Divergent Effects of p47phox Phosphorylation at S303-4 or S379 on Tumor Necrosis Factor-α Signaling via TRAF4 and MAPK in Endothelial Cells
Objectives—To define the mechanism of p47phox phosphorylation in regulating endothelial cell response to tumor necrosis factor-α (TNFα) stimulation.
Methods and Results—We replaced 11 serines (303-4, 310, 315, 320, 328, 345, 348, 359, 370, and 379) with alanines and investigated their effects on TNFα (100 U/mL, 30 minutes)–induced acute O2.− production and mitogen-activated protein kinase phosphorylation in endothelial cells. Seven constructs, S303-4A (double), S310A, S315A, S328A, S345A, S370A, and S379A, significantly reduced the O2.− production, and 4 of them (S328A, S345A, S370A, and S379A) also inhibited TNFα-induced extracellular-signal–regulated kinase (ERK) 1/2 phosphorylation. Blocking the phosphorylation of S303-4 and S379 inhibited most effectively TNFα-induced O2.− production. However, phosphorylation of S303-4 was not required for TNFα-induced p47phox membrane translocation and binding to TNF receptor–associated factor 4, ERK1/2 activation, and subsequent vascular cell adhesion molecule-1 expression. Knockout of p47phox or knockdown of TNF receptor–associated factor 4 using siRNA abolished TNFα-induced ERK1/2 phosphorylation, and inhibition of ERK1/2 activation significantly reduced the TNFα-induced vascular cell adhesion molecule-1 expression.
Conclusion—Phosphorylation of p47phox at different serine sites plays distinct roles in endothelial cell response to TNFα stimulation. Double serine (S303-4) phosphorylation is crucial for acute O2.− production, but is not involved in TNFα signaling through TNF receptor–associated factor 4 and ERK1/2. p47phox requires serine phosphorylation at distinct sites to support specific signaling events in response to TNFα.
- endothelial cells
- extracellular-signal–regulated kinase 1/2
- tumor necrosis factor-α
A multicomponent O2.−-generating NADPH oxidase 2 (Nox2) constitutively expressed in endothelial cells (EC) has been found to be a major source of endothelial reactive oxygen species (ROS) production in response to tumor necrosis factor-α (TNFα) stimulation.1 The Nox2 enzyme contains a cytochrome b558 (consisting of a Nox2 and a p22phox) and at least 4 regulatory subunits, including p40phox, p47phox, p67phox, and rac1. The phosphorylation of p47phox has been found to be a prerequisite for TNFα-induced endothelial ROS production, and knockout (KO) of p47phox severely compromises endothelial ROS response to TNFα stimulation.2–5
The p47phox has been shown to possess 11 serine phosphorylation sites, S303, S304, S310, S315, S320, S328, S345, S348, S359, S370, and S379, clustered within the carboxyl terminus, and flanked by basic amino acids.6 Phosphorylation of these serines neutralizes the strong positive charge in the region resulting in conformational changes to activate p47phox.6 Activated p47phox is able to bind to the p22phox/Nox2 complex and trigger O2.− production.6 Among these serines, the phosphorylation of double serines, S303-4 has been reported to weaken intramolecular interaction and to switch on the p47phox conformational change.7 However, the importance of these p47phox serines was mostly investigated in neutrophils or in cell-free systems. Little is known about their roles in the regulation of EC function. The p47phox also binds to signaling molecules, but the role of phosphorylation of the individual serine of p47phox in mediating TNFα signaling remains unknown.
The proinflammatory cytokine, TNFα, is a potent activator of endothelial Nox2 enzyme, and is involved in the pathogenesis of many diseases such as atherosclerosis, inflammation, and heart failure.8 TNFα signaling via its receptors requires the presence of TNF receptor–associated factors (TRAFs), which interact with downstream signaling molecules, such as mitogen-activated protein kinases (MAPKs) to mediate TNFα actions in cells.9 Previously, we and others have reported that p47phox is able to physically interact with the fourth member of the TRAF family (TRAF4) in response to TNFα stimulation of EC.10,11 However, the mechanism that controls p47phox interaction with TRAF4 and TNFα signaling remains poorly understood. In this study, we replaced 11 serines of human p47phox, S303, S304, S310, S315, S320, S328, S345, S348, S359, S370, and S379, with alanines that cannot be phosphorylated. We examined the effects of these mutations on TNFα-induced acute ROS production and signaling through TRAF4 and MAPKs, and the subsequent endothelial vascular cell adhesion molecule (VCAM)-1 expression. We report for the first time that phosphorylation of p47phox at different serine sites plays distinct roles in the regulation of TNFα-induced ROS productions and signaling via TRAF4 and MAPKs in EC.
Materials and Methods
Coronary Microvascular Endothelial Cell Isolation and Cell Culture
p47phox KO mice on a 129Sv background were obtained from the European Mouse Mutant Archive, and backcrossed to C57BL/6J for 10 generations. All studies were performed in accordance with protocols approved by the Home Office under the Animals (Scientific Procedures) Act 1986 UK. Coronary microvascular endothelial cells (CMEC) were isolated from the hearts of 10- to 12-week-old p47phox KO mice as described previously12 and used at passage 2. A mouse lymph node microvascular endothelial cell line (SVEC4-10) was obtained from the American Type Culture Collection. COS7phox cells with stable expression of Nox2 and p22phox were kindly provided by Dr M. C. Dinauer (Washington University School of Medicine, St. Louis, MO). The COS7phox cells were cotransfected with p67phox plus 1 of the p47phox mutations to examine the effects of p47phox mutations as described previously.13
Human p47phox cDNA (GenBank: AF330627.1), a kind gift from Dr F. Wientjes (University College London, UK), was cloned into pcDNA3.1/Zeo vector and used for the site-directed mutagenesis. Site-directed mutagenesis was performed according to the manufacturer’s instruction using the QuikChange Multi Site-Directed Mutagenesis kit (Agilent Technologies).
Gene Transfection and TNFα Stimulation
The gene transfection was performed as described previously.14 Transfection efficiency was 65% to 70% as checked using β-galac-tosidase reporter plasmid. After 48 hours of transfection, cells were treated with either vehicle or TNFα (100 U/mL) for 30 minutes in 5% FCS/DMEM and harvested for further experiments. For investigating VCAM-1 expression, cells were stimulated with TNFα for 12 hours.
Preparation of Membrane Fraction
Cellular membrane fraction was prepared as described previously.10 The nuclear, mitochondrial, and submitochondrial fractions were removed by differential ultracentrifugation. The membrane-enriched fraction was then collected by centrifugation for 60 minutes at 100 000g. After washing, the final pellets were resuspended in buffer and used as the membrane fractions.
Measurement of ROS Production
Endothelial O2.− production was measured using 3 independent assays. (1) NADPH (100 μmol/L)-dependent O2.− production by EC homogenates measured by lucigenin (5 μmol/L) chemiluminescence. The specificity of the assay was confirmed by adding superoxide dismutase (200 U/mL). (2) The O2.− production in intact EC was detected using dihydroethidium (25 μmol/L) fluorescence.15,16 Superoxide-specific product 2-hydroxyethidium was then detected at 580 nm (emission) and 480 nm (excitation) and measured using a JASCO high-performance liquid chromatography system. (3) The intracellular ROS production by adherent cells was detected by 2,7-dichlorodihydrofluorescein fluorescence microscopy. Tiron (10 mmol/L), a cell membrane–permeable O2.− scavenger, was used to verify the detection of O2.−.
Immunoprecipitation and Immunoblotting
Immunoblotting experiments were performed as described previously.17 The protein extract from human phagocytic U937 cells after stimulation with phorbol myristate acetate was used as a positive control for the detection of NADPH oxidase subunits, and α-tubulin detected in the same sample was used as loading control. Coimmunoprecipitation of p47phox with TRAF4 was performed as described previously.18 Subsequent immunoblotting was performed using either phosphoserine monoclonal antibody for p47phox phosphorylation or antibody to TRAF4 or p47phox for their association. For the quantification of phos-p47phox or TRAF4 coimmunoprecipitated down with p47phox, the levels of total p47phox immunoprecipitated down were first checked by prerun, and the final gel loading was calculated and justified to achieve equal amount of total p47phox between samples.
Immunofluorescence Confocal Microscopy
Cell preparation and immunofluorescence microscopy were performed as described previously using cells cultured onto 4-well chamber slides.10 Antibody binding was detected by extravidin-fluorescein isothiocyanate (green) or streptavidin-Cy3 (red). Normal rabbit or goat IgG (5 µg/mL) was used instead of primary antibody as negative controls. Images were acquired at 0.5 µm using a Zeiss LS510 confocal microscopy system.
Data were presented as means±SD of results taken from at least 3 independent cell cultures per condition. In the case of CMEC, each isolation used 6 mice per group, and the data presented were the mean from at least 3 isolations. Comparisons were made by 1-way ANOVA with Bonferroni test analysis, and P<0.05 was considered statistically significant.
Distinct Roles of p47phox Phosphorylation at Different Serine Sites in TNFα-Induced O2.− Production and ERK1/2 Activation
We replaced 11 serines with alanines, ie, S303-4A (double), S310A, S315A, S320A, S328A, S345A, S348A, S359A, S370A, and S379A, to block the phosphorylation sites, and examined their effects on TNFα-induced NADPH-dependent O2.− generation and extracellular-signal–regulated kinase (ERK) 1/2 phosphorylation in SVEC4-10 cells after gene transfection. Compared with vector-transfected controls, there was no significant difference between cells transfected with different constructs in their basal (without TNFα) levels of O2.− production and ERK1/2 phosphorylation (Figure 1, top). However, when the cells were stimulated with TNFα (100 U/mL, for 30 minutes), 7 mutations, S303-4A, S310A, S315A, S328A, S345A, S370A, and S379A, reduced significantly (P<0.05), whereas others, S320A, S348A, and S359A, had no significant effect on the levels of O2.− production (Figure 1A, left middle) compared with vector-transfected controls. TNFα-induced O2.− production was abolished in the presence of superoxide dismutase which confirmed the assay specificity (Figure 1A, lower left). Four mutations, S328A, S345A, S370A, and S379A, inhibited significantly (P<0.05), whereas the others had no significant effect on ERK1/2 phosphorylation (Figure 1B, middle and bottom). The double mutation, S303-4A, showed the greatest inhibitory effect on TNFα-induced O2.− production, but had no significant effect on acute ERK1/2 phosphorylation, whereas S379A was the most effective mutation to inhibit both.
Divergent Effects of S303-4A and S379A on TNFα-Induced O2.− Production and MAPK Phosphorylation
To further investigate the differences between S303-4A and S379A, we generated a triple mutation, S303-4A plus S379A (S303-4/379A), and examined 3 mutations in parallel on TNFα-induced acute O2.− production using 3 independent methods, ie, lucigenin chemiluminescence in cell homogenates (Figure 2A), dihydroethidium fluorescence in intact cells measured by high-performance liquid chromatography (Figure 2B), and dichlorodihydrofluorescein fluorescence in intact adherent EC detected by fluorescence microscopy (Figure II in the online-only Data Supplement). Compared with vehicle-treated cells, TNFα significantly increased the levels of O2.− production by cells transfected with either vector or wild-type (WT) p47phox cDNA, but not in cells transfected with S303-4A, S379A, or S303-4/379A mutations (Figure 2A and 2B). However, when we looked at the levels of TNFα-induced MAPK phosphorylation (Figure 2C), we found significant increases in the levels of ERK1/2, and p38MAPK phosphorylation in cells transfected with empty vector, WT p47phox or S303/4A, but not in cells transfected with S379A or S303-4/379A. The phos-jun N-terminal kinase bands were weak and could be seen only in cells transfected with vector or WT p47phox. The levels of total ERK1/2, p38MAPK, and jun N-terminal kinase remained without significant changes after TNFα stimulation.
The potential effects of S303-4A and S379A on the expression of Nox2 subunits, ie, p47phox, p40phox, p67phox, p22phox, Nox2, and rac1, were examined by immunoblot (Figure III in the online-only Data Supplement). Compared with vector-transfected controls, the p47phox expression was significantly increased in cells transfected with either WT p47phox or p47phox mutations, which confirmed the transfection efficiency. There was no significant effect of these mutations on the levels of expression of other Nox2 subunits except a slight but significant decrease in p40phox expression in cells transfected with WT p47phox and p47phox mutations, which may be due to the compensation between the p47phox and p40phox.17 We also examined the mRNA levels of Nox1, Nox2, and Nox4 by quantitative real-time polymerase chain reaction (Figure IV in the online-only Data Supplement) and found no significant effect of p47phox mutations on the levels of these Noxes. However, TNFα stimulation induced significant increases in the mRNA levels of Nox2 but not Nox1 or Nox4.
Opposing Effects of S303-4A and S379A on TNFα-Induced p47phox Membrane Translocation and Binding to TRAF4
p47phox phosphorylation and membrane translocation have been reported as critical steps in EC ROS response to TNFα stimulation.10 To investigate this, we prepared the EC membrane fractions and examined the p47phox membrane translocation (Figure 3A) and NADPH-dependent O2.− production (Figure 3B). In cells transfected with control vector or WT p47phox, TNFα significantly increased the levels of p47phox membrane translocation and O2.− production in the membrane fractions. Compared with WT p47phox-transfected cells, S303-4A had no inhibitory effect on p47phox membrane translocation, but significantly inhibited TNFα-induced O2.− production, whereas S379A or triple mutation S303-4/379A inhibited both (Figure 3A and 3B). We then examined potential effects of S303-4A and S379A on TNFα-induced p47phox association with TRAF4 by confocal immunofluorescence (Figure 3C). The p47phox was labeled by a rabbit polyclonal antibody and detected by fluorescein isothiocyanate (green color), and TRAF4 was labeled by a goat polyclonal antibody and detected by Cy3 (red color). In vehicle-treated vector-transfected cells, p47phox and TRAF4 were mainly detected in the perinuclear region showing an eccentric distribution pattern. There was some colocalization of the 2 molecules mainly in the perinuclear region as indicated in yellow in the merged image. TNFα induced p47phox and TRAF4 plasma membrane translocation and association in vector-transfected cells and the distribution was even across the cell. Compared with vector-transfected cells, S303-4A did not inhibit p47phox and TRAF4 association around the plasma membrane. However, the distribution pattern was patchy and uneven across the cell. S379A completely inhibited p47phox membrane translocation and association with TRAF4.
The opposing effect of S303-4A versus S379A on TNFα-induced p47phox membrane translocation and association with TRAF4 was further examined by coimmunoprecipitation. Total p47phox in the membrane fractions was immunoprecipitated down and detected for serine phosphorylation using a phos-serine antibody, or for the presence of TRAF4 using an antibody to TRAF4. To compare the levels of phos-p47phox or TRAF4 coimmunoprecipitated down with p47phox, the gel loading was justified according to the pre-run results to achieve an equal level of total p47phox between samples. In the parallel experiments, TRAF4 was immunoprecipitated down and detected for the presence of p47phox, and the total TRAF4 detected from the same sample was used as loading controls (Figure 3D). Compared with vehicle-treated cells, TNFα stimulation significantly increased the levels of p47phox phosphorylation and association with TRAF4 in cells transfected with control vector or WT p47phox or S303-4A but not in cells transfected with S379A or triple mutation S303-4/379A.
Effects of TRAF4 siRNA on TNFα-Induced ROS Production and ERK1/2 Phosphorylation
To ascertain whether TRAF4 played a key role in the ERK1/2 phosphorylation observed in cells transfected with S303-4A, we knocked down TRAF4 with specific siRNA, and examined the O2.− production (Figure 4A) and ERK1/2 phosphorylation (Figure 4B). Compared with cells treated with scrambled siRNA, knockdown of TRAF4 by siRNA did not inhibit TNFα-induced O2.− production. However, knockdown of TRAF4 completely abolished TNFα-induced acute ERK1/2 phosphorylation. The successful knockdown of TRAF4 protein was confirmed by Western blot. The levels of total ERK1/2 were unchanged in all samples.
TNFα-Induced ROS Production and ERK1/2 Phosphorylation in p47phox KO CMEC and COS7phox Cells
The distinct effects of S303-4A and S379A on TNFα-induced acute ROS production and ERK1/2 phosphorylation were further examined using CMEC isolated from p47phox KO mice (Figure 5, left) and p47phox-deficient COS7phox cells (Figure 5, right). Successful transfection was confirmed by p47phox Western blot (Figure 5A, left). Compared with vehicle-treated cells, TNFα significantly increased the O2.− production by p47phox KO CMEC or COS7phox cells transfected with WT p47phox, but not in cells transfected with empty vector, or S303-4A or S379A (Figure 5A). TNFα-induced ERK1/2 phosphorylation was seen in cells transfected with WT p47phox or S303-4A, but not in cells transfected with vector or S379A (Figure 5B).
The role of protein kinase C in TNFα-induced p47phox phosphorylation was investigated by preincubating CMEC with a pan protein kinase C inhibitor bisindolylmaleimide I (10 µmol/L for 30 minutes). Total p47phox was immunoprecipitated down from cell homogenates and detected for the levels of serine phosphorylation (Figure 5C). No p47phox phosphorylation was detected in p47phox KO CMEC transfected with an empty vector. TNFα significantly increased the levels of p47phox phosphorylation in cells transfected with WT p47phox and S303-4A, and this was significantly inhibited in the presence of bisindolylmaleimide I.
Effects of S303-4A and S379A on TNFα-Induced p47phox Association With TRAF4, p47phox Membrane Translocation, and VCAM-1 Expression in p47phox KO CMEC
The difference between S303-4A and S379A in TNFα-induced p47phox association with TRAF4 was further examined in p47phox KO CMEC after gene transfection and TNFα (30 minutes) stimulation. Total p47phox was immunoprecipitated down and detected for the presence of TRAF4 (Figure 6A). Compared with cells transfected with an empty vector, TRAF4 was detected in the p47phox immunoprecipitates of cells transfected with WT p47phox and S303-4A mutation, and this was significantly inhibited in cells transfected with S379A.
The pathophysiological significance of S303-4A and S379A in TNFα-induced endothelial dysfunction was investigated by looking at the VCAM-1 membrane expression in relationship to p47phox membrane translocation in p47phox KO CMEC. Membrane fractions were prepared from p47phox KO CMEC after gene transfection and 12 hours of TNFα stimulation (Figure 6B). Compared with cells transfected with an empty vector, p47phox membrane translocation and VCAM-1 expression were detected in p47phox KO cells transfected with WT p47phox or S303-4A, and these were significantly inhibited in cells transfected with S379A. The relationship between the levels of ERK1/2 phosphorylation and VCAM-1 expression was examined in the whole cell homogenates of p47phox KO cells after 12 hours of TNFα stimulation (Figure 6C). Compared with vector-transfected cells, VCAM-1 expression was significantly increased in cells transfected with WT p47phox. TNFα-induced VCAM-1 expression was significantly reduced in the presence of an inhibitor of ERK1/2 activation (U0126), and this was further confirmed by confocal microscopy. VCAM-1 was labeled by fluorescein isothiocyanate (green), and the nuclei were labeled by propidium iodide (PI, red) (Figure 6D).
It is well established that the phosphorylation of serine residues in the polybasic region at the carboxyl terminus of p47phox plays an important role in Nox2 assembly and activation. However, very few studies have looked at the contribution of individual serine in the regulation of endothelial function, in particular in endothelial response to TNFα stimulation. Here, we reported that the phosphorylation of p47phox at different serine sites plays distinct roles in regulating endothelial response to TNFα stimulation. We replaced 11 serines with alanines to block the phosphorylation sites, and found that the phosphorylation of serines 303-4 (double), 310, 315, 328, 345, 370, and 379 was required for EC acute O2.− response to TNFα stimulation. However, not all serines were involved in endothelial O2.− production, and substitution of S320, S348, and S359 with alanines had no significant effects on the levels of O2.− production. Our finding is in line with a previous study showing that phosphorylation of some serines was unnecessary, or even inhibitory to Nox2 activation, although the inhibitory serine sites were not identified in this previous study.19
MAPKs, in particular ERK1/2, are crucial to TNFα signaling pathways, and the time course of TNFα-induced ERK1/2 activation bears a remarkable resemblance to the time course of O2.− production in EC.10,20 It has been shown previously that phosphorylation of p47phox is a key step in TNFα-induced MAPK activation, and KO of p47phox abolishes MAPK response to TNFα stimulation.10,20 In the current study, we provide further evidence that the phosphorylation of 4 serine residues, S328, S345, S370, and S379, of p47phox is involved in mediating acute ERK1/2 response to TNFα stimulation. However, we find that the phosphorylation of double serines (S303-4), which is critical for O2.− production, is not involved in TNFα-induced acute ERK1/2 activation. The differential effects of S303-4A on TNFα-induced O2.− production and ERK1/2 phosphorylation imply that O2.− production is not necessary for p47phox to mediate ERK1/2 response to TNFα stimulation in EC, and that another mechanism may be involved.
The TRAF family contains 6 scaffold proteins that link the TNF receptors to signaling cascades including the MAPKs.21,22 TRAF4 is predominately intracellularly located but can translocate and bind to the TNF receptors through its C-TRAF domain.23 Very recently, TRAF4 has been found to be a highly mobile shuttle protein and to potentiate ERK1/2 phosphorylation in proliferating epithelial cells.24 TRAF4 and p47phox can interact with each other through their C-terminals,11 and the phosphorylation of p47phox promotes its association with TRAF4 and ERK1/2 activation.10 In the current study, we find that phosphorylation of S303-4 is not required for p47phox and TRAF4 interaction and ERK1/2 activation in response to TNFα stimulation. The crucial role of TRAF4 in TNFα-induced ERK1/2 phosphorylation was clearly demonstrated by knockdown of TRAF4 using siRNA.
The inhibitory effects of S303-4A and S379A on TNFα-induced O2.− production have been evaluated using 3 independent methods, ie, lucigenin chemiluminescence, dihydroethidium fluorescence high-performance liquid chromatography, and dichlorodihydrofluorescein fluorescence microscopy. The results confirm that abolishing the phosphorylation of S303-4 or S379 indeed severely compromises acute O2.− production by EC. However, there is a distinct difference between S303-4A and S379A phosphorylation on acute TNFα-induced p47phox membrane translocation, binding to TRAF4 and subsequent ERK1/2 phosphorylation in EC. For example, abolishing S303-4 phosphorylation has no significant effect, whereas elimination of S379 phosphorylation led to an immediate failure of p47phox membrane translocation and association with TRAF4 and ERK1/2 phosphorylation. The difference between S303-4A and S379A has been further confirmed using p47phox KO CMEC and the p47phox-deficient COS7phox cells. Interestingly, phosphorylation of S379 had been shown previously to induce a switch between the 2 SH3 domains of the p47phox, which is a key step required for p47phox membrane translocation and interactions with other proteins, and single substitution of S379 almost abolished leukocyte Nox2 activity.25,26 Our data provide novel evidence of a key role for S379 phosphorylation in regulating the action of p47phox in EC.
Vascular inflammation is a critical step in the development of cardiovascular disorders such as atherosclerosis and heart failure. Leukocyte adhesion is primarily mediated by endothelial surface expression of adhesion molecules, ie, VCAM-1. Although the levels of VCAM-1 expression have been suggested to be associated with the levels of ROS generation,27 in the current study we find that blocking S303-4 phosphorylation inhibits O2.− production, but has no significant effect on endothelial VCAM-1 expression. The pathophysiological relevance of this is that increased O2.− production is not a prerequisite for p47phox to mediate TNFα signaling, VCAM-1 expression, and acute inflammation, which may imply that nonspecific antioxidant therapy is not effective in those conditions. However, an inhibitor of ERK1/2 activation, such as U0126, significantly reduces TNFα-induced VCAM-1 expression. An inhibitor of protein kinase C, such as bisindolylmaleimide I (a pan protein kinase C inhibitor), completely inhibits TNFα-induced p47phox phosphorylation in p47phox KO CMEC transfected with WT p47phox or S303-4A or S379A.
The double serines 303-4 are located within the p47phox intramolecular SH3-binding site (amino acids 286–314) and are extensively phosphorylated during Nox2 activation in neutrophils.5 It has been shown that the oxidase is still active if only 1 serine 303 or 304 is converted to alanine.26 However, if both serines 303 and 304 are converted to alanines, Nox2 activity is abolished but it does not affect the phosphorylation of the remaining serines of p47phox and its plasma membrane translocation.28,29 Therefore, phosphorylation of S303-4 has been suggested to disrupt p47phox intramolecular inhibition and allow p22phox binding.5 This also explains why S303-4 phosphorylation is crucial for EC ROS production. However, we find that S303-4 phosphorylation is not necessary for TNFα-induced acute ERK1/2 activation and VCAM-1 expression. It is possible that TNFα-induced p47phox membrane translocation and association with TRAF4 happen before the phosphorylation of S303-4, and the interaction between p47phox/TRAF4 is sufficient to induce acute ERK1/2 activation and VCAM-1 expression. More detailed investigation is required to fully discover the role of p47phox in TNFα-induced endothelial VCAM-1 expression and inflammation.
In conclusion, we report for the first time that phosphorylation of p47phox at different serine residues plays distinct roles in mediating TNFα-induced endothelial acute O2.− production and signaling. Phosphorylation of S379 is necessary for acute O2.− production, ERK1/2 activation, and subsequent VCAM-1 expression. However, phosphorylation of S303-4 is essential for acute O2.− production, but not for p47phox membrane translocation, binding to TRAF4, ERK1/2 activation, and subsequent VCAM-1 expression. p47phox requires serine phosphorylation at distinct sites to support specific signaling events in response to TNFα.
Sources of Funding
This work is supported by the Wellcome Trust (grant 07863/Z/05/Z).
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.112.247775/-/DC1.
- Received November 30, 2011.
- Accepted March 14, 2012.
- © 2012 American Heart Association, Inc.
- Pagano PJ
- Li JM,
- Mullen AM,
- Yun S,
- Wientjes F,
- Brouns GY,
- Thrasher AJ,
- Shah AM
- Park L,
- Anrather J,
- Zhou P,
- Frys K,
- Wang G,
- Iadecola C
- Ago T,
- Nunoi H,
- Ito T,
- Sumimoto H
- Mathew SJ,
- Haubert D,
- Krönke M,
- Leptin M
- Li JM,
- Fan LM,
- Christie MR,
- Shah AM
- Xu YC,
- Wu RF,
- Gu Y,
- Yang YS,
- Yang MC,
- Nwariaku FE,
- Terada LS
- Price MO,
- McPhail LC,
- Lambeth JD,
- Han CH,
- Knaus UG,
- Dinauer MC
- Fernandes DC,
- Wosniak J Jr,
- Pescatore LA,
- Bertoline MA,
- Liberman M,
- Laurindo FR,
- Santos CX
- Zhao H,
- Joseph J,
- Fales HM,
- Sokoloski EA,
- Levine RL,
- Vasquez-Vivar J,
- Kalyanaraman B
- Fan LM,
- Teng L,
- Li JM
- Li JM,
- Shah AM
- Chandel NS,
- Schumacker PT,
- Arch RH
- Massenet C,
- Chenavas S,
- Cohen-Addad C,
- Dagher MC,
- Brandolin G,
- Pebay-Peyroula E,
- Fieschi F
- Inanami O,
- Johnson JL,
- McAdara JK,
- Benna JE,
- Faust LR,
- Newburger PE,
- Babior BM
- Johnson JL,
- Park JW,
- Benna JE,
- Faust LP,
- Inanami O,
- Babior BM