J Biol Chem, Vol. 274, Issue 46, 32596-32602, November 12, 1999

Hydrogen Peroxide Activates p70^S6k Signaling Pathway^*

Gyu-Un Bae, Dong-Wan Seo, Hyoung-Keun Kwon, Hoi Young Lee^§, Sungyoul Hong^¶, Zee-Won Lee, Kwon-Soo Ha, Hyang-Woo Lee, and Jeung-Whan Han^**

From the  Department of Biochemistry, College of Pharmacy, and the ^¶ Department of Genetic Engineering, College of Life Science and Natural Resources, Sungkyunkwan University, Suwon 440-746, Korea, the ^§ Department of Pharmacology, College of Medicine, Konyang University, Nonsan 320-711, Korea, and the  Biomolecule Research Group, Korea Basic Science Institute, Taejon 305-333, Korea

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

We investigated a possible role of reactive oxygen species (ROS) in p70^S6k activation, which plays an important role in the progression of cells from G₀/G₁ to S phase of the cell cycle by translational up-regulation of a family of mRNA transcripts that encode for components of the protein synthetic machinery. Treatment of mouse epidermal cell JB6 with H₂O₂ generated extracellularly by glucose/glucose oxidase led to the activation of p70^S6k and p90^Rsk and to phosphorylation of p42^MAPK/p44^MAPK. The activation of p70^S6k and p90^Rsk was dose-dependent and transient, maximal activities being in extracts treated for 15 and 30 min, respectively. Further characterization of ROS-induced activation of p70^S6k using specific inhibitors for p70^S6k signaling pathway, rapamycin, and wortmannin revealed that ROS acted upstream of the rapamycin-sensitive component FRAP/RAFT and wortmannin-sensitive component phosphatidylinositol 3-kinase, because both inhibitors caused the inhibition of ROS-induced p70^S6k activity. In addition, Ca²⁺ chelation also inhibited ROS-induced activation of p70^S6k, indicating that Ca²⁺ is a mediator of p70^S6k activation by ROS. However, down-regulation of 12-O-tetradecanoylphorbol-13-acetate (TPA)-responsive protein kinase C (PKC) by chronic pretreatment with TPA or a specific PKC inhibitor Ro-31-8220 did not block the activation of p70^S6k by ROS, indicating that the activation of TPA-responsive PKC was not required for stimulation of p70^S6k activity by H₂O₂ in JB6 cells. Exposure of JB6 cells to platelet-derived growth factor or epidermal growth factor led to a rapid increase in H₂O₂, phosphorylation, and activation of p70^S6k, which were antagonized by the pretreatment of catalase. Taken together, the results suggest that ROS act as a messenger in growth factor-induced p70^S6k signaling pathway.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Addition of mitogens to quiescent cells leads to activation of a number of specific protein kinases. Among the kinases affected are two families of growth factor-regulated serine/threonine kinases that phosphorylate the 40 S ribosomal protein S6 in vitro. One of these, referred to as p90^Rsk (1, 2), has been shown to lie on a signaling pathway that includes p21^Ras, raf-1, MEK1, and mitogen-activated protein kinase (MAPK)¹ and is a direct target of MAPK (3). The other family consists of two enzymes termed p70^S6k and p85^S6k (4, 5) which represent two isoforms of the same kinase that are encoded by a common gene and are identical except for a 23-amino acid extension at the amino terminus of p85^S6k (6, 7). Unlike p90^Rsk, p70^S6k and p85^S6k have been shown to reside on a novel p21^Ras-independent mitogenic signaling pathway (8, 9) that bifurcates at the level of the receptor from the p21^Ras pathway (9, 10). The p70^S6k is cytoplasmic, whereas the amino-terminal extension of p85^S6k targets it to the nucleus (11). The major substrate of the kinase in both compartments of the cell appears to be the 40 S ribosomal protein S6 (12), whose multiple phosphorylation in the cytosol has been implicated in the selective translational up-regulation of a family of mRNA transcripts that contain polypyrimidine tract at their 5' transcriptional start site (13, 14). Consistent with this finding, inhibition of p70^S6k activation by microinjection of neutralizing antibodies into cells (15) or by treatment of cells with the inhibitors of p70^S6k, the immunosuppressant rapamycin (16-18), severely impedes cell cycle progression.

Although p70^S6k is activated by numerous stimuli, including growth factors, cytokines, phorbolesters, oncogenic products, Ca²⁺, inhibitors of protein synthesis (7, 19), and hormones such as angiotensin II (20), the signal transduction pathway that mediates p70^S6k is poorly understood. This pathway bifurcates at a growth factor receptor docking site that is distinct from that of the p21^Ras/MAPK pathway (9). Many studies, including point mutational analysis of platelet-derived growth factor (PDGF) receptor (10), and the effect of various mutants of phosphatidylinositol 3-kinase (PI3K) (21) and specific inhibitor wortmannin for PI3K on p70^S6k activity (10, 22), have suggested that PI3K is an upstream mediator of p70^S6k activity. Recent studies favor a model in which protein kinase B lies below PI3K and upstream of p70^S6k (23, 24), although its function as a regulator for p70^S6k signaling pathway has been challenged (25, 26). Protein kinase B is activated through a wortmannin-sensitive phosphorylation. Recently, the kinase responsible for phosphorylation has been characterized and has been designated phosphoinositide-dependent protein kinase PDK1 (27). Rapamycin inhibits the activity of FKBP12-rapamycin-associated protein (FRAP/RAFT) kinase by binding to its cognate binding protein, FK506-binding protein (FKBP12), thus inhibiting p70^S6k stimulation (28). FRAP/RAFT appears to lies either on downstream of PI3K or on a parallel pathway, because TPA-induced activation of p70^S6k is insensitive to specific inhibitor for PI3K, wortmannin, which blocks growth factor-induced stimulation of PI3K and p70^S6k, but still blocked by rapamycin (29). However, the precise mechanism by which FRAP or PI3K regulates p70^S6k signaling pathway remains to be elucidated. This uncertainty appears to be due to the complex nature of the p70^S6k activation mechanism, which requires multiple hierarchical phosphorylation by several protein kinases (29, 30). Recently, one of them has been identified as phosphoinositide-dependent protein kinase PDK1 which activates p70^S6k through phosphorylation at Thr²²⁹ (27).

Emerging evidence suggests that reactive oxygen species (ROS) at low concentration may function as signaling intermediators of cellular responses (31). Recently, the production of ROS has been reported in a wide variety of cell types, including vascular smooth muscle cells (32), chondrocytes (33, 34), and fibroblasts (35), and is stimulated by a variety of stimuli, including cytokines such as transforming growth factor interleukin-1 (36) and tumor necrosis factor (33, 37), peptide growth factors PDGF (32, 33), epidermal growth factor (EGF) (38), and basic fibroblast growth factor (33), and G-protein-coupled receptors such as angiotensin II (39) and lysophosphatidic acid (40, 41). The increase in production of ROS following various external stimuli has been associated with various cellular processes, such as cell proliferation (32) or apoptosis (42, 43). However, the mechanisms through which ROS act are still poorly understood. Recently, ROS have been shown to stimulate signaling pathways implicated in growth factor and cytokine effects through activating their important components, such as MAPK, extracellular-regulated protein kinase (32, 44), and c-Jun NH₂-terminal kinase (34), and transcription factors such as NF-B (43, 45) and AP-1 (33). In addition, ROS have been observed to induce the anchorage-independent growth in soft agar of mouse epidermal cell JB6 (46), the competence-related proto-oncogenes c-fos and c-myc in JB6 (47), and the phosphorylation of ribosomal protein S6 in vitro (48). These observations prompted us to investigate a possible involvement of ROS in the growth factor-stimulated p70^S6k and p90^Rsk activation pathway, with particular focus on the differential regulation of both pathways by ROS. Our findings support a pivotal role of ROS for p70^S6k signaling pathway.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Cell Culture and Preparation of Cell Extracts-- Balb/c mouse epidermal cells (JB6) were grown at 37 °C in modified Eagle's medium supplemented with 8% fetal bovine serum in a humidified 5% CO₂. 80% confluent cells were made quiescent by culturing for 24 h in modified Eagle's medium (MEM) containing no bovine serum. Serum-starved cells were treated at 37 °C with following stimuli: 0.33 unit/ml glucose/glucose oxidase or 500 nM TPA for 15 min, and 5 ng/ml PDGF or 50 nM EGF for 20 min. Alternatively, cells were pretreated for the indicated times at 37 °C with serum-free MEM containing 2 mM EGTA, 2, 10, and 20 µM BAPTA-AM, 30, 300, and 3000 unit/ml Aspergillus niger catalase, 500 nM wortmannin, 5 nM rapamycin, or 5 µM PKC inhibitor Ro-31-8220 before stimulation. In some experiments, cells were pretreated for 24 h with 5 µM TPA. The controls were carried out by incubating the cells for the corresponding period in serum-free MEM containing dimethyl sulfoxide instead of the agents. Following stimulation, the cells were rinsed twice with a ice-cold wash buffer solution containing 50 mM Tris-HCl (pH 7.5), 120 mM NaCl, 20 mM NaF, 1 mM EDTA, 5 mM EGTA, 15 mM sodium pyrophosphate, 30 mM p-nitrophenyl phosphate, 1 mM benzamidine, and 0.1 mM phenylmethylsulfonyl fluoride and then extracted in the same buffer containing 1% Nonidet P-40. Cell extracts were collected with a plastic scraper, homogenized, and cleared by centrifugation at 4 °C for 15 min at 15,000 × g. Protein concentration was measured by the method of Bradford, with bovine serum albumin as the standard. Aliquots of the supernatant were frozen in liquid nitrogen and stored at 70 °C.

Immunoprecipitation and S6 Kinases Activity Assay in Vitro-- p70^S6k and p90^Rsk were immunoprecipitated by incubating 20 µg (total 200 µl) of protein/assay of cell extract to an antibody directed to the carboxyl-terminal 18 residues of p70^S6k and the carboxyl-terminal 21 residues of p90^Rsk for 2 h incubation at 4 °C, respectively. Immunoprecipitation was facilitated by the addition of protein A-Sepharose (20 µl) for 30 min at 4 °C on a shaking plate. The beads were then washed twice at 4 °C with an extraction buffer containing 50 mM Tris-HCl (pH 7.5), 120 mM NaCl, 20 mM NaF, 1 mM EDTA, 5 mM EGTA, 15 mM sodium pyrophosphate, 30 mM p-nitrophenyl phosphate, 1 mM benzamidine, 0.1 mM phenylmethylsulfonyl fluoride, and 1% Nonidet P-40 and once with a dilution buffer containing 50 mM MOPS (pH 7.0), 1 mM dithiothreitol, 5 mM MgCl₂, 10 mM p-nitrophenyl phosphate, and 1% Nonidet P-40. S6 kinase (p70^S6k and p90^Rsk) activity was assayed using the S6 peptide as substrate by incubating the immunoprecipitated S6 kinases for 30 min in 25 µl of reaction mixture containing 50 mM MOPS (pH 7.0), 5 mM MgCl₂, 1 mM dithiothreitol, 10 mM p-nitrophenyl phosphate, 0.1% Nonidet P-40, 0.6 µM protein kinase inhibitor, and 12 µM ATP (plus 0.75 µCi of [-³²P]ATP). The reaction was terminated by the addition of 10 µl of 100 mM EDTA (pH 7.0). Following a brief centrifugation, the supernatant was spotted on P-81 paper. Unincorporated [-³²P]ATP was eliminated by three 10-min washes in 5% phosphoric acid, and phosphorylated S6 peptide bound to the paper was counted. The assays were carried out in duplicate. The results were calculated as units of S6 kinase activity/mg of protein lysate and expressed as the fold of increase over the control value. One unit of activity represents the transfer of 1 pmol of ³²P_i into S6 peptide/min under the assay condition.

Immunoblotting-- Cell lysates were boiled in Laemmli sample buffer for 3 min. Cell lysates containing 50 µg of total protein were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) on 15% slab gels for the determination of p70^S6k and p42^MAPK/44^MAPK, and proteins were transferred to polyvinylidene difluoride membranes. Membranes were blocked for 30 min in PBS containing 0.1% Tween 20 and 5% (w/v) dry skim milk powder and incubated overnight with anti-p70^S6k or anti-p42^MAPK/44^MAPK antisera. The membranes were then washed with PBS containing 0.1% Tween 20 and incubated for 2 h with an anti-rabbit secondary antibody. Bound antibodies were detected with the enhanced amplified alkaline phosphatase immunoblot system (Bio-Rad).

Measurement of Intracellular H₂O₂-- Cells were grown on coverslips for 2 days and serum-free MEM for 1 day. The serum-starved cells were stabilized in serum-free MEM without phenol red for at least 30 min and stimulated with 5 ng/ml PDGF for different times. Sometimes, cells were preincubated with A. niger catalase (300 or 3000 unit/ml) for 20 min before treatment with PDGF or EGF for 5 min, respectively. For the last 10 min of stimulation, 5 µM H₂DCFDA was added to monitor intracellular H₂O₂. Then the cells were immediately observed with a laser scanning confocal microscope (Carl Zeiss LSM 410). The images of samples excited by a 488 nm argon laser were filtered by a long pass 515 nm filter. Thirty cells were randomly selected from three independent experiments, and dichlorofluorescein fluorescence intensities of treated cells were compared with those of unstimulated cells (36, 38).

Measurement of Intracellular Catalase Activity-- Catalase activity from nonstimulated or stimulated JB6 cell lysates was assayed by a slightly modified method of Aebi (49). Briefly, serum-starved JB6 cells (1 × 10⁷ cells/ml) were incubated with A. niger catalase (3000 unit/ml) for the indicated times. After incubation, cells were washed twice in PBS, trypsinized, and homogenized in extraction buffer containing 50 mM Tris-HCl (pH 7.5), 120 mM NaCl, 20 mM NaF, 1 mM EDTA, 5 mM EGTA, 15 mM sodium pyrophosphate, 1 mM benzamidine, 0.1 mM phenylmethylsulfonyl fluoride, and 1% Nonidet P-40. Cell extracts were cleared by centrifugation at 15,000 × g for 15 min at 4 °C. Catalase activity was assayed using H₂O₂ (10 mM) as a substrate by incubating the catalase in the supernatants for 30 s in 1 ml of reaction mixture containing 50 mM sodium phosphate and potassium phosphate (pH 7.0). Activity was monitored as the decrease in absorbance following decomposition of H₂O₂ at 240 nm. Assays were carried out in duplicates, and the results were expressed as units of catalase activity/µg of protein lysate.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Activation of p42^MAPK/p44^MAPK and p90^Rsk in Response to H₂O₂-- Recently, peptide growth factors such as PDGF, basic fibroblast growth factor, and EGF have been demonstrated to produce ROS. One of the targets of ROS appears to be p42^MAPK/p44^MAPK (3) which is upstream kinase of p90^Rsk. On the other hand, an earlier observation revealed that exposure of mouse epidermal cells JB6 to ROS could activate a protein kinase in cellular extracts which phoshorylated ribosomal protein S6 in vitro (48). These studies led to an assumption that p90^Rsk, the downstream kinase of p42^MAPK/p44^MAPK, might be responsible for the induction of S6 phosphorylation in vitro. To examine the possibility, JB6 cells were arrested in G₀, and the p42^MAPK/p44^MAPK phosphorylation and p90^Rsk activity were measured after treatment of the cells with ROS generated extracellularly by glucose/glucose oxidase (G/GO) producing H₂O₂. As shown in Fig. 1, G/GO treatment led to a dose-dependent phosphorylation of p42^MAPK/p44^MAPK and activation of p90^Rsk, as measured by its slower migration on Western blots of one-dimensional SDS-PAGE and immune complex kinase assay (Fig. 1, A and B), respectively. The increase in p42^MAPK/p44^MAPK phosphorylation following ROS treatment was followed by the activation of p90^Rsk. The increase in p90^Rsk activity was transient after active oxygen treatment, with a peak of S6 phosphorylation occurring after 30 min. A slow inactivation of p90^Rsk then ensued within 60 min (Fig. 1D). However, the increase in p42^MAPK/p44^MAPK phosphorylation was apparent as early as 5 min after the H₂O₂ treatment and sustained for more than 30 min (Fig. 1C). This time discrepancy indicated that the activation of p42^MAPK/p44^MAPK preceded activation of p90^Rsk located in the downstream position. Taken together, these data demonstrate conclusively that H₂O₂ activates p42^MAPK/p44^MAPK and p90^Rsk in mouse epidermal cell JB6. Consistent with the data presented herein, p42^MAPK/p44^MAPK activation in response to H₂O₂ has also been reported for other cell types (32, 44).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Activation of MAPK and p90Rsk in response to H2O2 induced by G/GO in JB6 cells. A, serum-starved JB6 cells were stimulated with the indicated concentrations of G/GO for 15 min. Cells were then harvested. The changes of MAPK activity in G/GO-stimulated cell lysates were monitored as slower migration of a family of bands on SDS-PAGE and immunoblot analysis as described under "Experimental Procedures." B, p90Rsk in the cell lysates was immunoprecipitated by an antibody directed to the carboxyl-terminal 21 residues of p90Rsk and assayed for S6 kinase activity as described under "Experimental Procedures." The p90Rsk activity was expressed relative to that of unstimulated cells. Con, unstimulated cells. C, to examine the time course of MAPK, serum-starved JB6 cells were incubated for different times with G/GO (0.33 unit/ml). Detailed experimental conditions are described under "Experimental Procedures." D, p90Rsk in the cell lysates stimulated with G/GO (0.33 unit/ml) was assayed to determine time course. The results shown represent the means ± S.E. of three determinations.

H₂O₂-induced p70^S6k Activation-- The kinetics of p90^Rsk activation in JB6 cells following ROS exposure was not exactly consistent with that of total S6 kinase activation in cellular extracts (48), suggesting involvement of another family of S6 kinase, p70^S6k, which is activated by virtually all mitogenic stimuli, including growth factors, cytokines, phorbol esters, and oncogenic products (7). Indeed, treatment of mouse epidermal JB6 cells with G/GO caused a rapid phosphorylaton and activation of p70^S6k, as judged by the slower migration of a family of bands on SDS-PAGE and by immune complex kinase assay, respectively (Fig. 2). The S6 phosphorylation increased dose-dependently (Fig. 2, C and D). The maximal activity found in the extracts of cells treated for 15 min was 6-7-fold higher than basal levels. Within 60 min, the activity and phosphorylation decreased substantially (Fig. 2, A and B). These results are similar to those found with growth factors or serum-treated various cell lines (50). Taken together, these data indicate that H₂O₂ activates p70^S6k as well as p42^MAPK/p44^MAPK/p90^Rsk.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of H2O2 generated by G/GO on p70S6k in JB6 cells. A, extracts (50 µg of total protein) from quiescent cells or cells stimulated with G/GO (0.33 unit/ml) were subjected to SDS-PAGE and immunoblot analysis using specific antibody against the p70S6k as described under "Experimental Procedures." B, p70S6k in the cell lysates was immunoprecipitated by an antibody directed to the carboxyl-terminal 18 residues of p70S6k and assayed for S6 kinase activity as described under "Experimental Procedures." The p70S6k activity was expressed as the fold of increase over control value. C, serum-starved JB6 cells were incubated for 15 min with different concentrations of G/GO, and cell lysates were subjected to Western blot analysis. D, p70S6k activity in the cell lysates stimulated with different concentrations of G/GO was assayed after immunoprecipitation. The results shown represent the means ± S.E. of three separate experiments. Con, unstimulated cells.

Effect of Rapamycin and Wortmannin on H₂O₂-stimulated p70^S6k Activity-- Using a variety of receptor mutants (9, 10), a constitutively activated PI3K (21), and PI3K inhibitors wortmannin and LY294002, PI3K has been shown to be a signaling component that proximately exists in receptor tyrosine kinase on p70^S6k signaling pathway (10, 22). Another crucial regulator of p70^S6k is FRAP/RAFT, the direct target of rapamycin in mammals (28). FRAP/RAFT is a large molecular weight protein a homologous to PI3K with rapamycin-sensitive protein kinase activity (28). Immunosuppressant rapamycin inhibits an upstream kinase FRAP/RAFT near p70^S6k through formation of a complex with FKBP12. Because of pivotal roles of PI3K and FRAP/RAFT in the regulation of p70^S6k in response to growth factors (21, 51), their involvement in the stimulation of p70^S6k by H₂O₂ were examined using selective inhibitors of p70^S6k, rapamycin and wortmannin. Pretreatment of the cells with 5 nM rapamycin resulted in p70^S6k activities below the control levels, and wortmannin pretreatment inhibited the H₂O₂-stimulated increase in p70^S6k activity by about 70% (Fig. 3A). The inhibitory effects of both agents were paralleled by an increase in the mobility of the kinase, as assessed by the gel mobility immunoblotting (Fig. 3B). Under these conditions, both agents had no effect on p90^Rsk activation and p42^MAPK/p44^MAPK phosphorylation (data not shown), as expected from a previous report (16). The inhibitor studies indicated that FRAP and PI3K were required for the stimulation of 70^S6k by H₂O₂. This was consistent with data that showed the involvement of PI3K and FRAP/RAFT in growth factor-induced 70^S6k signaling pathway (21, 51).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of rapamycin or wortmannin on p70S6k activity in G/GO-stimulated JB6 cells. JB6 cells were incubated in the absence or presence of G/GO (0.33 unit/ml) for 15 min following pretreatment for 30 min with either 5 nM rapamycin or 500 nM wortmannin. Cell lysates were subjected either to immunoprecipitation assay for p70S6k activity (A) or to Western blot analysis for p70S6k phosphorylation (B) as described in the legend to Fig. 2.

Differential Calcium Requirements for H₂O₂-induced p70^S6k and MAPK/p90^Rsk Activation-- Previous studies have shown that incubation of several cells with compounds that increased intracellular calcium (the ionophore A23187, thapsigargin) stimulated p70^S6k (20) and that PDGF-induced p70^S6k activation was ablated by Ca²⁺ chelation (26), indicating a possible role of Ca²⁺ in regulation of p70^S6k. We therefore examined involvement of Ca²⁺ in H₂O₂ activation of p70^S6k. Although incubation of JB6 cells with extracellular Ca²⁺ chelator EGTA (2 mM) inhibited the H₂O₂-stimulated p70^S6k activity by about 60% (Fig. 4A), the cell-permeant Ca²⁺ chelator BAPTA-AM inhibited the activity dose-dependently with 100% inhibition occurring at 10 µM BAPTA-AM (Fig. 4A). Chelation of extracellular Ca²⁺ with EGTA only partially affected the H₂O₂-stimulated p70^S6k activity, further supporting the thesis that activation of p70^S6k by H₂O₂ was dependent on intracellular Ca²⁺ rather than extracellular Ca²⁺. The inhibition by Ca²⁺ chelator on H₂O₂-induced activation of p70^S6k was accompanied with inhibition of p70^S6k phosphorylation, as monitored with collapse of slow migration of a family of bands into a single band (Fig. 4B). In comparison, preincubation of JB6 cells with EGTA or BAPTA-AM did not inhibit p90^Rsk activity (Fig. 4C). Furthermore, BAPTA-AM stimulated the activation of p90^Rsk by H₂O₂ in a dose-dependent manner (Fig. 4C), similar to the results obtained by p42^MAPK/p44^MAPK mobility shift (Fig. 4B). These results indicated that H₂O₂ stimulated p70^S6k signaling pathway by increasing intracellular Ca²⁺ level, and intracellular Ca²⁺ does not play a substantial role in regulating p42^MAPK/p44^MAPK and p90^Rsk in JB6 cells. Consistent with this result, we noted that exogenous H₂O₂ or H₂O₂ produced endogenously by growth factors caused increase in cytosolic [Ca²⁺].² Similarly, H₂O₂ stimulation of Ca²⁺ release has also been observed in Rat-2 fibroblast (40).

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Ca2+ requirement for the activation of p70S6k, p42MAPK/p44MAPK, and p90Rsk by H2O2. Cells were pretreated with EGTA (2 mM) to chelate extracellular Ca2+ for 30 min or BAPTA-AM (2, 10, and 20 µM) for 15 min and then incubated with G/GO (0.33 unit/ml) for 15 min. Both kinase assay (p70S6k and p90Rsk, A and C) and Western blot analysis (p70S6k and p42MAPK/p44MAPK; B) were carried out as described above. The results shown represent the means ± S.E. of three separate experiments.

Effects of PKC on H₂O₂-induced p70^S6k Activation-- PKC comprises a large family of multiple isoforms that exhibit distinct properties, including sensitivities to calcium and the phorbol ester family of tumor promoters. The isoforms are divided into three subgroups; conventional, Ca²⁺-responsive PKC, the novel, Ca²⁺-unresponsive PKC, and the atypical, Ca²⁺- and TPA-unresponsive PKC (52). Recently, p70^S6k has been shown to be regulated by TPA-responsive (16, 29) and -unresponsive PKC isoforms (53, 54). In addition, it has been reported that PKC isoforms, including , I, and  of conventional PKC,  and  of novel PKC, and  of atypical PKC, are tyrosine phosphorylated and catalytically activated by treatment of cells with H₂O₂ (55). To further characterize the effect of H₂O₂ on p70^S6k, we focused on the TPA-responsive PKC isoforms. As shown in Fig. 5, PKC activity was down-regulated by chronic pretreatment with 5 µM TPA for 24 h and then stimulated with H₂O₂ or TPA. The activation and phosphorylation of p70^S6k by H₂O₂ was not significantly inhibited by this pretreatment, although PKC was not completely but significantly down-regulated by TPA pretreatment as evidenced by a slight activation of p70^S6k activity over the activity of quiescent cells (Fig. 5A). Nevertheless, because of the uncertain specificity of the TPA down-regulation, the study was extended by using a specific PKC inhibitor Ro-31-8220, which acts as an ATP-competitive inhibitors of PKC and has been extensively used for studying the role of PKC in cell signaling (56, 57). Preincubation of the cells with Ro-31-8220 completely inhibited the stimulation of p70^S6k by TPA, suggesting that the conditions used were sufficient to down-regulate TPA-responsive PKC isoforms (Fig. 5B). However, Ro-31-8220 down-regulation of PKC did not attenuate the phosphorylation and activation of p70^S6k by H₂O₂ compared with that induced by TPA (Fig. 5B). Despite the fact that various PKC isoforms are phosphorylated and activated in response to H₂O₂ (55) and that the TPA-responsive PKC activate p70^S6k in other cell types (16, 29) and in mouse epidermal cell JB6, our results did not support a role of TPA-responsive PKC in the regulation of p70^S6k by H₂O₂. We could not, however, rule out the possibility that other PKC isoforms less sensitive or nonsensitive to Ro-31-8220 could contribute to the H₂O₂ stimulation of p70^S6k.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Effect of down-regulation of PKC on H2O2 stimulation of p70S6k in JB6 cells. After pretreatment either with 5 µM TPA for 24 h (A) or 5 µM Ro-31-8220 for 20 min (B), cells were stimulated with 500 nM TPA or G/GO (0.33 unit/ml) as described "Experimental Procedures." Cell lysates were subjected either to Western blot analysis for p70S6k phosphorylation or immunoprecipitation assay for p70S6k activity. The results shown represent the means ± S.E. of three determinations. DMSO, dimethyl sulfoxide.

Generation of H₂O₂ in JB6 Cell by PDGF or EGF and Its Effect on p70^S6k Activity-- By measuring the intracellular generation of ROS with oxidation of the peroxide-sensitive fluorophore 2',7'-dichlorofluorescein (DCF), we then examined whether ROS could be generated in JB6 cells by PDGF or EGF stimulation. Microfluorometric study with laser scanning confocal microscopy revealed that exposure of quiescent JB6 cells to PDGF-AB isoform (5 ng/ml) or EGF (50 nM) resulted in a rapid increase in DCF fluorescence by 2-fold (Fig. 6). The increase in DCF fluorescence by PDGF was transient with a maximal 2-fold increase over the basal levels at 5 min and then returned to the basal levels within 30 min (Fig. 6). The time course of DCF fluorescence by PDGF was similar to that by EGF (data not shown). To further characterize the role of H₂O₂ in PDGF- or EGF-induced p70^S6k activation in JB6 cells, we reduced H₂O₂ intracellularly generated by PDGF or EGF with catalase. Catalase has been shown to be taken up across the plasma membrane of vascular smooth muscle cells by an unknown mechanism (32). To decrease cytosolic concentration of H₂O₂, JB6 cells were preincubated with catalase (3000 units/ml), and preincubation of catalase led to time-dependent increase in catalase activity (Fig. 7A). The maximal activity found in the extracts of the cells treated for 4 h was about 5-fold over the basal levels. The increase in catalase activity was not brought about by its nonspecific binding to the outer membrane of JB6 cells, because the enzymatic activity of catalase-loaded cells was resistant to proteinase K treatment (Fig. 7A, inset). The PDGF- or EGF-stimulated increase in DCF fluorescence was abolished by the preincubation with catalase, suggesting that the increase in DCF fluorescence was mainly due to H₂O₂ and that the catalase activity taken up into JB6 cells was sufficient to decrease H₂O₂ intracellularly generated by PDGF or EGF (Fig. 7B). We next investigated the effect of catalase on PDGF or EGF stimulation of p70^S6k. Treatment of quiescent JB6 cells with PDGF or EGF led to a rapid phosphorylation of p70^S6k, as evidenced by a slower migration of a family of bands on SDS-PAGE and immunoblot analysis. However, preincubation of the cells with catalase severely impeded PDGF or EGF stimulation of p70^S6k, indicating involvement of H₂O₂ in PDGF- or EGF-induced p70^S6k activation in JB6 cells (Fig. 7C).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Production of H2O2 in JB6 cells by PDGF. JB6 cells were grown on coverslips for 2 days and serum-free MEM for 1 day. The serum-starved cells were stabilized in serum-free MEM without phenol red for at least 30 min and stimulated with 5 ng/ml PDGF for the indicated times. ROS generation was measured by DCF fluorescence as described under "Experimental Procedures." Thirty cells were randomly selected from three independent experiments. Data shown represent relative fluorescence intensities of treated cells to those of unstimulated cells.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Uptake of catalase into JB6 cells and its effect on H2O2 generation and p70S6k activation by PDGF or EGF. A, time-dependent intracellular catalase activity following exogenous treatment of catalase to JB6 cells. Serum-starved JB6 cells (1 × 107 cells/ml) were incubated with catalase (3000 unit/ml) for various times. Changes of catalase activity in the cell lysates were monitored as described under "Experimental Procedures." Inset, cells incubated with extracellular catalase (3000 units/ml) for 2 h were collected and then exposed to proteinase K (1 mg/ml) in solution for 20 min. After 20 min, cells were quickly settled down and washed with PBS. The cells were subsequently lysed in extraction buffer. Residual catalase activity was measured as described above. The catalase activity in the cell lysates loaded with exogenous catalase was expressed as folds of increase over that of unloaded cells. Specific activity of catalase in the unloaded cells was 5.8 ± 0.4 milliunits/µg protein. The results shown represent the means ± S.E. of three determinations. B, inhibitory effect of catalase on H2O2 production by growth factors. Cells were preincubated with catalase (300 or 3000 units/ml) for 20 min before treatment with PDGF (5 ng/ml) or EGF (50 nM) for 5 min. ROS generation was measured by DCF fluorescence as described under "Experimental Procedures." Thirty cells were randomly selected from three independent experiments. Data shown represent relative fluorescence intensities of treated cells to those of unstimulated cells. C, inhibitory effect of catalase on PDGF- or EGF-stimulated p70S6k activity. Serum-starved JB6 cells (1 × 106 cells/ml) were preincubated in the absence or presence of A. niger catalase (300 or 3000 units/ml) for 20 min before treatment with PDGF (5 ng/ml) or EGF (50 nM) for 20 min. Extracts (50 µg of total protein) from quiescent or growth factor-stimulated cells were subjected to SDS-PAGE and immunoblot analysis using specific antibody against the p70S6k as described under "Experimental Procedures."

In summary, we identified the p70^S6k signaling pathway as a novel pathway regulated by ROS. Exposure of mouse epidermal cell JB6 to extracellular H₂O₂ produced by G/GO was sufficient to activate p70^S6k. ROS activation of p70^S6k required Ca²⁺, PI3K, and FRAP/RAFT but not TPA-responsive PKC. Furthermore, stimulation of endogenous H₂O₂ production by growth factor PDGF or EGF also stimulated p70^S6k. Using catalase, we showed that H₂O₂ mediated activation of p70^S6k by PDGF or EGF. Thus, our studies identified a serine/threonine kinase with a demonstrated role in cell proliferation that was regulated by H₂O₂. Recently, ROS has been shown to have a carcinogenic potential and is associated with tumor promotion (58, 59). Also, it acts as essential intracellular second messengers for several cytokines and growth factors (32, 33, 36-38), thus emphasizing the importance of this pathway in growth control. An identification of H₂O₂ as a regulator of p70^S6k contributes to an expanding list of molecules on the p70^S6k signaling pathway that have pivotal roles in growth control and oncogenesis.

	FOOTNOTES

^* This work was supported in part by Grant 971-0708-073-1 from Korea Science and Engineering Foundation and by funds from the Seok Chun Research Fund of Sungkyunkwan University.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^** To whom correspondence should be addressed. Tel.: 82-331-290-7716; Fax: 82-331-292-8800; E-mail: jhhan@yurim.skku.ac.kr.

² G.-U. Bae, K.-S. Ha, H.-W. Lee, and J.-W. Han, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: MAPK, mitogen-activated protein kinase; ROS, reactive oxygen species; PI3K, phosphatidylinositol 3-kinase; FKBP, FK506-binding protein; FRAP, FK506-rapamycin-associated protein; RAFT, rapamycin and FKBP12 target; PKC, protein kinase C; MEM, modified Eagle's medium; PDGF, platelet-derived growth factor; EGF, epidermal growth factor; TPA, 12-O-tetradecanoylphorbol-13-acetate; BAPTA-AM, bis-(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid tetraacetoxy-methyl ester; MOPS, 4-morpholinepropanesulfonic acid; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline; G/GO, glucose/glucose oxidase; DCF, 2',7'-dichlorofluorescein.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES