Cytokine-induced Activation of Nuclear Factor-κB Is Inhibited by Hydrogen Peroxide through Oxidative Inactivation of IκB Kinase*

Rapid activation of the IκB kinase (IKK) complex is considered an obligatory step in the activation of nuclear factor-κB (NF-κB) in response to diverse stimuli. Since oxidants have been implicated in the regulation of NF-κB, the focus of the present study was the activation of IKK by tumor necrosis factor α (TNFα) in the presence or absence of hydrogen peroxide (H2O2). Exposure of mouse alveolar epithelial cells to H2O2 was not sufficient to activate IKK, degrade IκBα, or activate NF-κB. In contrast, TNFα induced IKK activity rapidly and transiently resulting in IκBα degradation and NF-κB activation. Importantly, in the presence of H2O2, the ability of TNFα to induce IKK activity was markedly decreased and resulted in prevention of IκBα degradation and NF-κB activation. Neither tyrosine kinases nor phosphatidylinositol 3-kinases, known regulators of NF-κB by oxidants, were involved in IKK inhibition by H2O2. Direct addition of H2O2 to the immunoprecipitated IKK complex inhibited enzyme activity. Inhibition of IKK activity by H2O2 was associated with direct oxidation of cysteine residues present in the IKK complex and occurred only in enzymatically active IKK. In contrast to previously published observations, our findings demonstrate that the oxidant H2O2 reduces NF-κB activation by inhibiting activated IKK activity.

Rapid activation of the IB kinase (IKK) complex is considered an obligatory step in the activation of nuclear factor-B (NF-B) in response to diverse stimuli. Since oxidants have been implicated in the regulation of NF-B, the focus of the present study was the activation of IKK by tumor necrosis factor ␣ (TNF␣) in the presence or absence of hydrogen peroxide (H 2 O 2 ). Exposure of mouse alveolar epithelial cells to H 2 O 2 was not sufficient to activate IKK, degrade IB␣, or activate NF-B. In contrast, TNF␣ induced IKK activity rapidly and transiently resulting in IB␣ degradation and NF-B activation. Importantly, in the presence of H 2

O 2 , the ability of TNF␣ to induce IKK activity was markedly decreased and resulted in prevention of IB␣ degradation and NF-B activation. Neither tyrosine kinases nor phosphatidylinositol 3-kinases, known regulators of NF-B by oxidants, were involved in IKK inhibition by H 2 O 2 . Direct addition of H 2 O 2 to the immunoprecipitated IKK complex inhibited enzyme activity. Inhibition of IKK activity by H 2 O 2 was associated with direct oxidation of cysteine residues present in the IKK complex and occurred only in enzymatically active IKK. In contrast to previously published observations, our findings demonstrate that the oxidant H 2 O 2 reduces NF-B activation by inhibiting activated IKK activity.
Oxidative stress plays a role in the pathogenesis of many inflammatory diseases. For instance, diseases of the lung such as asthma or chronic obstructive pulmonary disease, are accompanied by the presence of oxidants (1)(2)(3). Reactive oxygen and nitrogen species are capable of causing oxidative damage to macromolecules including lipid peroxidation, DNA damage, and protein modifications such as oxidation of cysteine residues (4,5). After exposure, the cellular response to oxidants depends on the induction of signaling cascades and activation of transcription factors like nuclear factor-B (NF-B) 1 and activating protein-1.
The transcription factor NF-B is known to be responsive to oxidative stress (6,7). NF-B plays a pivotal role in the development of chronic inflammation, cell survival, and proliferation (8 -10). In unstimulated cells, NF-B is sequestered in the cytoplasm through interaction with IB inhibitory proteins. Exposure to a number of stimuli (e.g. cytokines, lipopolysaccharide, and viruses) will result in the phosphorylation, ubiquitination, and degradation of IB, liberating the NF-B dimers (7)(8)(9). The NF-B complex translocates into the nucleus and transcriptionally activates target genes by binding to responsive elements in the DNA called B motifs. Phosphorylation of the inhibitory protein IB occurs at specific serine residues (11) by IB kinases (IKKs) (12).
IKK is present as a large (700 -900 kDa) complex composed of two catalytic subunits IKK␣ and IKK␤ and a regulatory subunit, IKK␥ (13). The IKK complex serves as a point of convergence for both positive and negative NF-B regulators (14). Upstream kinases including mitogen-activated protein kinase/extracellular signal-regulated kinase kinase kinase 1 (MEKK1), NF-B-inducing kinase, and protein kinase B (Akt) activate the IKK complex through mechanisms not fully understood but result in the phosphorylation of two serine residues in the activation loop of IKK␣ (serines 176 and 180) and IKK␤ (serines 177 and 181) (13)(14)(15). This activation occurs rapidly and is followed by a series of autophosphorylation events of serine residues present in the carboxyl-terminal tail of the IKK␤ subunit. When a critical number of serine residues are phosphorylated, the complex becomes inactive (13) as was demonstrated by replacement of 9 -10 C-terminal serines residues by alanines, which resulted in a complex that was active four times longer (14). Since the IKK complex is an important mediator in the positive and negative regulation of NF-B, it requires stringent regulation achieved through rapid activation and inactivation.
An alternative activation pathway of NF-B exists, specifically important for activation by oxidants, including pervanadate, hypoxia/reoxygenation, and UV-C (16), in which the NF-B complex was liberated upon tyrosine phosphorylation of IB␣ (16 -20). This alternative pathway involves phosphorylation of tyrosine 42 instead of serines 32 and 36 (17,21), is in some systems (17) but not in others (21) associated with degradation of IB␣, and does not appear to involve the activation of IKK (22). The interaction between the p85 subunit of phosphatidylinositol 3-kinase (PI3K) and tyrosine-phosphorylated IB␣ may serve to dissociate NF-B from IB␣ without degradation (21). Importantly oxidants, including H 2 O 2 , are known to activate the PI3K pathway (23,24). The precise mechanism by which oxidants activate NF-B remains to be elucidated, but it is becoming clear from the available data that oxidant-induced NF-B activation is highly cell type-dependent (25), pointing perhaps to a lack of a uniform mechanism of activation (26).
During pulmonary inflammatory disease states, epithelial cells are simultaneously exposed to both oxidants and inflammatory cytokines. However, most studies to date have not considered simultaneous exposure of cells to oxidants and cytokines. Furthermore, no studies have investigated the effects of oxidants on IKK activation in lung epithelial cells, an important target for inhaled or inflammatory cell-derived oxidants. Therefore, we investigated the effects of TNF␣ and H 2 O 2 on IKK activation in lung epithelial cells. Our data demonstrates that H 2 O 2 by itself does not activate NF-B in a line of alveolar type II cells. However, H 2 O 2 decreased the ability of TNF␣ to induce IKK activation, IB␣ degradation, and NF-B activation by direct oxidation of cysteine residues in the activated IKK complex. In contrast to other observations, our studies demonstrate a negative role for H 2 O 2 in the activation of NF-B by TNF␣ due to oxidative inactivation of the IKK complex.

EXPERIMENTAL PROCEDURES
Cell Culture and Reagents-Spontaneously transformed alveolar type II cells (C10) were kindly provided by Dr. Alvin Malkinson (27). The C10 cells were propagated in CRML-1066 medium containing 50 units/ml penicillin, 50 g/ml streptomycin, 2 mM L-glutamine, and 10% fetal bovine serum, all from Life Technologies, Inc. Rat alveolar type II epithelial (RLE) cells were cultured as described previously (20). For the experiments, cells were plated onto 60-mm dishes and grown to 70 -90% confluency. At least 1 h before adding the test agents the cells were switched to phenol red-free Dulbecco's modified Eagle's medium/ F12 containing 50 units/ml penicillin, 50 g/ml streptomycin, and 0.5% fetal bovine serum (Life Technologies, Inc.). Murine recombinant TNF␣ was purchased from Calbiochem. Herbimycin A was obtained from Life Technologies, Inc., LY294002 and H 2 O 2 from Sigma, and glucose oxidase (GOx) from Roche Molecular Biochemicals. N-(Biotinoyl)-NЈ-(iodoacetyl)ethylenediamine (BIAM) was purchased from Molecular Probes (Eugene, OR). LY294002, herbimycin A, and BIAM were dissolved in dimethyl sulfoxide (Me 2 SO) and used at a concentration of 0.1% Me 2 SO. IB␣, IKK␥, p65, p50, and c-Jun N-terminal kinase (JNK) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), phosphotyrosine, phospho-Akt (pAkt), and Akt antibodies were obtained from Upstate Biotechnology (Lake Placid, NY), the ␤-actin antibody was obtained from Sigma, and horseradish peroxidase-conjugated streptavidin was purchased from Roche Molecular Biochemicals. All experiments were performed in duplicate and repeated at least three times.
Kinase Assays-C10 cells were exposed to the test agents individually or simultaneously. At times ranging from 5 min to 24 h, cells were transferred to ice, washed once with cold phosphate-buffered saline, and lysed in lysis buffer, pH 6.5 (50 mM HEPES, 150 mM NaCl, 1 mM EDTA, 2 mM MgCl 2 , 10 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, 0.1% Nonidet P-40, 10 g/ml leupeptin, 1% aprotinin, 0.5 mM DTT, and 100 M NaF). Lysates were cleared by centrifugation at 14,000 rpm at 4°C for 10 min. Protein concentrations were determined with the Bradford method (D c Protein Assay, Bio-Rad). IKK or JNK were immunoprecipitated from 200 g of protein with an IKK␥ or JNK antibody at 4°C for 1.5 h using protein A-agarose beads (Life Technologies, Inc.). Precipitates were washed once with lysis buffer and once with kinase buffer containing 20 mM HEPES, 20 mM ␤-glycerophosphate, 1 mM MnCl 2 , 5 mM MgCl 2 , 2 mM NaF, and 1 mM DTT for the IKK assay and 20 mM HEPES, 20 mM ␤-glycerophosphate, 20 mM MgCL 2 , 2 mM DTT, and 0.1 mM Na 3 VO 4 for the JNK assay. The kinase reaction was performed using 1 g of glutathione S-transferase (GST)-IB␣-(1-54) kindly provided by Dr. Rosa Ten (Mayo Clinic, Rochester, MN) or GST-c-Jun (New England Biolabs, Beverly, MA) and 5 Ci of [␥-32 P] adenosine triphosphate at 30°C for 30 min. Reactions were stopped by the addition of 2ϫ Laemmli sample buffer (2% SDS, 10% glycerol, 0.1 M DTT, and 0.01% bromphenol blue). Samples were boiled and stored at Ϫ20°C. Proteins were separated on a 15% polyacrylamide gel, and gels were dried and examined by autoradiography. Results were quantitated by phosphorimaging or densitometry.
Western Blot-At selected times following exposure to test agents, cells were washed once with cold phosphate-buffered saline and lysed in 1ϫ Laemmli sample buffer. Samples were boiled, and 50 l was loaded on a 15% polyacrylamide gel. Proteins were transferred to nitrocellulose (Schleicher & Schuell), and membranes were subsequently blocked in 5% milk in Tris-buffered saline (TBS). pAkt and Akt were detected according to the manufacturer's instructions. Levels of IB␣ and IKK␥ were detected according to the following protocol: membranes blocked overnight in TBS/milk were washed two times for 15 min in TBS containing 0.05% Tween 20 and incubated with the primary antibody against IB␣ or IKK␥ for 1 h at 4°C. Membranes were washed three times for 20 min in TBS/Tween and incubated with a peroxidaseconjugated secondary antibody for 1 h at room temperature. After a 30-min wash with TBS/Tween, conjugated peroxidase was detected by chemiluminescence according to the manufacturer's instructions (Kirkegaard and Perry Laboratories, Gaithersburg, MD).
Gel Mobility Shift Assays-To determine DNA binding activity of NF-B, complexes binding to a radiolabeled double-stranded oligonucleotide containing a NF-B consensus sequence were analyzed (Promega, Madison, WI). Nuclear extracts and gel shift assays were prepared as described previously (28). Four micrograms of nuclear protein was used per binding reaction, and protein-DNA complexes were resolved on a 5% polyacrylamide gel in 0.25ϫ Tris borate-EDTA buffer at 120 V for 2 h. Gels were dried and exposed to film (X-Omat Blue XB-1, Kodak). To determine the subcomposition of the complexes, supershift reactions were performed by preincubation of the nuclear extracts with antibodies specific to the p65 or p50 subunits of NF-B.
Construction of a C10 Cell Line Stably Expressing NF-B Luciferase-Plasmid 6 B-tk-luc was kindly provided by Dr. Patrick Baeuerle (Micromet, Martinsried, Germany). C10 cells were transfected by electroporation (20) with the plasmid 6 B-tk-luc in the presence of the plasmid PSV-2Neo (Promega) conferring neomycin resistance using a ratio of 15:1, respectively. Twenty-four hours post-transfection, the medium was replaced with cell culture medium containing 100 g/ml geneticin (Life Technologies, Inc.). After the appearance of colonies in 2-3 weeks, dishes were trypsinized, and pools of neomycin-resistant cells were propagated. The NF-B inducibility of the 6 B-tk-lucexpressing cells was demonstrated using TNF␣ as a positive control. 2 Prior to the addition of test agents, cultures were switched to phenol red-free Dulbecco's modified Eagle's medium/F12 medium containing 0.5% serum, 50 units/ml penicillin, and 50 g/ml streptomycin for a minimum of 1 h. Luciferase units were corrected for total protein amounts.
In Vitro Kinase Assay-Cells were left untreated or were exposed to TNF␣ for 5 min, and a kinase assay was performed as described above with the following modifications. To assess the redox sensitivity of immunoprecipitated, active IKK, the concentrations of DTT present in the lysis and kinase buffers were adjusted to concentrations ranging from 100 to 500 M. In addition, the immunoprecipitated IKK complex from TNF␣-stimulated cells was subsequently exposed to H 2 O 2 for 5 min in lysis buffer prior to assessment of kinase activity. Western blots detecting levels of immunoprecipitated IKK␥ were performed to establish loading differences.
Assessment of Cysteine Oxidation-Cells were left untreated or were exposed to TNF␣ for 5 min and harvested in lysis buffer for immunoprecipitation of the IKK complex as described above. Cysteine oxidation was assessed using BIAM, which selectively recognizes reduced cysteines, according to the protocol of Kim et al. (29) with the following modifications. After isolation the IKK complex was exposed to 500 M H 2 O 2 for 5 min in a buffer containing 50 mM MES, 100 mM NaCl, 50 M phenylmethylsulfonyl fluoride, 1.0 g/ml leupeptin, 1.0 g/ml aprotinin, and 0.5% Triton X-100. The buffer was made fresh, rendered free of O 2 by bubbling with N 2 at a low flow rate for 1 h, and used at two different pH levels (pH ϭ 6.5 and pH ϭ 8.5). After incubation with H 2 O 2 , the protein A-agarose beads containing the IKK complex were centrifuged and exposed to 20 M BIAM for 15 min. The reaction was terminated by addition of 20 mM ␤-mercaptoethanol. The beads were centrifuged briefly, and 1ϫ Laemmli sample buffer was added. Western blots were performed as described above using horseradish peroxidaseconjugated streptavidin to detect BIAM and IKK␥ as a loading control.
Assessment of Intracellular H 2 O 2 -C10 cells were preincubated for 30 min with dichlorofluorescein (DCF) diacetate (Molecular Probes) before 10 ng/ml TNF␣ was added for 5 min. Cells were washed once with phosphate-buffered saline, harvested by trypsinization, centrifuged, and resuspended in Hanks' balanced salt solution (Life Technologies, Inc.) to a density of 500,000cells/ml. DCF oxidation was detected by flow cytometry.

IKK Activation in Response to TNF␣ and H 2 O 2 -
In a number of cell lines TNF␣ rapidly and transiently activates the IKK complex, resulting in NF-B activation and thus NF-B-regulated gene transcription. In our studies, a rapid increase (2 min) of IKK activity was observed in the C10 cells in which a maximal activity was reached at 5 min that returned toward baseline in 30 min (Fig. 1A). Since H 2 O 2 has been described to activate NF-B in various cell types (6, 7), we determined whether in C10 cells H 2 O 2 was able to activate IKK. As is demonstrated in Fig. 1B, H 2 O 2 added as a bolus of 200 or 500 M did not induce IKK activation in a time frame ranging from 5 to 90 min. Examination of time points up to 24 h failed to demonstrate H 2 O 2 -induced IKK activity. 3 We next determined whether continuous production of H 2 O 2 by the addition of GOx to the medium would activate IKK. GOx uses glucose in the medium to produce H 2 O 2 and when adding 5 units/ml GOx to the medium for 2 h, steady state levels of H 2 O 2 of ϳ500 M were measured (data not shown) (30). Similar to results obtained with bolus H 2 O 2 , no increases in the enzymatic activity of IKK were observed in cells treated with GOx (Fig. 1C), and following 90 min of treatment with GOx, the IKK activity was decreased in comparison to the sham controls. Western blots demonstrated an equal presence of IKK␥ indicating that the changes in kinase activity are not due to loading differences or differences in the amounts of immunoprecipitated IKK complex (data not shown). To establish that H 2 O 2 acts differently on various signal transduction pathways we examined the effect of bolus H 2 O 2 on JNK. In contrast to the lack of effect on IKK activity, the addition of 200 M H 2 O 2 resulted in the activation of JNK (Fig. 1D) in C10 cells, indicating that H 2 O 2 acts in a specific fashion on different kinases.
During inflammatory conditions that exist in many disease states, oxidants and cytokines are present simultaneously. We therefore determined whether H 2 O 2 could alter the ability of TNF␣ to induce IKK. While TNF␣ rapidly activated IKK, the presence of 200 M H 2 O 2 decreased the ability of TNF␣ to activate IKK ( Fig. 2A). Similar results were observed when cells were exposed to TNF␣ and GOx (Fig. 2B). Like TNF␣, interleukin-1␤ was able to rapidly activate IKK (Fig. 2C). The ability of interleukin-1␤ to activate IKK was also diminished in the presence of H 2 O 2 (Fig. 2C). These results demonstrate that  /ml). A, cells were exposed to TNF␣ and H 2 O 2 simultaneously for 5 min after which proteins were collected, and an IKK assay was performed using the substrate GST-IB␣- (1-54). B, cells were exposed to TNF␣ and GOx simultaneously for 5 min, and IKK activity was determined. C, cells were exposed to interleukin-1␤ and H 2 O 2 simultaneously for 5 min after which proteins were collected, and IKK activity was measured. D, RLE cells were exposed to TNF␣ and H 2 O 2 separately or simultaneously for 5 min after which proteins were collected, and an IKK assay was performed.

FIG. 1. Effects of TNF␣ (10 ng/ml), H 2 O 2 (200 and 500 M), and GOx (5 units/ml) on IKK and JNK activity.
A, C10 cells were exposed to TNF␣ for the indicated times after which cells were lysed. The IKK complex was immunoprecipitated from 200 g of protein, and kinase assays were performed using the substrate GST-IB␣- (1-54). B, C10 cells were treated with H 2 O 2 for variable times after which the kinase activity of the IKK complex was determined. C, C10 cells were exposed to GOx for variable times after which the kinase activity of the IKK complex was determined. D, C10 cells were exposed to H 2 O 2 for variable times, and JNK activity was determined using GST-c-Jun as a substrate. IKK activity induced by two different stimuli that activate distinct receptors was altered in a similar fashion by H 2 O 2 . Since in a previous study (20) an increased NF-B luciferase activity was observed in RLE cells, IKK activity was determined in RLE cells after exposure to H 2 O 2 in the absence or presence of TNF␣ (Fig. 2D). H 2 O 2 did not activate the IKK complex in contrast to TNF␣. However, when TNF␣ and H 2 O 2 were added simultaneously, an inhibition of the TNF␣-induced IKK activity was observed in the RLE cells as seen in the C10 cells. This finding clearly demonstrates that there is a discrepancy between regulation of IKK activity and NF-B-dependent transcriptional activity in the RLE cells exposed to H 2 O 2 in contrast to a coordinated regulation of NF-B activity in C10 cells.
IB␣ Degradation, NF-B Binding to DNA, and NF-B-induced Luciferase Activity-Based on the observation that H 2 O 2 decreased the ability of TNF␣ to induce IKK activation at 5 min, we speculated that this would also be reflected in an attenuated degradation of IB␣, binding of NF-B to DNA, and NF-B-dependent luciferase reporter gene activity. As is demonstrated in Fig. 3A, TNF␣ decreased the abundance of IB␣ after 15 min of exposure, indicative of IB␣ degradation. However, in the presence of H 2 O 2 , the TNF␣-induced degradation of IB␣ (Fig. 3A) was prevented. These results indicate that the decrease in TNF␣-induced IKK activation in the presence of H 2 O 2 culminated in a loss of IB␣ phosphorylation and degradation. In support of the lack of effects on IKK activity, H 2 O 2 by itself did not affect the levels of IB␣. A similar inhibitory effect of H 2 O 2 was observed when TNF␣-induced NF-B binding to DNA was investigated. Results in Fig. 3B demonstrate decreased binding of p65/p50 and p50/p50 to the NF-B-responsive element in cells exposed simultaneously to TNF␣ and H 2 O 2 for 15 min compared with the binding seen with TNF␣ alone. Antibodies against the p65 and p50 subunits of NF-B modified the electrophoretic mobility, indicating that the probe bound complexes consisted of p65 and p50 (Fig. 3B). No change in complex composition was observed when H 2 O 2 was added to TNF␣-exposed cells. To confirm that the reduced binding of the NF-B complex to DNA that occurred in the presence of H 2 O 2 and TNF␣ was reflected in a decreased transcriptional activation potential, we assessed C10 cells stably transfected with a NF-B-dependent luciferase reporter construct. Results in Fig.  3C demonstrate that in the presence of H 2 O 2 , the TNF␣-induced luciferase reporter activity was repressed following 6 h of exposure.
Involvement of the PI3K/Akt Pathway and Tyrosine Kinases-After establishing that H 2 O 2 decreased the TNF␣-induced IKK activity, we next determined whether the oxidant effect was mediated through a tyrosine kinase or PI3K/Akt signaling pathway, which are known to be induced by oxidants (23,31) and important in the activation of NF-B (31, 32). Addition of H 2 O 2 to the cells resulted in the phosphorylation of Akt, the downstream kinase of PI3K, within 5 min (Fig. 4A). Preincubation of the cells (30 min) with the PI3K inhibitor LY294002 blocked the H 2 O 2 -induced phosphorylation of Akt, indicating that this event was regulated by PI3K. We next determined whether LY294002 could reverse the inhibitory effect of H 2 O 2 on TNF␣-induced IKK activity. As is demonstrated in Fig. 4B, LY294002 did not restore the TNF␣-induced IKK activity in the presence of H 2 O 2 but instead led to a complete inhibition of the enzymatic activity of IKK. Thus, it appears that the residual TNF␣-induced IKK activity observed in the presence of H 2 O 2 may be due to the activation of Akt. Although a number of investigators have demonstrated the activation of tyrosine kinases in response to oxidants (17), the addition of H 2 O 2 to C10 cells for 5 min did not result in tyrosine phosphorylation of multiple proteins (Fig. 4C). Although tyrosine phosphorylation may occur at later time points following exposure to H 2 O 2 , these were not investigated since the inhibition of TNF␣-induced IKK activity by H 2 O 2 occurred as early as 5 min. In support of this observation, the tyrosine kinase inhibitor herbimycin A did not reverse the inhibitory effects of H 2 O 2 on TNF␣-induced IKK activity (Fig. 4D). Me 2 SO (0.1%) was used as a vehicle control and did not affect IKK or NF-B  ). A, cells were exposed for 5 and 15 min to TNF␣ (10 ng/ml) and/or H 2 O 2 after which a Western blot was performed for IB␣. B, nuclei were isolated from cells exposed for 15 min to TNF␣ (1.0 ng/ml) and H 2 O 2 simultaneously for the evaluation of binding of NF-B to its consensus DNA sequence using an electrophoretic mobility shift assay. Antibodies (Ab) directed against p65 and p50 verified their presence in the DNA binding complex. C, luciferase activity was determined in C10 cells stably transfected with an NF-B-driven reporter construct 6 h following the addition of TNF␣ (10 ng/ml) and/or H 2 O 2 .
activation. Collectively our data demonstrate that the inhibitory effect of H 2 O 2 on the TNF␣-induced activation of IKK occurred independently of Akt/PI3K or tyrosine kinase activities.
In Vitro Kinase Assay with H 2 O 2 -To investigate whether the IKK complex is sensitive to oxidation, cells were exposed to TNF␣ for 5 min, and the active IKK was immunoprecipitated, exposed to H 2 O 2 , and subjected to kinase activity assays. Since the presence of 500 M of the reducing agent DTT may buffer the effects of H 2 O 2 we performed the kinase assay in the presence of varying concentrations of DTT in the presence or absence of H 2 O 2 . As is demonstrated in Fig. 5A, DTT concentrations of 250 or 500 M were necessary to achieve maximal kinase activity of IKK (Fig. 5A) following exposure to TNF␣. When the concentration of DTT was reduced to 100 M, a dramatic decrease in kinase activity of IKK was observed. The addition of 200 or 500 M H 2 O 2 to the immunoprecipitated IKK complex resulted in a dose-dependent reduction in kinase activity. The decrease in IKK activity induced by H 2 O 2 occurred to the greatest extent in the presence of lower DTT concentrations as expected (Fig. 5A). Importantly the loss in IKK activity in the presence of H 2 O 2 was not due to dissociation of IKK␥ from the immunoprecipitated complex as demonstrated by the equal presence of IKK␥ protein in the immunoprecipitates (Fig.  5B). Western blots for IKK␣ and IKK␤ could not be performed since the corresponding 85-and 87-kDa bands were masked by the heavy chain of the IKK␥ antibody used for immunoprecipitation. In summary, our results illustrate that H 2 O 2 -induced oxidation of the IKK complex directly abolishes the kinase activity, which may account for the inhibitory effects of H 2 O 2 observed in intact cells after exposure to TNF␣. Our findings also demonstrate that the presence of 500 M DTT in the lysis and kinase buffers prevents oxidation due to H 2 O 2 and potentially air handling and ensures maximal kinase activity.
After establishing that the presence of 500 M DTT prevents oxidative inactivation of IKK, we next addressed the reversibility of oxidative inactivation. The active immunoprecipitated IKK complex was treated with 200 M H 2 O 2 in the presence of lysis buffer containing 100 M DTT to maximize oxidation as demonstrated in Fig. 5A. Samples were subsequently handled in kinase buffers containing increasing concentrations of DTT for evaluation of IKK enzymatic activity. Fig. 5C demonstrates that ϳ10% of the IKK enzymatic activity can be restored within 30 min by increasing the DTT concentration from 100 to 500 M. These findings clearly demonstrate that once IKK is oxidized within the cell by H 2 O 2 , subsequent manipulation of the sample in buffers containing DTT only partially reverses the oxidative inactivation of IKK within the 30-min time frame investigated here.
Determination of Oxidant-sensitive Cysteine Residues in the IKK Complex-After establishing that the activity of IKK was directly inhibited by H 2 O 2 we determined whether cysteine oxidation occurred following treatment with H 2 O 2 . As BIAM selectively reacts with reduced cysteine residues, oxidation by H 2 O 2 will diminish the binding of streptavidin. Since the reaction of BIAM with reduced cysteine residues is pK a -dependent, two different pH levels were studied (pH ϭ 6.5 and pH ϭ 8.5). As demonstrated in Fig. 6A, a 5-min exposure of the immunoprecipitated IKK complex to H 2 O 2 resulted in the reduced binding of streptavidin peroxidase, strongly suggesting the oxidation of cysteine residues. Importantly cysteine oxidation was only observed in IKK immunoprecipitates from TNF␣exposed cells but not sham controls. These findings demonstrate that H 2 O 2 is capable of causing cysteine oxidation only in enzymatically active IKK. The difference in BIAM-streptavidin reactivity in H 2 O 2 versus sham groups was not due to loading differences as demonstrated by the equal amounts of IKK␥ (Fig. 6A). Since the IKK complex is sensitive to oxidation and exposure of cells to TNF␣ has been demonstrated to result in intracellular production of H 2 O 2 (33, 34) the lack of IKK oxidation in TNF␣-exposed cells following labeling with BIAM was surprising. We therefore checked whether exposure of cells to TNF␣ for 5 min caused an increase in the intracellular production of H 2 O 2 by analyzing the oxidation of DCF using flow cytometry. As shown in Fig. 6B, no increases in DCF oxidation were observed in cells exposed to 10 ng/ml TNF␣ for 5 min compared with sham controls. As a positive control, cells exposed to a bolus of H 2 O 2 displayed a marked oxidation of DCF. These results demonstrate that the concentration of intracellular H 2 O 2 produced in response to TNF␣ may not be sufficient to cause IKK oxidation. DISCUSSION The transcription factor NF-B is an important regulator of immune responses. It is induced by a wide variety of stresses and transcriptionally activates many cytokine and chemokine A, cells were exposed to 500 M H 2 O 2 in the presence or absence of LY294002, and a Western blot was performed to visualize pAkt and total Akt. B, IKK enzymatic activity in cells exposed for 5 min to TNF␣ and H 2 O 2 in the presence or absence of LY294002. C, Western blot analysis examining phosphotyrosine (pTyr)-containing proteins following exposure to H 2 O 2 in the presence or absence of herbimycin A. ␤-Actin was determined as a loading control. D, IKK enzymatic activity in cells exposed for 5 min to TNF␣ and H 2 O 2 in the presence or absence of herbimycin A. DMSO, Me 2 SO. genes, genes important in proliferation, matrix degradation, and prevention of apoptosis. Oxidative stress has been thought to play a critical role in its activation based upon many observations that demonstrate that cytokine-induced NF-B can be prevented following treatment with antioxidants or metal chelators. The role of oxidants per se in the activation of NF-B has been the subject of considerable debate (7,26). It appears that the ability of H 2 O 2 to induce NF-B depends on the cell type being investigated and may be linked to the levels of antioxidants present in those cells (7,25,26). However, even within one cell type marked variations in the ability of H 2 O 2 to activate NF-B are apparent. For instance, among investigated T cell lines, Wurzburg T cells but not Jurkat T lymphocytes are responsive to H 2 O 2 (for a review, see Ref. 26). In support of this variability, our laboratory demonstrated in a line of RLE cells that H 2 O 2 caused activation of an NF-B-driven luciferase reporter gene (20) in contrast to our present observations obtained in the mouse counterpart, C10 cells. The differences in the response to H 2 O 2 between C10 and RLE cells could be due to differences in culture conditions, antioxidant defenses, or time frames that were analyzed. The increases in NF-Bdependent luciferase activity in RLE cells treated with H 2 O 2 occurred without the preceding activation of IKK (Fig. 2D) or the degradation of IB␣ (20). This observation is consistent with findings by other groups demonstrating that the PI3K/Akt pathway, induced by oxidants (23,24), is able to activate NF-B by dissociating tyrosine 42-phosphorylated IB␣ from NF-B without degrading IB␣ (17,18,21,22). Other studies have demonstrated that tyrosine phosphorylation can also prevent signal-induced degradation of IB␣ and binding of NF-B to DNA (35,36). In contrast to the aforementioned studies, phosphorylation of tyrosine 42 of IB␣ induced by H 2 O 2 or pervanadate can in fact result in degradation of IB␣ (22,37), illustrating the conflicting nature of the studies published to date. In addition to the regulation of IKK and IB␣ by oxidants, oxidants also affect NF-B-dependent transcription. Phospho  6. A, oxidation of cysteine residues in the IKK complex by H 2 O 2 (500 M). Cells were exposed to TNF␣ for 5 min, and the IKK complex was isolated by immunoprecipitation. Binding of BIAM to the immunoprecipitated IKK complex from TNF␣-treated cells was detected with horseradish peroxidase-streptavidin. IKK␥ levels were assessed as a loading control. B, assessment of oxidation of DCF-DA in C10 cells exposed for 5 min to TNF␣ or H 2 O 2 by flow cytometry. histone deacetylases might be equally sensitive to oxidants, and different histone deacetylases might be expressed in different cell lines cells.
The IKK complex is an important point of convergence used by many different stimuli to activate NF-B. Upstream kinases in the NF-B pathway that include MEKK1 and NF-B-inducing kinase phosphorylate IKK␣ on serines 176 and 180 and IKK␤ on serines 177 and 181 in their kinase domain. Mutation of these serine residues to alanines leads to inactivation of the IKK complex and inhibits the activation of NF-B. Importantly IKK is also an important target for negative regulation of NF-B (41)(42)(43)(44)(45)(46). Of the two catalytic subunits of the IKK complex, IKK␤ has the major role in responding to pro-inflammatory stimuli (5,14) and is sensitive to inactivation by aspirin, salicylate, cyclopentenone prostaglandins, and the thiol-reactive metal arsenite (5,26,47). Since no information exists about the direct effects of H 2 O 2 on the regulation of IKK, we investigated in the present study whether this oxidant could directly affect the enzymatic activity. In support of previous observations demonstrating a lack of degradation of IB␣ in response to oxidative stress, our present data also reveal a lack of activation of IKK in cells treated with bolus H 2 O 2 or H 2 O 2 generated continuously by GOx.
Despite the lack of effect of H 2 O 2 itself on IKK activity, H 2 O 2 affected signal-induced IKK activation by causing a marked decrease in enzyme activity. This effect was independent of Akt or tyrosine kinase activity and was also observed when H 2 O 2 was added to immunoprecipitated, active IKK. These results strongly suggest that H 2 O 2 is capable of directly inactivating IKK due to the redox sensitivity of the complex. This redox sensitivity was further demonstrated in experiments using BIAM (29), which detects oxidation of cysteine residues. Marked cysteine oxidation of the enzymatically active IKK complex occurred following treatment with H 2 O 2 . We detected a remarkable difference in cysteine oxidation between sham and TNF␣-treated cells exposed to H 2 O 2 , indicating that only the active IKK complex is sensitive to H 2 O 2 -induced cysteine oxidation. This explains the lack of inhibition of H 2 O 2 on baseline IKK activity as observed in this study.
At present, we do not know the cysteine site(s) of the IKK complex that may be oxidatively modified by H 2 O 2 . One likely candidate is cysteine 179 in the kinase domain of IKK␤. Recent studies demonstrated that that the cyclopentenone prostaglandin 15-deoxy-⌬ 12-14 -prostaglandin J 2 covalently modified cysteine 179 of IKK␤ decreasing its enzymatic activity and consequently blocking NF-B activation (41). Similarly arsenite, which is reactive toward vicinal thiols, binds cysteine 179 of IKK␤, inhibiting enzymatic activity of the IKK complex (43). Consequently a mutant of IKK␤, in which cysteine 179 is replaced with alanine, is refractory to inhibition by arsenite or 15-deoxy-⌬ 12-14 -prostaglandin J 2 . Cysteine 179 is located between serines 177 and 181 in the kinase domain. Oxidative modification of this amino acid might alter the conformation of the complex and may prevent upstream kinases from phosphorylating the serines 177 and 181, thereby preventing activation of the complex. Therefore, cysteine 179 of IKK␤ also appears to be a likely candidate for oxidation and enzymatic inactivation by H 2 O 2 .
The decrease in TNF␣-induced IKK activity observed in the presence of H 2 O 2 translated into a diminished degradation of IB␣, decreased NF-B binding to DNA, and a decrease in the NF-B-dependent transcriptional activation. These results suggest that under inflammatory conditions where H 2 O 2 is formed NF-B activation could become depressed. The ramifications of this oxidant effect are unclear. The decreased expression of pro-inflammatory mediators may be associated with a dampened inflammatory response in response to proinflammatory stimuli, which could be beneficial to disorders associated with chronic inflammation. Alternatively the decreased ability to activate NF-B in the presence of H 2 O 2 may be associated with an enhancement of apoptosis when inflammatory cytokines are present. We currently are investigating the ramifications to pulmonary epithelium following combined exposures to cytokines and oxidants.