Oxidative Stress Reprograms Lipopolysaccharide Signaling via Src Kinase-dependent Pathway in RAW 264.7 Macrophage Cell Line*

Oxidative stress generated during ischemia/reperfusion injury has been shown to augment cellular responsiveness. Whereas oxidants are themselves known to induce several intracellular signaling cascades, their effect on signaling pathways initiated by other inflammatory stimuli remains poorly elucidated. Previous work has suggested that oxidants are able to prime alveolar macrophages for increased NF-κB translocation in response to treatment with lipopolysaccharide (LPS). Because oxidants are known to stimulate the Src family of tyrosine kinases, we hypothesized that the oxidants might contribute to augmented NF-κB translocation by LPS via the involvement of Src family kinases. To model macrophage priming in vitro, the murine macrophage cell line, RAW 264.7, was first incubated with various oxidants and then exposed to low dose LPS. These studies show that oxidant stress is able to augment macrophage responsiveness to LPS as evidenced by earlier and increased NF-κB translocation. Inhibition of the Src family kinases by either pharmacological inhibition using PP2 or through a molecular approach by cell transfection with Csk was found to prevent the augmented LPS-induced NF-κB translocation caused by oxidants. Interestingly, while Src kinase inhibition was able to prevent the LPS-induced NF-κB translocation in oxidant-treated macrophages, this strategy had no effect on NF-κB translocation caused by LPS in the absence of oxidants. These findings suggested that oxidative stress might divert LPS signaling along an alternative signaling pathway. Further studies demonstrated that the Src-dependent pathway induced by oxidant pretreatment involved the activation of phosphatidylinositol 3-kinase. Involvement of this pathway appeared to be independent of traditional LPS signaling. Together, these studies provide a novel potential mechanism whereby oxidants might prime alveolar macrophages for altered responsiveness to subsequent inflammatory stimuli and suggest different cellular targets for immunomodulation following ischemia/reperfusion.

Civilian trauma remains a leading cause of disability and mortality in North American society (1). Late deaths are related to the development of multiple organ failure in up to two-thirds of these patients. Many studies have suggested that the global ischemia-reperfusion resulting from resuscitated hemorrhagic shock predisposes to organ injury by priming for an exaggerated immune response to a delayed inflammatory stimulus (2). The synergistic effect of sequential stimuli has been coined the "two-hit" hypothesis for the development of organ injury in trauma patients (2). We have previously reported an animal model wherein antecedent resuscitated shock was shown to prime for increased LPS 1 -induced lung injury by augmenting alveolar macrophage NF-B nuclear translocation and gene transcription of the chemokine, cytokine-induced neutrophil chemoattractant (CINC), the rodent orthologue of interleukin-8. The exaggerated generation of CINC by alveolar macrophages in this model was primarily responsible for increased lung neutrophilia and consequent lung injury in animals exposed to the combined stimuli of resuscitated shock followed by LPS (3).
Ischemia/reperfusion produces oxidative stress through the irreversible conversion of xanthine dehydrogenase to xanthine oxidase (XO) during ischemia with consequent production of oxygen-free radicals and H 2 O 2 during reperfusion (4,5). Several lines of evidence suggest a role for oxidative stress as contributing to activation of hematopoietic cells. In the in vitro setting, oxidants have been shown to induce activation of multiple signaling pathways in hematopoietic cells. The best studied of these involves signaling cascades culminating in NF-B translocation and induction of transcription of NF-B-dependent genes. Early reports documented the ability of oxidants to induce NF-B translocation and also defined the role of oxidant stress in NF-B translocation by various inflammatory stimuli including LPS and tumor necrosis factor. In vivo studies have generally supported a role for oxidative stress in the induction of NF-B translocation following ischemia/reperfusion. Schwartz et al. (6) reported that inhibition of xanthine oxidase generation or activity using a tungsten-enriched diet or allopurinol, respectively, prevented induction of NF-B-dependent gene expression such as tumor necrosis factor in a murine hemorrhagic shock model. The potential priming role of oxidant stress generated by ischemia/reperfusion is well demonstrated in our previous reports. In these studies, neither shock/ resuscitation nor low dose LPS was able to induce significant NF-B translocation in alveolar macrophages, whereas the combined presence of these two stimuli culminated in a robust response, far greater than the additive effect of the separate responses. Importantly, supplementation of resuscitation fluid with the antioxidant N-acetylcysteine prevented the synergistic rise in NF-B translocation (3). These observations are consistent with a role for oxidative stress in the priming of macrophages for increased responsiveness to a delayed second inflammatory stimulus. However, whereas oxidative stress is known to activate a number of signaling pathways in inflammatory cells, the alterations in cell signaling leading to enhanced responsiveness following exposure to oxidants have not been elucidated.
The Src family of tyrosine kinases are involved in many signal transduction pathways. The Src tyrosine kinases have been shown to be activated by oxidant stress such as UV irradiation as well as H 2 O 2 (7,8). Relevant to the activation of NF-B and induction of NF-B-dependent proinflammatory genes, Src has also been shown to phosphorylate inhibitory B (IB) on tyrosine residues, an event that activates NF-B without inducing IB degradation itself (9). In the present studies, we tested the hypothesis that oxidant stress might contribute to augmented NF-B translocation in response to LPS through involvement of members of the Src family kinases. To examine this hypothesis, an in vitro culture system was established wherein macrophages were sequentially exposed to an oxidative stress followed by LPS as a means of modeling the in vivo "two-hit hypothesis" of cellular activation and organ injury following shock/resuscitation. The major findings in the present studies are that Src family kinases are not only involved in macrophage priming by oxidants, but also that antecedent oxidant stress reprograms LPS-NF-B signaling such that it changes from a Src-independent pathway to one that is Src-dependent.

EXPERIMENTAL PROCEDURES
Buffers and Reagents-Hydrogen peroxide (30%) (H 2 O 2 ) was purchased from University of Toronto Medstore (Toronto, ON). Antibodies against p-Lyn, c-Fgr, p-Hck, p-phosphoinositol (PI) 3-kinase, and IB␣ were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The polyclonal anti-IB␣, anti-phosphotyrosine-specific monoclonal antibody (G410), PI 3-kinase antibody (clone UB93-3), and an in vitro tyrosine kinase assay kit were purchased from Upstate Biochemical Institute (Lake Placid, NY). An inhibitor of the Src family of proteintyrosine kinases, PP2, was purchased from Calbiochem, Inc. (San Diego, CA). The PI substrate was purchased from Avanti Polar Lipids, Inc. (Alabaster, AL). The Protein A-Sepharose beads, horseradish peroxidase-conjugated donkey anti-rabbit IgG secondary antibody, and ECL were purchased from Amersham Biosciences. The 50ϫ protease inhibitor mixture was purchased from BD Pharmingen (San Jose, CA). The TLC plates and ATP were obtained from Sigma. Endotoxin-free Dulbecco's modified Eagle's medium and Hanks' balanced salt solution were purchased from Invitrogen; fetal calf serum was from Hyclone.
Cell Preparation-The murine macrophage cell line, RAW 264.7 (ATCC), was cultured in Dulbecco's modified Eagle's medium supplemented with 100 units/ml penicillin, 100 g/ml streptomycin (P/S), and 10% fetal calf serum. Cells were maintained at 37°C in a humidified atmosphere of 5% CO 2 .
Cell Activation-Experiments were carried out in Hanks' balanced salt solution, 2% fetal calf serum, 2 mM glucose. The cells were exposed to XO (50 milliunits/ml), LPS (0.1 g/ml), or a combination of pretreatment with XO for 2 h and then LPS (Escherichia coli O111:B4) treatment ranging from 0.5 min to 1 h. Alternatively, the oxidant H 2 O 2 (100 M) was used instead of the XO prior to LPS exposure. Reactions were stopped by placing the cells on ice. In inhibition studies, RAW cells were preincubated in the presence of 10 M PP2 for 10 min at 37°C.
Transient Transfections-Twenty-four hours after plating the cells on culture dishes, cotransfections were performed using green fluorescent protein (GFP) pFR-hrGFP plasmid (Stratagene) and Csk plasmid. The Csk plasmid was a kind gift from Dr. I. Mucsi (Semmelweis University, Budapest, Hungary). The full-length coding sequence of murine Csk was subcloned into pcDNA3 vector using HindIII and XbaI restric-tion enzymes. Transient transfection with the corresponding plasmids was performed using Superfect reagent (Qiagen Ltd.) according to the manufacturer's instructions. Routinely, cells were transfected with 4 g of plasmid DNA per well (for 6-well plates). The DNA was diluted in OptiMEM medium (Invitrogen). The ratio of total plasmid DNA to Superfect reagent was 4 g to 20 l, respectively. After 3 h of incubation the medium containing the remaining complexes was removed and replaced with fresh growth medium with 10% fetal calf serum and P/S. Stable Transfection of Cell Line-Stable transfection of RAW 264.7 cell lines expressing Csk were kindly provided by Dr. Z. Honda (University of Tokyo, Japan) (10). The cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum, P/S, and neomycin antibiotic G418 (Invitrogen).
Northern Blot Analysis-Total RNA was extracted from cells using the guanidium-isothiocyanate method (11). Hybridization was conducted using a [␥-32 P]dATP end-labeled 30-base oligonucleotide probe for CINC with the sequence 5Ј-GCGGCATCACCTTCAAACTCTGGAT-GTTCT-3Ј, which is complementary to nucleotides 134 -164 of CINC cDNA (kindly provided by Dr. Timothy S. Blackwell, Vanderbilt University School of Medicine, Nashville, TN) (12). Blots were then washed under conditions of high stringency, and specific mRNA bands were detected by autoradiography in the presence of intensifying screens. Blots were stripped and reprobed for glyceraldehyde-3-phosphate dehydrogenase, which is a ubiquitously expressed housekeeping gene to control for loading. Expression of mRNA was quantitated using a Phos-phorImager and accompanying ImageQuant software (Amersham Biosciences), and was normalized to the glyceraldehyde-3-phosphate dehydrogenase signal.
Immunoprecipitation-After treatments, cells were lysed for 5 min on ice with buffer containing 100 mM NaCl, 30 mM Hepes, 20 mM NaF, 1 mM EGTA, and 1% Triton X-100 to which protease mixture inhibitor had been added. Cell lysates were then centrifuged at 13,000 ϫ g for 10 min, and the supernatants were then collected and precleared with protein A-Sepharose beads. Equal amounts of cellular proteins were immunocomplexed with 2 g of rabbit polyclonal anti-IB␣, anti-Hck, anti-Fgr, or anti-Lyn antibody for 1 h rotating at 4°C. Subsequently, 30 g of protein A-Sepharose beads were added to the samples and incubated at 4°C for 1 h. The resulting immune complexes were washed three times with the same lysis buffer and then resuspended in 25 l of Laemmli buffer.
Western Blot Analysis-Cells were pelleted and lysed in ice-cold cell lysis buffer. Whole cell lysates were prepared with 2ϫ Laemmli, 0.1 M dithiothreitol buffer followed by immediate boiling at 100°C for 5 min. Cytosolic fractions were isolated with 1% Triton X-100, 150 mM NaCl, 10 mM Tris-HCl (pH 7.4), 2 mM sodium orthovanadate, 10 g/ml leupeptin, 50 mM NaF, 5 mM EDTA, 1 mM EGTA, and 1 mM phenylmethylsulfonyl fluoride. Postnuclear supernatants were collected following centrifugation at 13,000 ϫ g for 5 min and diluted with 2ϫ Laemmli buffer, 0.1 M dithiothreitol followed by immediate boiling for 5 min. Lysates prepared from 100,000 cells were separated on 12.5% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA). Blots were then probed with 1:1000 dilution of anti-phosphotyrosine-specific monoclonal antibody (G410), anti-IB␣ antibody, or anti-phosphospecific antibody for 1 h at room temperature. Following incubation with the horseradish peroxidase-conjugated donkey IgG secondary antibody at 1:3000 dilution for 1 h at room temperature, blots were developed using an ECL-based system.
Tyrosine Kinase Assay-Immunoprecipitation of Hck was performed as previously described. The tyrosine kinase assay kit (non-radioactive) was used. Biotinylated substrate peptide containing poly(Glu 4 -Tyr) was incubated with the tyrosine kinase, in the presence of non-radioactive ATP. PP2 (10 M) was added to the kinase reaction. The reaction was then placed in a microplate coated with streptavidin. Horseradish peroxidase-conjugated monoclonal anti-phosphotyrosine antibody was used to detect the phosphorylated substrate. Chemiluminescence was used for colorimetric detection with a spectrophotometric plate reader set at wavelength of 450 nm. Reference wells were used to determine the amount of peptide/well. PI 3-Kinase Activity Assay-After treatments, cells were lysed with 600 l of ice-cold lysis buffer containing 100 mM NaCl, 30 mM Hepes, 20 mM NaF, 1 mM EGTA, and 1% Triton X-100 to which protease mixture inhibitor had been added. Samples were left on ice for 10 min then cellular debris was removed by centrifugation at 13,000 ϫ g for 10 min at 4°C. The lysates were further incubated overnight with 5 l of PI 3-kinase antibody at 4°C and then coupled to 20 l of protein A-Sepharose beads for 2 h at 4°C on a rotator. The immunoprecipitates were washed three times with buffer containing 50 mM Hepes (pH 7.25), 5 mM EDTA, 5 mM MgCl 2 , and 0.1 M Na 3 VO 4 . Reaction was then performed as previously described (13). Briefly the immunoprecipitates were resuspended in 30 l of the same buffer to which 1ϫ protease mixture inhibitor has been added in a reaction mixture containing the 30 l of beads in 70 M ATP, 20 Ci of [␥-32 P]ATP, and 5 g of L-␣-phosphatidylinositol, mixed gently, and incubated for 20 min at 37°C. The reaction was stopped with 20 l of 6 N HCl and the organic layer was extracted with 160 l of CHCl 3 /CH 3 OH (1:1) and separated on a silica gel thin layer chromatography plate in solvent containing CHCl 3 /CH 3 OH/H 2 O/NH 4 OH (60:47:11.3:2). Radiolabeled phosphatidylinositol phosphates were visualized by autoradiography on X-Omat film.
Electrophoretic Mobility Shift Assay (EMSA)-The probe used for EMSA is a 30-bp double-stranded construct (5Ј-CCTGTGCTCCGG-GAATTTCCCTGGCCTGGA-3Ј) corresponding to a sequence (Ϫ72 to Ϫ42) in the CINC-proximal promoter region containing the NF-B motif (12). End labeling was performed by T4 kinase in the presence of [␥-32 P]dATP. Labeled oligonucleotides were purified on a Sephadex G-50M column (Amersham Biosciences).
An aliquot of 5 g of nuclear protein was incubated with the labeled double-stranded probe (ϳ50,000 cpm) in the presence of 5 g of nonspecific blocker, poly(dI-dC) in binding buffer (10 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM EDTA, 0.2% Nonidet P-40, and 0.5 mM dithiothreitol) at 25°C for 20 min. Specific competition was performed by adding 100 ng of unlabeled double-stranded CINC oligonucleotide probe to the nuclear extract from the sample with the greatest nuclear binding. The mixture was separated by electrophoresis on a 5% polyacrylamide gel in 1ϫ Tris glycine/EDTA buffer. Gels were vacuum dried and subjected to autoradiography.
Immunofluorescence Microcopy-RAW 264.7 cells were allowed to adhere to autoclaved glass coverslips overnight at 37°C in 5% CO 2 , incubated in previously outline experimental conditions. Cells were fixed for 30 min in PBS supplemented with 2% paraformaldehyde. The coverslips were washed three times with PBS for 10 min each, permeabilized with 0.2% Triton X-100 in PBS for 5 min, and then blocked with 5% bovine serum albumin in PBS for 30 min at room temperature. The samples were stained with goat anti-p65 polyclonal primary antibodies (1:50 dilution in PBS) for 1 h at room temperature, washed three times with PBS for 5 min each, and incubated with fluorescently labeled Alexa 555 donkey anti-goat IgG secondary antibodies (1:400 dilution in PBS) (Molecular Probes Inc., Eugene, OR) for 1 h at room temperature. Nuclei were counterstained with 4Ј,6Ј-diamidino-2-phenylindole chromosomal staining (Molecular Probes Inc.).
The coverslips were mounted on glass slides using DAKO solution, from Dakocytomation (Carpinteria, CA). The staining was visualized using a Nikon TE200 fluorescence microscope (ϫ100 objective) coupled to a Orca 100 camera (Hamamatsu Photonics, Hamamatsu, Japan) driven by Simple PCI software (Compix Inc., Imaging Systems, Cranberry Township, PA).
Immunofluorescence data analysis for single cell assays involved calculating a ratio of translocated to non-translocated cells, with an average of 100 cells counted for each group. The percentage of transfected cells was determined by counting the number of GFP transfected to non-GFP transfected cells.
Statistical Analysis-Data are presented as the mean Ϯ S.E. of three independent studies. When blots are shown, they are representative of at least three independent studies. For statistical comparison, the means were compared using analysis of variance by Student's t test (SPSS Software, Chicago, IL). A probability value of less than 0.05 was considered significant. RAW 264.7 cells were exposed to either increasing doses (M) of H 2 O 2 , LPS (0.01 g/ml), or the combination of pretreatment with H 2 O 2 for 1 h then LPS exposure. C, XO primes for increased LPS-induced NF-B translocation. RAW 264.7 cells were exposed to XO (50 milliunits/ml), LPS (0.1 g/ml), or the combination of pretreatment with XO for 2 h and then LPS treatment. Electrophoretic mobility shift assays were carried out with an oligonucleotide probe containing an NF-B-binding sequence to the CINC promoter site. EMSA was performed at the given LPS exposure times. D, XO primes for increased LPS responsiveness. RAW 264.7 cells were either untreated or exposed to XO (100 milliunits/ml), LPS (0.1 g/ml), or pretreated 2 h with XO then stimulated by LPS. The cells were examined for CINC mRNA by Northern blot. Glyceraldehyde-3-phosphate dehydrogenase (G3PDH) mRNA is identified as a control for RNA loading. A representative of three independent experiments is shown.

Effect of Oxidative
higher doses of H 2 O 2 alone caused modest NF-B translocation in this model, whereas 1000 M resulted in cytotoxicity. The dose of LPS chosen in these studies was based on its ability to induce relatively slow and modest changes in NF-B translocation, as a means of highlighting the ability of oxidants to augment this activation. Xanthine oxidase is a proximal generator of oxidants following ischemia/reperfusion and was also tested for its ability to augment LPS responsiveness. In Fig.  1C, cells were exposed to XO (50 milliunits/ml) alone, LPS (0.1 g/ml) alone, or the combination of pretreatment with XO followed by LPS treatment. Whereas XO pretreatment alone induced little NF-B translocation, it markedly augmented the responsiveness to LPS at all time points studied. XO doses from 25 to 500 milliunits/ml were studied and optimal priming occurred at the 100 milliunits/ml concentration (data not shown).
Because CINC gene expression is dependent on NF-B translocation and its transcription is augmented in vivo by shock plus LPS (3), we evaluated a functional correlate of the increased NF-B translocation in this model by following the expression of the chemokine CINC. Pretreatment of cells with XO significantly enhanced levels of CINC mRNA expression following LPS treatment compared with LPS alone (Fig. 1D). The levels of CINC mRNA following XO treatment alone were modestly increased compared with control levels. Together, these studies established a cellular model of oxidant priming for increased LPS responsiveness, representing an in vitro correlate of the priming of alveolar macrophages observed after shock/resuscitation in vivo. Based on the dose-response studies, we chose to use 100 M H 2 O 2 as the priming dose of oxidative stress in subsequent experiments.
Oxidants Augment Tyrosine Phosphorylation of the Src Family Members-To determine whether members of the Src family of kinases might contribute to oxidant-induced signaling, we evaluated tyrosine phosphorylation as a measure of their activation. The three major Src family members expressed in the macrophage are Hck, Fgr, and Lyn (15). H 2 O 2 (100 M) induced phosphotyrosine accumulation in all three of these kinases within 10 min following exposure (Fig. 2). Therefore, the ability of oxidative stress to cause tyrosine phosphorylation of the three Src kinases is consistent with their potential contribution to the activation of downstream signaling (16).
To examine the role of Src kinases in the oxidant-induced priming of LPS signaling, we utilized the agent PP2, a potent and selective pharmacological inhibitor of the Src family of tyrosine kinases (17). An in vitro kinase assay was performed to confirm the ability for PP2 to prevent the activation of Src kinase. Hck kinase activity was totally blocked by the addition of 10 M PP2 (n ϭ 2 studies). Ten min pretreatment with PP2 significantly attenuated H 2 O 2 /LPS-induced translocation of NF-B seen at 30 min (Fig. 3). The inactive analogue of PP2, PP3 had no inhibitory effect (data not shown). In addition, pretreatment of macrophages with PP2 prior to treatment with LPS alone did not inhibit LPS-induced NF-B translocation and was occasionally mildly stimulatory, as illustrated in Fig.  3. These findings imply that, in agreement with previous data, Src kinases are not required for NF-B translocation induced by LPS in the absence of oxidative stress (18). Together, these studies suggest that oxidant pretreatment may alter the LPS signaling pathway such that it critically involves Src kinases.
H 2 O 2 Augments LPS-induced p65 Nuclear Translocation-As an alternative approach to examining the effect of oxidant stress on LPS signaling, we used immunofluorescence (IF) microscopy to investigate nuclear translocation of p65 as a surrogate for NF-B translocation. Fig. 4A shows a time course of the effect of H 2 O 2 on LPS-induced p65 nuclear translocation where 4Ј,6Ј-diamidino-2-phenylindole staining was used to verify nuclear localization of p65. Nuclear p65 translocation was observed by 5-20 min in response to LPS plus H 2 O 2 pretreatment, whereas neither H 2 O 2 nor LPS alone induced translocation at these time points. By 30 to 45 min, p65 translocation was observed in both LPS-and H 2 O 2 /LPS-treated cells. These findings reproduce those observed for the EMSA, in that the combination of H 2 O 2 /LPS induced an earlier translocation of p65 compared with LPS alone. However, they differ somewhat insofar as the EMSA also demonstrated an increased magnitude of nuclear translocation, whereas the immunofluorescence studies did not. This is likely a result of the fact that the degree of p65 translocation detected by immunofluorescence is poorly quantitated, compared with translocation of binding proteins in the EMSA. Nevertheless, the major finding that oxidant exposure primes for earlier LPS-stimulated NF-B translocation is common to both methods.
Having demonstrated the ability of this model to mimic the findings observed in the cell populations, several approaches were taken to substantiate the contributions of the Src family kinases in this pathway. First, cells were treated with or without the Src inhibitor, PP2. Neither control cells nor control cells treated with PP2 alone showed evidence of cytoplasmic localization of p65 (Fig. 4B, upper two panels). However, PP2 was able to inhibit LPS-induced NF-B translocation in cells preexposed to H 2 O 2 (Fig. 4B, lower two panels). At this early time point (t ϭ 15 min after LPS), there was no p65 translocation in cells treated with either H 2 O 2 or LPS alone (Fig. 4A) and there

FIG. 3. PP2 inhibits H 2 O 2 /LPS-induced NF-B translocation.
Cells were exposed to H 2 O 2 (100 milliunits/ml), LPS (0.01 g/ml), or the combination of pretreatment with H 2 O 2 for 1 h followed by LPS exposure. Pretreatment with the Src family inhibitor PP2 (10 M) for 10 min prior to treatment conditions was also carried out. A representative EMSA at t ϭ 30 min of LPS exposure is shown. EMSA was performed with an oligonucleotide probe containing an NF-B binding sequence to the CINC promoter site. DAPI, 4Ј,6Ј-diamidino-2-phenylindole.  Fig. 4A. was no effect of PP2 (data not shown). We also employed a molecular approach wherein cells were transfected with the C-terminal Src kinase (Csk), an endogenous inhibitor of Src family kinases. Co-transfection with GFP was used to identify transfected cells. As illustrated above, in Fig. 5 at t ϭ 15 min, p65 translocation had occurred in cells treated with H 2 O 2 /LPS. Co-transfection had no effect in control, H 2 O 2 -, and LPStreated cells (data not shown). At this early time point, Csk transfection completely prevented nuclear translocation of p65 induced by H 2 O 2 plus LPS (Fig. 5).
As noted above (Fig. 4A), at later time points following LPS stimulation, cells pretreated with or without H 2 O 2 exhibited comparable p65 translocation (Ͼ80% of cells). To investigate the role of Src family kinases on LPS-induced downstream signaling after prolonged LPS stimulation, with or without prior oxidative stress, we evaluated the effect of Src inhibition on p65 translocation. In these studies, stable transfection of RAW cells with cytoplasmic Csk was used. After 45 min of LPS treatment, p65 translocation had occurred in Ͼ80% of cells whether pretreated with either H 2 O 2 or vehicle (Fig. 6). Although Csk transfection prevented nuclear translocation in cells treated with H 2 O 2 plus LPS, it had no effect on LPSinduced p65 nuclear translocation in the absence of H 2 O 2 pretreatment. Cell counting confirmed these findings: Csk transfection of cells exposed to H 2 O 2 plus LPS caused a 96.5 Ϯ 2.73% reduction in the number of cells with p65 translocation (mean Ϯ S.E., n ϭ 3 studies with 150 cells counted per study), compared with a slight increase (7.46 Ϯ 2.9%, mean Ϯ S.E., n ϭ 3 studies with 150 cells counted per study) observed in Csk-transfected cells treated with LPS alone. In aggregate, these studies suggest that exposure of cells to oxidant stress causes an alteration in LPS signaling for NF-B, such that it becomes dependent on activation of Src family kinases.
Oxidant Stress Primes for Increased PI 3-Kinase Activity in a Src-dependent Manner-Recent reports have shown that the PI 3-kinase pathway may be a downstream effector of cellular response to oxidative stress (19). During its activation, this kinase is phosphorylated on tyrosine residues, and thus might serve as a possible substrate for the Src family kinases (20). We hypothesized that LPS signaling following oxidant stress involves the PI 3-kinase pathway and is activated by Src kinases. We first examined tyrosine phosphorylation of PI 3-kinase following LPS stimulation with and without pre-exposure to H 2 O 2 . As shown in Fig. 7A, H 2 O 2 /LPS exhibited an increased phosphotyrosine accumulation by 5 min compared with LPS alone. This rise was prevented by treatment with the Src tyrosine kinase inhibitor PP2. To confirm that this was associated with increased activation of PI 3-kinase, kinase assay was performed by immunoprecipitating the p85 subunit of PI 3-kinase and incubating with phosphatidylinositol and [␥-32 P]dATP. The combination of H 2 O 2 plus LPS caused a marked increase in PI 3-kinase activity at 5 and 15 min of LPS stimulation compared with either LPS or H 2 O 2 alone. The Src family kinase inhibitor, PP2, was able to prevent this activation at these time points (Fig. 7B). As reported by others (21,22), at higher doses, LPS alone (0.5-1.0 g/ml) was able to cause activation of PI 3-kinase, but again this activation was not inhibited by PP2 (data not shown). Furthermore, consistent with the idea that activation of Src family kinases is upstream of PI 3-kinase following oxidant stress, phosphorylation of Fgr by LPS following H 2 O 2 is potentiated (Fig. 7C). These studies suggested that the PI 3-kinase pathway is involved in the downstream signaling following Src activation. Because inhibition of the Src family inhibited NF-B translocation, we examined whether inhibition of the PI 3-kinase pathway might have similar effects. Immunofluorescence studies showed complete translocation of p65 by 45 min for both LPS and H 2 O 2 plus LPS-treated cells (Fig. 7D). Consistent with the differential involvement of PI 3-kinase in LPS signaling in oxidanttreated cells, wortmannin had little effect on LPS-stimulated p65 translocation (10.5 Ϯ 7.31% decrease, mean Ϯ S.E., n ϭ 3 studies with 100 cells counted in each study) while causing a 52.7 Ϯ 9.25% reduction of p65 translocation in H 2 O 2 /LPS-treated cells (mean Ϯ S.E., n ϭ 3 studies with 100 cells counted in each study).

FIG. 5. Csk prevents the earlier oxidant-induced NF-B translocation.
Cells were co-transfected with Csk and GFP. Cells were exposed to H 2 O 2 (100 milliunits/ml) for 1 h followed by LPS (0.01 g/ml) for 15 min. The upper panel displays GFP-transfected cells (green), found in the right column. The middle panel displays IF staining for anti-p65 NF-B antibody (red). The lower panel displays dual staining for p65 (red) and 4Ј,6Ј-diamidino-2-phenylindole (DAPI) staining for nuclei (blue). Translocation is represented by a purple because the co-localization of the red and blue within the nucleus. Csk transfection prevented the earlier p65 translocation in cells exposed to H 2 O 2 /LPS. Csk transfection had no effect on control, H 2 O 2 , or LPS alone (data not shown)

Reprogrammed LPS Signaling Does Not Involve Tyrosine
Phosphorylation of IB␣-NF-B nuclear translocation is known to be regulated through its cytoplasmic binding to IB. One recent report demonstrated that oxidant stress related to hypoxia/reoxygenation was able to induce NF-B activation through a c-Src-dependent tyrosine phosphorylation of IB␣ in HeLa cells (23). To examine whether this pathway might be involved in the Src-dependent LPS signaling in RAW cells following H 2 O 2 exposure, tyrosine phosphorylation of IB␣ was studied. As shown in Fig. 8A, whereas both LPS alone and H 2 O 2 plus LPS induced a modest degree of phosphotyrosine accumulation on IB␣, there was no difference between the two groups. Furthermore, consistent with the idea that NF-B translocation was augmented through a pathway involving serine phosphorylation of IB␣, its degradation was augmented in cells exposed to oxidant stress plus LPS compared with LPS alone (Fig. 8B). DISCUSSION The contribution of oxidative stress to intracellular signaling pathways has become a common theme of investigation in the area of infection/inflammation. In addition to their ability to directly activate various signaling cascades, oxidants have also been shown to prime cells for an augmented response to a second inflammatory stimulus. This concept is highly relevant to the consideration of patients with critical illness, where oxidants generated during ischemia/reperfusion have been shown to heighten the inflammatory response and consequent organ injury to a delayed stimulus (2). In the present studies, we have modeled this two-hit process in vitro to investigate the cellular mechanisms whereby oxidative stress might exert this priming effect. The major finding in this article is that antecedent oxidative stress reprograms the LPS signaling pathway leading to NF-B translocation such that it involves activation of Src family kinase members. Activation of PI 3-kinase appears to be a consequence of Src activation and is clearly involved in the downstream signaling events leading to NF-B translocation. Consistent with this notion others have shown that members of the Src family can induce tyrosine phosphorylation of the p85 subunit leading to an enhancement of PI 3-kinase activity (20,24). Taken together, our studies elucidate a novel signaling pathway associated with oxidant-induced priming of macrophages and suggest consideration of alternative strategies directed at inhibiting cell activation under conditions of ischemia/reperfusion.
The diversion of LPS signaling into the PI 3-kinase pathway appears to be a critical component of the reprogramming that occurs after oxidant exposure. Whereas the mechanism of this effect is unknown, one recent report suggested that LPS-induced ceramide generation was capable of activating PI 3-ki- FIG. 7. Role of PI 3-kinase in NF-B translocation following oxidant stress. Oxidants augment the LPS-induced PI 3-kinase activity in a Src dependent manner. RAW cells were pretreated with H 2 O 2 (100 M) for 1 h, LPS (0.01 g/ml), or the combination of H 2 O 2 and LPS for the indicated times. Cells were pretreated with 10 M PP2 10 min prior to the oxidant exposure. A, whole cell lysates were separated by SDS-PAGE and immunoblotted with an antibody against the phosphorylated p85 subunit of PI 3-kinase. B, PI 3-kinase was immunoprecipitated from the whole cell lysates using an antibody specific for the p85 regulatory unit. Kinase activity was determined by evaluating the phosphorylation of PI substrate. C, H 2 O 2 augments the LPS-induced Src family activation. Whole cell lysates were immunoprecipitated for Fgr and then assessed for tyrosine phosphorylation using immunoblotting with the anti-phosphotyrosine antibody 4G10. The blots were then stripped and probed for anti-Fgr to demonstrate equal protein loading. D, differential inhibition of p65 translocation by PI 3-kinase inhibition with LPS versus H 2 O 2 and LPS. Cells were pretreated with wortmannin (500 nM) for 30 min prior to the addition of H 2 O 2 or LPS. IF staining for the p65 NF-B subunit was preformed after LPS incubation for 30 min. All experiments were repeated three times, and data from one representative experiment are shown.
FIG. 8. Role of IB␣ in the oxidant-induced priming of the macrophage. A, oxidants did not prime for increased tyrosine phosphorylation. Cells were either exposed to LPS (0.01 g/ml) or H 2 O 2 (100 M) 1 h prior to LPS treatment for 1, 5, 10, and 20 min. Whole cell lysates were prepared and subjected to immunoprecipitation with IB␣ antibody. Precipitates were then analyzed by Western blot using an antibody specific for phosphotyrosine. B, oxidants primes for rapid IB␣ protein degradation. Cells were exposed to LPS (0.01 g/ml), H 2 O 2 (100 M), or the H 2 O 2 (100 M) 1 h prior the LPS treatment for 1, 5, 10, 20, and 30 min. Whole cell lysates were separated by SDS-PAGE and immunoblotted with an antibody against IB␣. A representative of three independent experiments is shown. nase (22) and may thus be considered a candidate molecule in the priming process. Ceramide is generated from membrane sphingomyelin through the action of sphingomyelinase, an enzyme that resides downstream of phosphatidylcholine-specific phospholipase. A role for ceramide in the priming process is indirectly supported by the observations regarding the susceptibility of phosphatidylcholine-specific phospholipase to activation by oxidants. First, in alveolar epithelial cells, phosphatidylcholine-specific phospholipase activity was shown to be preferentially activated by oxidants compared with other phospholipases (25). Second, Giron-Calle and colleagues (26) showed that H 2 O 2 pretreatment of alveolar macrophages was able to prime alveolar macrophages for enhanced respiratory burst activity in response to a subsequent agonist exposure, an effect that was prevented by the inhibitor of phosphatidylcholine-specific phospholipase, D609. Finally, the Src-dependent activation of PI 3-kinase in the present studies would also be consistent with a role for ceramide in the priming process. In colonic smooth muscle cells, ceramide-induced PI 3-kinase activation was found to be associated with tyrosine phosphorylation of pp60 src and to be inhibited by a Src kinase inhibitor (27). Together with the data presented here, these reports suggest a mechanism whereby oxidants might divert LPS signaling such that the induction of ceramide results in Src kinasedependent activation of PI 3-kinase and downstream NF-B nuclear translocation.
One interesting finding in the present experiments was the fact that oxidant pretreatment not only reprogrammed LPS signaling to involve Src family kinases, but did so to the exclusion of the traditional Src-independent LPS signaling cascade. Specifically, strategies of Src inhibition in oxidant-treated cells caused near complete inhibition of NF-B translocation, with no apparent signaling though the Src-independent cascade. This would suggest that oxidant pretreatment might have the capacity to completely divert the LPS signal away from the more traditional cascade to NF-B activation. Recent studies have demonstrated the requirement for aggregation of LPS receptors, namely the CD14 and TLR4 receptor, within lipid rafts to initiate downstream LPS signals (28). One might speculate that oxidative stress alters lipid rafts such that different molecules are recruited to the LPS signaling complex and initiate alternate pathways. In this regard, Parat and colleagues (29) recently demonstrated that oxidative stress was able to inhibit calveolin-1 palmitoylation and trafficking in endothelial cells. Because calveolae may play a role in sequestration of glycosylphosphatidylinositol-linked proteins such as CD14, inhibition of calveolin-1 by oxidant stress may interfere with classical LPS signaling (i.e. via the CD14/TLR4 mediated pathway) (30). Whether LPS signaling following oxidative stress occurs through the CD14-TLR4 complex with diversion into a Src kinase-dependent pathway or through alternate surface receptors for LPS requires further investigation.
Dissociation of IB␣ from NF-B is required for the translocation of the latter into the nucleus. Two distinct phosphorylation patterns of IB␣, either on serine or tyrosine residues, have been shown to lead to NF-B translocation. One recent report showed that tyrosine phosphorylation of IB␣ was pivotal in mediating NF-B translocation after hypoxia/reoxygenation in HeLa cells and that oxidative stress was important in mediating this effect (23). By contrast, the present studies suggest that the downstream effect of oxidant treatment followed by LPS culminated in an effect on IB␣ that was mediated by serine phosphorylation. First, both LPS alone and H 2 O 2 /LPS induced equivalent degrees of tyrosine phosphorylation on IB␣. Second, H 2 O 2 /LPS induced rapid degradation of IB␣, a finding more consistent with induction of serine phos-phorylation of IB␣ than with tyrosine phosphorylation, because the latter has been reported to cause dissociation from NF-B, but without degradation (31). This conclusion would be consistent with the contribution of PI 3-kinase-mediated signaling, because IB kinase is a well described substrate of the Akt, one of the potential downstream mediators of PI 3-kinase (32).
The present studies provide a novel mechanism for the priming effects of oxidative stress, namely through altering signaling through a Src kinase-dependent pathway involving activation of PI 3-kinase. The potential importance of this pathway was suggested recently in studies using an experimental model of lung injury following endotoxemia where PI 3-kinase knockout animals exhibited reduced lung injury, neutrophil sequestration, and NF-B sequestration (33). Together, these findings suggest that alternative therapeutic strategies aimed at modulating oxidant stress following shock/resuscitation and/or specifically defined signaling pathways might minimize the predisposition of trauma victims for developing organ injury.