Activation of the JAK-STAT pathway by reactive oxygen species

¹ Pulmonary and Critical Care Division, Tupper Research Institute, New England Medical Center, Boston 02111; and ² Department of Physiology, Tufts University School of Medicine, Boston, Massachusetts 02111

	ABSTRACT

Top Abstract Introduction Materials & Methods Results Discussion References

Reactive oxygen species (ROS) play an important role in the pathogenesis of many human diseases, including the acute respiratory distress syndrome, Parkinson's disease, pulmonary fibrosis, and Alzheimer's disease. In mammalian cells, several genes known to be induced during the immediate early response to growth factors, including the protooncogenes c-fos and c-myc, have also been shown to be induced by ROS. We show that members of the STAT family of transcription factors, including STAT1 and STAT3, are activated in fibroblasts and A-431 carcinoma cells in response to H₂O₂. This activation occurs within 5 min, can be inhibited by antioxidants, and does not require protein synthesis. STAT activation in these cell lines is oxidant specific and does not occur in response to superoxide- or nitric oxide-generating stimuli. Buthionine sulfoximine, which depletes intracellular glutathione, also activates the STAT pathway. Moreover, H₂O₂ stimulates the activity of the known STAT kinases JAK2 and TYK2. Activation of STATs by platelet-derived growth factor (PDGF) is significantly inhibited by N-acetyl-L-cysteine and diphenylene iodonium, indicating that ROS production contributes to STAT activation in response to PDGF. These findings indicate that the JAK-STAT pathway responds to intracellular ROS and that PDGF uses ROS as a second messenger to regulate STAT activation.

hydrogen peroxide; platelet-derived growth factor; TYK2

	INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

OXIDATIVE DAMAGE contributes to the pathogenesis of several human diseases, including atherosclerosis, acute respiratory distress syndrome (ARDS), Alzheimer's disease, and pulmonary fibrosis (4, 28, 42). Virtually every known organism has evolved specific mechanisms to protect itself from oxidative damage. Mechanisms for reaction to specific oxidative conditions have been identified in bacteria, yeast, and mammalian cells. An important part of the cellular defense to oxidative stress is the specific induction of new gene expression in response to specific oxidative stressors. Some of the genes induced in response to oxidative stress such as superoxide dismutase and glutathione peroxidase are directly involved in the neutralization of oxygen radicals and their precursors (42, 61). Others, such as p21^CIP1/WAF1, c-fos, and the caspases, are involved in the control of cell proliferation and apoptosis in response to oxidative stress (81). The cell must be able to detect these oxidative stressors and then specifically activate transcription factors to upregulate the expression of these genes. Indeed, the molecular detection and response of cells to reactive oxygen species (ROS) are likely to be among the earliest evolved second-messenger systems. Moreover, there is substantial evidence that ROS are not only pathological but, in many instances, are utilized as second messengers by the cell in response to growth factors and cytokines (72, 85, 87, 89).

Thus it is not surprising that some of the transcription factors known to be involved in immediate early gene expression are also regulated by ROS. One of the best characterized examples of this is nuclear factor-B (NF-B). Before activation, NF-B is held in the cytoplasm through a complex with another factor called I-B (3). In response to peroxide or ultraviolet (UV) treatment of many cell types, the I-B factor is degraded, releasing NF-B to translocate to the nucleus, where it can specifically activate transcription (35, 56, 57, 72). Moreover, ROS or peroxides appear to be the second messenger used by cells to activate NF-B, inasmuch as antioxidants prevent the activation of NF-B in response to a variety of cytokines (72). Recent data indicate that ROS are also important mediators of the response to the platelet-derived growth factor (PDGF) (85).

The signal transducer and activator of transcription (STAT) factors were originally described as growth factor- and interferon-inducible DNA binding complexes (32, 51). Subsequently, the STAT factors have been shown to be induced by a wide variety of growth factors and cytokines (39). These factors participate in the regulation of many genes, including the c-fos protooncogene, caspases, and the cell cycle regulator p21^CIP1/WAF1, which can also respond to oxidative stress (11, 12, 30, 49, 75, 94, 100). The STAT factors are unique, in that they are phosphorylated on tyrosine residues in response to a variety of growth factors and cytokines (25, 55, 66, 68, 69, 77, 80). On phosphorylation, STATs dimerize via SH2-phosphotyrosine interactions and become competent to bind DNA (76). Before phosphorylation the STATs are found in the cytoplasm and translocate to the nucleus on activation (69, 78). In response to interferon treatment, the Janus kinase (JAK) family of kinases is required for STAT phosphorylation (59, 91, 95). Here we report that STAT1 and STAT3 factors are activated on treatment of Rat-1 fibroblasts with H₂O₂ and that this activation is sensitive to antioxidants. Moreover, H₂O₂ stimulates the activity of the STAT kinases JAK2 and TYK2. Finally, we find that ROS play an important role in the activation of STATs by PDGF.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Cell culture. Rat-1 fibroblasts were maintained in DMEM with 10% fetal bovine calf serum (FCS). NIH/3T3 and A-431 cells were maintained in DMEM-10% bovine calf serum (CS). For experiments, cells were made quiescent by growth to confluence and incubation in DMEM-0.5% FCS or CS for 48 h. Iron-free medium was used for all stimulations to minimize ROS generated by the Haber-Weiss reaction (52). For stimulations, epidermal growth factor (EGF) and PDGF-BB were obtained from Collaborative Biomedical (Waltham, MA). Catalase (bovine liver), glucose oxidase, H₂O₂, buthionine sulfoximine (BSO), diphenyl iodinium (DPI) chloride, N-acetyl-L-cysteine (NAC), pyrrolidine dithiocarbamate (PDTC), N-nitro-L-arginine, vanadate, vanadyl sulfate, nitrosoglutathione, isosorbide dinitrite, sodium nitroprusside, and reduced glutathione were obtained from Sigma Chemical (St. Louis, MO).

Antibodies. The anti-STAT1C rabbit antiserum was made against a peptide with the unique COOH-terminal 37 amino acids of murine STAT1 protein. The anti-NH₂-terminal STAT1 antibody was obtained from Transduction Laboratories. STAT3 antibody was a polyclonal rabbit antiserum made against another unique sequence in the COOH-terminal region of human STAT3 (amino acids 688-700, QEHPEADPGSAAP). The antiphosphotyrosine STAT3 antibody was obtained from New England Biolabs (Beverly, MA). The JAK1, JAK2, TYK2, and phosphotyrosine (4G10) antibodies were obtained from Upstate Biotechnology (Lake Placid, NY). The anti--PGDF receptor antibody was from Santa Cruz Biotechnology (Santa Cruz, CA).

Preparation of nuclear extracts. After the indicated treatments, cells were rinsed three times with ice-cold PBS. PBS containing 1 mM Na₃VO₄ and 5 mM NaF was added to each dish, and the cells were scraped from the dish and pelleted at 1,500 rpm for 10 min at 4°C. Cells were resuspended in the same buffer and pelleted as described above. Then the pellet was resuspended in 0.8 ml of ice-cold hypotonic buffer, transferred to microfuge tubes, and allowed to swell on ice for 15-30 min. The lysate was vortexed vigorously for 1 min, and the nuclei were pelleted (14,000 g for 30 s). The nuclear pellets were resuspended in 100-150 µl of high-salt buffer and rotated at 4°C for 30 min. The extracted proteins were separated from residual nuclei (14,000 g for 10 min), and the supernatant was quick frozen in a dry ice-methanol bath. The buffer compositions were as follows: 1) 10 mM HEPES (pH 7.9), 10 mM KCl, 1.5 mM MgCl₂, 1 mM EDTA, 1 mM EGTA, 1 mM Na₃VO₄, 1 mM Na₄P₂O₇, 20 mM NaF, 1 mM dithiothreitol (DTT), 0.5 mM phenylmethylsulfonyl fluoride (PMSF), and 1 µg/ml each of leupeptin, antipain, pepstatin, and chymostatin (hypotonic buffer) and 2) 20 mM HEPES (pH 7.9), 420 mM NaCl, 1.5 mM MgCl₂, 25% glycerol, 1 mM EDTA, 1 mM EGTA, 1 mM Na₃VO₄, 1 mM Na₄P₂O₇, 20 mM NaF, 1 mM DTT, 0.5 mM PMSF, and 1 µg/ml each of leupeptin, antipain, pepstatin, and chymostatin (high-salt buffer).

Electrophoretic mobility-shift assays. Nuclear extracts (12 µg) were added to ³²P-end-labeled oligonucleotide m67-SIE probe (94) (~30,000 cpm, ~5 fmol), and the mixture was incubated in binding buffer for 30 min at 30°C. Final binding reactions (20 µl) contained 14 mM HEPES, pH 7.9, 85 mM NaCl, 10 mM KCl, 0.3 mM MgCl₂, 1.25 mM DTT, 0.25 mM EDTA, 0.2 mM EGTA, 15% glycerol, 50 µg/ml poly(dI-dC) · poly(dI-dC), and 250 µg/ml acetylated BSA. Binding reactions were electrophoresed through 5% polyacrylamide gels (39:1 acrylamide-bis) containing 2.5% glycerol in 0.5× Tris-borate-EDTA buffer at room temperature. The gel was then dried and exposed to X-ray film.

Transient transfections. Murine NIH/3T3 fibroblasts were grown in DMEM with 10% CS. For transient transfection assays, the calcium phosphate method was used (calcium phosphate transfection kit, 5 Prime 3 Prime). The reporter construct had six copies of the high-affinity SIE site upstream of a minimal TK-luciferase gene and was a gift of Richard Jove (88). For reporter assays, NIH/3T3 cells were maintained for 30-40 h in medium containing 0.5% CS after transfection and stimulated with H₂O₂ for 6 h before harvest. One microgram of reporter construct and 100 ng of pRL-TK normalization plasmid were used per 35-mm dish. Dual luciferase assay was carried out following the manufacturer's recommendation (Promega). All transfection experiments were performed in duplicate, and results were normalized to the expression of the Renilla luciferase transfection control.

Immunoprecipitation and Western blotting. Quiescent Rat-1 cells were treated for 10 min with 1 mM H₂O₂, 100 ng/ml interferon-, or 100 ng/ml EGF, lysed in lysis buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 2 mM EGTA, 2 mM Na₃VO₄, 1 mM PMSF, and 5 µg/ml of the protease inhibitors leupeptin, pepstatin, and antipain), and incubated on ice for 30 min. The lysate was then centrifuged for 10 min at 1,000 g at 4°C. The supernatant was removed, and the protein concentration was determined via a Bradford assay. Lysate (500 µg of protein) was incubated with JAK2, 4G10, or anti-PDGF--receptor antibodies as specified (UBI) at 4°C overnight. The immune complexes were then captured with 100 µl of a slurry of protein A-agarose beads for 2 h. The beads were collected by centrifugation, resuspended in 2× Laemmli sample buffer, and boiled before they were run on an 8% SDS polyacrylamide gel and transferred to nitrocellulose. The membrane was then blotted with anti-JAK or 4G10 antibodies, depending on the experiment, and processed via chemiluminescence (Amersham).

Kinase assays. Whole cell extracts of quiescent Rat-1 cells that were treated with H₂O₂ for 5 min were immunoprecipitated as outlined above on 100 µl of a slurry of protein A-agarose beads with anti-JAK2 antibody. The beads were then washed with kinase buffer (50 mM NaCl, 5 mM MgCl₂, 5 mM MnCl₂, 0.1 mM Na₃VO₄, and 10 mM HEPES) before they were incubated with 0.25 mCi/ml of [-³²P]ATP for 30 min at room temperature. The beads were washed with PBS and processed as described above before they were run on an SDS polyacrylamide gel. The gel was dried, exposed to X-ray film, and quantitated on a phosphorimager.

Cell permeabilization. Quiescent Rat-1 fibroblasts were permeabilized by a modification of the method of Bernier et al. (5). Cells were incubated for 20 min in a digitonin permeabilization buffer (15 µg/ml digitonin, 20 mM Tris, pH 7.5, 125 mM KCl, 5 mM NaCl, 10 mM MgCl₂, 11 mM glucose, and 0.1% albumin) and the specified reagents, such as vanadate. Permeabilization was assayed by trypan blue staining. Cells were then processed as outlined above for nuclear extracts, and band shift assays were performed. The permeabilization buffer was modified from the method of Bernier et al. by decreasing the digitonin and using a 1:4 dilution for the final concentration.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Activation of STATs by ROS. To investigate whether members of the STAT family of transcription factors are activated by ROS, Rat-1 fibroblasts were treated with 1 mM H₂O₂ for various lengths of time. Rat-1 fibroblasts were initially examined because they have a relatively low background of constitutive STAT binding activity and have a robust activation of STAT factors in response to PDGF. Nuclear extracts were prepared from the H₂O₂-treated cells, and the extracts were analyzed by electrophoretic mobility shift assay, with the high-affinity c-fos SIE site as a probe. This sequence binds with high affinity to most STATs (73, 94). Within 5 min of exposure to H₂O₂, there is a rapid activation of DNA binding activity (Fig. 1A). The induction is biphasic, with a decrease in activity observed at 15 min and a return to peak levels by 30 min. By 2 h the levels of H₂O₂-induced DNA binding activity returned to near-basal levels. The biphasic response was observed in repeated time course experiments, although the timing of the first decrease in binding varies from 15 to 30 min. A dose-response experiment indicates that STAT activation by H₂O₂ is first detected at 100 µM (see complex B) and increases with doses up to 1 mM (Fig. 1B). Three distinct complexes (A, B, and C) are induced, with the lowermost band (complex C) being the faintest. The magnitude of induction is as much as 14-fold, as measured by phosphorimage analysis. For comparison, the induction of STATs by PDGF is shown in Fig. 1. At optimum doses of PDGF and H₂O₂, PDGF gives a two- to threefold greater induction of STAT DNA binding than does H₂O₂. As would be expected, generation of H₂O₂ by glucose and glucose oxidase in the culture medium also induces the STAT DNA binding (Fig. 2A). The rapid induction of STAT binding by H₂O₂ is inhibited by catalase treatment and does not require new protein synthesis, as it is not blocked by cycloheximide (Fig. 2B). These results indicate that H₂O₂ is the active agent for STAT induction and that the induction is not secondary to an earlier transcriptional response to H₂O₂.

View larger version (93K): [in this window] [in a new window]   Fig. 1.   H2O2 activates signal transducer and activator of transcription (STAT) in Rat-1 cells. A: time course of STAT activation by H2O2 in Rat-1 cells. Confluent, quiescent Rat-1 cells were treated with 1 mM H2O2 for indicated periods of time or with 25 ng/ml platelet-derived growth factor (PDGF) for 15 min and then lysed, and nuclear proteins were extracted and assayed for band shift with an SIE probe. B: dose response of STATs to H2O2 treatment for Rat-1 fibroblasts. Indicated concentrations of H2O2 were added to confluent, quiescent fibroblasts for 10 min, and extracts were prepared for band shift gels. Probe was high-affinity SIE (94).

View larger version (114K): [in this window] [in a new window]   Fig. 2.   Activation of STATs is specific for H2O2. A: glucose oxidase (GO) activates STATs. Confluent, quiescent (Q) Rat-1 fibroblasts were treated for 10 min with indicated concentrations of glucose oxidase, then extracts were prepared for band shift analysis by using a high-affinity SIE probe. FP, free probe. B: catalase inhibits H2O2 activation of STATs. Confluent, quiescent Rat-1 fibroblasts were preincubated with 10 µg/ml cycloheximide (CH) or 3,000 U/ml catalase before addition of 1 mM H2O2 for 10 min. Extracts were then prepared for band shift analysis by using a high-affinity SIE probe.

The activation of the SIE binding activity by H₂O₂ is rapid and specific and has the binding and specificity properties characteristic of the STAT proteins. Inclusion of 100-fold molar excess of oligonucleotides of the interferon-stimulated response element, the cAMP response element, or high-affinity c-fos SIE STAT binding site indicates that only the high-affinity SIE site is able to compete for all three bands (Fig. 3). At least three distinct STAT binding complexes are observed, sometimes referred to as bands A, B, and C (C being the fastest-migrating and, in this case, the faintest complex) (68). In the case of PDGF, there is evidence that the uppermost band (A) is primarily composed of an STAT3 homodimer, the lowermost band (C) of an STAT1 homodimer, and the middle band (B) of a heterodimer of the two (78). Thus, by the position of migration through the gel, H₂O₂ induces in fibroblasts complexes likely to be STAT1 and STAT3, with the two STAT3-containing complexes predominating. To confirm the identity of these SIE binding complexes, antibody supershift analysis was performed (Fig. 3). The addition of anti-STAT3 antiserum to the binding reaction supershifts and diminishes the upper complex and slightly diminishes band B but leaves the lower complex unaltered. The addition of anti-STAT1 antibody, made against a unique COOH-terminal peptide of STAT1, supershifts the lowermost and weakest SIE binding complex (band C) and diminishes the middle band (band B). Thus H₂O₂ induces at least STAT3 and weakly STAT1 in Rat-1 fibroblasts. Because anti-STAT3 antibody only weakly diminishes band B but strongly reacts with band A, it is likely that there are other STATs induced by H₂O₂ or that alternate forms of STAT3 are present in complex B. STAT3-, which has a truncated COOH terminus, would not be recognized by this antibody (9). Alternatively, STAT3 may be poorly recognized by the antibody when it is in the heterodimeric complex. STAT5 and STAT6 can be induced by PDGF in fibroblasts and run near the same position in the gel (62, 90). However, antibodies to STAT5b and STAT6 failed to shift or disrupt the complexes (data not shown).

View larger version (64K): [in this window] [in a new window]   Fig. 3.   Induction of STAT1 and STAT3 by H2O2 in Rat-1 fibroblasts. Extracts of quiescent Rat-1 cells or cells treated for 10 min with 1 mM H2O2 were assayed using a high-affinity SIE probe on band shift gels. Binding reactions were performed in presence or absence of indicated antibodies or competitor DNAs. Indicated antisera or normal rabbit serum (NRS) was incubated with 10 µl (12 µg protein) of nuclear extracts for 30 min on ice at a dilution of 1:10 before binding reaction. Then probe was added, resulting in a final volume of 20 µl. Competitor DNA sequences for the SIE, interferon-stimulated response element (ISRE) (51) and cAMP response element (CRE) (22) were used at a concentration 100 times greater than that of 32P-labeled probe.

To further confirm the activation of STAT3 by H₂O₂, we examined the tyrosine phosphorylation state of tyrosine-705 of STAT3 with an anti-STAT3 phosphotyrosine-specific antibody by Western blotting after H₂O₂ treatment. Phosphorylation of tyrosine-705 is known to be responsible for STAT3 dimerization and activation of DNA binding (97, 103). This antibody recognizes phospho-Y705-STAT3 as opposed to STAT3 or phosphotyrosine alone. There is an increase in STAT3 tyrosine phosphorylation after H₂O₂ (Fig. 4), consistent with the induction of STAT3 DNA binding activity. The same blot was stripped and reblotted with an anti-STAT3 antibody. This experiment confirmed that the increases in tyrosine phosphorylation observed are not due to increases in the levels of STAT3 protein. The two STAT3 bands observed are likely due to differential phosphorylation of STAT3 on serine-727 residue, which regulates STAT3 transactivation activity (13, 96, 97).

View larger version (41K): [in this window] [in a new window]   Fig. 4.   STAT3 is phosphorylated on tyrosine in response to H2O2. Quiescent Rat-1 cells were treated with 1 mM H2O2 for indicated times. Total cellular extract was obtained and run on an 8% SDS polyacrylamide gel. STAT3 proteins [phosphorylated (P-STAT3) and nonphosphorylated (STAT3)] were detected using a PhosphoPlus STAT3 (Tyr-705) antibody (New England Biolabs) or an STAT3 antibody and chemiluminescence detection.

To further characterize the activation of STATs by ROS, we attempted to induce STAT activity in Rat-1 cells by using other types of ROS (data not shown). When cells were treated for as long as 1 h with superoxide-generating stimuli, such as xanthine and xanthine oxidase (0.6 U/ml), menadione (100 µM), or paraquat (200 µM), no or very little induction of STAT complexes was observed. Thus STAT activation in Rat-1 cells appears analogous to the activation of NF-B, in that it is activated by peroxides and not superoxides (71). Although superoxide is broken down into H₂O₂, the concentrations of H₂O₂ generated are apparently not sufficient for STAT induction. Consistent with this hypothesis, it has been found previously and we have confirmed that xanthine oxidase and menadione are less efficient generators of H₂O₂ than glucose oxidase (10, 24). Very high levels of menadione and xanthine/xanthine oxidase will induce small but detectable levels of STAT DNA binding (data not shown).

Other types of ROS also failed to activate STATs. Stimulation of Rat-1 cells with nitric oxide donors, such as isosorbide dinitrite (5 mM), sodium nitroprusside (0.5 mM), or nitrosoglutathione (0.2 mM), failed to activate STAT DNA binding (data not shown). Heavy metals such as ⁵²Fe can produce hydroxyl radicals via the Haber-Weiss reaction but failed to induce the JAK-STAT pathway. In addition, UV irradiation (60 J/m²), previously shown to cause oxidative stress and activate Jun kinase and the EGF receptor (21, 36, 67), fails to stimulate activation of STAT DNA binding in Rat-1 fibroblasts. Heat shock, a more general cell stressor, also failed to induce the STAT pathway (data not shown). Thus the STAT pathway is activated by specific types of ROS and is not a general response to cellular stress. The failure of these cellular stresses to induce STATs is not due to acute toxicity of these agents. Although some of these treatments will lead to toxicity or apoptosis after several hours of treatment, the cells remained intact during the short time of exposure needed to induce the STAT pathway.

Antioxidants inhibit STAT induction by ROS and PDGF. To assess whether the activation of the STAT pathway by H₂O₂ was secondary to oxidative stress, we performed experiments with Rat-1 cells using H₂O₂ and various antioxidants (Fig. 5A). By themselves, the antioxidants NAC, PDTC, and glutathione fail to induce STAT activity in quiescent cells. However, the activation of the STATs by H₂O₂ was inhibited by each of these antioxidants. In contrast, BSO, a glutathione-depleting agent, led to the induction of STAT DNA binding (Fig. 5B). Figure 5C shows that the DNA binding activity induced by BSO is specifically competed away by unlabeled high-affinity STAT binding oligonucleotide. These results indicate that specific types of oxidative stress lead to the activation of STAT proteins.

View larger version (80K): [in this window] [in a new window]   Fig. 5.   Effects of antioxidants and buthionine sulfoximine (BSO) on induction of STAT pathway by H2O2. A: confluent, quiescent Rat-1 fibroblasts were treated with 20 mM N-acetyl-L-cysteine (NAC), 100 µM pyrrolidine dithiocarbamate (PDTC), or 1 mM glutathione (GSH) alone or in combination with 1 mM H2O2 before nuclear extracts were prepared and analyzed as described in Fig. 1 legend. When cells were treated with H2O2, antioxidants were added for 90 min before incubation with H2O2 for 15 min. B: quiescent Rat-1 cells were treated for 5 h with 2 mM BSO or solvent (DMSO), then extracts were prepared for band shift analysis by using a high-affinity SIE probe. C: extracts from BSO-treated lanes from B were run on band shift gels with and without a 100× excess of unlabeled SIE oligonucleotides as a competitor DNA (comp). D: activation of STATs in A-431 cells by oxidative stress. Confluent, quiescent A-431 cells were treated with 100 ng/ml epidermal growth factor (EGF) for 15 min or indicated concentrations of H2O2 for 30 min before extracts were prepared for band shift analysis by using a high-affinity SIE probe. In indicated lane, cells were treated with 2 mM BSO for 12 h before harvest.

The induction of STATs by oxidative stress is not limited to Rat-1 cells. Other fibroblasts such as NIH/3T3 and Balb/c 3T3 cells activate STATs in response to H₂O₂ (data not shown). In addition, H₂O₂ and BSO activate STATs in the epidermally derived A-431 cell line (Fig. 5D). However, in these cells the activation requires higher concentrations of H₂O₂, and the response is less robust. This may indicate that A-431 cells have higher levels of antioxidant enzymes than fibroblasts. This would not be surprising, since many tumor cells have greater resistance to oxidative stress (63).

Interestingly, antioxidants reduce the level of STAT activation by PDGF. NAC reduced the activation of STATs by PDGF by 40%, as measured by phosphorimager analysis (Fig. 6A). This suggests that intracellular oxidation in response to PDGF contributes to the overall levels of STAT induction. Intracellular generation of ROS in response to ligands such as PDGF and tumor growth factor- is believed to be mediated by the activity of a membrane-bound NADH oxidase (41, 48, 54, 87). This enzyme complex is known to be inhibited by DPI. Therefore, we tested whether STAT activation by PDGF could be inhibited by DPI (Fig. 6B). DPI inhibited STAT activation in a dose-responsive manner by PDGF to >80% at 100 µM, as determined by densitometry. DPI was consistently a better inhibitor than NAC, suggesting that the targets of the reactive oxygen products produced in response to PDGF may be colocalized with the NADH oxidase. Because DPI is a flavoprotein inhibitor and not completely specific for the NADH oxidase, we investigated whether inhibitors of other ROS-generating reactions might inhibit STAT activation by PDGF. Inhibitors of nitric oxide synthase (N^G-nitro-L-arginine methyl ester), xanthine oxidase (allopurinol), mitochondrial electron transport (KCN), and microsomal P-450 (methoxypsoralen) were much less effective than DPI in inhibiting STAT induction by PDGF (Fig. 6C). KCN did inhibit 25% of the STAT induction, probably by virtue of reducing the overall ROS load on the cell by inhibiting respiration. These findings suggest that the NADH oxidase is an important intermediary enzyme for the generation of ROS and subsequent activation of STATs by PDGF.

View larger version (94K): [in this window] [in a new window]   Fig. 6.   Antioxidants inhibit PDGF activation of STATs. A: confluent, quiescent Rat-1 fibroblasts were pretreated for 6.5 h with 20 mM NAC before addition of indicated concentrations of PDGF. Extracts were then prepared for band shift analysis by using a high-affinity SIE probe. B: quiescent Rat-1 cells were treated for 15 min with 25 ng/ml of PDGF with or without 30 min of pretreatment with indicated concentrations of diphenylene iodonium (DPI). Extracts were then prepared for band shift analysis by use of high-affinity SIE probe. C: specific inhibition of STAT activation by 20 µM DPI. Confluent, quiescent Rat-1 fibroblasts were pretreated with indicated flavoprotein inhibitors [250 µM methoxypsoralen (MePs), 200 µM allopurinol (AP), 1 mM N-monomethyl-L-arginine (LMNA), and 1 mM KCN] for 30 min and treated with 25 ng/ml PDGF for 15 min, then extracts were prepared for band shift analysis by use of a high-affinity SIE probe.

To examine whether STAT activation by H₂O₂ results in transcriptional activation by STATs as well, we determined whether H₂O₂ could activate transcription of a reporter gene with only STAT binding sites upstream of the promoter. Thus a TK-luciferase reporter gene with or without six high-affinity SIE/STAT binding sites upstream was transfected into NIH/3T3 cells and stimulated for 6 h with 250 µM H₂O₂. H₂O₂ produced a ninefold induction of transcription of the reporter that was dependent on the presence of the STAT binding sites (Fig. 7). Thus H₂O₂ activates STAT DNA binding activity and transcriptional activation.

View larger version (9K): [in this window] [in a new window]   Fig. 7.   Activation of a STAT-dependent promoter by H2O2. NIH/3T3 cells were transfected with TK-luciferase (TK-Luc) or 6XSIE/TK-luciferase (SIE-Luc) reporter constructs. Cells were serum starved for 36 h in 0.5% calf serum, then stimulated for 6 h with 250 µM H2O2 in DMEM. Cell extracts were then processed for luciferase activity. Relative luciferase activities are from an average of 2 experiments. All transfections were performed in duplicate for each experiment.

Activation of JAK2 and TYK2 by H₂O₂. Phosphorylation of the STATs on a specific tyrosine residue is required for the dimerization of the STATs via an SH2 domain-phosphotyrosine interaction and appears to be sufficient to activate STAT DNA binding (76, 77). Thus H₂O₂ activates a cellular tyrosine kinase that phosphorylates the STAT proteins or inactivates a STAT phosphatase. H₂O₂ has previously been shown to stimulate tyrosine phosphorylation in some cells as well as to inhibit tyrosine phosphatases (26, 33, 60). Although the JAK family of kinases have been shown to be necessary for STAT activation by interferons, recent data from our laboratory and others indicate that other tyrosine kinases may also activate STATs (50, 102).

We initially investigated whether any of the JAK family of kinases is activated by H₂O₂ treatment. Quiescent Rat-1 fibroblasts were treated for various times with H₂O₂, and cellular extracts were immunoprecipitated with antiphosphotyrosine monoclonal antibody (4G10) and then processed for Western blotting with an anti-JAK2 antibody. Figure 8A shows an increase in tyrosine phosphorylation of JAK2 detectable at 1 min of stimulation, with a peak of autophosphorylation at 5'. To determine whether the tyrosine phosphorylation of JAK2 was indicative of increased kinase activity of the enzyme, the activity of the enzyme was measured by immune complex kinase assay. JAK2 was immunoprecipitated from extracts of H₂O₂-treated quiescent cells and incubated in the presence of [-³²P]ATP for 5 min. The resulting extracts were then analyzed on SDS polyacrylamide gels for autophosphorylation of JAK2 and total ³²P incorporation into protein. Figure 8B shows that H₂O₂ caused an increase in in vitro autophosphorylation activity of the JAK2. Quantitation of total ³²P incorporation by phosphorimage analysis indicates a fourfold increase in JAK2 kinase activity (Fig. 8C). Thus, not only is JAK2 phosphorylated on tyrosine in response to H₂O₂ treatment, but JAK2 kinase activity is increased as well. Given the strong evidence that JAK kinases are in vivo STAT kinases, this result would suggest that STAT activation by H₂O₂ is mediated at least in part by JAK2. We also examined the activation of JAK1 and TYK2 by H₂O₂. We found no evidence of JAK1 activation, but TYK2 does become phosphorylated on tyrosine in response to H₂O₂ (Fig. 8D), and its kinase activity is activated sixfold (data not show). The expression of the other JAK family member, JAK3, is limited to hematopoietic cells and was not investigated (98). Thus JAK2 and TYK2 respond to H₂O₂.

View larger version (37K): [in this window] [in a new window]   Fig. 8.   Activation of JAK2 and TYK2 by H2O2. A: quiescent Rat-1 cells were treated with 1 mM H2O2 for times indicated, and whole cell extracts were prepared. Immunoprecipitation with antiphosphotyrosine antibody (4G10) was performed, and extracts were resolved on an SDS polyacrylamide gel. A Western blot analysis was performed using anti-JAK2 antibody as indicated and processed via chemiluminescence. B: quiescent Rat-1 cells were treated for indicated times with 1 mM H2O2, and extracts were immunoprecipitated on protein A-agarose beads with anti-JAK2 antibody. Immunoprecipitates were then incubated with [-32P]ATP, electrophoresed through an SDS polyacrylamide gel, and exposed to X-ray film. C: quantitation of kinase assay in B by phosphorimager analysis. D: whole cell extracts from quiescent Rat-1 cells were treated with 1 mM H2O2 for 5 min, extracts were precipitated with anti-TYK2 or JAKI antibody, and immunoprecipitates were processed for Western blots with antiphosphotyrosine antibody (4G10) and chemiluminescence detection.

There is one report that treatment of mesangial cells with 5 mM H₂O₂ activates the PDGF receptor, and it is known that PDGF activates JAK kinases as well as STATs (27). Therefore, we investigated whether H₂O₂ might activate the JAK-STAT pathway through activation of the PDGF receptor in Rat-1 cells. We have found that Rat-1 cells treated with 1 mM H₂O₂ fail to induce phosphorylation of the PDGF- receptor on tyrosine (data not shown). We have also found that EGF only very weakly induces STATs in Rat-1 cells (probably because of low receptor number), so that activation of the EGF receptor does not account for the induction of STATs or JAKs by H₂O₂.

STATs are activated by phosphatase inhibitors. There is evidence that JAK kinases are regulated by phosphatases (15, 17, 19, 43, 46, 65, 101). Previously, it was shown that the pervanadate derivative of vanadate, which is a potent phosphatase inhibitor, can induce the STAT pathway in permeabilized cells (31, 38). H₂O₂ has also been shown to inactivate phosphatases in vitro and in vivo (33, 84). Thus we sought to compare H₂O₂ induction of STATs with pervanadate induction of STATs. This was done in permeabilized cells to facilitate entry of vanadates into the cells, inasmuch as some reports indicate that vanadate does not readily cross the cell membrane (31, 34). Digitonin-permeabilized cells were treated with H₂O₂, vanadyl sulfate, orthovanadate, or pervanadate for 30 min and analyzed for STAT activation (Fig. 9). Vanadate induction of STATs was relatively weak under these conditions, even though it is also a phosphatase inhibitor. Pervanadate is a more potent phosphatase inhibitor than other vanadates, since it inhibits phosphatases by a covalent and irreversible mechanism (37). STAT activation by pervanadate is quantitatively and qualitatively different from H₂O₂ induction of STAT. Figure 9 shows that pervanadate is a very powerful inducer of STAT activation in permeabilized cells (the most potent inducer of STATs we have observed). The same experiments were carried out in unpermeabilized cells with identical results (data not shown), and subsequent reports indicate that vanadate easily enters some cell types (37). However, much more prominent in the pervanadate-treated cell extracts than in the H₂O₂-treated extracts is the lower STAT1 band. Thus, if H₂O₂ is working to activate the STAT pathway by inhibition of phosphatases, then it may be acting only on a subset of the phosphatases that are inhibited by pervanadate.

View larger version (98K): [in this window] [in a new window]   Fig. 9.   Induction of STATs by pervanadate in permeabilized cells. Rat-1 cells were permeabilized using digitonin and treated with 500 µM orthovanadate (oV) or 500 µM vanadyl sulfate (VS) alone or in combination with indicated concentrations of H2O2. Band shift assays were performed using a high-affinity SIE probe. Lanes 10 and 11 are lighter exposures of lanes 8 and 9.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Recent work in many laboratories has shown the activation of critical signal transduction pathways by ROS, such as the NF-B, the JNK-AP-1, and the ras/rac mitogen-activated protein kinase pathway (23, 36, 41, 72, 100). Although the STAT family of transcription factors is known to be activated by most cytokines and growth factors, we have shown that, in addition to these agonists, STAT family members can also be activated by oxidative stress. In particular, we have shown in Rat-1 fibroblasts that the activation of STAT3 and STAT1 in response to H₂O₂ occurs within minutes and is independent of new protein synthesis. The activation of STATs by H₂O₂ also results in STAT-dependent activation of transcription of a luciferase reporter gene. This phenomenon does not appear to be a generalized cell damage response, inasmuch as other stimuli, such as superoxide, UV irradiation, heavy metals, and heat shock, do not activate STATs. In addition, induction of STATs by H₂O₂ occurs in A-431 epithelial cells as well as in other fibroblasts.

Induction of STATs by H₂O₂ was observed with extracellular concentrations ranging from 100 µM-1 mM H₂O₂. These concentrations are similar to the amounts of H₂O₂ needed to mimic the levels of intracellular H₂O₂ generated by PDGF treatment (85). Although intracellular H₂O₂ levels vary significantly between tissues and cellular conditions, some tissues such as the lens of the eye appear to have basal levels of H₂O₂ as high as 25 µM (29). Thus the response of STATs to H₂O₂ could well be a physiological response. Moreover, in several pathological states, such as asthma and ARDS, significant increases in the amounts of H₂O₂ can be measured, and these concentrations can be >1 mM in some cases (20, 53, 70).

The activation of STATs by H₂O₂ could be inhibited by antioxidants, indicating that activation is dependent on ROS. The importance of cellular redox state is indicated by the observation that depletion of reduced glutathione with BSO also leads to STAT activation. In fact, the types of oxidative stress that lead to STAT activation are very similar to those that lead to activation of NF-B (72, 82).

Not only can exogenous H₂O₂ activate STATs, but our findings also implicate ROS in growth factor activation of STATs. Recently, it has been shown that PDGF increases intracellular H₂O₂ concentrations in vascular smooth muscle cells and that treatments with antioxidants such as NAC or catalase transfection can block some downstream PDGF signaling pathways, such as the mitogen-activated protein kinase pathway (85). Here we find that PDGF activation of the STAT pathway can be significantly inhibited by NAC and DPI, indicating that ROS are required for maximal STAT signaling by PDGF. The incomplete inhibition by antioxidants of the induction of STATs by PDGF may be a reflection of the fact that STATs may be phosphorylated by multiple tyrosine kinases that are activated by PDGF, including SRC kinases, JAK kinases, and the PDGF receptor kinase (18, 50, 92). An interesting possibility is that PDGF activation of the STAT pathway requires simultaneous activation of tyrosine kinases and the inhibition of phosphatases.

The mechanism of H₂O₂ activation of STAT proteins likely involves activation of the JAK2 and TYK2 kinases. The JAK kinases are required for the activation of STATs by interferons (59, 91, 95). We have found that the STATs are phosphorylated in response to H₂O₂ and that the JAK2 and TYK2 kinases undergo phosphorylation and activation. However, it has previously been shown that H₂O₂ can also activate other kinases, such as some of the members of the SRC kinase family (60). Thus the activation of STATs by H₂O₂ is likely to be mediated by multiple kinases as appears to be the case for the response to PDGF (92).

We have found that the induction of STATs by H₂O₂ follows a biphasic time course. A similar biphasic STAT response has been noted for the response of STATs to interleukin-6, ANG II, and prolactin (47, 83, 99). In none of these cases is the basis or the function of the biphasic response understood. One possibility is that the second induction involves an exchange of repressor forms of STATs for activating ones. This appears to be an important mode of STAT regulation in the slime mold Dictyostelium (44). Likewise, in mammals STAT isoforms that are naturally occurring dominant negatives are expressed (14). The biphasic time course of STAT activation implies the induction of a biphasic kinase cascade. Such a kinase cascade has previously been noted for the S6 ribosomal kinases (86). The activation of JAK2 and TYK2 that we have observed only accounts for the first phase of induction. We have yet to identify the kinases responsible for the second phase of STAT induction.

There are several possibilities for how H₂O₂ might activate intracellular kinases. One possibility is that H₂O₂ activates receptors on the outside of the cell, as has been found for UV light activation of the JNK pathway (64). Although this is a possibility since many cytokine and growth factors activate JAKs, UV treatment of Rat-1 cells at least fails to elicit the STAT response. In addition, it appears unlikely that H₂O₂ directly activates receptors from the outside of the cell or by direct membrane perturbation, inasmuch as we have found that non-cell-permeant peroxides such as cumene hydroperoxide fail to induce STAT activation (93; data not shown). Moreover, we have found that the concentrations of H₂O₂ used in our experiments are insufficient to activate the PDGF- receptor kinase.

Another possible mechanism for activation of JAK2 by H₂O₂ is inactivation of protein tyrosine phosphatases (PTPs). Tyrosine phosphorylation of proteins is dependent on the balance between kinases and PTPs within a cell. Our data with the PTP inhibitors vanadate and pervanadate suggest that inhibition of PTPs results in STAT activation. The much greater activation of STATs by pervanadate than by vanadate or H₂O₂ alone is consistent with recent data in the literature (17, 31). Pervanadate is a significantly better inhibitor of PTP, because it irreversibly oxidizes a critical cysteine in the catalytic domain, whereas vanadate inhibits only competitively (37). In addition, in vitro and in vivo data indicate that pervanadate is a more effective inhibitor of phosphatases than ROS alone (33, 34, 37). It is possible that in vivo oxidation of this critical cysteine may, in fact, be an in vivo mechanism of downregulating PTPs. Interestingly, the STAT pathway has been shown to be positively and negatively regulated by phosphatases (16, 19, 46). Identification of the phosphatases involved in STAT regulation in Rat-1 fibroblasts is likely to be crucial to understanding the mechanism of H₂O₂ induction of STAT signaling.

In contrast, Haque et al. (31) found that cell lines that were mutant for JAK1, JAK2, and TYK2 are still able to activate STATs after pervanadate treatment. This result does not rule out a JAK2- or TYK2-dependent mechanism for H₂O₂ induction of the STAT pathway, since only a single JAK family kinase is defective in each of these mutant cell lines. Given the intensity of the pervanadate response, it is likely that pervanadate activates multiple JAK kinases or other STAT kinases and that mutations in individual kinases would therefore fail to inhibit STAT activation. This is also the case for STAT activation by growth factors (18, 50, 92).

The biologic roles of the STAT factors are beginning to be understood. It is clear that STATs participate in lactation and mammary gland development, resistance to viral infections, cell cycle regulation, apoptosis, and cellular transformation. STAT3, for instance, has been shown to be important for transformation by v-src (8, 88). The importance of STATs is also evident from knockout mice. Knockouts of STAT2 and STAT3 result in embryonic lethality. STAT1 knockout mice are defective in response to interferons, and STAT5 knockout mice have deficiencies in reproduction (for review see Ref. 14).

The fact that almost all cytokines activate the JAK-STAT pathway suggests that STATs play important roles in the inflammatory process (40). Interestingly, these are the same situations where high levels of ROS are generated. Moreover, the production of H₂O₂ appears to be an important second messenger in many vital cellular processes, including growth factor signaling, cell proliferation, apoptosis, and cellular transformation (2, 85, 87) in addition to contributing to the development of diseases such as ARDS and Alzheimer's (4, 20, 53). Our data suggest that the STATs are intermediates in ROS-mediated gene expression. Although induction of STATs by H₂O₂ may not be the only mechanism of STAT induction by PDGF, any ligand that increases intracellular H₂O₂ could result in the activation of the STAT pathway. Potential target genes for ROS-induced STATs are genes that have known STAT binding sites in their promoters and include genes involved in cell growth regulation, such as c-fos and c-myc (45, 58, 75, 81, 94), genes involved in antioxidant defense (79), such as rat SOD1 (7), and genes involved in apoptosis, such as the caspases (6, 11, 30, 49). We are currently investigating the possibility that the STATs are required for the regulation of these genes by ROS.

	ACKNOWLEDGEMENTS

We thank Jay Lee, E. J. Kulbokas, Deepa Bhavsar, and Susan Spear for help with these experiments.

	FOOTNOTES

This work was supported by National Institutes of Health Grants R01-GM-51551 and R01-GM-54304 (B. H. Cochran) and K08-HL-03547 (A. R. Simon).

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. §1734 solely to indicate this fact.

Address for reprint requests: B. H. Cochran, Dept. of Physiology, Tufts University School of Medicine, 136 Harrison Ave., Boston, MA 02111.

Received 28 April 1998; accepted in final form 29 August 1998.

	REFERENCES

Top Abstract Introduction Materials & Methods Results Discussion References

1.   Amstad, P. A., G. Krupitza, and P. A. Cerutti. Mechanism of c-fos induction by active oxygen. Cancer Res. 52: 3952-3960, 1992[Abstract/Free Full Text].

2.   Bae, Y. S., S. W. Kang, M. S. Seo, I. C. Baines, E. Tekle, P. B. Chock, and S. G. Rhee. Epidermal growth factor (EGF)-induced generation of hydrogen peroxide. Role in EGF receptor-mediated tyrosine phosphorylation. J. Biol. Chem. 272: 217-221, 1997[Abstract/Free Full Text].

3.   Baeuerle, P. A., and D. Baltimore. I-B: a specific inhibitor of the NF-B transcription factor. Science 242: 540-546, 1988[Abstract/Free Full Text].

4.   Behl, C., J. B. Davis, R. Lesley, and D. Schubert. Hydrogen peroxide mediates amyloid beta protein toxicity. Cell 77: 817-827, 1994[Medline].

5.   Bernier, M., A. Liotta, H. Kole, D. Shock, and J. Roth. Dynamic regulation of intact and C-terminal truncated insulin receptor phosphorylation in permeabilized cells. Biochemistry 33: 4343-4351, 1994[Medline].

6.   Bladier, C., E. J. Wolvetang, P. Hutchinson, J. B. de Haan, and I. Kola. Response of a primary human fibroblast cell line to H₂O₂: senescence-like growth arrest or apoptosis? Cell Growth Differ. 8: 589-598, 1997[Abstract].

7.   Bonni, A., D. A. Frank, C. Schindler, and M. E. Greenberg. Characterization of a pathway for ciliary neurotrophic factor signaling to the nucleus. Science 262: 1575-1579, 1993[Abstract/Free Full Text].

8.   Bromberg, J. F., C. M. Horvath, D. Besser, W. W. Lathem, and J. E. Darnell, Jr. Stat3 activation is required for cellular transformation by v-src. Mol. Cell. Biol. 18: 2553-2558, 1998[Abstract/Free Full Text].

9.   Caldenhoven, E., T. B. van Dijk, R. Solari, J. Armstrong, J. A. M. Raaijmakers, J. W. J. Lammers, L. Koenderman, and R. P. de Groot. STAT3, a splice variant of transcription factor STAT3, is a dominant negative regulator of transcription. J. Biol. Chem. 271: 13221-13227, 1996[Abstract/Free Full Text].

10.   Chang, M., M. Shi, and H. J. Forman. Exogenous glutathione protects endothelial cells from menadione toxicity. Am. J. Physiol. 262 (Lung Cell. Mol. Physiol. 6): L637-L643, 1992[Abstract/Free Full Text].

11.   Chin, Y. E., M. Kitagawa, K. Kuida, R. A. Flavell, and X. Y. Fu. Activation of the STAT signaling pathway can cause expression of caspase 1 and apoptosis. Mol. Cell. Biol. 17: 5328-5337, 1997[Abstract].

12.   Chin, Y. E., M. Kitagawa, W.-C. S. Su, Z.-H. You, Y. Iwamoto, and X.-Y. Fu. Cell growth arrest and induction of cyclin-dependent kinase inhibitor p21^WAF1/CIP1 mediated by STAT1. Science 272: 719-722, 1996[Abstract].

13.   Chung, J., E. Uchida, T. C. Grammer, and J. Blenis. STAT3 serine phosphorylation by ERK-dependent and -independent pathways negatively modulates its tyrosine phosphorylation. Mol. Cell. Biol. 17: 6508-6516, 1997[Abstract].

14.   Darnell, J. E., Jr. STATs and gene regulation. Science 277: 1630-1635, 1997[Abstract/Free Full Text].

15.   David, M., H. E. Chen, S. Goelz, A. C. Larner, and B. G. Neel. Differential regulation of the / interferon-stimulated Jak/Stat pathway by the SH2 domain-containing tyrosine phosphatase SHPTP1. Mol. Cell. Biol. 15: 7050-7058, 1995[Abstract].

16.   David, M., P. M. Grimley, D. S. Finbloom, and A. C. Larner. A nuclear tyrosine phosphatase downregulates interferon-induced gene expression. Mol. Cell. Biol. 13: 7515-7521, 1993[Abstract/Free Full Text].

17.   David, M., G. Romero, Z. Y. Zhang, J. E. Dixon, and A. C. Larner. In vitro activation of the transcription factor ISGF3 by interferon- involves a membrane-associated tyrosine phosphatase and tyrosine kinase. J. Biol. Chem. 268: 6593-6599, 1993[Abstract/Free Full Text].

18.   David, M., L. Wong, R. Flavell, S. A. Thompson, A. Wells, A. C. Larner, and G. R. Johnson. STAT activation by epidermal growth factor (EGF) and amphiregulin. Requirement for the EGF receptor kinase but not for tyrosine phosphorylation sites or JAK1. J. Biol. Chem. 271: 9185-9188, 1996[Abstract/Free Full Text].

19.   David, M., G. Zhou, R. Pine, J. E. Dixon, and A. C. Larner. The SH2 domain-containing tyrosine phosphatase PTP1D is required for interferon /-induced gene expression. J. Biol. Chem. 271: 15862-15865, 1996[Abstract/Free Full Text].

20.   Dekhuijzen, P. N., K. K. Aben, I. Dekker, L. P. Aarts, P. L. Wielders, C. L. van Herwaarden, and A. Bast. Increased exhalation of hydrogen peroxide in patients with stable and unstable chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 154: 813-816, 1996[Abstract].

21.   Derijard, B., M. Hibi, I. H. Wu, T. Barrett, B. Su, T. Deng, M. Karin, and R. J. Davis. JNK1: a protein kinase stimulated by UV light and Ha-Ras that binds and phosphorylates the c-Jun activation domain. Cell 76: 1025-1037, 1994[Medline].

22.   Deutch, P. J., J. P. Hoeffler, J. L. Jameson, J. C. Lin, and J. F. Habener. Structural determinants for transcriptional activation by cAMP-responsive DNA elements. J. Biol. Chem. 263: 18466-18472, 1988[Abstract/Free Full Text].

23.   Devary, Y., R. A. Gottlieb, T. Smeal, and M. Karin. The mammalian ultraviolet response is triggered by activation of Src tyrosine kinases. Cell 71: 1081-1091, 1992[Medline].

24.   Feldman, C., R. Anderson, K. Kanthakumar, A. Vargas, P. J. Cole, and R. Wilson. Oxidant-mediated ciliary dysfunction in human respiratory epithelium. Free Radic. Biol. Med. 17: 1-10, 1994[Medline].

25.   Fu, X.-Y., and J.-J. Zhang. Transcription factor p91 interacts with the epidermal growth factor receptor and mediates activation of the c-fos gene promoter. Cell 74: 1135-1145, 1993[Medline].

26.   Gamou, S., and N. Shimizu. Hydrogen peroxide preferentially enhances the tyrosine phosphorylation of epidermal growth factor receptor. FEBS Lett. 357: 161-164, 1995[Medline].

27.   Gonzalez-Rubio, M., S. Voit, D. Rodriguez-Puyol, M. Weber, and M. Marx. Oxidative stress induces tyrosine phosphorylation of PDGF- and - receptors and pp60c-src in mesangial cells. Kidney Int. 50: 164-173, 1996[Medline].

28.  Halliwell, B. Reactive oxygen species in living systems: source, biochemistry, and role in human disease. Am. J. Med. 91, Suppl. 3C: 14S-22S, 1991.

29.   Halliwell, B., and J. M. C. Gutteridge. Free Radicals in Biology and Medicine (2nd ed.). Oxford, UK: Oxford University Press, 1989.

30.   Hampton, M. B., and S. Orrenius. Dual regulation of caspase activity by hydrogen peroxide: implications for apoptosis. FEBS Lett. 414: 552-556, 1997[Medline].

31.   Haque, S. J., V. Flati, A. Deb, and B. R. Williams. Roles of protein-tyrosine phosphatases in Stat1 -mediated cell signaling. J. Biol. Chem. 270: 25709-25714, 1995[Abstract/Free Full Text].

32.   Hayes, T. E., A. M. Kitchen, and B. H. Cochran. Inducible binding of a factor of the c-fos regulatory region. Proc. Natl. Acad. Sci. USA 84: 1272-1276, 1987[Abstract/Free Full Text].

33.   Hecht, D., and Y. Zick. Selective inhibition of protein tyrosine phosphatase activities by H₂O₂ and vanadate in vitro. Biochem. Biophys. Res. Commun. 188: 773-779, 1992[Medline].

34.   Heffetz, D., I. Bushkin, R. Dror, and Y. Zick. The insulinomimetic agents H₂O₂ and vanadate stimulate protein tyrosine phosphorylation in intact cells. J. Biol. Chem. 265: 2896-2902, 1990[Abstract/Free Full Text].

35.   Henkel, T., T. Machleidt, I. Alkalay, M. Kronke, N. Y. Ben, and P. A. Baeuerle. Rapid proteolysis of I-B- is necessary for activation of transcription factor NF-B. Nature 365: 182-185, 1993[Medline].

36.   Hibi, M., A. Lin, T. Smeal, A. Minden, and M. Karin. Identification of an oncoprotein- and UV-responsive protein kinase that binds and potentiates the c-Jun activation domain. Genes Dev. 7: 2135-2148, 1993[Abstract/Free Full Text].

37.   Huyer, G., S. Liu, J. Kelly, J. Moffat, P. Payette, B. Kennedy, G. Tsaprailis, M. J. Gresser, and C. Ramachandran. Mechanism of inhibition of protein-tyrosine phosphatases by vanadate and pervanadate. J. Biol. Chem. 272: 843-851, 1997[Abstract/Free Full Text].

38.   Igarashi, K., M. David, D. S. Finbloom, and A. C. Larner. In vitro activation of a transcription factor by  interferon requires a membrane-associated tyrosine kinase and is mimicked by vanadate. Mol. Cell. Biol. 13: 3984-3989, 1993[Abstract/Free Full Text].

39.   Ihle, J. STATs: signal transducers and activators of transcription. Cell 84: 331-334, 1996[Medline].

40.   Ihle, J. N., B. A. Witthuhn, F. W. Quelle, K. Yamamoto, W. E. Thierfelder, B. Kreider, and O. Silvennoinen. Signaling by the cytokine receptor superfamily: JAKs and STATs. Trends Biochem. Sci. 19: 222-227, 1994[Medline].

41.   Irani, K., Y. Xia, J. L. Zweier, S. J. Sollott, C. J. Der, E. R. Fearon, M. Sundaresan, T. Finkel, and P. J. Goldschmidt-Clermont. Mitogenic signaling mediated by oxidants in Ras-transformed fibroblasts. Science 275: 1649-1652, 1997[Abstract/Free Full Text].

42.   Janssen, Y. M. W., B. V. Houten, P. J. A. Borm, and B. T. Mossman. Biology of disease: cell and tissue responses to oxidative damage. Lab. Invest. 69: 261-273, 1993[Medline].

43.   Jiao, H., K. Berrada, W. Yang, M. Tabrizi, L. C. Platanias, and T. Yi. Direct association with and dephosphorylation of Jak2 kinase by the SH2-domain-containing protein tyrosine phosphatase SHP-1. Mol. Cell. Biol. 16: 6985-6992, 1996[Abstract].

44.   Kawata, T., A. Shevchenko, M. Fukuzawa, K. A. Jermyn, N. F. Totty, N. V. Zhukovskaya, A. E. Sterling, M. Mann, and J. G. Williams. SH2 signaling in a lower eukaryote: a STAT protein that regulates stalk cell differentiation in Dictyostelium. Cell 89: 909-916, 1997[Medline].

45.   Kim, D.-W., V. Cheriyath, A. L. Roy, and B. H. Cochran. TFII-I enhances activation of the c-fos promoter through interactions with upstream elements. Mol. Cell. Biol. 18: 3310-3320, 1998[Abstract/Free Full Text].

46.   Klingmuller, U., U. Lorenz, L. C. Cantley, B. G. Neel, and H. F. Lodish. Specific recruitment of SH-PTP1 to the erythropoietin receptor causes inactivation of JAK2 and termination of proliferative signals. Cell 80: 729-738, 1995[Medline].

47.   Kodama, H., K. Fukuda, J. Pan, S. Makino, M. Sano, T. Takahashi, S. Hori, and S. Ogawa. Biphasic activation of the JAK/STAT pathway by angiotensin II in rat cardiomyocytes. Circ. Res. 82: 244-250, 1998[Abstract/Free Full Text].

48.   Krieger-Brauer, H. I., and H. Kather. Human fat cells possess a plasma membrane-bound H₂O₂-generating system that is activated by insulin via a mechanism bypassing the receptor kinase. J. Clin. Invest. 89: 1006-1013, 1992.

49.   Kumar, A., M. Commane, T. W. Flickinger, C. M. Horvath, and G. R. Stark. Defective TNF--induced apoptosis in STAT1-null cells due to low constitutive levels of caspases. Science 278: 1630-1632, 1997[Abstract/Free Full Text].

50.   Leaman, D. W., S. Pisharody, T. W. Flickinger, M. A. Commane, J. Schlessinger, I. M. Kerr, D. E. Levy, and G. R. Stark. Roles of JAKs in activation of STATs and stimulation of c-fos gene expression by epidermal growth factor. Mol. Cell. Biol. 16: 369-375, 1996[Abstract].

51.   Levy, D. E., D. S. Kessler, R. Pine, N. Reich, and J. E. Darnell, Jr. Interferon-induced nuclear factors that bind a shared promoter element correlate with positive and negative transcriptional control. Genes Dev. 2: 383-393, 1988[Abstract/Free Full Text].

52.   Lubec, G. The hydroxyl radical: from chemistry to human disease. J. Invest. Med. 44: 324-346, 1996[Medline].

53.   Mathru, M., M. W. Rooney, D. J. Dries, L. J. Hirsch, L. Barnes, and M. J. Tobin. Urine hydrogen peroxide during adult respiratory distress syndrome in patients with and without sepsis. Chest 105: 232-236, 1994[Abstract/Free Full Text].

54.   Meier, B., A. R. Cross, J. T. Hancock, F. J. Kaup, and O. T. Jones. Identification of a superoxide-generating NADPH oxidase system in human fibroblasts. Biochem. J. 275: 241-245, 1991.

55.   Meyer, D. J., G. Campbell, B. H. Cochran, L. Argetsinger, A. C. Larner, D. S. Finbloom, C. Carter-Su, and J. Schwartz. Growth hormone induces a DNA-binding factor related to the interferon-stimulated 91-kDa transcription factor. J. Biol. Chem. 269: 4701-4704, 1993[Abstract/Free Full Text].

56.   Meyer, M., H. L. Pahl, and P. A. Baeuerle. Regulation of the transcription factors NF-B and AP-1 by redox changes. Chem. Biol. Interact. 91: 91-100, 1994[Medline].

57.   Meyer, M., R. Schreck, and P. A. Baeuerle. H₂O₂ and antioxidants have opposite effects on activation of NF-B and AP-1 in intact cells: AP-1 as secondary antioxidant-responsive factor. EMBO J. 12: 2005-2015, 1993[Medline].

58.   Minami, M., M. Inoue, S. Wei, K. Takeda, M. Matsumoto, T. Kishimoto, and S. Akira. STAT3 activation is a critical step in gp130-mediated terminal differentiation and growth arrest of a myeloid cell line. Proc. Natl. Acad. Sci. USA 93: 3963-3966, 1996[Abstract/Free Full Text].

59.   Müller, M., J. Briscoe, C. Laxton, D. Guschin, A. Ziemiecki, O. Silvennoinen, A. G. Harpur, G. Barbieri, B. A. Witthuhn, C. Schindler, S. Pellegrini, A. F. Wilks, J. N. Ihle, G. R. Stark, and I. M. Kerr. The protein tyrosine kinase JAK1 complements defects in interferon-/ and - signal transduction. Nature 366: 129-135, 1993[Medline].

60.   Nakamura, K., T. Hori, N. Sato, K. Sugie, T. Kawakami, and J. Yodoi. Redox regulation of an src family protein tyrosine kinase p56lck in T cells. Oncogene 8: 3133-3139, 1993[Medline].

61.   Pahl, H. L., and P. A. Baeuerle. Oxygen and the control of gene expression. Bioessays 16: 497-502, 1994[Medline].

62.   Patel, B. K., L. M. Wang, C. C. Lee, W. G. Taylor, J. H. Pierce, and W. J. LaRochelle. Stat6 and Jak1 are common elements in platelet-derived growth factor and interleukin-4 signal transduction pathways in NIH 3T3 fibroblasts. J. Biol. Chem. 271: 22175-22182, 1996[Abstract/Free Full Text].

63.   Porta, C., M. Moroni, P. Guallini, C. Torri, and F. Marzatico. Antioxidant enzymatic system and free radicals pathway in two different human cancer cell lines. Anticancer Res. 16: 2741-2747, 1996[Medline].

64.   Rosette, C., and M. Karin. Ultraviolet light and osmotic stress: activation of the JNK cascade through multiple growth factor and cytokine receptors. Science 274: 1194-1197, 1996[Abstract/Free Full Text].

65.   Ruff, S. J., K. Chen, and S. Cohen. Peroxovanadate induces tyrosine phosphorylation of multiple signaling proteins in mouse liver and kidney. J. Biol. Chem. 272: 1263-1267, 1997[Abstract/Free Full Text].

66.   Ruff-Jamison, S., K. Chen, and S. Cohen. Induction by EGF and interferon- of tyrosine phosphorylated DNA binding proteins in mouse liver nuclei. Science 261: 1733-1736, 1993[Abstract/Free Full Text].

67.   Sachsenmaier, C., P. A. Radler, R. Zinck, A. Nordheim, P. Herrlich, and H. J. Rahmsdorf. Involvement of growth factor receptors in the mammalian UVC response. Cell 78: 963-972, 1994[Medline].

68.   Sadowski, H. B., K. Shuai, J. E. Darnell, Jr., and M. Z. Gilman. A common nuclear signal transduction pathway activated by growth factor and cytokine receptors. Science 261: 1739-1744, 1993[Abstract/Free Full Text].

69.   Schindler, C., K. Shuai, V. R. Prezioso, and J. E. J. Darnell. Interferon-dependent tyrosine phosphorylation of a latent cytoplasmic transcription factor. Science 257: 809-813, 1992[Abstract/Free Full Text].

70.   Schraufstatter, I., and C. Cochrane. Oxidants: types, sources, and mechanisms of injury. In: The Lung: Scientific Foundations, edited by R. Crystal, and J. B. West. New York: Raven, 1991, p. 1803-1810.

71.   Schreck, R., K. Albermann, and P. A. Baeuerle. Nuclear factor-B: an oxidative stress-responsive transcription factor of eukaryotic cells. Free Radic. Res. Commun. 17: 221-237, 1992[Medline].

72.   Schreck, R., P. Rieber, and P. A. Baeuerle. Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF-B transcription factor and HIV-1. EMBO J. 10: 2247-2258, 1991[Medline].

73.   Seidel, H. M., L. H. Milocco, P. Lamb, J. J. Darnell, R. B. Stein, and J. Rosen. Spacing of palindromic half sites as a determinant of selective STAT (signal transducers and activators of transcription) DNA binding and transcriptional activity. Proc. Natl. Acad. Sci. USA 92: 3041-3045, 1995[Abstract/Free Full Text].

74.   Shibanuma, M., T. Kuroki, and K. Nose. Induction of DNA replication and expression of proto-oncogene c-myc and c-fos in quiescent Balb/3T3 cells by xanthine/xanthine oxidase. Oncogene 3: 17-21, 1988.

75.   Shibanuma, M., T. Kuroki, and K. Nose. Stimulation by hydrogen peroxide of DNA synthesis, competence family gene expression and phosphorylation of a specific protein in quiescent Balb/3T3 cells. Oncogene 5: 1025-1032, 1990[Medline].

76.   Shuai, K., C. M. Horvath, L. H. Huang, S. A. Qureshi, D. Cowburn, and J. J. Darnell. Interferon activation of the transcription factor Stat91 involves dimerization through SH2-phosphotyrosyl peptide interactions. Cell 76: 821-828, 1994[Medline].

77.   Shuai, K., G. R. Stark, I. M. Kerr, and J. E. Darnell, Jr. A single phosphotyrosine residue of stat91 required for gene activation by interferon-. Science 261: 1744-1746, 1993[Abstract/Free Full Text].

78.   Shuai, K., A. Ziemiecki, A. F. Wilks, A. G. Harpur, H. B. Sadowski, M. Z. Gilman, and J. E. Darnell. Polypeptide signalling to the nucleus through tyrosine phosphorylation of Jak and Stat proteins. Nature 366: 580-583, 1993[Medline].

79.   Shull, S., N. H. Heintz, M. Periasamy, M. Manohar, Y. M. Janssen, J. P. Marsh, and B. T. Mossman. Differential regulation of antioxidant enzymes in response to oxidants. J. Biol. Chem. 266: 24398-24403, 1991[Abstract/Free Full Text].

80.   Silvennoinen, O., C. Schindler, J. Schlessinger, and D. E. Levy. Ras-independent growth factor signaling by transcription factor tyrosine phosphorylation. Science 261: 1736-1739, 1993[Abstract/Free Full Text].

81.   Simon, A. R., B. L. Fanburg, and B. H. Cochran. Oxidative stress and cell proliferation. In: Oxygen, Gene Expression, and Cellular Proliferation, edited by L. B. Clerch, and D. J. Massaro. New York: Dekker, 1997, p. 123-138.

82.   Staal, F. J. J., M. Roederer, and L. A. Herzenberg. Intracellular thiols regulate the activation of NF-B and transcription of human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 87: 9943-9949, 1990[Abstract/Free Full Text].

83.   Stevens, A. M., Y. F. Wang, K. A. Sieger, H. F. Lu, and L. Y. Yu-Lee. Biphasic transcriptional regulation of the interferon regulatory factor-1 gene by prolactin: involvement of -interferon-activated sequence and Stat-related proteins. Mol. Endocrinol. 9: 513-525, 1995[Abstract/Free Full Text].

84.   Sullivan, S. G., D. T. Chiu, M. Errasfa, and J. M. Wang. Effects of H₂O₂ on protein tyrosine phosphatase activity in HER 14 cells. Free Radic. Biol. Med. 16: 399-403, 1994[Medline].

85.   Sundaresan, M., Z.-X. Yu, V. J. Ferrans, K. Irani, and T. Finkel. Requirement for generation of H₂O₂ for platelet-derived growth factor signal transduction. Science 270: 296-299, 1995[Abstract/Free Full Text].

86.   Sweet, L. J., D. A. Alcorta, and R. L. Erikson. Two distinct enzymes contribute to biphasic S6 phosphorylation in serum-stimulated chicken embryo fibroblasts. Mol. Cell. Biol. 10: 2787-2792, 1990[Abstract/Free Full Text].

87.   Thannickal, V. J., P. M. Hassoun, A. C. White, and B. L. Fanburg. Enhanced rate of H₂O₂ release from bovine pulmonary artery endothelial cells induced by TGF-₁. Am. J. Physiol. 265 (Lung Cell. Mol. Physiol. 9): L622-L626, 1993[Abstract/Free Full Text].

88.   Turkson, J., T. Bowman, R. Garcia, E. Caldenhoven, R. P. De Groot, and R. Jove. Stat3 activation by Src induces specific gene regulation and is required for cell transformation. Mol. Cell. Biol. 18: 2545-2552, 1998[Abstract/Free Full Text].

89.   Umans, J. G., and R. Levi. Nitric oxide in the regulation of blood flow and arterial pressure. Annu. Rev. Physiol. 57: 771-790, 1995[Medline].

90.   Valgeirsdottir, S., K. Paukku, O. Silvennoinen, C. H. Heldin, and L. Claesson-Welsh. Activation of Stat5 by platelet-derived growth factor (PDGF) is dependent on phosphorylation sites in PDGF- receptor juxtamembrane and kinase insert domains. Oncogene 16: 505-515, 1998[Medline].

91.   Velazquez, L., M. Fellous, G. R. Stark, and S. Pellegrini. A protein tyrosine kinase in the interferon-/ signaling pathway. Cell 70: 313-322, 1992[Medline].

92.   Vignais, M. L., H. B. Sadowski, D. Watling, N. C. Rogers, and M. Gilman. Platelet-derived growth factor induces phosphorylation of multiple JAK family kinases and STAT proteins. Mol. Cell. Biol. 16: 1759-1769, 1996[Abstract].

93.   Vroegop, S., D. Decker, and S. Buxser. Localization of damage induced by reactive oxygen species in cultured cells. Free Radic. Biol. Med. 18: 141-151, 1995[Medline].

94.   Wagner, B. J., T. H. Hayes, C. J. Hoban, and B. H. Cochran. The SIF binding element confers sis/PDGF inducibility onto the c-fos promoter. EMBO J. 9: 4477-4484, 1990[Medline].

95.   Watling, D., D. Guschin, M. Muller, O. Silvennionen, B. A. Witthuhn, J. N. Ihle, G. R. Stark, and I. M. Kerr. Complementation of a mutant cell is defective in the interferon- signal transduction pathway by the protein tyrosine kinase Jak2. Nature 366: 166-170, 1993[Medline].

96.   Wen, Z., and J. E. Darnell, Jr. Mapping of Stat3 serine phosphorylation to a single residue (727) and evidence that serine phosphorylation has no influence on DNA binding of Stat1 and Stat3. Nucleic Acids Res. 25: 2062-2067, 1997[Abstract/Free Full Text].

97.   Wen, Z., Z. Zhong, and J. E. Darnell, Jr. Maximal activation of transcription by Stat1 and Stat3 requires both tyrosine and serine phosphorylation. Cell 82: 241-250, 1995[Medline].

98.   Witthuhn, B. A., O. Silvennoinen, O. Miura, K. S. Lai, C. Cwik, E. T. Liu, and J. N. Ihle. Involvement of the Jak-3 Janus kinase in signalling by interleukins 2 and 4 in lymphoid and myeloid cells. Nature 370: 153-157, 1994[Medline].

99.   Wu, Y. Y., and R. A. Bradshaw. Induction of neurite outgrowth by interleukin-6 is accompanied by activation of Stat3 signaling pathway in a variant PC12 cell (E2) line. J. Biol. Chem. 271: 13023-13032, 1996[Abstract/Free Full Text].

100.   Xanthoudakis, S., G. Miao, F. Wang, Y. C. Pan, and T. Curran. Redox activation of Fos-Jun DNA binding activity is mediated by a DNA repair enzyme. EMBO J. 11: 3323-3335, 1992[Medline].

101.   Yin, T., R. Shen, G. S. Feng, and Y. C. Yang. Molecular characterization of specific interactions between SHP-2 phosphatase and JAK tyrosine kinases. J. Biol. Chem. 272: 1032-1037, 1997[Abstract/Free Full Text].

102.   Yu, C., D. Meyer, G. Campbell, A. Larner, C. Carter-Su, J. Schwartz, and R. Jove. Enhanced DNA-binding activity of a Stat3-related protein in cell transformed by the Src oncoprotein. Science 269: 81-83, 1995[Abstract/Free Full Text].

103.   Zhong, Z., Z. Wen, and J. J. Darnell. Stat3: an STAT family member activated by tyrosine phosphorylation in response to epidermal growth factor and interleukin-6. Science 264: 95-98, 1994[Abstract/Free Full Text].