Activation of the mitogen-activated protein kinase signaling pathway in neutrophils. Role of oxidants.

In addition to their role in bacterial killing, reactive oxygen intermediates (ROI) produced by the NADPH oxidase may participate in the regulation of intracellular pathways. We have recently demonstrated that ROI produced by the oxidase regulate tyrosine phosphorylation in neutrophils, possibly by alterations in the cellular redox state. The purpose of the present study was to characterize the identities of certain of the redox-sensitive tyrosine-phosphorylated substrates and the significance of the increased phosphorylation. As a prominent 42-44-kDa phosphorylated band was noted in oxidant-treated cells, we investigated the possible phosphorylation and activation of mitogen-activated protein (MAP) kinase under these conditions. Immunoprecipitation of MAP kinase followed by immunoblotting with anti-phosphotyrosine antibodies indicated that a 42-44-kDa polypeptide was tyrosine-phosphorylated in response to treatment of cells, either with the oxidizing agent diamide or with H2O2 in cells where catalase was inhibited. Using an in vitro renaturation assay with myelin basic protein as the substrate, oxidant-induced stimulation of kinase activity of a 42-44-kDa band was observed in both whole cell extracts and in MAP kinase immunoprecipitates. The mechanism of redox-sensitive activation of MAP kinase was examined. First, exposure of cells to oxidants caused a significant increase in the activity of MEK (the putative activator of MAP kinase), as determined by an in vitro kinase assay using recombinant catalytically inactive glutathione S-transferase-MAP kinase as the substrate. Additionally, oxidant treatment of cells resulted in inhibition of the activity of CD45, a protein tyrosine phosphatase known to dephosphorylate and inactivate MAP kinase. We conclude that oxidant treatment of neutrophils can activate MAP kinase by stimulating its tyrosine and (presumably) threonine phosphorylation via MEK activation, a response that may be potentiated by inhibition of MAP kinase dephosphorylation by phosphatases such as CD45.

In addition to their role in bacterial killing, reactive oxygen intermediates (ROI) produced by the NADPH oxidase may participate in the regulation of intracellular pathways. We have recently demonstrated that ROI produced by the oxidase regulate tyrosine phosphorylation in neutrophils, possibly by alterations in the cellular redox state. The purpose of the present study was to characterize the identities of certain of the redox-sensitive tyrosine-phosphorylated substrates and the significance of the increased phosphorylation. As a prominent 4244-kDa phosphorylated band was noted in oxidanttreated cells, we investigated the possible phosphorylation and activation of mitogen-activated protein (MAP) kinase under these conditions. Immunoprecipitation of MAP kinase followed by immunoblotting with anti-phosphotyrosine antibodies indicated that a 4244-kDa polypeptide was tyrosine-phosphorylated in response to treatment of cells, either with the oxidizing agent diamide or with H,O, in cells where catalase was inhibited. Using an in vitro renaturation assay with myelin basic protein as the substrate, oxidant-induced stimulation of kinase activity of a 4244-kDa band was observed in both whole cell extracts and in MAP kinase immunoprecipitates. The mechanism of redox-sensitive activation of MAP kinase was examined. First, exposure of cells to oxidants caused a significant increase in the activity of MEK (the putative activator of MAP kinase), as determined by an in vitro kinase assay using recombinant catalytically inactive glutathione S-transferase-MAP kinase as the substrate. Additionally, oxidant treatment of cells resulted in inhibition of the activity of CD45, a protein tyrosine phosphatase known to dephosphorylate and inactivate MAP kinase. We conclude that oxidant treatment of neutrophils can activate MAP kinase by stimulating its tyrosine and (presumably) threonine Medical Research Council of Canada (to G. P. D., D. R., and S. G.), the *This work was supported in part by operating grants from the Ontario Thoracic Society (to G. P. D.), and the National Sanitarium Association (to G. P. D.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 8 Recipient of a Canadian Thoracic Society Fellowship Award and a National Council of Scientific and Technological Development Award (Brazil). phosphorylation via MEK activation, a response that may be potentiated by inhibition of MAP kinase dephosphorylation by phosphatases such as CD45.
When exposed t o invading microorganisms and inflammatory mediators, neutrophils exhibit a variety of rapid and coordinated responses including cytoskeletal rearrangement, motility, exocytosis of secretory granules, and activation of the NADPH oxidase (reviewed by Sha'afi and Molski (1988)). The latter results in the production of reactive oxygen intermediates (ROI),' a process termed the respiratory burst.
Tyrosine phosphorylation may play a crucial role in the regulation of effector functions of neutrophils (Grinstein and Furuya, 1991). Several soluble activating agents including Nformyl-methionyl-leucyl-phenylalanine (fMLP: Huang et ul. (19881, Berkow and Dodson (1990)), tumor necrosis factor (Akimaru et al., 19921, granulocyte-monocyte colony stimulating factor (Gomez-Cambronero et a l . , McColl et al., 19911, phorbol 12-myristate 13-acetate (Berkow and Dodson, 19901, and platelet activating factor (Gomez-Cambronero et al., 1991) have been shown to induce tyrosine phosphorylation in neutrophils. While the identity and functional role of many of the proteins that are tyrosine-phosphorylated in response to soluble activating agents remain incompletely understood, recent studies have demonstrated that members of the MAP kinase family can be phosphorylated on tyrosine residues and activated in response to the chemoattractant fMLP (Grinstein and Furuya, 1992;Torres et al., 1993) or by granulocyte-monocyte colony stimulating factor (Gomez-Cambronero et al., 1992). MAP kinases were originally described as serine/ threonine kinases that are activated by various growth factors and tumor promoters in mammalian cultured cells and are now thought to function as key intermediaries in signaling processes leading to cellular growth and differentiation (reviewed by Nishida and Gotoh (1993) and Blenis (1993)).
Reactive oxygen intermediates produced by phagocytes have traditionally been viewed as potent microbicidal agents (Cross and Jones, 1989), and, in pathological circumstances, excess ROI production may result in host cell damage. Recent evidence from studies in several cell types suggest that ROI may also function as signaling molecules. These studies include the demonstration that H,O, activates the transcription factor fMLP, formyl-methionyl-leucyl-phenylalanine; PAGE, polyacrylamide The abbreviations used are: ROI, reactive oxygen intermediates; gel electrophoresis; PMSF, phenylmethylsufonyl fluoride; Dm, dithiothreitol; MBP, myelin basic protein; MAP kinase, mitogen-activated togen-activated protein kinase or Extracellular signal-regulated E-protein kinase; ERK, extracellular signal-regulated kinase; MEK, Minase; p44mnpk, ERK1. NFKB, which controls the synthesis of a variety of cytokines (Schreck et al., 19911, and, more recently, that scavengers of free radicals can suppress tumor necrosis factor-a-induced expression of the JE/MCP-1 and CSF-1 genes, which regulate the expression of the monocyte chemoattractant protein and the monocyte colony stimulating factor, respectively (Satriano et al., 1993). Additionally, endogenously produced oxidants have been implicated in the activation of the protein tyrosine kinases ltk (Bauskin et al., 1991) and ~7 2 "~~ (Schieven et al., 1993) in lymphocytes and in the inhibition of protein tyrosine phosphatases in lymphocytes and macrophages (Hecht and Zick, 1992;Zor et al., 1993).
In a previous study, we provided evidence that tyrosine phosphorylation in neutrophils was sensitive to oxidants (Fialkow et al., 1993). The purpose of the current study was to characterize the identities of the redox-sensitive tyrosine-phosphorylated substrates and the physiological significance of the increased phosphorylation. As a prominent 4244-kDa phosphorylated band was noted in oxidant-treated cells, we investigated the possibility that phosphorylation and activation of MAP kinase ( M , = 42,000-44,OOO) occurred under these conditions. In this report, we provide evidence that oxidant treatment of cells does indeed result in tyrosine phosphorylation and activation of MAP kinase. Our studies suggest that activation of MAP kinase by oxidants is mediated in part by activation of MEK, the putative upstream activator of MAP kinase (reviewed by Anh et al. (1992) and Crews and Erikson (1993)). Additionally, MAP kinase activation by oxidants may be potentiated by inhibition of cysteine-containing (Tonks et al., 1988) tyrosine phosphatases such as CD45 that are capable of dephosphorylating and inactivating MAP kinase (Anderson et al., 1990).
For experiments using MLP, neutrophils were suspended in bicarbonate-free RPMI 1640 (Sigma), buffered to pH 7.3 with 25 IILM sodium Hepes. Anti-phosphotyrosine antibodies were from two sources: (i) affinity-purified polyclonal anti-phosphotyrosine was from Transduction Laboratories (Lexington, KY) and (ii) affinity-purified monoclonal antiphosphotyrosine (4G10 hybridoma) was from Upstate Biotechnology (Lake Placid, NY). Both antibodies gave essentially identical results on Western blots. A monoclonal anti-MAP kinase antibody (raised against a sequence corresponding to amino acids 325-345 of ERKl from Zymed Laboratories, San Francisco, CA) was used for immunoblotting. Antibodies used for immunoprecipitation of MAP kinase were from two different sources. (i) MAP kinase antiserum was kindly provided by John Blenis (Harvard Medical School, Boston, MA). This antiserum immunoprecipitates 42-kDa, 44-kDa, and 45-kDa MAP kinase polypeptides. (ii) Antibodies 956/837, a mixture of polyclonal antibodies raised to the sequence 352-367 of rat ERKl that recognizes ERKl (44 kDa), ERK2 (42 kDa), and ERK4 (45 kDa) as characterized by Boulton and Cobb (1991) were from Santa Cruz Biotechnology (Santa Cruz, CA). ProteinNG-agarose beads were from Santa Cruz Biotechnology or from Oncogene Sciences (Uniondale, NY). Protein A-Sepharose was from Pharmacia Biotech (Uppsala, Sweden). Myelin basic protein was kindly provided by Dr. Mario Moscarello (Dept. of Biochemistry, Hospital for Sick Children, Toronto, Canada). Anti-MEK antibodies (raised against the N terminus peptide PKKKPTPIQLNPNPEY coupled to KLH with BDB) for immunoprecipitation and immunoblotting were kindly provided by Dr. Gilles L'Allemain (Centre de Biochimie, Centre National de la Recherche Scientifique, Universite de Nice, Nice, France). The cDNA for the recombinant catalytically inactive p44"apk-glutathione S-transferase MAP kinase (kinase-MAP kinase) and was kindly provided by R. Erikson, Harvard University. The cDNA was expressed in Escherichia coli as a glutathione S-transferase-fusion protein and purified as described (Ausubel et al., 1990). Monoclonal 9.4 anti-CDM antibody (HB 10508) used for immunoblotting and GAP 8.3 (HB 12) used for CD45 immunoprecipitation were isolated from the corresponding hybridomas obtained from the American Type Tissue Collection (ATCC, Rockville, MD). Anti-Shc antibodies used for immunoprecipitation and immunoblotting were kindly provided by Jane McGlade and Tony Pawson (Division of Molecular and Developmental Biology, Samuel Lunenfeld Research Institute, Toronto, Ontario). The anti-Shc antibody recognizes three proteins of the Shc family of 46, 52, and 66 kDa (Pelicci et al., 1993). The anti-Grb2 antibodies were provided by Joseph Schlessinger, New York University Medical Center.

Cell Isolation
Human neutrophils (>98% pure) were isolated from citrated whole blood obtained by venipuncture, using dextran sedimentation and discontinuous plasma-Percoll gradients as described previously (Haslett et al., 1985). The separation procedure required 2 h, and the cells were used immediately after isolation for the experiments described. The functional integrity and nonactivated state of neutrophils isolated in this manner have been validated extensively in previous publications (Haslett et al., 1985;Downey et al., 1992).

HL-60 Culture
HL-60 cells (American Type Culture Collection) were grown in RPMI 1640 medium (University of Toronto Media Preparation Service, Toronto, Ontario) supplemented with 10% heat-inactivated fetal bovine serum (Life Technologies, Inc.), L-glutamine (2 mM), streptomycin (100 units/ml), and penicillin (100 mg/ml). The cells were passaged at starting densities of 2.5-3.5 x lo5 cells/ml and maintained in suspension culture at 37 "C in an air atmosphere containing 5% CO,. The cultures were passaged every 3-4 days so that the cell density did not exceed 1.5 x lo6 cells/ml. Granulocytic differentiation was induced by treatment with 1.25% (v/v) dimethyl sulfoxide for 6-7 days. Differentiation was confirmed by analysis of cell size using the Coulter Counter and by staining characteristics using a modified Wright-Giemsa (Diff-Quick TM, American Scientific Products, McGraw Park, IL). The cells were centrifuged and resuspended in RPMI 1640 at a density of 8 x lo6 celL'ml for use in experiments.

SDS-PAGE and Immunoblotting
For SDS-PAGE of whole cell extracts, 5 x lo6 cells were sedimented, resuspended in 25 pl of phosphate-buffered saline containing 1 mM PMSF and 125 1. 11 of boiling 2% SDS sample buffer, and boiled for an additional 10 min. 35 p1 of each sample and molecular weight standards were then subjected to electrophoresis in the presence of SDS on a 4-20% polyacrylamide gradient gel (Novex Experimental Technology, San Diego, CAI according to the method of Laemmli (1970). Following SDS-PAGE, the samples and molecular weight standards were electrophoretically transferred to nitrocellulose paper and incubated overnight at 4 "C in a blocking solution containing 0.25% gelatin, in 10% ethanolamine and 0.1 M Tris, pH 9.0. The blot was then incubated with buffer (0.25% gelatin, 0.05% Nonidet P-40 in 0.15 M NaCI, 5 mM EDTA, and 50 mM Tris, pH 7.5) containing the relevant antibodies for 2 h while shaking at room temperature. The primary antibodies were used at the following concentrations: the anti-phosphotyrosine antibody 4G10 or the polyclonal anti-phosphotyrosine antibody from Transduction Laboratories and anti-MEK antibody were used at 1:5000 dilution. The monoclonal anti-MAP kinase was used at 1:2500 dilution. The anti-Shc and anti-Grb2 antibodies were used a t 1:200 dilution. The monoclonal 9.4 anti-CD45 antibody was used at 1:lOOO dilution. Blots were then washed three times with the same buffer used for incubation with primary antibody. Except for the experiments with anti-Shc antibodies, all blots were incubated with horseradish peroxidase-conjugated antimouse or anti-rabbit IgG at 15000 dilution for 1 h a t room temperature, washed three additional times, and developed using a n enhanced chemiluminescence system (Amersham). For immunoblottings of Shc immunoprecipitates, lZ5I-labeled goat anti-mouse or '251-labeled Protein A were used as the secondary antibodies. After washing three times as above, the blots were dried and exposed to autoradiography with Kodak X-Omat AR film (Eastman) with intensifying screens at -70 "C. Autoradiograms were scanned using a PDI model DNA 35 scanner and the Discovery series 1D gel analysis software.

Immunoprecipitation
For these studies, neutrophils were first incubated with 2.5 mM diisopropyl fluorophosphate for 30 min at 21 "C to inactivate proteases and thus minimize protein degradation (Rotrosen and Leto, 1990).
MAP Kinase Immunoprecipitation-Neutrophils (2 x 1O7cells) were suspended in 100 pl of lysis buffer A (150 mM NaCl, 5 mM EGTA, 5 mM EDTA, 10 mM sodium pyrophosphate, 10 mM NaF, 1 mM sodium vanadate, 1 mM PMSF, 10 pg/ml aprotinin, 10 pg/ml leupeptin, 1 p~ pepstatin, 10% glycerol, 25 mM Tris-HC1, pH 7.4) containing 1% SDS and boiled for 5 min. Next, 900 pl of ice-cold lysis buf€er A, without SDS, but containing 1% Triton X-100, 0.5% Nonidet P-40, and 1% sodium deoxycholate was added. The lysate was sonicated and sedimented in an Eppendorf Microfuge tube for 15 min at 4 "C. The supernatant was incubated with 4 pg of anti-MAP kinase antibody (Santa Cruz Biotechnology) or 7.5 p1 of antiserum (provided by Dr. J. Blenis) for 2 h at 4 "C followed by incubation with 40 pl of protein A/G-agarose (previously blocked in bovine serum albumin and washed twice with lysis buffer A) overnight at 4 "C. The immune complexes were washed twice with 25 m Tris buffer, pH 7.4, containing 2 mM EDTA and 150 mM NaCl, boiled for 5 min in Laemmli sample buffer, and rapidly sedimented, and the supernatant was analyzed by SDS-PAGE as described above.
MEKImmunopreczpitation-Neutrophils (2 x 10') were suspended in 1 ml of lysis buffer B (50 mM Tris-HC1, pH 7.5, 150 mM NaC1, 10% glycerol, 1 mM EGTA, 10 mM sodium pyrophosphate, 10 m NaF, 1 m~ sodium vanadate, 1 mM PMSF, 10 pg/ml aprotinin, 10 pdml leupeptin, 1 )uvl pepstatin). The lysate was sonicated and sedimented in a Microfuge for 10 min. The anti-MEK antibodies were first coupled to Protein A-Sepharose (3 pl of antibodyl40 pl of a 50% suspension of washed Protein A beads) by incubating the mixture for 4 h a t 4 "C. Cell lysates were then incubated with 43 p1 of the anti-MEK antibody/ Protein A-Sepharose mixture for 2 h at 4 "C. The immune complexes were washed six times with lysis buffer B and once with a kinase buffer containing 50 mM Tris, pH 8.0, 3 mM magnesium acetate, 5 mM DTT, 1 mM EGTA, 0.1 mg/ml ovalbumin and assayed for MEK activity as described below. Additionally, aliquots were analyzed by SDS-PAGE and immunoblotting to confirm the effectiveness of the immunoprecipitation procedure.

CD45 Immunoprecipztation-Neutrophils
(3 x lo7 cells) were lysed in 1 ml of lysis buffer C (phosphate-buffered saline with 0.2% Nonidet P-40, 2 m PMSF, 0.5 mM benzamidine, 10 pg/ml aprotinin, 10 pg/ml leupeptin) a t 4 "C for 10 min, and insoluble material was removed by centrifugation in a Microfuge. The supernatant was incubated with 10 pl of anti-CD45 (GAP 8.3) for 2 h at 4 "C followed by incubation with anti-mouse IgG (whole molecule)-agarose overnight a t 4 "C. The immune complexes were washed six times with lysis buffer C and once with a phosphatase buffer containing 25 mM Hepes, pH 7.8,5 mM EDTA and then assayed for tyrosine phosphatase activity as described below. It is important to note that both the lysis and phosphatase assay buffers were previously degassed and then deoxygenated by bubbling with nitrogen for 10 min in order to minimize the oxidation of the enzyme after firmed by cell surface labeling (Cole et al., 1987) with NHS-LC-biotin cell lysis. The effectiveness of CD45 immunoprecipitation was con-(Pierce) according to the manufacturer's instructions and by immunoblotting with the 9.4 monoclonal anti-CD45 antibody.
Shc Immunoprecipitation-Neutrophils (2 x lo7) were suspended in 1 ml of lysis buffer D (20 mM Tris-HC1, pH 7.5, 150 m~ NaCI, 10% glycerol, 0.1% Triton X-100,l mM EGTA, 10 mM sodium pyrophosphate, 10 m~ NaF, 1 m~ sodium vanadate, 1 mM PMSF, 10 pdml aprotinin, 10 pdml leupeptin, 1 p~ pepstatin). The lysate was sonicated and sedimented in a Microfuge for 10 min. Anti-Shc antibodies were first bound to Protein A-Sepharose (2 pl of antibody/40 p1 of a 50% suspension of washed Protein A-Sepharose) by incubating the mixture for 4 h a t 4 "C. Cell lysates were then incubated with 42 pl of the anti-Shc antibodyprotein A-Sepharose mixture for 2 h a t 4 "C. The immune complexes were washed three times with lysis buffer D, resuspended in Laemmli buffer, and boiled for 5 min. Samples were analyzed by SDS-PAGE and immunoblotting with either anti-Shc, anti-Grb2, or antiphosphotyrosine antibodies.
Blot Stripping and Reprobing Blots were stripped of the primary antibody-secondary antibody complex by incubating them in stripping buffer (100 mM 2-mercaptoethanol, 2% SDS, 62.5 mM Tris-HC1, pH 6.7) for 30 min at 50 "C. The blots were incubated with blocking buffer and reprobed with a different antibody according to the procedures described above.
Myelin Basic Protein (MBP) Kinase Assay in SDS-Polyacrylamide Gels MAP kinase activity was determined using a gel renaturation assay with MBP as the substrate (Kameshita and Fujisama, 1989). 10% polyacrylamide minigels were cast with 0.5 mg/ml MBP added to the gel prior to polymerization. After electrophoresis, the SDS was removed by washing the gels twice for 30 min each with 100 ml of 20% isopropyl alcohol in 50 mM Tris-HC1, pH 8.0, followed by a 1-h incubation with 100 ml of 50 m Tris-HCl, pH 8.0, 5 mM 2-mercaptoethanol. Gels were incubated twice for 30 min each in 100 ml of 6 M guanidine HCI, 5 mM 2-mercaptoethanol, 50 mM Tris-HC1, pH 8.0, and subsequently renatured by incubation with five changes (250 ml each) of 5 mM 2-mercaptoethanol, 0.04% Tween 40, 50 mM Tris-HC1, pH 8.0, over 12-18 h a t 4 "C. The kinase assay was camed out by incubation of the gel with 25 ml of 10 m Hepes, pH 8.0, 2 mM dithiothreitol (DTT), 0.1 m~ EGTA, 5 mM MgCl,, 25 p~ ATP, and 25 pCi of [Y-~*PIATP (1 pCi/pl) for 1 h a t room temperature. The gels were washed thoroughly with 100 ml of 5% trichloroacetic acid (w/v) containing 1% sodium pyrophosphate until the washes contained minimal radioactivity (usually t%8 changes). Dried gels were exposed to Kodak X-OMAT film for 8-12 h at -85 "C prior to development.
MEK Activity Assay MEK immunoprecipitates were obtained from neutrophils as described above. 30 pl of each immunoprecipitate was assayed for the ability to phosphorylate a catalytically inactive p44"'"Pk-glutathione Stransferase fusion protein (kinase-MAPK) by mixing with 20 pl of kinase buffer (50 mM Tris, pH 8.0,3 m~ magnesium acetate, 5 mM DTT, 1 m EGTA, 0.1 mg/ml ovalbumin, 25 p~ ATP, 5 pCi of [y-32P]ATP, containing 10 pl (0.35 mg/ml) of recombinant MAP kinase). After a 30-min incubation a t 30 "C, the reaction was terminated by the addition of boiling Laemmli sample buffer, boiled for 5 min, and the supernatant was analyzed by SDS-PAGE and autoradiography.
Protein nrosine Phosphatase Activity Protein tyrosine phosphatase activity of the CD45 immunoprecipitates was measured as described (Mustelin, 1989) using free O-phosphotyrosine as the substrate. Briefly, the immunoprecipitates were incubated with assay buffer (25 mM Hepes, pH 7.8, 5 mM EDTA) and 10 mM L-0-phosphotyrosine for 20 min a t 37 "C, and the reaction was stopped by the addition of trichloroacetic acid. 200 pl of the supernatant was diluted in 300 pl of H,O and incubated with 0.5 ml of a solution containing 2 volumes of H,O, 1 volume of 6 N sulfuric acid, 1 volume of 2.5% ammonium molybdenate, and 1 volume of 10% ascorbic acid for 1-3 h and subsequently assayed for the concentration of the hydrolysis product (free phosphate) by measurement of the absorbance at 750 nm. KH,PO, was used to generate a standard curve. DTT was not included in the assay buffer to preclude reversal of the effects of prior oxidant treatment (Tonks et al., 1988. It should be noted that greater than 90% of the protein phosphatase activity of the immunoprecipitates was inhibited by 1 mM sodium vanadate which is characteristic of CD45. All experiments were performed at least three times with blood from different donors. Results are presented as representative autoradiograms or mean S.E. Where indicated, data were subjected to analysis of variance for repeated measures with correction for multiple comparisons (Scheffe).

RESULTS AND DISCUSSION
To study the effects of oxidants on cellular tyrosine phosphorylation, neutrophils were exposed to two membrane-pemeant oxidants. In agreement with earlier observations (Fialkow e t al., 1993), treatment of cells with the thiol-oxidizing agent diamide (Kosower and Kosower, 1987)  ECL reagents was shortened, as to define more clearly the approximate molecular weights of the polypeptides that were tyrosine-phosphorylated respectively; lane 4, cells were preincubated with 50 mhf aminotriazole for 60 min and then exposed to 1 m M H,O, for 10 min, prior to SDS-PAGE and Western analysis. C, diamide-induced tyrosine phosphorylation is inhibited by N-acetyl+cysteine and dithiothreitol. Autoradiogram of an immunoblot using a polyclonal anti-phosphotyrosine antibody. Lane I, untreated neutrophils; lane 2, cells incubated with 1 m M diamide for 20 min a t 37 "C; lane 3, cells were preincubated with 30 mM N-acetyl-1.-cysteine (NAC) for 15 min and subsequently exposed to 1 m%% diamide for 20 min; lanes 4 and 5, cells incubated with 10 mM DTT for 10 min (lane 4 ) or 20 min (lane 5) and subsequently incubated with diamide for 20 min, and the reaction was stopped by boiling in sample buffer as described under "Experimental Procedures." D, tyrosine phosphorylation of MAP kinase After the indicated treatments, neutrophils were lysed and immunoprecipitation of MAP kinase was carried out as described under "Experimental in response to oxidants as determined by immunoprecipitation. Autoradiogram of an immunoblot with a monoclonal anti-phosphotyrosine antibody.
Procedures." Aliquots of the immunoprecipitates were used for immunoblotting. Lane I, untreated neutrophils; lane 2, cells incubated with 1 mhl diamide for 20 min a t 37 "C; lane 3, neutrophils pretreated with 50 mM aminotriazole for 60 min and then exposed to 1 m M H,O, for 10 min.
whether the 42-44-kDa tyrosine-phosphorylated polypeptide was a member of the MAP kinase family, the nitrocellulose blot was stripped and reprobed with a monoclonal anti-MAP kinase antibody. Fig. L4, right panel, illustrates that the tyrosinephosphorylated 42-44-kDa protein co-migrated with MAP kinase, suggesting that it might be a member of the MAP kinase family. To ensure that the enhanced tyrosine phosphorylation was related to a change in the cellular redox potential, the effects of the antioxidants N-acetyl-L-cysteine and dithiothrei-to1 on tyrosine phosphorylation were studied. Fig. 1C illustrates that the increased tyrosine phosphorylation induced by diamide was completely abrogated by prior incubation of cells by either of the antioxidants, thus confirming that the enhanced tyrosine phosphorylation was mediated by alterations in the cellular redox state.
The above data suggest, but do not demonstrate conclusively, that a member of the MAP kinase family becomes tyrosinephosphorylated in oxidant-treated neutrophils. To characterize further the identity of the 42-44-kDa protein, extracts from resting and oxidant-treated cells were lysed and subjected to immunoprecipitation with a polyclonal anti-MAP kinase antibody. The immunoprecipitates were analyzed by gel electrophoresis and probed with a monoclonal anti-MAP kinase antibody to confirm the effectiveness of the immunoprecipitation (not shown). More importantly, stripping and reprobing the blot with a monoclonal anti-phosphotyrosine antibody demonstrated tyrosine phosphorylation of MAP kinase following treatment of cells with diamide or H,Odaminotriazole (Fig.  10). The similarity of the 42-44-kDa polypeptide and MAP kinase mobility in SDS-PAGE gels in combination with immunoprecipitation and immunoblotting provide strong evidence that the tyrosine-phosphorylated 42-44-kDa polypeptide is, a t least in part, a member of the MAP kinase family. MAP kinases are unique because their activation requires phosphorylation on both tyrosine and threonine residues (Anderson et al., 1990). To determine if, in addition to tyrosine phosphorylation, oxidant treatment of cells induced activation of MAP kinase, an in vitro gel renaturation assay was em- ployed. Treatment of cells with diamide induced a significant increase in the kinase activity of a 42-44-kDa protein in whole cell extracts (Fig. 2 A ) . An even more marked induction of kinase activity was observed in extracts from neutrophils treated with H,OJaminotriazole. The results shown in Fig. 2A also illustrate that, in several experiments, MAP kinase activity was noted to be elevated in cells incubated with aminotriazole alone. Similar observations were noted when examining whole cell tyrosine phosphorylation following treatment with aminotriazole alone (not shown). A plausible explanation for these observations is that inhibition of catalase by treatment of cells with aminotriazole allowed H,O,, produced by constitutive activity of oxidant-generating systems, to attain a concentration sufficient to induce tyrosine phosphorylation of cellular proteins including MAP kinase. Additionally, Fig. 2A illustrates that oxidant treatment of cells lead to increased kinase activity of a polypeptide of M, = 55,000-60,000, suggesting that oxidants can regulate the activity of kinases other than MAP kinase which are also capable of phosphorylating MBP. It is noteworthy that the effects of diamide and H,OJ aminotriazole on the enhancement of MBP-kinase activity appear to be in reverse order of potency when compared to their effect on tyrosine phosphorylation of the 42-44-kDa polypeptide (refer to Fig. lA, left panel) as determined by densitometric analysis. Specifically, from perusal of the studies of whole cell tyrosine phosphorylation (Fig. l A , left panel), it is apparent that diamide was more potent than H,OJaminotriazole in the induction of tyrosine phosphorylation of the 42-44-kDa polypeptide(s). In contrast, Fig. 2A illustrates that treatment with H,OJaminotriazole was more potent than diamide in the ability to increase MBP-kinase activity. There are at least three possible explanations for this observation: (i) as MAP kinase activation requires both tyrosine and threonine phosphorylation, one can presume that oxidants also increase phosphorylation on threonine residues of MAP kinase. Hence, it is possible that diamide and H,OJaminotriazole have different potencies in enhancing threonine phosphorylation on MAP kinase; (ii) these two compounds may have different effects on upstream activators or inhibitors of MAP kinase; (iii) as whole cell extracts were analyzed by the in vitro gel renaturation assay, it is possible that a kinase distinct from (but co-migrating with) MAP kinase could have been activated by H,OJaminotriazole but not by diamide, thus contributing to the increased MBPkinase activity. Confirmation that the observed increase in MBP-kinase activity of the 42-44-kDa protein was due to activation of MAP kinase (and not of other kinases capable of phosphorylating MBP of similar molecular size) was obtained by demonstrating increased MBP-kinase activity in MAP kinase immunoprecipitates of neutrophils treated with oxidants (Fig. 2B). Furthermore, analysis of MAP kinase immunoprecipitates (as compared to whole cell extracts) demonstrated that treatment of cells with H,OJaminotriazole was more potent than diamide in the induction of MBP-kinase activity as determined by densitometric analysis. Taken together, these results provide strong evidence that a member of the MAP kinase family is activated by oxidants.
Recent studies investigating the signaling pathways linking activation of cell surface receptors with downstream MAP kinase activation have disclosed a complex network of kinases. As discussed above, phosphorylation of MAP kinase by a mixed function kinase termed MEK (also called MAP kinase kinase) results in MAP kinase activation (reviewed by Blenis (1993) and Crews and Erikson (1993)). Furthermore, phosphorylation of both tyrosine and threonine residues of MAP kinase appears to be required for its full activation. While MAP kinases can also undergo autophosphorylation on tyrosine residues alone leading to activation, the level of this activity is much less than that observed after combined serine and threonine phosphorylation (Seger et al., 1991;Crews et al., 1991). Additionally, the kinetics of autophosphorylation and autoactivation are much too slow to account for MAP kinase stimulation in vivo (Pelech and Sanghera, 1992). Thus, it is unlikely that autophosphorylation and autoactivation account for the increased MAP kinase activity observed in oxidant-treated cells. Accordingly, we explored the possibility that oxidant treatment of cells activated MEK, the putative upstream activator of MAP kinase. MEK was immunoprecipitated from resting and oxidant-treated neutrophils, and the immunoprecipitates were assayed for their ability to phosphorylate recombinant MAP kinase. It should be noted that a kinase-defective MAP kinase recombinant protein was used as the substrate in these experiments to circumvent autophosphorylation that might complicate interpretation of the results. In resting neutrophils, MEK activity was found to be minimal. However, treatment of cells with diamide or H,OJ aminotriazole induced a marked increase in MEK activity, as judged by increased "P incorporation into recombinant MAP kinase (Fig. 3). To determine the potency of oxidants in MEK activation, we compared the effects of oxidants to those of the soluble activating agent fMLP. In agreement with previous observations (Grinstein et al., 19941, treatment of cells with cytochalasin B followed by fMLP caused a marked increase in MEK activity. The effect of fMLP on MEK activation was slightly greater than that of H,Odaminotriazole or diamide (1.17 and 1.28 times higher, respectively) as determined by densitometric analysis of the autoradiograms.
The results illustrated in Fig. 3 also demonstrate that MEK is capable of autophosphorylation following treatment of cells with either diamide or with H,O~aminotriazole. Three different human MEK cDNAs (MEK1, MEK2, and MEK3) have been identified to date Zheng and Guan, 1993a, 199313). Studies examining the properties of these ME& have demonstrated that only MEKl and -2 can be activated by autophosphorylation. However, as for MAP kinase, phosphorylation of ME& by upstream activators may be required to achieve full activity (Zheng and Guan, 1993b). Hence, it is unlikely that the increased MEK activity (with respect to MAP kinase phosphorylation and activation) observed in oxidanttreated cells is accounted for by MEK autophosphorylation alone; a more plausible explanation is that oxidants increase MEK activity indirectly via activation of certain upstream regulatory elements. The identity and mechanisms of regulation of some of the elements upstream of MEK and MAP kinase have been elucidated recently (reviewed by McCormick (1993)). More specifically, receptor activation initiates a cascade of events involving protein-protein interactions between the adaptor protein Grb2 and the guanine nucleotide exchange factor h-Sosl which leads to activation of the low molecular weight GTPase ras. This leads in turn to activation of raf, MEK, and MAP kinase (McCormick, 1993). Shc, an SH2-containing protooncogene, has been shown to be phosphorylated on tyrosine residues which promotes its interaction with Grb2 (Rozakis-Adcock et al., 1992).
As discussed above, whole cell extracts of oxidant-treated neutrophils demonstrated prominent tyrosine-phosphorylated polypeptides of the approximate molecular mass of Shc isoforms (46, 52, and 66 kDa: Pelicci et al., 1993). Accordingly, we investigated whether oxidant treatment of neutrophils induced tyrosine phosphorylation of Shc and binding to Grb2 which Autoradiogram of an immunoblot analysis of Shc immunoprecipitates using a monoclonal anti-phosphotyrosine antibody. After the indicated treatments, neutrophils were lysed, and immunoprecipitation of Shc was carried out as described under "Experimental Procedures." Aliquots of the immunoprecipitates were used for immunoblotting with an anti-phosphotyrosine antibody. Lane 1, untreated neutrophils; lane 2, cells incubated with 1 mM diamide for 20 min a t 37 "C. To exclude the possibility that tyrosine phosphorylation of p52 Shc in resting cells was due to cell activation during the isolation procedures, cultured HL 60 cells were studied. In experiments not illustrated, Shc was immunoprecipitated from lysates of dimethyl sulfoxidedifferentiated HL 60 cells. Even in these (presumably unactivated) cultured cells, p52 She was found to be phosphorylated on tyrosine. might account for downstream activation of MEK. Lysates from resting and oxidant-treated neutrophils were subjected to immunoprecipitation with Shc antiserum and analyzed by SDS-PAGE and immunoblotting with either anti-Shc antibodies (to confirm the effectiveness of the immunoprecipitation) or with anti-phosphotyrosine antibodies.
These anti-Shc antibodies recognize three isoforms of Shc: 46, 52, and 66 kDa (Pelicci et al., 1993). Immunoblotting of the immunoprecipitates with these same antibodies revealed the presence of a polypeptide of approximately 54 kDa, corresponding to the 52-kDa isoform of Shc (not shown). The observation that this isoform of Shc in hematopoetic cells migrates a t this higher molecular weight has been reported by others (Cutler et al., 1993). Immunoblot analysis with monoclonal anti-phosphotyrosine antibodies revealed detectable tyrosine phosphorylation of the 52-kDa Shc even in resting neutrophils (Fig. 4, lane 1 ). However, treatment of cells with diamide failed to increase the level of tyrosine phosphorylation of Shc (Fig. 4, lane 2 ) . Furthermore, in neither resting nor oxidant-treated cells did Grb2 associate with Shc as determined by immunoblot analysis of the Shc immunoprecipitates with anti-Grb2 antibodies. These results argue against the involvement of Shc in oxidant-induced activation of the MEK-MAP kinase pathway in neutrophils and suggest that alternate upstream elements in this pathway may be activated by changes in the redox state of neutrophils, a possibility that is currently under investigation.
Protein phosphatases such as HVH-1 (Zheng and Guan, 1993c), MKP-1 (Sun et al., 1993), PAC-1 (Ward et al., 1994), and CD45 (Anderson et al., 1990) have been shown to dephosphorylate and inactivate MAP kinase. Importantly, many protein tyrosine phosphatases are dependent on conserved cysteine groups in their first catalytic domain for full catalytic activity (Tonks et al., 1988;Streuli et al., 1989) suggesting that their activity might be negatively regulated by oxidants. As CD45 is a major leukocyte tyrosine phosphatase and is known to have an absolute requirement for the presence of sulfhydryl compounds for activity (presumably for maintaining the conserved cysteine residue in the first catalytic domain in a reduced state :  Tonks et al., 1990), we examined the possibility that oxidant treatment of neutrophils inhibited CD45 phosphatase activity, Neutrophils were exposed to oxidants or buffer control as indicated. Cells were incubated with 1 mM diamide for 20 min a t 37 "C or preincubated with 50 mal aminotriazole for 60 min and subsequently exposed to 1 mM H,O, for 10 min. After treatment, cells were lysed, and CD45 immunoprecipitation was carried out as described under "Experimental Procedures." Tyrosine phosphatase activity of the CD45 immunoprecipitates was then measured using free 0-phosphotyrosine as a substrate, as described under "Experimental Procedures." Data are representative of a t least four experiments done in duplicate. Vertical bars indicate standard error of the mean. Asterisks indicate p < 0.01 with respect to the control after normalization for interexperiment variability, determined by analysis of variance for repeated measures with correction for multiple comparisons (Sheffe). H, N-acetyl-L-cysteine prevents diamideinduced inhibition of CD45. Neutrophils were exposed to oxidants or buffer control as indicated. Cells were incubated with 1 mM diamide for 20 min a t 37 "C. Where indicated, diamide-treated neutrophils were preincubated with 30 m M N-acetyl-1.-cysteine for 15 min. After treatment, cells were lysed and CD45 immunoprecipitation was carried out as described under "Experimental Procedures." Tyrosine phosphatase activity of the CD45 immunoprecipitates was then measured using free 0-phosphotyrosine as a substrate as described under "Experimental Procedures," Data are representative of a t least four experiments done in duplicate. Vertical bars indicate standard error of the mean. Asterisks indicate p < 0.01 with respect to the control after normalization for interexperiment variability, determined by analysis of variance for repeated measures with correction for multiple comparisons (Sheffe). It should be noted that the values for the phosphatase activity in control and oxidant-treated samples were higher than those reported in A. The difference observed reflects the fact that for the experiments reported in R , a new lot of primary antibody and Protein A-Sepharose beads were used for immunoprecipitation. However, the relative inhibition by diamide of CD45 phosphatase activity was similar in the two sets of experiments ( 4 6 4 versus 51% inhibition in A and R, respectively). thus contributing to prolongation of the activation of MAP kinase. CD45 immunoprecipitates from control and oxidanttreated neutrophils were compared for their ability to hydrolyze 0-phosphotyrosine. Treatment of neutrophils with H,OJ aminotriazole ( Fig. 5A) or diamide (Fig. 5, A and B 1 resulted in 41 and 46% inhibition of CD45 phosphatase activity, respectively, when compared with untreated cells ( p < 0.01 by analysis of variance). As described under "Experimental Procedures,'' dithiothreitol was not included in the assay buffer used for these phosphatase assays. This was done to preclude reversal of the effects of prior oxidant treatment (Tonks et al., 1988. Fig. 5B also illustrates that the inhibitory effect of diamide on CD45 phosphatase activity was abrogated by prior exposure of cells to N-acetyl-L-cysteine, suggesting that inhibi- and 30 min a t 37 "C. After treatment, cells were lysed and CD45 immunoprecipitation was carried out as described under "Experimental Procedures." Tyrosine phosphatase activity of the CD45 immunoprecipitates was then measured using free 0-phosphotyrosine as a substrate, as described under "Experimental Procedures." Data are representative of at least four experiments done in duplicate. Vertical bars indicate standard error of the mean. Values for 2, 5, 10,20, and 30 min are significantly different from control as determined by analysis of variance with correction for multiple comparisons. B, time course of the effect of diamide on MBP-kinase activity in whole cell extracts. Neutrophils were incubated with 1 mM diamide or buffer control for the indicated times, and MBP-kinase activity was determined in whole cell extracts using a gel renaturation assay with MBP as the substrate, as described under "Experimental Procedures." C, time course of MBPkinase activity in MAP kinase immunoprecipitates of diamide-treated neutrophils. Neutrophils were incubated with 1 mM diamide or buffer control for the indicated times. After treatment, cells were lysed, and immunoprecipitation of MAP kinase was carried out as described under "Experimental Procedures." Aliquots of the immunoprecipitates were then assayed for MBP-kinase activity by a gel renaturation assay as described under "Experimental Procedures." phils and in neutrophils from a patient with chronic granulomatous disease. Neutrophils from a normal donor and from an X-linked, gp91'h"r-deficient chronic granulomatous disease patient were treated with or without fMLP as indicated. After treatment, MAP kinase antibody and Protein NG-agarose were used to immunoprecipitate MAP kinase. Aliquots of the immunoprecipitates were then assayed for MBPkinase activity by a gel renaturation assay as described under "Experimental Procedures." Lanes I and 2, neutrophils from the chronic granulomatous disease patient; lanes 3 and 4, neutrophils from a normal donor; lanes I and 3, untreated neutrophils; lunes 2 and 4, cells preincubated with 5 p~ cytochalasin B for 2 min and subsequently stimulated with M fMLP for 2.5 min. The increased MBP-kinase activity of a 42-44-kDa protein is indicated. B , fMLP does not inhibit CD45 activity. Neutrophils were preincubated with 5 p~ cytochalasin B for 2 min and subsequently stimulated with M fMLP for the indicated times. Control indicates untreated cells. After treatment, cells were lysed and CD45 immunoprecipitation was carried out as described under "Experimental Procedures." Tyrosine phosphatase activity of the CD45 immunoprecipitates was then measured using free O-phosphotyrosine as a substrate as described under "Experimental Procedures." Data are representative of al least four experiments done in duplicate. Vertical burs indicate standard error of the mean. Incubation of neutrophils with fMLP for shorter times (1, 2, and 3 min) also did not have a significant inhibitory effect on CD45 activity (not shown).
tion of phosphatase activity was in fact due to the oxidant properties of these compounds.
To evaluate more completely the relationship between oxidant-induced inhibition of CD45 and activation of MAP kinase observed under these oxidizing conditions, we compared the kinetics of inhibition of CD45 with the kinetics of increased tyrosine phosphorylation and activation of MAP kinase. For these studies, diamide was used as the oxidizing agent. Fig. 6A illustrates that diamide-induced inhibition of CD45 was maximal between 10 and 20 min (65% and 64% inhibition, respectively). By way of comparison, diamide-induced tyrosine phosphorylation of MAP kinase, as determined by immunoprecipitation followed by immunoblotting with anti-phosphotyrosine antibodies, was maximal a t 20 min. Densitometric analysis demonstrated that the absorbance of the 42-kDa band was 0.03 in control cells compared with 0.12,0.58, and 0.38 after lo-, 20-, and 30-min diamide treatment, respectively (these data are from one of two experiments with similar results). In additional experiments, the time course of diamide-induced increase in MBP-kinase activity was analyzed in both whole cells extracts and MAP kinase immunoprecipitates using a gel renaturation assay with MBP as the substrate, as described under "Experimental Procedures.'' Fig. 6B illustrates that an increase in MBP-kinase activity in whole cell extracts was detectable between 2 and 5 min reaching a maximum between 20 and 30 min. Similarly, analysis of MBP-kinase activity in MAP kinase immunoprecipitates from oxidant-treated neutrophils demonstrated a maximal increase a t 20 min that paralleled tyrosine phosphorylation (Fig. 6C). In summary, the kinetics of diamide-induced tyrosine phosphorylation and activation of MAP kinase correlated with the kinetics of inhibition of CD45, suggesting that inhibition of the phosphatase might contribute to the prolongation of kinase activation. However, other phosphatases such as HVH-1 (Zheng and Guan, 1993~) and MKP-1 (Sun et al., 19931, which were not measured in this study, might also be inhibited by oxidant treatment and thus contribute to the prolonged activation of MAP kinase. Taken together, these results support the notion that the activity of protein tyrosine phosphatases such as CD45 can be negatively regulated by oxidants, as has been suggested by other investigators (Monteiro et al., 1991;Heffetz et al., 1990;Hecht and Zick, 1992) and that, in addition to activation of MEK, activation of MAP kinase by oxidants may be facilitated by inhibition of cysteine-containing protein tyrosine phosphatases such as CD45.
To this juncture, we have studied the effects of exogenous oxidizing agents such as diamide and H,O, on the MEK-MAP kinase pathway. I t was of interest to determine if production of endogenous oxidants could similarly modulate this pathway. Agents that activate neutrophils via surface receptors such as N-formyl-methionyl-leucyl phenylalanine (fMLP) result in increased activity of the NADPH oxidase and enhanced tyrosine phosphorylation of many polypeptides (Huang et al., 1988;Berkow and Dodson, 1990) including MAP kinase (Grinstein and Furuya, 1992;Torres et al., 1993). It is plausible that the effect of fMLP on MAP kinase activation might be, at least in part, dependent on endogenously produced ROI. To characterize more completely the role of endogenous ROI on MAP kinase activation, we compared the effects of fMLP in cells from a normal donor with those from a patient with chronic granulomatous disease who is known to be deficient in the production of ROI. Fig. 7A illustrates that fMLP-induced activation of MAP kinase was similar in chronic granulomatous disease neutrophils a s compared to neutrophils from a normal donor. These results suggest that although MAP kinase activation can be induced by direct exposure of cells to oxidizing conditions, receptor-mediated stimuli such as fMLP can activate MAP kinase via a mechanism that is independent of the production of ROI by the NADPH oxidase.
To determine if the activity of CD45 could be regulated by receptor-mediated stimuli, we studied the effects of treatment with fMLP. Fig. 7B illustrates that treatment of neutrophils with fMLP for up to 10 min did not alter phosphatase activity in CD45 immunoprecipitates. Thus, although CD45 can be inhibited by exposure of cells to oxidizing conditions, production of endogenous ROI by receptor-mediated stimuli such as fMLP does not appear to influence the activity of CD45. However, other protein phosphatases such as MKP-1 that may regulate MAP kinase activity could potentially be regulated by chemoat-tractants, a possibility that is currently under investigation.
In summary, the experiments described herein provide evidence that MAP kinase can be phosphorylated and activated in response to alterations of the redox state of cells induced by treatment with oxidants. These effects appear to be the result of a combination of activation of MEK and inhibition of cysteine-containing protein tyrosine phosphatases such as CD45. The latter effect would inhibit dephosphorylation of MAP kinase and contribute to prolongation of MAP kinase activation. Our data also suggest that while alterations in the redox state of the cell by addition of exogenous oxidizing agents are sufficient to activate the MEK-MAP kinase pathway, receptormediated activation of this pathway appears to proceed independently of the production of endogenous ROI. However, in the context of the milieu of an acute inflammatory response, antioxidant defenses may be overwhelmed which could create an environment that may allow endogenous ROI to influence this signaling pathway. Furthermore, we speculate that since ROI are released extracellularly and can pass across plasma membranes (Halliwell and Gutteridge, 1985) these oxidants have the potential to influence intracellular pathways in vicinal cells such as other leukocytes, fibroblasts, and endothelial and epithelial cells.