Participation of the calcium/calmodulin-dependent kinases in hydrogen peroxide-induced Ikappa B phosphorylation in human T lymphocytes.

NF-kappaB is an important transcription factor that has a role in a variety of responses such as inflammation, oncogenesis, apoptosis, and viral replication. Oxidative stress is well known to induce the activation of NF-kappaB. Cells can be exposed to either endogenously produced oxidants or oxidants produced by surrounding cells. In addition, ischemia reperfusion and certain cancer therapies such as chemotherapy and photodynamic therapy are thought to result in oxygen radical production. Because of the important role that NF-kappaB has in multiple responses, it is critical to determine the mechanisms by which oxidative stress induces NF-kappaB activity. We report that the calmodulin antagonist W-7 and the calcium/calmodulin-dependent (CaM) kinase inhibitors KN-93 and K252a, can block oxidative stress-induced IkappaB phosphorylation in Jurkat T lymphocytes. Furthermore, KN-93 but not KN-92 can block hydrogen peroxide-induced Akt and IKK phosphorylation. In addition, we found that expression of a kinase-dead CaM-KIV construct in two cell lines inhibits IkappaB phosphorylation or degradation and that expression of CaM-KIV augments hydrogen peroxide-induced IkappaB phosphorylation and degradation. Although the CaM kinases appear to be required for this response, increases in intracellular calcium do not appear to be required. These results identify the CaM kinases as potential targets that can be used to minimize NF-kappaB activation in response to oxidative stress.

NF-B transcription factor is a heterodimer composed of p65 and p50 Rel family members (1,2). NF-B is in an inactive state when bound to IB (inhibitor of B). NF-B bound to IB is sequestered in the cytoplasm of the cell were it cannot act as a transcriptional regulator (3). This is because the binding of IB to NF-B masks the nuclear localization region on NF-B (4). Phosphorylation of IB targets it for ubiquitination and proteolysis (5)(6)(7). Following IB degradation, the nuclear localization region of NF-B is revealed and it translocates to the nucleus where it acts as a transcription factor (2,7). One complex that is known to phosphorylate IB is IB kinase (IKK) 1 (6). Akt, MEKK (mitogen-activated or extracellular-regulated kinase kinase), and NF-B-inducing kinase have been reported to induce the phosphorylation of IKK (8,9). There are also other pathways that lead to IB phosphorylation. Ultraviolet light is reported to induce NF-B activation in the absence of IKK activation (10). Although a great deal has been learned about the Rel family of transcription factors, researchers are only just beginning to understand the upstream regulation of these important proteins (11).
Reactive oxygen intermediates, such as hydrogen peroxide, are also known to induce NF-B activity in many cell types (12)(13)(14). In addition to reactive oxygen intermediates, NF-B activation has been reported following exposure of T lymphocytes to calcium ionophores, the tumor promoter PMA, or antibodies to the T-cell receptor (15). Understanding the signaling pathways induced by oxygen radicals, and the role that these pathways have in cell survival and death, could allow one to inhibit the survival pathways thereby promoting cell death. Oxygen radicals are also known to induce the production of several inflammatory cytokines that have NF-B responsive promoters (16 -18). Alternatively, one may be able to constrain the oxidant-induced production of these inflammatory mediators with CaM kinase inhibitors.
The CaM kinases are a group of related kinases that are activated in response to increased calcium levels (reviewed in Ref. 19). The primary members of this family which demonstrate broad substrate specificity are CaM kinase I, II, and IV. The other members of this family are dedicated to the phosphorylation of a single substrate (i.e. myosin light chain kinase, phosphorylase kinase, and elongation factor 2 kinase/CaM-KIII). CaM-KI is broadly distributed, CaM-KII is found in 1 The abbreviations used are: IKK, inhibitor of B kinase; CaM, calcium/calmodulin-dependent kinase; IB, inhibitor of B; NF-B, nuclear factor of B; PMA, phorbol 12-myristate 13-acetate; BAPTA-AM, 1,2-bis(O-aminophenoxy)ethane-N,N,NЈ,NЈ-tetraacetic acid tetra(acetoxymethyl)ester; ERK, extracellular signal-regulated kinase; FCS, fetal calf serum; Me 2 SO, dimethyl sulfoxide; PBS, phosphate-buffered saline. many tissues but is expressed at the highest level in neural tissue, whereas CaM-KIV is found primarily in brain, testes, and T lymphocytes. To elicit maximal activation of CaM-KII and CaM-KIV, both calmodulin-calcium complexes and phosphorylation events are thought to be required. However, the mechanism by which these two enzymes are phosphorylated differs. CaM-KII reportedly undergoes autophosphorylation upon calcium/calmodulin binding, whereas CaM-KIV is phosphorylated by the kinase CaM-KK (20). CaM-KK is activated by calcium/calmodulin complexes as well, but the potential for other means of regulation has been suggested (21). Because CaM-KII and CaM-KK phosphorylate Akt (22,23) and Akt is reported to activate IKK, it is conceivable that the activation of the CaM kinases could have a role in NF-B activation. In this report, we demonstrate that the CaM kinases also appear to have a role in hydrogen peroxide-induced IB phosphorylation.

EXPERIMENTAL PROCEDURES
Cells and Reagents-The Jurkat cell line was obtained from ATCC (Rockville, MD). The Wurzburg cell line was obtained through the kindness of Dr. Leonard Herzenberg (Stanford University, Palo Alto, CA). These cells were cultured in RPMI 1640 with 5% fetal calf serum, 2 mM glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin. Human peripheral blood lymphocytes were isolated from platelet-depleted blood by centrifugation over a Histopaque gradient. T lymphocytes were isolated as previously described by rosetting with sheep red blood cells (24,25). The isolated primary T cells were rested overnight in RPMI 1640 with 5% fetal calf serum, 2 mM glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin prior to use. The COS-7 and 293T cell lines were cultured in Dulbecco's modified Eagle's medium with 10% fetal calf serum, 2 mM glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin. Anti-IB␣, anti-phospho-IB␣(Ser 32/36 ), antiphospho-IKK(Ser 180/181 ), anti-Akt, and anti-phospho-Akt(Thr 308 ) were purchased from Cell Signaling Technology (Beverly, MA). Anti-phospho-ERK (Thr 183 and Tyr 185 ) was purchased from Promega (Madison, WI). KN-92, KN-93, K252a, and BAPTA-AM were purchased from Calbiochem (San Diego, CA) and dissolved in Me 2 SO. Fluo-3 was purchased from Molecular Probes (Eugene, OR) and dissolved in Me 2 SO. W-7 was purchased from Sigma and dissolved in Me 2 SO.
Sample Preparation for Immunoblot Analysis-Cells were washed and resuspended in serum-free RPMI 1640. One ml, containing 1.25 ϫ 10 6 cells, was added to microcentrifuge tubes and placed at 37°C for at least 1 h prior to the start of the experiment. Cells were stimulated with hydrogen peroxide. Following stimulation for various times, the tubes were centrifuged in a microcentrifuge for 30 s, the supernatants were removed, cell pellets were resuspended in 110 l of cold lysis buffer (25 mM Tris-HCl, pH 7.4, 50 mM NaCl, 0.5% sodium deoxycholate, 2% IGEPAL, 0.2% SDS, 1 mM phenylmethylsulfonyl fluoride, 50 g/ml aprotinin, 50 M leupeptin, 0.5 mM Na 3 VO 4 ) and placed on ice for 15 min. Lysates were centrifuged for 10 min at 14,000 rpm in an Eppendorf microcentrifuge, supernatants (98 l) were removed and mixed with 42 l of 3.3ϫ sample buffer (200 mM Tris-HCl, pH 6.8, 33% glycerol, 6.6% SDS, 16.6% 2-mercaptoethanol, 0.04% bromphenol blue). Samples were boiled (5 min) and frozen.
Measurement of Intracellular Calcium-To determine the levels of intracellular calcium induced following stimulation, cells were loaded with Fluo-3. This was done by incubating 5 ϫ 10 6 cells/ml for 30 min with 10 M Fluo-3 at 37°C in RPMI 1640 (serum-and phenol red-free). Following labeling, cells were washed and resuspended at 5 ϫ 10 6 cells/ml in Hanks' balanced salt solution. Cells were stimulated with OKT3 (5 g/ml) cross-linked with rabbit anti-mouse Ig (5 g/ml), or with hydrogen peroxide (1 mM) in the presence or absence of EGTA. To determine changes in intracellular calcium, fluorescence was monitored in 200 l of cells (1 ϫ 10 6 ) in black opaque 96-well Corning microtiter plates on a Bio-Tek FL600 fluorescence absorbance spectophotometer , or a combination of both. The cells were then stimulated for 30 min with hydrogen peroxide (10 mM), cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using an antibody against IB␣ (Cell Signaling Technology, Beverly, MA) or phospho-ERK (Promega). Immunoblotting was performed as previously described (40, 54 -57).
(Winooski, VT). The wavelength for excitation was 485 nm and emission was measured at 530 nm.
Transfection of CaM-KIV-Transient transfections were performed using either a truncated activable form of CaM-KIV, a kinase-dead CaM-KIV, or the empty vector pSG5 (26,27). 293T cells were transfected by calcium phosphate precipitation. Briefly, 293T cells (grown to 60 -70% confluence) were washed and placed in fresh media (10 ml). Twenty five g of plasmid in 2.5 ml of 250 mM CaCl 2 was added to an equal volume of 2 ϫ HBS (280 mM NaCl, 50 mM HEPES, 1.5 mM Na 2 HPO 4 , 10 mM KCl, 10 mM glucose) under constant agitation. This mixture was added to cells resulting in a final volume of 15 ml. Cells were then incubated for 10 h at 37°C. Following this incubation, cells were washed and placed back into culture. Cells were then harvested 36 h after transfection and treated as indicated. COS-7 cells (5 ϫ 10 6 cells per transfection) were transfected with 25 g of plasmid DNA by electroporation at 400 kV and 500 microfarads using a Bio-Rad Gene Pulser II apparatus (Bio-Rad). COS-7 cells were harvested 36 h after transfection and treated as indicated.
Measurement of Intracellular Glutathione Levels-Cells (1.25 ϫ 10 6 cells/ml) were treated with KN-93, KN-92, or Me 2 SO for 30 min at 37°C. The cells were then stimulated for 30 min with hydrogen peroxide at 37°C, the cells were then pelleted, the supernatant was removed, and the cellular pellet was resuspended in 100 l of 4% sulfosalicylic acid. The acidified samples were allowed to sit 1 h at room temperature and then frozen at Ϫ70°C until they were assayed. Immediately prior to assaying the samples, they were diluted such that when assayed they had a final concentration of 1% 5-sulfosalicyclic acid. For determination of reduced GSH, 100 l of sample was treated with 3.9 l of a 1:4 dilution triethanolamine prior to assay to raise the pH to 7.0. For determination of GSSG, 100 l of sample was treated with 3.9 l of 2-vinylpyridine prior to the addition of 3.9 l of a 1:4 dilution triethanolamine. Samples were then assayed using the enzymatic recycling assay described by Baker et al. (28). Briefly, 50 l of sample or standard was mixed with 100 l of buffer (47.2 mM sodium phosphate, 472 M EDTA, 226 M 5,5Ј-dithiobis(nitrobenzoic acid), 313 M NADPH, 20 units of GSH reductase, pH 7.5) and the rate of the change in the optical density at 405 nm was determined over the linear part of the assay using a fluorescence/absorbance plate reader (FL600; Bio-Tek).

RESULTS AND DISCUSSION
We have found that treatment of Jurkat cells with hydrogen peroxide results in the phosphorylation of IB␣ (Fig. 1A). The shift in mobility noted on these immunoblots is reported to be because of phosphorylation of IB (29,30). Unlike what was reported in other cell types, we have found that treatment of the Jurkat cell line with hydrogen peroxide does not result in IB degradation. This is in agreement with the findings of others that demonstrate hydrogen peroxide treatment of Jurkat cells fails to induce NF-B activity (31). Treatment of the Wurzburg cell line (a Jurkat cell derivative; Fig. 1B) and primary T lymphocytes (T lymphocytes isolated from the peripheral blood by erythrocyte rosetting; Fig. 1C) with hydrogen peroxide did result in IB degradation.
It has been known for some time that Jurkat T cells are refractory in the ability to activate the transcription factor NF-B in response to hydrogen peroxide when compared with the European Jurkat cell line derivative, Wurzburg (31). Because IB is phosphorylated in these experiments and the degradation following IB phosphorylation is the last step leading to NF-B activation, we would have been surprised to see degradation occurring in Jurkat cells. The reason for the difference in the ability to degrade NF-B between these two closely related cell lines (Jurkat and Wurzburg) is not known. Phosphorylation of IB could be noted with 1 mM hydrogen peroxide in Jurkat cells.
Although Jurkat cells require higher levels of hydrogen peroxide than other cell types to induce IB phosphorylation, the advantage of using these cells is that they do not degrade IB or activate NF-B in response to oxidative stress. When IB is degraded, it allows NF-B to move to the nucleus were it can  KN-93 (1 M). The cells were then stimulated for 30 min with hydrogen peroxide (10 mM), cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using an antibody against IB␣. Immunoblotting was performed as previously described (40, 54 -57).

FIG. 3. EGTA blocks hydrogen peroxide-induced increases in intracellular calcium.
Jurkat cells were loaded with Fluo-3 (10 M), incubated with or without the addition of EGTA (2 mM) for 30 min, and stimulated with hydrogen peroxide (1 mM) or OKT3 (5 g/ml) crosslinked with rabbit anti-mouse Ig (5 g/ml). Fluo-3 fluorescence was measured using a Bio-Tek fluorescent microtiter plate reader (FL600) over the indicated time period (seconds).
induce the transcription of genes with NF-B-binding sites. One of the genes induced by NF-B encodes for IB (32). This could complicate studies examining IB phosphorylation especially if the treatments altered the kinetics of the response. Transfection of other genes or the use of inhibitors that may have multiple roles both upstream and downstream of IB could potentially lead to a misinterpretation of results, especially if they result in IB transcription. Thus, there is an advantage to using the Jurkat cell line for experiments to identify the pathways leading to IB phosphorylation and the activation of this important transcription factor.
Using both the Jurkat cell line and the Wurzburg cell line together could provide an understanding of the mechanisms that control IB degradation. Identifying the mechanisms that control IB degradation would identify potential targets which could be used to regulate NF-B activation. Methods to prevent the degradation of IB could be used to sensitize some cancer cells to various cancer treatments.
Because hydrogen peroxide is reported to cause a calcium flux in Jurkat T lymphocytes (33) and increases in intracellular calcium have been shown to induce (or synergize with other factors to induce) NF-B activity in T cells and other cell types (32, 34 -39), it was determined whether increases in intracellular calcium were involved in oxidative stressinduced IB phosphorylation. It was found that EGTA, BAPTA-AM, nor a combination of both blocked IB phosphorylation induced by oxidative stress (Fig. 1, A-C). We previously published (40) that increases in intracellular calcium would induce ERK activation in Jurkat cells and primary T lymphocytes. EGTA alone or BAPTA-AM in conjunction with EGTA blocked ionomycin-induced ERK activation in parallel experiments (Fig. 1D). These results suggest that increases in intracellular calcium are not required for oxidative stressinduced IB phosphorylation. Concurrent to these experiments, we performed experiments using the CaM kinase inhibitor KN-93 and the inactive analog of KN-93, KN-92. We found that KN-93, but not the inactive analog KN-92, blocked hydrogen peroxide-induced IB phosphorylation in both Jurkat and primary T lymphocytes (Fig. 2).
To assure that hydrogen peroxide-induced calcium flux was blocked in these cells, we treated Jurkat cells for 30 min with 2 mM EGTA (similar to Fig. 1). We then stimulated the cells with 1 mM hydrogen peroxide. We found, as has been previously reported, that both antibodies to the T cell receptor (OKT3) and hydrogen peroxide induced a calcium flux in these cells (Fig. 3) (24,33). Treatment of the Jurkat cells with EGTA, prior to stimulation with hydrogen peroxide, blocked increases in intracellular calcium (Fig. 3). These data demonstrate that we were able to block increases in intracellular calcium into these cells with EGTA.
Another possible explanation for the results obtained using KN-93 is that it is acting as an antioxidant. To test this, we measured the levels of oxidized glutathione in cells in the presence of KN-93, as glutathione is the major antioxidant within cells (41). Following an oxidative stress, one would expect to detect more glutathione in the oxidized form. This is exactly what was observed following hydrogen peroxide stimulation of Jurkat cells (Fig. 4). The addition of KN-93 did not significantly alter the level of oxidized glutathione in either unstimulated or hydrogen peroxide-stimulated Jurkat cells. These data indicate that KN-93 is not acting as an antioxidant.
We also performed experiments using a second CaM kinase inhibitor, K252a. This inhibitor, although it does not have a good negative control associated with it like KN-93, is structurally dissimilar to KN-93. We found that K252a (IC 50 ϭ 1.8 nM for CaM-kinase), like KN-93, blocked hydrogen peroxideinduced IB phosphorylation (Fig. 5). The ability of multiple CaM kinase inhibitors to block this response strengthens the argument that these inhibitors are acting via CaM kinase and not nonspecifically.
Although an increase in intracellular calcium is not required for IB phosphorylation in response to hydrogen peroxide, the CaM kinases appear to be required. These were surprising results and suggested that in response to hydrogen peroxide calcium-independent activation of the CaM kinases was occurring. In a recent article by Hughes et al. (51) it was demonstrated that PMA, which is not known to increase levels of intracellular calcium in T lymphocytes, induced NF-B activation in a CaM kinase-dependent man- The cells were harvested and the level of oxidized glutathione within these cells was determined using an enzymatic recycling assay as previously described (28). FIG. 5. K252a blocks IB phosphorylation induced by hydrogen peroxide. Jurkat cells were washed and resuspended in RPMI 1640 containing 5% FCS. The cells were warmed to 37°C and treated for 30 min with the indicated concentration of K252a. The cells were then stimulated for 30 min with hydrogen peroxide (10 mM), cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using an antibody against IB␣. Immunoblotting was performed as previously described (40, 54 -57). The cells were then stimulated for 30 min with PMA (100 nM), cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using an antibody against IB␣. Immunoblotting was performed as previously described (40, 54 -57). ner. We have repeated some of these experiments and we too found that KN-93, but not KN-92, inhibited IB degradation in response to PMA (Fig. 6).
The mechanism of activation of CaM-KII and CaM-IV is thought to differ, however, both reportedly require calmodulincalcium complexes. For maximal activity CaM-KII and CaM-KIV are also reported to require phosphorylation. The mechanism by which these two enzymes are phosphorylated differs. CaM-KII reportedly undergoes autophosphorylation upon calcium/calmodulin binding and CaM-KIV is phosphorylated by the kinase CaM-KK (20). CaM-KK is reportedly activated by calcium/calmodulin complexes as well, but the potential for other means of regulation has been suggested (21). Phosphorylation of CaM-KIV by CaM-KK occurs in the pseudosubstrate subdomain that is contained in a regulatory domain and interacts with the catalytic domain of the kinase. Once phosphorylated these kinases retain catalytic activity even after calcium levels decrease. The phosphatases PP1 and PP2A have been reported to cleave the phosphate group off of CaM-KII rendering it inactive. The phosphatase PP2A has been reported to cleave the phosphate off of CaM-KIV (42). It is known that hydrogen peroxide and okadaic acid will inhibit PP2A and PP1 (43)(44)(45)(46)(47)(48). Inhibition of this phosphatase could conceptually re-sult in the activation of the CaM-KIV in the absence of increases in intracellular calcium.
We found that incubation of the Jurkat cells for 30 min with the calmodulin antagonist, W-7, inhibits IB phosphorylation (Fig. 7). These results suggest that calmodulin is needed for the effects but not calcium. It is possible that calmodulin perhaps undergoes some other poststimulatory modification(s) that induces the binding of calmodulin to the CaM kinases. There is some evidence that indicates calmodulin can be phosphorylated on certain residues following stimulation of certain cells with epidermal growth factor. On the other hand, we have preliminary evidence that CaM-KK is tyrosine phosphorylated following hydrogen peroxide stimulation of these cells. This too could potentially influence the ability of calmodulin to bind CaM kinases or directly influence the activity of the enzyme (49,50). We do know that CaM kinase activity in these cells is increased following hydrogen peroxide treatment in the absence of any detectable calcium flux. 2 To determine whether CaM-KIV could potentially have a role in hydrogen peroxide-induced IB phosphorylation or degradation, we transiently transfected COS-7 and 293T cells with a truncated activable form or an truncated inactive form of CaM-KIV that was previously shown to act as a dominantnegative inhibitor of CaM-KIV (26,27). Cellular transfection was monitored by a FLAG epitope placed on the constructs and also by the appearance of a CaM-KIV reactive protein (at the correct molecular weight for the truncated protein) in the construct transfected but not the vector-transfected cells (data not shown). We found that when CaM-KIV was transfected into 293T cells, a slight reproducible enhancement of the relative level of IB was found in the phosphorylated (shifted) form (Fig. 8). In contrast, when the dominant-negative form of CaM-KIV was transfected into 293T cells, a slight diminishment of the relative level of the shifted form of IB could be noted. We FIG. 7. The calmodulin antagonist W-7 inhibits hydrogen peroxide-induced IB phosphorylation. Jurkat cells were washed and resuspended in RPMI 1640 containing 5% FCS. The cells were warmed to 37°C and treated for 30 min with 40 M W-7. The cells were then stimulated for 30 min with hydrogen peroxide (10 mM), cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using an antibody against IB␣. Immunoblotting was performed as previously described (40, 54 -57).  (26,27). Cells were harvested 48 h post-transfection and stimulated with 10 mM hydrogen peroxide for 30 min. Lysates were prepared and immunoblotted with an antibody against IB␣. Transfection of the constructs was monitored by immunoblotting for both the FLAG epitope and the truncated protein. Immunoblotting was performed as previously described (40, 54 -57) .   FIG. 9. Hydrogen peroxide induces the phosphorylation of Akt. Jurkat cells were washed and resuspended in RPMI 1640 containing 5% FCS. The cells were warmed to 37°C and treated for the indicated time with hydrogen peroxide or ionomycin. Cellular lysates were prepared and subjected to immunoblot analysis using an antibody against total Akt (A) (Cell Signaling Technology), phospho-T308-Akt (B) (Cell Signaling Technology), and IB␣ (C). Immunoblotting was performed as previously described (40, 54 -57).
also examined the effect of these constructs on IB degradation in COS-7 cells (Fig. 8). In a separate set of immunoblots, we found that COS-7 cells do not express detectable levels of CaM-KIV, but do express detectable levels of CaM-KK (data not shown). When we transfected dominant-negative CaM-KIV into these cells, hydrogen peroxide-induced IB degradation did not appear to be inhibited. This result is expected if these cells do not express endogenous CaM-KIV. If the activable form of CaM-KIV is trans-fected into these cells a slight reproducible increase in the degradation of IB can be noted. These data suggest that hydrogen peroxide could potentially act directly on CaM-KIV or upstream at CaM-KK. The fact that complete inhibition of IB phosphorylation could not be demonstrated in these cells with kinase-dead CaM-KIV could reflect either the presence of a second CaM kinase participating in this pathway or only a partial transfection of our population of cells. In support of the former hypothesis, Hughes et al. (51) demonstrated that transfection of cells with a kinase-dead CaM-KII inhibited PMA-induced IB activation. Both CaM-KK and CaM-KII have been shown to phosphorylate Akt, so it is possible that both the CaM-KIV and CaM-KII pathways participate in this response.
To determine whether Akt and IKK could act downstream of the CaM kinases, we determined if Akt and IKK were phosphorylated in response to hydrogen peroxide and if the phosphorylation of these kinases could be blocked by KN-93 but not KN-92. The phosphorylation of Akt on threonine 308 was very , cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using an antibody against total Akt (A) or phospho-T308-Akt (B). Immunoblotting was performed as previously described (40, 54 -57). The cells were then stimulated for 30 min with hydrogen peroxide (10 mM) or ionomycin (500 nM), cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using antibodies against phospho-T308-Akt (A), IB␣ (B), or phospho-p38 (C). Immunoblotting was performed as previously described (40, 54 -57). Jurkat cells were washed and resuspended in RPMI 1640 containing 5% FCS. The cells were warmed to 37°C and treated for the indicated time with hydrogen peroxide (1 mM) or PMA (100 nM). Cellular lysates were prepared and subjected to immunoblot analysis using an antibody against total IB or phospho-IB(Ser 32/36 ). Immunoblotting was performed as previously described (40, 54 -57). rapid, occurring within minutes, and preceded the kinetics of IB phosphorylation (Fig. 9). Ionomycin treatment alone did not induce the phosphorylation of Akt on this residue and the level of total Akt protein did not change in response to hydrogen peroxide or ionomycin (Fig. 9). Thus, although the CaM kinases may be required for this response they are not sufficient in themselves to induce Akt activation. In preliminary experiments we have found that hydrogen peroxide also induces the phosphorylation of Akt on serine 473 (data not shown). Akt does appear to be activated in response to hydrogen peroxide because the Forkhead protein, which is thought to be immediately downstream of Akt in a kinase cascade, is phosphorylated as determined using phospho-FKHR antibodies (Cell Signaling Technology; data not shown). To determine whether hydrogen peroxide-induced phosphorylation of Akt was potentially mediated via the calcium-independent activation of the CaM kinases, we treated cells with the CaM kinase inhibitor KN-93, EGTA, BAPTA-AM, or a combination of EGTA and BAPTA-AM (Figs. 10 and 11). Similar to the hydrogen peroxide-induced phosphorylation of IB, we found that phosphorylation of Akt on residue threonine 308 was independent of a calcium flux (Fig. 10) and could be inhibited by KN-93 FIG. 14. Dose response of hydrogen peroxide-induced cell death and Akt activation. A, Jurkat cells were washed and resuspended in RPMI 1640 containing 5% FCS. The cells were warmed to 37°C and then stimulated for 30 min with the indicated concentration of hydrogen peroxide, cellular lysates were prepared, and the lysates were subjected to immunoblot analysis using an antibody against phospho-T308-Akt. Immunoblotting was performed as previously described (40, 54 -57). B-D, cells were treated with the indicated concentration of hydrogen peroxide and examined for apoptosis by Annexin-V (B, 2.5 h), propidium iodide (C, 2.5 h), or subdiploid levels of DNA using propidium iodide (D, 5 h). In B and C untreated cells are shown by the dotted lines and treated cells by the solid lines in the histograms. (Fig. 11). These results suggest that a CaM kinase may be involved in the hydrogen peroxide-induced phosphorylation of Akt. Although KN-93 is reported to inhibit CaM-KII and CaM-KIV (52), its ability to inhibit the related CaM-KK has not been reported. Thus, it appears that a CaM kinase is mediating these effects, although it is unknown which CaM kinase is involved. Because Akt is reported to induce the phosphorylation of IKK we determined the kinetics and dose response of IKK phosphorylation following stimulation with hydrogen peroxide. We found, using a phosphospecific IKK antibody that hydrogen peroxide rapidly induced the phosphorylation of IKK (Fig. 12). In addition, IKK phosphorylation by hydrogen peroxide was inhibited by KN-93, but not KN-92, indicating the involvement of the CaM kinases in this response. Phosphorylation of IKK could be noted at 500 M concentrations of hydrogen peroxide (Fig. 12C). We examined the phosphorylation of IB using an antibody that detects IB when phosphorylated on serines 32 and 36, the two sites of phosphorylation that occur because of IKK activation. Using this antibody we found that IB was phosphorylated on serines 32 and 36 (Fig. 13).
Using the antibody toward phosphorylated Akt (one of our most sensitive indicators of the activation of this pathway) we found that levels as low as 100 M would induce the detectable phosphorylation of Akt (Fig. 14A). The apparent differences of the concentrations of hydrogen peroxide required to induce the phosphorylation of Akt, IKK, and IB may be because of differences in the sensitivity of the antibodies. Many of the factors that induce apoptosis also induce NF-B activation. To determine whether the effects of hydrogen peroxide were being mediated via a decrease in the structural integrity of the cell we performed Annexin-V/propidium iodide staining and propidium iodide subdiploid staining of hydrogen peroxide-treated Jurkat cells at 2.5 and 5 h, respectively. We found that only the highest concentration of hydrogen peroxide induced the signs of apoptosis and that apoptosis of the cells could not be noted at lower concentrations that still induced this pathway (Fig. 14B). When we used trypan blue to assess the viability of the cells, we found that although some toxicity was noted at the 10 mM dose of hydrogen peroxide, at 5 mM concentrations and below no toxicity could be noted (data not shown). This is similar to what was recently reported with primary human T lymphocytes in which no toxicity was noted at concentrations as high as 10 mM (53).
In this report, we have demonstrated that hydrogen peroxideinduced IB phosphorylation occurs via a CaM kinase-dependent manner. These results suggest that modulators of the CaM kinase pathways could be used to influence NF-B activation that occurs because of oxidative stress. Because it is speculated that oxygen radicals are involved in the inflammatory processes that can give rise to atherosclerosis, inhibition of the CaM kinases may benefit these individuals. Alternatively, because NF-B is thought to have anti-apoptotic properties, this type of inhibition could sensitize cancer cells to certain chemotherapeutic agents which act through the generation of oxygen radicals.