Epidermal growth factor receptor-dependent Akt activation by oxidative stress enhances cell survival.

The serine/threonine kinase Akt (also known as protein kinase B) is activated in response to various stimuli by a mechanism involving phosphoinositide 3-kinase (PI3-K). Akt provides a survival signal that protects cells from apoptosis induced by growth factor withdrawal, but its function in other forms of stress is less clear. Here we investigated the role of PI3-K/Akt during the cellular response to oxidant injury. H(2)O(2) treatment elevated Akt activity in multiple cell types in a time- (5-30 min) and dose (400 microM-2 mm)-dependent manner. Expression of a dominant negative mutant of p85 (regulatory component of PI3-K) and treatment with inhibitors of PI3-K (wortmannin and LY294002) prevented H(2)O(2)-induced Akt activation. Akt activation by H(2)O(2) also depended on epidermal growth factor receptor (EGFR) signaling; H(2)O(2) treatment led to EGFR phosphorylation, and inhibition of EGFR activation prevented Akt activation by H(2)O(2). As H(2)O(2) causes apoptosis of HeLa cells, we investigated whether alterations of PI3-K/Akt signaling would affect this response. Wortmannin and LY294002 treatment significantly enhanced H(2)O(2)-induced apoptosis, whereas expression of exogenous myristoylated Akt (an activated form) inhibited cell death. Constitutive expression of v-Akt likewise enhanced survival of H(2)O(2)-treated NIH3T3 cells. These results suggest that H(2)O(2) activates Akt via an EGFR/PI3-K-dependent pathway and that elevated Akt activity confers protection against oxidative stress-induced apoptosis.

Oxidative stress poses a major threat to organisms living in an aerobic environment and is believed to play a causative role in many disease processes. Cells respond to oxidant injury with the activation of multiple signal transduction pathways that serve to coordinate the cellular response and ultimately determine the outcome. Depending on the particular stimulus encountered or the cell type involved, the response can range from proliferation and transformation, to growth arrest or cell death (1)(2)(3). Among the major signaling pathways and/or key mediators known to influence survival of cells subjected to oxidant injury are the phosphorylation cascades leading to activation of mitogen-activated protein kinases (MAPK) 1 including extracel-lular signal-regulated kinase (ERK), c-Jun N-terminal kinase (JNK) and p38 (4 -6), NF-B (7), and the tumor suppressor protein p53 (8).
The serine/threonine kinase Akt, also known as protein kinase B, was originally identified as the cellular homolog of the v-akt oncogene (9). It is activated via a phosphoinositide 3-kinase (PI3-K)-dependent signaling pathway when cells or tissues are exposed to growth factors, insulin, certain cytokines, and integrin-linked extracellular stimuli (10 -16). Akt has received widespread attention as an important anti-apoptotic protein through which various survival signals suppress cell death induced by growth factor withdrawal, cell cycle discordance, and detachment of cells from their extracellular matrix (17)(18)(19)(20)(21)(22)(23). However, its potential role in influencing cell fate during other conditions of stress is less clear.
Several reports have shown that Akt can be activated by some stresses such as heat, hyperosmotic stress, H 2 O 2 , cadmium chloride (CdCl 2 ), and sodium arsenite (24 -27). However, other studies have provided evidence that apoptosis-inducing stresses including ceramide, hyperosmotic stress, UVC, and ionizing radiation result in down-regulation of the PI3-K/Akt pathway (28 -30). Little is known regarding the mechanisms involved in altering Akt activity during stress or the functional significance of such changes in Akt activity. Indeed, the high concentrations of certain of the agents utilized to modulate Akt activity in many of the studies noted above raises concerns about the biologic relevance of the observations.
In the present study we sought to explore the mechanisms involved in activation of Akt during the cellular response to oxidant injury and to determine its influence on cell survival. By using H 2 O 2 as a model oxidant, we show that sub-millimolar concentrations lead to activation of Akt in a variety of cell types, and we further provide genetic, enzymatic, and pharmacological evidence indicating that this occurs through an epidermal growth factor receptor (EGFR)/PI3-K-dependent signaling pathway. Most importantly, using various strategies to inhibit or enhance Akt activity, we present findings suggesting a pivotal role for the PI3-K/Akt pathway in promoting cell survival following oxidant injury.

EXPERIMENTAL PROCEDURES
Materials-Hydrogen peroxide (H 2 O 2 ), phosphatidylinositol (PI), and phosphatidylinositol 3-phosphate (PIP) were purchased from Sigma. Histone H2B was from Roche Molecular Biochemicals. Wortmannin and LY294002 were from Calbiochem. Anti-hemagglutinin (HA) monoclonal antibody, 12CA5, was from Roche Molecular Biochemicals. The anti-Akt1, anti-ERK2, anti-JNK1, and anti-EGFR polyclonal antibodies were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The antiactive MAPK and anti-active JNK polyclonal antibodies were from * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Promega (Madison, WI). The anti-rat PI3-K and anti-phosphotyrosine antibodies were from Upstate Biotechnology, Inc. (Lake Placid, NY). The plasmid expressing hemagglutinin epitope (HA)-tagged Akt has been described (11). The plasmid expressing a dominant-negative mutant form of the p85 subunit of PI3-K (⌬p85), which lacks a binding site for the p110 catalytic subunit, was kindly provided by Drs. Wataru Ogawa and Masato Kasuga (31). The plasmid that expresses a constitutively active mutant form of p110 (pcDNA3-p110-CAAX) was a gift from Dr. Julian Downward (32).
Transfections and Adenovirus Infections-For transient transfection assays, cells were plated into 60-mm dishes and transfected with DNA using LipofectAMINE (Life Technologies, Inc.) according to the manufacturer's instructions. On the following day the transfected cells were placed in serum-free medium and left overnight prior to treatment with H 2 O 2 or other agents.
For adenovirus infections, a recombinant adenovirus construct expressing constitutively active myristoylated Akt (Ad.Akt) and a control adenovirus lacking the Akt insert (Ad.null) were utilized. For infection, HeLa cells were incubated with the adenovirus at 100 plaque-forming units/cell in serum-free growth medium for 1 h. The virus-containing medium was then removed and replaced with fresh medium containing 10% fetal bovine serum. We have demonstrated that these conditions result in infection of Ͼ95% HeLa cells (33). Sixteen h after infection, cells (4 ϫ 10 5 ) were replated in 60-mm dishes. Fourteen h later, cells were exposed to H 2 O 2 .
Akt, ERK, and JNK Kinase Assays-Akt activity was measured as described previously (11). After treatment with H 2 O 2 or other agents, cells were rinsed with ice-cold phosphate-buffered saline. Cells were lysed for 20 min at 4°C in lysis buffer (20 mM Tris-HCl, pH 7.4, 137 mM NaCl, 1 mM Na 3 VO 4 , 20 mM NaF, 1 mM NaPP i , 1% Nonidet P-40, 10% glycerol, 1 mM phenylmethylsulfonyl fluoride, and 5 g/ml each of aprotinin and leupeptin), and insoluble material was removed by centrifugation. Endogenous or HA-tagged Akt was immunoprecipitated with goat polyclonal anti-Akt or mouse monoclonal anti-HA (12CA5) antibodies, respectively, and 30 l of 50% slurry of protein G-Sepharose (Amersham Pharmacia Biotech) for 4 h. The beads containing the immunoprecipitates were washed three times with lysis buffer, once with ice-cold water, and then once with Akt kinase buffer (20 mM Hepes, pH 7.4, 10 mM MgCl 2 , 10 mM MnCl 2 ). To measure kinase activity, immunoprecipitated Akt was added to 30 l of kinase buffer containing 10 Ci of [␥-32 P]ATP and 2 g of histone H2B, the substrate, as reported previously (6). 32 P-Labeled protein was separated by 15% SDS-PAGE and detected by autoradiography, and the relative intensity of labeling was quantified with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). Akt Ser 473 phosphorylation was assessed by immunoblotting with antibody specific for Akt phosphorylated on Ser 473 (New England Biolabs Inc., Beverly, MA). ERK and JNK kinase activities were measured in a similar fashion using myelin basic protein and GST-c-Jun fusion protein as substrates, respectively.
PI3-K Assay-PI3-K enzymatic assays were performed as described previously with some modification (34). Briefly, after stimulation with H 2 O 2 or insulin, cells were washed twice with PBS containing 1 mM CaCl 2 , 1 mM MgCl 2 , and 100 M Na 3 VO 4 . Cells were lysed in 1 ml of lysis buffer containing 20 mM Tris-HCl, pH 7.4, 137 mM NaCl, 1 mM CaCl 2 , 1 mM MgCl 2 , 0.1 mM Na 3 VO 4 , 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 5 g/ml aprotinin, and 5 g/ml leupeptin. Five hundred g of cell lysate was mixed with 5 l of a polyclonal antibody directed against the p85 subunit of PI3-K and 30 l of 50% slurry of protein G-Sepharose beads. The mixture was incubated for 4 h at 4°C and then sequentially washed as follows: twice with the lysis buffer; twice with PBS containing 1% Nonidet P-40; twice with buffer containing 10 mM Tris, pH 7.4, 5 mM LiCl, and 0.1 mM Na 3 VO 4 ; and twice with kinase reaction buffer containing 7 mM Tris, pH 7.6, 125 mM NaCl, 15 mM MgCl 2 , and 0.1 mM Na 3 VO 4 . PI3-K activity was measured using phosphatidylinositol (PI) as a substrate. The reactions were carried out in a 60-l volume containing 7 mM Tris, pH 7.6, 125 mM NaCl, 15 mM MgCl 2 , and 0.1 mM Na 3 VO 4 , 20 M ATP, 20 g of PI, and 30 Ci of [␥-32 P]ATP. After 30 min of incubation at 30°C, the reactions were stopped by addition of 20 l of 6 N HCl. Phospholipids were extracted with 160 l of CHCl 3 /MeOH (1:1, v/v), and equal volume aliquots from the organic phase were resolved by thin layer chromatography on TLC Silica Gel 60 plates (Merck) in chloroform/methanol/water/ammonium hydroxide (60:47:11.3:2, v/v). Unlabeled phosphatidylinositol 3-phosphate (PIP) was run in parallel to determine the position of phosphorylated PI. The incorporation of 32 P into phosphatidylinositol 3-phosphate was determined using a PhosphorImager.
EGFR Immunoprecipitation and Western Blot Analysis-Cells were washed in ice-cold PBS containing 100 M Na 3 VO 4 and then lysed in buffer containing 20 mM Hepes, pH 7.4, 2 mM EGTA, 50 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , 5 mM NaF, 1% Triton X-100, 10% glycerol, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 5 g/ml leupeptin, 5 g/ml aprotinin. Five hundred g protein was incubated with 5 g of anti-EGFR antibody and 30 l of protein A-Sepharose for 4 h at 4°C. Immune complexes were washed four times with the same lysis buffer and resuspended in 2ϫ sample buffer. For Western analysis, samples were electrophoresed through 4 -12% NuPAGE Bis-Tris gels (NOVEX, San Diego, CA) and transferred to PVDF membranes (Millipore, Bedford, MA). Immunoblot analysis was carried out using the appropriate antibodies. Specific proteins were detected with the enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech).
Apoptosis and Colony Formation Assays--HeLa cells were assessed for apoptosis by staining with 4Ј6Ј-diamidino-2-phenylindole (DAPI) as described previously (6). Apoptotic cells were scored based on the presence of highly condensed or fragmented nuclei. That the DAPI staining reflects true apoptosis was confirmed by FACS analysis (6).
Long term survival of normal and v-Akt-expressing NIH3T3 cells following H 2 O 2 treatment was determined using a standard clonogenic assay. Cells were initially plated in 6-well cluster plates at a density of 30,000 cells/well. Following treatment, cells were trypsinized and serially diluted. Four dilutions ranging from 1:10 to 1:1,000,000 were prepared from each treatment. Colonies were allowed to grow for 10 -14 days and were then stained with crystal violet (0.1% crystal violet in 10% ethanol) and counted. Colony forming efficiency was calculated based on the number of colonies that grew and the number of cells plated into each well. Each colony forming efficiency is expressed relative to control (untreated) cells.

Characterization of Akt Activation by H 2 O 2 in Different Cell
Lines-To examine the effect of H 2 O 2 treatment on Akt activity, different cell types were exposed to various doses of the oxidant for 30 min, after which an immunocomplex kinase assay was used to assess Akt activity. The particular doses of H 2 O 2 used were based on the relative sensitivity of the cells to toxic effects of the oxidant. As shown in Fig. 1A, endogenous Akt activity was stimulated by H 2 O 2 treatment in all cell types, with the increase in activity ranging from 2-to 3-fold over untreated cells. UVC irradiation and several chemotherapeutic agents that are known to induce apoptosis were also examined for their abilities to activate Akt. Representative results shown in Fig. 1B indicate that, of the agents tested, only H 2 O 2 resulted in significant activation of Akt, and none of the other agents lowered Akt activity.
To examine the kinetics and dose-response relationships for Akt activation by H 2 O 2 treatment, NIH3T3 cells were transiently transfected with a hemagglutinin-tagged Akt (HA-Akt) expression vector, and 24 h later cells were treated either with 800 M H 2 O 2 for the indicated times or with various doses of H 2 O 2 for 30 min. HA-Akt was then immunoprecipitated from cells using an anti-HA antibody, and kinase activity was determined. Activation of the HA-Akt was evident within 5 min exposure to 800 M H 2 O 2 ( Fig. 2A). Maximum levels of Akt activity were seen 15-30 min after addition of H 2 O 2 , followed by a return to basal levels by 60 min. Akt activation was also dose-dependent; activity was increased approximately 2-fold with a dose of 400 M and further increased to levels ϳ20-fold higher than those seen in control cells with a dose of 2 mM H 2 O 2 (Fig. 2B). Western blot analysis with an anti-HA antibody was used to verify that equal amounts of HA-Akt protein were present in the immunoprecipitates regardless of the treatment conditions.
That the activation of Akt by H 2 O 2 was the result of oxidant injury was supported by experiments in which the ability of N-acetylcysteine (NAC) to block activation was assessed. NAC is a glutathione precursor and can enhance the antioxidant capacity of the cell both by acting directly as a scavenger of free radicals and by increasing glutathione levels (35). As shown in Fig. 2C, NAC treatment acted in a dose-dependent manner to inhibit Akt activation by H 2 O 2 in both HeLa and NIH3T3 cells. This effect is most likely attributed to the free radical scavenging effect of NAC, rather than its modulation of glutathione levels, as the effects of NAC were rapid and other agents such as diethyl maleate and buthionine sulfoximine which deplete cellular glutathione did not alone lead to Akt activation (results not shown).
Dependence of H 2 O 2 -induced Akt Activation on PI3-K-Several strategies were employed to evaluate the role of PI3-K in the activation of Akt by H 2 O 2 . PI3-K consists of a regulatory subunit, p85, and a catalytic subunit, p110. In response to growth factor and cytokine treatment, phosphorylation of p85 leads to activation of p110. A dominant negative mutant form of p85 (⌬p85), which lacks the binding site for p110, has previously been shown to inhibit growth factor-induced PI3-K activity (10,31). Therefore, if PI3-K were involved in the activation of Akt by H 2 O 2 , the dominant negative p85 mutant should block this activation. As shown in Fig. 3A, transfection of cells with ⌬p85 along with an HA-Akt expression plasmid completely inhibited HA-Akt activation by H 2 O 2 in NIH3T3 cells and decreased H 2 O 2 -induced HA-Akt activation by ϳ85% in HeLa cells. Further evidence for the involvement of PI3-K in H 2 O 2 -induced Akt activation was provided by treatment of cells with two potent inhibitors of PI3-K, wortmannin and LY294002. As shown, these agents abolished activation of Akt by H 2 O 2 in both NIH3T3 and HeLa cells (Fig. 3A). Fig. 3B shows the dose-dependent effect of wortmannin in NIH3T3 cells where even 5 nM of the inhibitor led to greater than 40% reduction in H 2 O 2 -induced Akt activation.
To assess directly whether PI3-K was activated in response to H 2 O 2 treatment, PI3-K was immunoprecipitated from H 2 O 2treated HeLa cells using an anti-p85 antibody and examined for kinase activity. As a positive control, insulin-treated HeLa cells were also examined. The results of a representative experiment are shown in Fig. 4. Although the effect of the oxidant was not as great as that seen with insulin treatment, a significant increase in PI3-K activity was observed with two different doses of H 2 O 2 and at two different time points.
Role of EGFR in Mediating H 2 O 2 -induced Akt Activation-We and others (36 -38) have previously provided evidence that growth factor receptors, and EGFR in particular, play an important role in mediating the activation of ERK MAPK in response to oxidant injury. Since PI3-K/Akt is also strongly activated by EGF stimulation, we investigated the possibility that EGFR plays a role in mediating Akt activation following H 2 O 2 treatment. To demonstrate phosphorylation of EGFR by H 2 O 2 treatment, EGFR was immunoprecipitated using an anti-EGFR antibody, and the precipitated proteins were then analyzed on Western blots using an anti-phosphotyrosine antibody. As shown in Fig. 5A 5B). The second strategy relied on the fact that continued presence of growth factors in the medium often results in down-regulation of receptor levels on the cell surface (36,39). Such receptor down-regulation is associated with transient refractoriness to subsequent stimulation. Therefore, we examined the influence of EGF pretreatment on the ability of H 2 O 2 to activate Akt. As shown in Fig. 5C, EGF pretreatment greatly inhibited activation of Akt in response to H 2 O 2 treatment. Taken together, these findings indicate that the EGFR plays an important role in mediating Akt activation by oxidants.
PI3-K/Akt Activation Inhibits H 2 O 2 -induced Apoptosis-We have previously shown (6) that H 2 O 2 treatment leads to death in a variety of other cell types and, in particular, apoptosis of HeLa cells. The importance of growth factor signaling pathways in contributing to survival is supported by the fact that suramin, a broad inhibitor of growth factor receptor, reduced survival of H 2 O 2 -treated cells (5). 2 In keeping with our previous findings, treatment of HeLa cells with the EGF receptor tyrosine kinase inhibitor AG1478 likewise resulted in enhanced cell death; the percentage of apoptotic cells seen 24 h following treatment with 600 M H 2 O 2 increased from 47 Ϯ 1.8% to 69.8 Ϯ 6.6% (n ϭ 3) in the presence of AG1478. Since the PI3-K/Akt pathway serves an anti-apoptotic function in some circumstances, and is activated downstream of the EGFR in the response to H 2 O 2 treatment, we investigated the influence of PI3-K/Akt activation on H 2 O 2 -induced cell death. First, we examined whether inhibiting activation of the pathway would alter survival of H 2 O 2 -treated cells. Pretreatment of HeLa cells with either 100 nM wortmannin or 25 M LY294002 (shown above to completely block Akt activation in response to H 2 O 2 treatment) prior to the addition of 600 M H 2 O 2 led to a significant increase in apoptosis (Fig. 6). Importantly, these agents were not cytotoxic for cells when given alone. These findings suggest that PI3-K activity is important for survival during the cellular response to H 2 O 2 .
Several groups have reported the involvement of PI3-kinase in the activation of ERK and JNK (40 -43). On the other hand, other studies have suggested that PI3-K may mediate its antiapoptotic effects via inhibiting JNK activation (44). We have previously demonstrated that both ERK and JNK are activated in response to H 2 O 2 treatment of HeLa cells and act in opposing directions to influence cell survival (5, 6); ERK was shown to inhibit apoptosis, whereas JNK promoted cell death. To rule out the possibility that the pro-survival effect of PI3-K observed in the present study might be attributed to effects on either of these pathways, we examined whether perturbation of PI3-K activity would affect ERK and JNK activities in HeLa cells. As shown in the top panel of negative mutant p85 (⌬p85) or constitutively active mutant p110 (p110-CAAX) in HeLa cells also failed to alter JNK activation by H 2 O 2 (Fig. 7A, lower panel). Similar results were obtained with H 2 O 2 -treated NIH3T3 cells when ERK and JNK activities were assessed using phospho-specific antibodies to recognize the activated forms of these proteins (Fig. 7B). Thus, the influence of PI3-K activity on cell survival is independent of the MAPK signaling pathways.
If Akt is a pro-survival signal during H 2 O 2 treatment, then constitutive elevation of Akt activity would be expected to inhibit apoptosis of cells following treatment with the oxidant. To test this possibility, HeLa cells were infected with an adenovirus directing expression of myristoylated Akt (Ad.Akt), a constitutively active form of the kinase, prior to treatment with H 2 O 2 . Control cells received a similar adenoviral construct lacking the Akt insert (Ad.null). That the Ad.Akt-infected cells contain elevated amounts of Akt protein is shown at the top of NIH3T3 cells were pretreated with wortmannin for 30 min before addition of 800 M H 2 O 2 . Thirty min later HA-tagged Akt was immunoprecipitated from the cells and assayed for kinase activity using H2B as substrate. Western blot analysis showed equal amounts of HA-Akt protein expression.

FIG. 4. H 2 O 2 treatment increases PI3-K activity in HeLa cells.
Cells were treated with H 2 O 2 (0.6 or 1.2 mM) or insulin (160 ng/ml) as indicated in the figure, after which they were lysed, and an anti-p85 PI3-K subunit antibody was used to immunoprecipitate the protein.
Kinase activity was assayed as described under "Experimental Procedures" using phosphatidylinositol as substrate, and the product, PIP, was resolved by thin layer chromatography. Unlabeled PIP was run in parallel to determine its position. Incorporation of 32 P into PIP was quantitated using a PhosphorImager.

FIG. 5. Activation of Akt in response to H 2 O 2 treatment is mediated via EGFR signaling. A, kinetics of EGFR phosphorylation in HeLa cells treated with 600 M H 2 O 2 . Lysates were prepared from
HeLa cells at the indicated times and subjected to immunoprecipitation (IP) with anti-EGFR antibody. Immunoprecipitates were then analyzed by immunoblotting (IB) using either anti-phosphotyrosine or anti-EGFR antibodies. B, inhibition of Akt phosphorylation by treatment with inhibitors of EGFR activation. Cells were treated with the indicated inhibitors or solvent prior to treatment with 600 M H 2 O 2 . Akt phosphorylation was assayed by Western blot analysis 30 min after treatment. C, pretreatment of cells with EGF (400 ng/ml) prevents activation of Akt in response to H 2 O 2 treatment. HeLa cells were pretreated with EGF for 1 h (ϩ) or not pretreated (Ϫ). They were then washed and incubated in serum-free medium for 2 h. The cells were further treated with 600 M H 2 O 2 for 30 min. Cell lysates were collected and measured for Akt activity by Western blot analysis using antiphospho-Akt antibody.
We also performed experiments to compare survival of NIH3T3 cells with NIH3T3 cells stably expressing v-Akt, a constitutively active form of the kinase. Detachment of these cells from the plates precluded the assessment of survival via DAPI staining as we had done for HeLa cells. Therefore, we employed a clonogenic assay to examine the long term survival of control and v-Akt-expressing cells following treatment with various doses of H 2 O 2 . The results, presented in Fig. 8B, demonstrate greater survival in cells expressing the v-Akt. These findings further support the view that Akt provides a prosurvival signal during the cellular response to oxidant injury.

DISCUSSION
The major findings of this study are that Akt is activated by biologically relevant doses of H 2 O 2 , that this activation occurs via an EGFR/PI3-K-dependent mechanism, and most importantly, that Akt provides a pro-survival signal that can protect cells against oxidative stress.
Akt is an established downstream target of PI3-K following stimulation of cells with growth factors, and a role for Akt in promoting cell growth and protecting against growth factor withdrawal-induced apoptosis is well established (17,18,20,22,45,46). Whether Akt activation occurs in response to other stressful stimuli is more controversial (24 -26, 47). Several studies have reported activation of Akt by H 2 O 2 as well as by other stresses that are known to exert their toxic effects, at least in part, through an oxidative stress mechanism (24 -26). In contrast, three recent studies have reported that ceramide and other stresses known to elevate cellular ceramide levels and induce apoptosis down-regulate Akt activity (28 -30). In one such study, Zundel and Giaccia (29) provided evidence to indicate that UVC, ionizing radiation, and sorbitol all inhibit Akt, through stress-induced increases in ceramide that lead to inhibition of PI3-K and Akt. However, Zhou et al. (28) suggested that ceramide inhibits Akt independently of PI3-K. Although H 2 O 2 was not examined in either of these studies, like the other stresses examined, it is known to induce sphingomyelin hydrolysis to generate ceramide (48) and, accordingly, would be expected to inhibit PI3-K and/or Akt. Clearly we did not observe such an effect, but rather we saw a significant activation of both PI3-K and Akt in H 2 O 2 -treated cells. It is also worth noting that, unlike the above-mentioned studies, we did not observe a down-regulation of Akt activity in response to UVC irradiation.
Several studies have inferred a role for PI3-K in the activation of Akt by H 2 O 2 based on the sensitivity of Akt activation to inhibition by wortmannin. Our studies have confirmed these observations with pharmacological inhibitors and have provided additional evidence for the dependence of H 2 O 2 -induced Akt activation on PI3-K using genetic mutant forms of p85 that prevent PI3-K activation. What then are the mechanisms involved in initiating PI3-K/Akt activation following H 2 O 2 ? Our study provides several pieces of evidence that support a role for EGFR in initiating the response. First, the EGFR was found to undergo rapid phosphorylation in response to H 2 O 2 treatment; the kinetics of activation and its attenuation coincided with that for elevated PI3-K activity and Akt activation. Second, treatment of cells with three different pharmacologic agents known to prevent EGFR signaling also inhibited Akt activation by H 2 O 2 . Third, pretreatment of cells with EGF to down-regulate the EGFR likewise prevented Akt activation in response to H 2 O 2 treatment. These findings are consistent with additional studies from our laboratory and others (36 -38), showing that the EGFR plays an important role in mediating activation of ERK MAPK in response to oxidative stress. However, and importantly, the Akt and ERK activations rely on independent signaling pathways, as treatment of cells with wortmannin (which completely prevents Akt activation) did not alter ERK activation (Fig. 7). Likewise, treatment of cells with PD98059, a specific MEK1/2 inhibitor, that prevents ERK activation did not interfere with Akt activation by H 2 O 2 (data not shown). A recent study has implicated focal adhesion kinase (FAK), a non-receptor tyrosine kinase, as the upstream mediator of PI3-K activation in T98 glioblastoma cells (27). It was shown that FAK underwent tyrosine phosphorylation in response to H 2 O 2 treatment and associated with PI3-K in a time frame consistent with Akt activation. In that study, however, the kinetics of FAK phosphorylation and Akt activation (earliest activation occurring at 1 h and peaking at 4 h) were significantly delayed relative to what we have observed in our studies for both NIH3T3 and HeLa cells (activation seen as early as 5 min, peaking at 15-30 min, and returning to basal levels within 1 h of treatment). It is possible that different receptor and membrane-associated nonreceptor tyrosine kinases can contribute to the initiation of the response, dependent on the cell type and/or circumstances. In this regard, it is worth noting that we have obtained preliminary evidence that the plateletderived growth factor receptor also undergoes phosphorylation in response to H 2 O 2 treatment, and inhibitors capable of preventing this phosphorylation also partially inhibited Akt activation. Additional experiments will be required to understand better the role of growth factor receptors in mediating the response.
Finally, and most importantly, our study has provided the first direct evidence that Akt activation contributes to the survival of H 2 O 2 -treated cells. Prior to our study, all reports of Akt activation in response to H 2 O 2 treatment involved H 2 O 2 concentrations in excess of 1 mM, with most experiments performed with 5-10 mM H 2 O 2 . The use of such high concentrations raises concerns regarding the biologic relevance of the response as such concentrations can be markedly toxic for cells, leading to rapid necrosis. We have demonstrated that concentrations as low as 400 M can result in activation of Akt. We had previously shown that H 2 O 2 treatment leads to apoptosis of HeLa cells in a dose-dependent manner over a range of concentrations from 400 M to 1.2 mM (6). That Akt activation occurs over a similar dose-response range supports its biologic relevance, and a role for Akt activation in protecting cells against H 2 O 2 -induced apoptosis was suggested by the finding that inhibition of Akt activation with PI3-kinase inhibitors (wortmannin and LY294002) and AG1478 (the EGFR tyrosine kinase inhibitor) led to enhanced apoptosis in response to H 2 O 2 (Fig. 6). Direct evidence for the ability of Akt to confer protection against H 2 O 2 -induced apoptosis was obtained in two different cell types using two different strategies to constitutively elevate Akt activity as follows: infection of HeLa cells with adenovirus expressing myristoylated Akt, and constitutive expression of v-Akt in NIH3T3 cells. In both model systems, elevated Akt activity was associated with enhanced resistance of cells to apoptosis following H 2 O 2 treatment.
An important question remaining is what are the downstream targets responsible for the protective influence of Akt? A number of different Akt targets have been identified that are believed to contribute to its anti-apoptotic function as follows: BAD, a proapoptotic member of the Bcl-2 family; caspase-9, a protease that functions as an initiator and effector of apoptosis; forkhead, a transcription factor believed to regulate the activity of other apoptosis-related genes; and IKK␣, a kinase involved in activation of NF-B (49 -53). The possible contribution of these downstream targets is currently under investigation. However, a role for IKK␣ appears to be ruled out by earlier studies from our laboratory (6) indicating that activation of NF-B does not play a role in influencing survival of H 2 O 2treated HeLa cells.
In conclusion, although the precise mechanisms through which the PI3-K/Akt signaling pathway acts to modulate the response to oxidants remains to be determined, our current findings provide strong support for a crucial role of the PI3-K/ Akt pathway in regulating cellular protection during the response to oxidative stress. It is important to remember, however, that this is only one of many signaling pathways activated in response to oxidant injury. Some of these are generally pro-survival (e.g. ERK and NF-B), whereas others appear to be pro-apoptotic (e.g. JNK and p53). Thus, the ultimate cellular outcome in a given cell type will reflect the relative balance between these various activities.