illinois College of Medicine, Chicago, Illinois 60612 and the Departments of Pediatrics and Environmental Medicine, University of Rochester School of Medicine, Rochester, New York 14642
Received for publication, August 10, 2005 , and in revised form, February 21, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
in mediating NADPH oxidase activation and reactive oxidant species (ROS) generation in endothelial cells (ECs) and of PI3K-mediated oxidant signaling in the mechanism of NF-
B activation and intercellular adhesion molecule (ICAM)-1 expression. We used lung microvascular ECs isolated from mice with targeted deletion of the p110 catalytic subunit of PI3K. Tumor necrosis factor (TNF) 
challenge of wild type ECs caused p110 translocation to the plasma membrane and phosphatidylinositol 1,4,5-trisphosphate production coupled to ROS production; however, this response was blocked in p110–/– ECs. ROS production was the result of TNF activation of Ser phosphorylation of NADPH oxidase subunit p47phox and its translocation to EC membranes. NADPH oxidase activation failed to occur in p110–/– ECs. Additionally, the TNF-activated NF-B binding to the ICAM-1 promoter, ICAM-1 protein expression, and PMN adhesion to ECs required functional PI3K. TNF challenge of p110–/– ECs failed to induce phosphorylation of PDK1 and activation of the atypical PKC isoform, PKC
. Thus, PI3K lies upstream of PKC in the endothelium, and its activation is crucial in signaling NADPH oxidase-dependent oxidant production and subsequent NF-B activation and ICAM-1 expression. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
, p110
, p110, and p110
, have been identified (1), and of these, p110 has distinct properties. Type 1A PI3Ks, p110, p110, and p110, associate with one of the five regulatory subunits: p50, p55, and p85 (products of alternative splicing of a single gene) and p55 and p85 (2). In contrast, type 1B PI3K (or PI3K), the catalytic subunit p110 binds to the p101 adaptor molecule (3) or the G-activated regulatory subunit p84 (4). Type 1A PI3Ks are activated by interactions with tyrosine-phosphorylated molecules, whereas p110 is activated by heterotrimeric G proteins G and G that bind to the pleckstrin homology domain found in the NH2-terminal region of PI3K (3, 5). p110 is also activated by pro-inflammatory cytokines such as TNF (6). Expression of PI3K is largely confined to leukocytes, and there is a growing appreciation of its important role in immunity and host defense (7–18). Studies also demonstrated the presence of the PI3K isoform in endothelial cells (ECs) (19, 20), but its function remains unclear.
PI3Ks catalyze the conversion of phosphatidylinositol 4,5-bisphosphate to phosphatidylinositol 3,4,5-trisphosphate (PIP3), which is involved in the recruitment and activation of a variety of regulatory proteins via interactions with their pleckstrin homology and phox homology domains (21). phox domains, present in two subunits of the NADPH oxidase complex, p47phox and p40phox, bind to phosphatidylinositol 3,4-bisphosphate and phosphatidylinositol trisphosphate (both breakdown products of PIP3) (21, 22). Degradation of PIP3 occurs by either PTEN (3'-phosphatase and tensin homolog deleted on chromosome 10) or SH2-containing phosphatidyl inositol phosphatases (SHIP-1 and SHIP-2) (7, 23, 24).
NADPH oxidase is a tightly regulated membrane-bound enzyme complex catalyzing the one-electron reduction of oxygen to superoxide with the simultaneous oxidation of cytosolic NADPH (25). We showed that TNF-induced oxidant generation in ECs requires activation of PKC (26, 27). PKC associates with and phosphorylates p47phox, and in turn promotes p47phox association with Nox2 to generate the active NADPH oxidase complex (28, 29). We also showed that PKC activation of NADPH oxidase was required for TNF-induced oxidant generation in ECs (28).
In the present study, we addressed a possible role for PI3K as an upstream regulator of PKC activation and thereby in mediating NADPH oxidase assembly and generating the oxidant signaling required for NF-B activation and ICAM-1 expression in ECs. Our results show that TNF induces PIP3 production and mediates the PI3K activation of PKC. We show that PI3K plays a crucial role in signaling the activation of NADPH oxidase required for NF-B activation and ICAM-1 expression in ECs.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
antibody (catalog numbers SC-7177 and 4252), ICAM-1 (Western blot, catalog number SC-8439; immunofluorescence, catalog number SC-1511), SHIP-2 (catalog number SC-14502), VE-cadherin (catalog number SC-6458), ICAM-1 fluorescein isothiocyanate (catalog number SC-18853), and actin (catalog number SC-1616) from Santa Cruz Biotechnology (Santa Cruz, CA) and Cell Signaling Technologies (Beverly, MA). Phospho-PDK1 (Ser241, catalog number 3061) and phospho-PKC–/– (human, mouse, and rat cross-reactivity) (Thr410/403, catalog number 9378) antibodies were obtained from Cell Signaling Technologies. Non-phospho-specific rabbit polyclonal PKC antibody (catalog number 9372) and rat monoclonal PKC antibody (catalog number ALX-804–042) were obtained from Cell Signaling Technologies and Alexis Biochemicals (San Diego, CA), respectively. Alexa Flour 594 and 488 secondary antibodies, carboxy-H2DCFDA (catalog number C-400) cell-permeant indicator for H2O2 that is retained by cells, mouse monoclonal antibody to PIP3 (catalog number A21328), TRIzol reagent, Taq DNA polymerase and Pro-Q Diamond phosphoprotein gel stain (catalog number 33300) were obtained from Invitrogen. 1,2-Dioctanoyl-sn-glycero-3-[phosphoinositol-3,4,5-trisphosphate] (tetra-ammonium salt) was obtained from Avanti Polar Lipids, Inc. (Alabaster, AL). Phospho-Ser antibody (catalog number 61-8100) was obtained from Zymed Laboratories (San Francisco, CA). Anti-p47phox antibody used in this study was a gift from B. Babior and S. Catz (Scripps Research Institute, La Jolla, CA). Fetal bovine serum was from Hyclone (Logan, UT). Endothelial growth medium (EGM-2) was obtained from BioWhittaker (Walkersville, MD). Primary human pulmonary artery endothelial cells (HPAECs) were obtained from Clonetics (La Jolla, CA). RAW 264.7 cells were obtained from the American Type Culture Collection (Manassas, VA). Recombinant human TNF was obtained from Promega (Madison, WI) and R & D Systems (Minneapolis, MN), and recombinant mouse TNF was obtained from Roche Applied Science.
Cell Culture—HPAECs were cultured in EBM2 (endothelial basal medium) complete medium in gelatin-coated flasks with bullet kit additives. Mouse lung vascular endothelial cells (MLVECs) from WT (C57BL/6) and p110–/– mice were cultured as described (28, 30). C57BL/6 WT mice were obtained from Jackson Laboratories and p110 knockout mice were provided by J. Penninger (Amgen Institute, Toronto, Canada). MLVECs were characterized by their cobblestone morphology, Factor VIII and VE-cadherin staining, and uptake of low density lipoprotein. RAW 264.7 cells are a macrophage-like, Abelson leukemia virus transformed cell line derived from BALB/c mice. The cells are grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 20 mM HEPES, and 2 mM L-glutamine at 37 °C in a humidified atmosphere with 5% CO2. Prior to assay, the medium is changed to Dulbecco's modified Eagle's medium supplemented with 20 mM HEPES, 2 mM L-glutamine, and 0.1 mg/ml bovine serum albumin for 2 h. After the addition of zymosan A (125 µg/ml), the cells are placed on ice, and the medium is aspirated at the appropriate time point. The cells are washed with 1.5 ml of ice-cold PBS solution, pelleted, and PBS-aspirated.
PKC Kinase Assay—PKC activity was assayed as described (32). The cell lysates were immunoprecipitated with an antibody against PKC using protein A/G conjugated to agarose. The immunocomplexes were washed twice with ice-cold PBS and once with kinase buffer (25 mM Tris-HCl, pH 7.4, 5 mM MgCl2, 0.5 mM EGTA, 1 mM dithiothreitol) and resuspended in 30 µl of kinase buffer containing 2.5 µg of histone H1, 0.5 mM cold ATP, and 20–30 µCi of [-32P]ATP. The reaction mixture was incubated for 20 min at room temperature, and the reaction terminated by the addition of SDS sample buffer. The proteins were analyzed by SDS-PAGE, and the phosphorylated form of histone H1 was detected by autoradiography.
Phosphatidylinositol Extraction—The procedures were conducted as described (31). Briefly, ice-cold 1:1 CHCl3:CH3OH is added to each cell pellet and vortexed for 1 min, samples are centrifuged at 6000 rpm for 5 min at 4 °C, and supernatant is discarded. The remaining cell pellet is suspended in 200 µl of 2:1CHCl3:CH3OH containing 0.25% 12 N HCl, vortexed for 5 min, and pulse-spun. To the supernatant 40 µl of 1 N HCl are added, vortexed for 15 s, and centrifuged to separate the phases. The solvent from the collected lower layer is evaporated in a vacuum centrifuge, and lipid film was rapidly redissolved in 55 µl of 1:1:0.3 CHCl3: CH3OH:H2O. Before analysis, 5 µl of 300 mM piperidine are added, and the sample is vortexed and pulse-spun.
MS Analysis of PIP3—Mass spectral analysis was performed on a Finnigan TSQ Quantum triple quadrupole mass spectrometer (Thermo-Finnigan, San Jose, CA) as described (31). The samples were analyzed at an infusion rate of 10 µl/min in negative ionization mode over the range of m/z 400–1200. Peaks corresponding to known PIP3s were fragmented and manually inspected for the presence of the identification peaks. A confirmed identification was achieved when key fragmentation peaks were larger than three times the signal to-noise ratio. The lower limit of detection using this method was reported to be less than 9 pmol/ml for 38:4 PIP3 (Avanti Polar Lipids).
Immunoblotting—ECs were washed with ice-cold Tris-buffered saline and lysed in 10 mmol/liter Tris-HCl, pH 7.5, 5 mM EDTA, 10 mM EGTA, 1 mM MgCl2, 50 µg/ml phenylmethylsulfonyl fluoride, and a mixture of protease inhibitors. The lysates were sonicated for 10 s and then ultracentrifuged at 100,000 x g for 1 h at 4 °C, and the supernatants were collected and designated as cytosolic fraction. To isolate the membrane fraction, the remaining pellet was resuspended in the above lysis buffer containing 1% Triton X-100, sonicated, and incubated for 30 min at 4 °C (28, 32). These lysates are microcentrifuged at 4 °C, and the supernatants were designated membrane fraction. Immunocomplexes were Western blotted as described (32). For analysis of Ser-p47phox phosphorylation, total cell lysates were immunoprecipitated with p47phox antibody and Western blotted with phospho-Ser antibody. For analysis of total p47phox phosphorylation, total cell lysates were immunoprecipitated with p47phox antibody and subjected to SDS-PAGE, and gels were stained with Pro-Q Diamond phosphoprotein gel stain (33).
Confocal Microscopy—HPAECs, grown on gelatin-coated coverslips, were treated as indicated, washed with HBSS, fixed in 4% paraformaldehyde, and blocked with 5% goat serum containing 0.2% bovine serum albumin, 0.01% NaN3, and 0.1% Triton X-100. Thereafter, the cells were incubated for 1 h at room temperature with 1 µg of the indicated primary antibody. After three washes in HBSS, 4 µg/ml secondary antibodies Alexa Fluor 488 and 594 (Molecular Probes, Eugene, OR) were added for an additional 2 h at room temperature. The cells were extensively washed in HBSS and mounted on glass slides with ProLong Antifade mounting medium (Molecular Probes), and the images were acquired with a Zeiss LSM 510 Meta confocal microscope. For p110-PKC co-localization studies, the cells were incubated with TNF, fixed, and co-incubated with a rat monoclonal antibody to PKC, a rabbit polyclonal antibody to p110 and secondary antibodies Alexa Fluor 488 and 594 IgG as described above. For PIP3 detection, WT MLVECs were incubated with TNF, fixed, and incubated with monoclonal antibody to PIP3 (34) and secondary Alexa Flour 594 IgM. Appropriate band filters were used to detect both proteins and PIP3. Fixed cells labeled with p110-PKC antibodies were optically sectioned into z-stacks (0.3-µm-thick confocal sections) with the pinhole set to 1 Airy unit. Quantification of co-localization between p110 and PKC was performed using the co-localization module of Zeiss LSM 510–3.2 software.
Alterations in Cell Surface ICAM-1—MLVECs from WT and p110–/– mice were grown on gelatin-coated coverslips, incubated with TNF, fixed in 3.7% formaldehyde, permeabilized in 0.4% Triton X-100/PBS, blocked with PBS containing 0.1% Triton X-100, 5% bovine serum albumin, 0.5% gelatin, and incubated for 16 h at 4 °C with ICAM-1 antibody, or for 1 h with anti-ICAM IgG-fluorescein isothiocyanate and goat polyclonal VE-cadherin antibodies. The cells were washed three times with PBS, incubated with Alexa Fluor 594 IgG, and mounted on glass slides with ProLong Antifade mounting medium (Molecular Probes). The images were acquired with a Zeiss LSM 510 Meta confocal microscope. Quantification of ICAM-1 staining intensity in VE-cadherin-stained plasma membranes was from three to four images/coverslip, with each image containing an average of eight cells using ImageJ software (National Institutes of Health, Bethesda, MD).
|
amplification were used for PCR (35, 36): forward, 5'-GCTTGAAAACCTGCAGAATTCTCAAC-3'; reverse, 5'-CGTCTTTCACAATCTCGATCATTCC-3'. Mouse specific primers designed to span exons 3 to 6 of the p110 gene were as follows: forward, 5'-AGAGAAGTATGACGTCAGTTCC-3'; reverse 5'-TTGAGCCATCGTTGTGGCATCC-3'. The cDNA obtained from RT-PCR was PCR-amplified using Invitrogen Platinum Taq, and the entire PCR product was sequenced in double strand and compared with the expected reference sequence. The human sequence is deposited in Gen-BankTM (accession number AY496423
[GenBank]
), and the mouse sequence is deposited in GenBankTM (accession number AY831679
[GenBank]
).
Oxidant Generation—Oxidant generation in MLVECs was measured as described (28). The cells were loaded with the fluorescent dye carboxy-H2DCFDA (10 µM; Molecular Probes) for 1 h. After treatment with TNF, the cells were washed twice with HBSS and fixed in 4% paraformaldehyde for 20 min at room temperature. The cultures were then viewed with fluorescence microscopy.
Electrophoretic Gel Mobility Shift Assay—Nuclear protein extracts were prepared as described with the addition of protease inhibitors (37). The extracted protein was quantified, aliquoted (40–50 µg/aliquot), and stored at –70 °C until use. The oligonucleotide was radiolabeled with T4 polynucleotide kinase. Nuclear protein (10–15 µg) was incubated for 15 min at room temperature with labeled oligonucleotide. Incubation mixtures were separated on a 5% nondenaturing polyacrylamide gel. The following oligonucleotides were used for gel shift analysis: ICAM-1 NF-B: 5'-AGCTTGGAAATTCCGGAGCTG-3' and Ig-B: 5'-AGTTGAGGGGACTTTCCCAGGC-3'. The oligonucleotide designated as ICAM-1 NF-B represents the 21-bp sequence of ICAM-1 promoter encompassing NF-B binding site located 223 bp upstream of translation initiation site. The Ig-B oligonucleotide contains the consensus NF-B binding site present in the immunoglobulin gene. Sequence motifs within the oligonucleotides are underlined.
PMN Adhesion Assay—PMN adhesion to endothelial cells was determined with the modifications as described below (38). Isolated MLVECs from WT and p110–/– mice were grown to confluence in 96-well gelatin-coated plates. Mouse PMNs were isolated from whole blood as described (38); PMN purity was >95%, and the viability assessed by trypan blue exclusion was >98%. PMNs labeled with calcein-AM (Molecular Probes) were added to MLVECs pretreated with TNF (40 ng/ml) for the times indicated at 37 °C. Naïve PMNs were incubated with MLVECs for 2 h and washed six times with EBM2 medium, and the fluorescence readings were obtained using the PTI spectrofluorometer (Photon Technology International, Monmouth Junction, NJ) with detection at 494 and 517 nm, respectively.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Expression in Endothelial Cells—We initially determined the expression profile of p110 mRNA and protein in primary human and mouse ECs. RT-PCR was carried out using a set of primers specific for human and mouse p110 mRNA on total RNA from HPAECs, human monocytic cell line THP-1, and MLVECs (35, 36). As shown in Fig. 1A, p110 mRNA was detected in HPAECs, and it was identical to THP-1 p110 mRNA with a 316-bp PCR fragment. Sequencing and BLAST analysis indicated that the PCR fragment was 100% identical to human p110 (5, 39). p110 was also detected in MLVECs with a 350-bp PCR fragment (Fig. 1B). Sequencing and BLAST analysis indicated that this PCR fragment was 100% identical to mouse p110 (39). We next determined the effects of TNF in inducing the membrane association of p110 because this is a requirement for activation of PI3K (40, 41). Western blotting showed that TNF exposure of HPAECs increased the membrane localization of p110 within 5 min, and the response remained elevated up to 30 min (Fig. 1, C and D). TNF challenge of MLVECs induced a similar p110 translocation to the membrane (data not shown).
p110 Co-localizes with PKC—To examine the relative distribution of p110 in endothelial cells, HPAECs were treated with TNF, fixed, and double stained for p110 and PKC using Alexa Flour 594 and 488 secondary antibodies, respectively (Fig. 2, A and B). In the absence of TNF (control), PKC and p110 were primarily localized in the cytoplasm and perinuclear regions. TNF induced accumulation of PKC and p110 at the plasma membrane (Fig. 2, A and B). To assess co-localization of PKC and p110, we optically sectioned cells into 0.3-µm sections, and the plasma membrane and perinuclear regions were selected for each cell. We observed a striking overlap of p110 and PKC after TNF exposure (Fig. 2C). Co-localization coefficient in control cells was 0.34 compared with 0.90 after TNF treatment. We also determined co-localization of p110 with p120-catenin, a juxta-EC plasma membrane protein. In control and TNF-treated cells, we did not observe any pattern of co-localization of p110 with p120-catenin (data not shown), suggesting the specificity of p110-PKC co-localization. In another experiment, we determined the association between p110 and PKC by immunoprecipitation using an antibody to PKC and Western blotting for p110 (Fig. 2D). Membranes isolated from control HPAECs (0 min) showed that p110 and PKC were found in the same immunocomplex, and TNF challenge increased the appearance of p110 in immunoprecipitates of PKC (Fig. 2D).
TNF Induces Activation of p110 and PIP3 Production in Endothelial Cells—Activation of PKB/Akt is dependent on PDK1 phosphorylation, which requires activation of PI3Ks (1). This produces the second messenger PIP3, which binds to the pleckstrin homology domain of PDK1, recruiting it to the plasma membrane where PDK1 is activated by phosphorylation of its activation loop residue Ser241 (42). To determine whether TNF induced membrane translocation of p110 in ECs and resulted in PI3K activation, we carried out immunoblot analysis of PDK1. Challenge of HPAECs with TNF induced the phosphorylation of the membrane-associated PDK1 within 5 min, and the response was maintained up to 30 min (Fig. 3A). We also observed that challenge of MLVECs with TNF resulted in phosphorylation of membrane-associated PDK-1 (Fig. 3B). However, this response was inhibited in p110–/– MLVECs (Fig. 3B), indicating that p110 is required for TNF-induced PDK-1 activation.
Fig. 4A shows confocal microscopy results of PIP3 staining before and after TNF challenge using a monoclonal antibody specific for PIP3 (34). TNF increased PIP3 staining within 1 min, and the response was sustained for 30 min and localized mainly in the cytoplasm and perinuclear regions. We used electrospray mass spectrometry (ESI-MS) to determine PIP3 lipid species produced in ECs. Fig. 4B shows a typical positive control ESI-MS profile of control and zymosan-treated RAW 264.7 macrophages demonstrating an increase in PIP3 with a atomic mass of 1101 m/z. Using an internal PIP3 standard (31), we determined that ESI-MS correctly identified PIP3 (data not shown). To confirm the confocal immunofluorescence results for PIP3, we determined whether TNF could stimulate PIP3 production as measured using ESI-MS. We observed that TNF induced the production of PIP3 with atomic masses of 1075 and 1101 (Fig. 4B), similar to those reported for zymosan-treated RAW 264.7 cells (31). We did not detect any effect of TNF on PIP3 production in the p110–/– MLVECs (Fig. 4B).
|
|
–/– mice were Western blotted for SHIP-2. In WT MLVECs, in the basal state, SHIP-2 was present in the membrane, and its distribution was not further altered after TNF challenge (Fig. 4C). In contrast, SHIP-2 was undetectable in membranes isolated from p110–/– ECs (Fig. 4C) but was present in the cytosolic fraction, indicating that p110 is responsible for EC membrane association of SHIP-2 in the basal state.
p110 Deletion Prevents TNF-induced Phosphorylation and Membrane Translocation of p47 phox—We determined the phosphorylation of p47phox, a requirement for the activation of NADPH oxidase (47, 48). TNF challenge of WT cells induced time-dependent phosphorylation of p47phox as assessed by Pro-Q Diamond staining; the phosphorylated form of p47phox increased 2.8-fold at 5 min after TNF challenge and was sustained for 30 min at the fold level (Fig. 5A). We also observed that TNF challenge of WT cells reduced the electrophoretic mobility of p47phox, indicative of increased phosphorylation. However, TNF-induced phosphorylation of p47phox was significantly reduced in p110–/– ECs, ranging from 1.2-fold at 5 min to 0.9-fold at 30 min (Fig. 5A). A number of earlier studies in leukocytes have shown that p47phox is phosphorylated on several serine residues located between amino acids 303 and 379, and p47phox serine phosphorylation is crucial for the assembly of NADPH oxidase in phagocytes (47); thus, we determined p47phox Ser phosphorylation using immunoprecipitation and Western blotting. We observed that TNF induced Ser phosphorylation of p47phox in WT cells similar to the phosphorylation shown above using Pro-Q Diamond staining (Fig. 5A), but this response was absent in p110–/– cells (Fig. 5, B and C).
|
on the membrane translocation of p47 phox, another requirement for activation of NADPH oxidase (47). TNF challenge of WT cells induced time-dependent translocation of p47phox (Fig. 5, D and E). However, TNF-induced p47phox translocation was significantly reduced in p110–/– ECs (Fig. 5, D and E).
We used the fluorescent redox-sensitive dye carboxy-H2DCFDA to determine the role of p110 in mediating TNF-induced oxidant generation. WT ECs challenged with TNF for 30 and 60 min showed marked oxidant generation, whereas the response was absent in p110–/– ECs (Fig. 6), consistent with the impairment of NADPH oxidase activation shown above.
Because PKC is required for signaling TNF-induced NADPH oxidase and NF-B activation in ECs (27, 28), we examined the possible relationship between p110 and PKC. TNF stimulation of WT ECs resulted in phosphorylation and membrane translocation of PKC within 5 min, and the response was sustained up to 30 min (Fig. 7, A and B). However, TNF failed to induce PKC phosphorylation and translocation in p110–/– ECs (Fig. 7, A and B), indicating that PI3K is required for TNF-induced activation phosphorylation and translocation of PKC. To determine whether p110 is required for TNF-induced PKC activation, PKC from control and TNF-treated WT and p110–/– MLVECs was used in an in vitro kinase assay. PKC activity was increased 7–8-fold after 5 min of TNF challenge (Fig. 7C), a response sustained for 30 min. In contrast, TNF did not increase PKC activity in ECs isolated from p110–/– mice (Fig. 7C), indicating that p110 is required for TNF-induced PKC activation.
p110 Regulates NF-B Activation and ICAM-1 Expression—Because p47phox membrane translocation is necessary for NADPH oxidase activation and oxidant signaling in ECs (47), we next determined the role for PI3K in signaling TNF-induced NF-B activation and ICAM-1 expression. MLVECs isolated from WT and p110–/– mice were challenged with TNF for various times, and the nuclear proteins were isolated. We observed that TNF induced NF-B DNA binding and ICAM-1 protein expression in WT ECs, whereas both responses were blocked in p110–/– ECs (Fig. 8).
Because ICAM-1 expression in ECs results in the firm adhesion of PMNs, we determined the role of p110 in the mechanism of TNF-induced ICAM-1 cell surface expression. We observed that TNF exposure of WT MLVECs induced ICAM-1 expression on the cell surface, whereas this was blocked in p110–/– MLCECs (Fig. 9A). Studies also determined the adhesion of WT mouse PMNs to WT or p110–/– MLVECs. We observed that TNF exposure of MLVECs induced adhesion of naïve PMNs to MLVECs, whereas adhesion of PMNs to p110–/– MLVECs similarly challenged with TNF was blocked (Fig. 9B).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
–/– PMNs failed to produce PIP3 (17) and activate Akt/protein kinase B (16) and showed impaired respiratory burst and motility responses (15). Because the described role of PI3K in immunity is largely restricted to leukocytes (15), in the present study we investigated the possibility that PI3K plays a potentially important role in regulating the immune response in the vessel wall, specifically the vessel wall lining ECs. We demonstrated that in response to stimulation with TNF, PI3K-induced oxidant signaling in ECs triggered by NADPH oxidase was critical in mediating the activation of NF-B and expression of ICAM-1. We also demonstrated that the mechanism of PI3K-induced NADPH oxidase activation involved the activation of the atypical PKC isoform, PKC, which has previously been shown to induce the phosphorylation of p47phox (28). This allows p47phox to associate with Nox2 (gp91phox) in the plasma membranes and thereby generate the active NADPH oxidase complex (28, 29).
|
in mediating NADPH oxidase activation in ECs, we have employed a combination of genetic and biochemical approaches. We first demonstrated that p110 mRNA and protein are indeed abundantly expressed in primary ECs, indicating that PI3K is not confined to the leukocytes as had been surmised. We also observed that TNF induced the membrane translocation of p110, indicative of PI3K activation, as well as phosphorylation of Ser241 on PDK1, the substrate of PI3K (42). The TNF-induced membrane localization of p110 followed the same time course as TNF-induced PIP3 production. Importantly, TNF-induced PDK1 phosphorylation and PIP3 production failed to occur in ECs from p110–/– mice, indicating the requirement of p110 in the mechanism of PI3K activation.
Studies have shown that phosphorylation of p47phox causes a conformational change that releases the complexed p47phox subunit and allows it to translocate to the membrane and associate with the membrane-bound Nox2 (47–49). We observed that TNF induced time-dependent phosphorylation and membrane association of p47phox, whereas these responses were absent in p110–/– MLVECs. We showed that the TNF-induced PKC Thr410 phosphorylation, and PKC membrane translocation and activation were also blocked in MLVECs from p110–/– mice. These findings demonstrate that p110 lies upstream of PKC, and p110 signals PKC activation.
|
induced the membrane translocation of PKC. We also observed that TNF increased the phosphorylated PKC in the membrane fraction of MLVECs, whereas the total amount of PKC recovered in membrane immunoprecipitates increased slightly. This finding suggests that the activated PKC is present in the membrane. Because the membrane association of PKC after TNF challenge decreased in a time-dependent manner, there is the possibility that PKC may have translocated back to the cytosol and degraded. However, it is important to note that the results showing the time course of the TNF-induced PIP3 production occurring between 15 and 30 min corresponds well with increased membrane association of phosphorylated PKC, further suggesting an important relationship between PI3K and PKC.
The role of the PI3K 
PKC signaling pathway in the mechanism of NADPH oxidase assembly in ECs seen in the present studies is consistent with the function of PIP3 in activating PKC (47, 51). The mechanism of PI3K activation of PKC involves PDK1, a kinase known to phosphorylate PKC (52). Studies have shown that PDK1 induced the activation of PKC in a PIP3-dependent manner and that PDK1 and PKC became associated in response to platelet-derived growth factor exposure (53). These findings are in agreement with our evidence that PI3K plays a crucial role in signaling PKC activation in ECs via PDK1.
|
–/– mice. SHIPs regulate the activation of several PKC and PI3K isoforms by controlling the phosphorylation of PIP3 (23). SHIPs mediate the inhibitory effect by localizing at sites of synthesis of PIP3 (46, 54). Because SHIP-2 was not associated with MLVEC membranes in the absence of p110, it appears that the PIP3 produced by PI3K is a requirement for the membrane association of SHIP-2. The failure of SHIP-2 to localize to the membrane in p110–/– ECs also raises the possibility that this may be another factor responsible for the impairment of PKC activation and phosphorylation of p47phox seen in p110–/– ECs.
To address the functional implications of PI3K in the mechanism of NF-B activation and ICAM-1 expression, we studied the effects of TNF in p110–/– MLVECs. Deletion of p110 prevented TNF-activated oxidant generation as well as NF-B activation and ICAM-1 expression. Because TNF mediates ICAM-1 expression in ECs and results in the adhesion of naïve PMNs to ECs, we determined whether PI3K-induced oxidant signaling is capable of promoting the binding of PMNs to MLVECs, the initial step in the transendothelial PMN migration response. We observed that adhesion of WT mouse PMNs to WT MLVECs was increased 
5-fold in response to 4 h of treatment with TNF; however, PMN failed to adhere to p110–/– MLVECs similarly challenged with TNF. This finding demonstrates that PI3K is essential for PMN adhesion to TNF-activated MLVECs; however, only a component of endothelial adhesivity activated by PI3K may be dependent on ICAM-1. Because E-selectin-mediated adhesion was also decreased in p110–/– mice (20), PI3K likely plays a critical role in the expression of multiple NF-B-regulated adhesive proteins, including ICAM-1 and E-selectin.
In summary, we show that PI3K is required for the TNF-induced oxidant generation in ECs, and it does so through activation of PKC. Data obtained using ECs from p110–/– mice showed that PKC was not activated in the absence of p110, and p47phox failed to translocate to EC membrane, thereby preventing NADPH oxidase-induced reactive oxidant species production. Deletion of p110 expression also prevented the TNF activation of NF-B and ICAM-1 expression, and the adhesion of PMNs to ECs. Because PI3K mediates endothelial adhesiveness through its ability to promote NADPH oxidase assembly required for the activation of NF-B and expression of ICAM-1, our results suggest a key role of PI3K in regulating vessel wall innate immune responses.
|
|
FOOTNOTES |
|---|
1 To whom correspondence should be addressed: Dept. of Pharmacology, University of Illinois College of Medicine, 835 S. Wolcott Ave., E403, Chicago, IL 60612. Tel.: 312-413-3428; Fax: 312-996-1225; E-mail. RFrey{at}uic.edu.
2 The abbreviations used are: PI3K, phosphatidylinositol 3-kinase; EC, endothelial cell; WT, wild type; MLVEC, mouse lung microvascular EC; PDK, phosphoinositide-dependent protein kinase; PKC, protein kinase C; PMN, polymorphonuclear leukocyte; TNF, tumor necrosis factor; SHIP, SH2-containing phosphatidylinositol phosphatase; ICAM, intercellular adhesion molecule; PIP3, phosphatidylinositol 3,4,5-trisphosphate; HPAEC, human pulmonary artery endothelial cell; PBS, phosphate-buffered saline; RT, reverse transcription; MS, mass spectrometry; ESI-MS, electrospray mass spectrometry; HBSS, Hanks' balanced salt solution.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
