Tyrosine Phosphatase Inhibition by Nitric Oxide

Home , MAPK phosphatase, Nitroarginine

Rapid and Delayed p42/p44 Mitogen-Activated Protein Kinase Activation by Nitric Oxide: The Role of Cyclic GMP and Tyrosine Phosphatase Inhibition1

Dagmar Callsen,* Josef Pfeilschifter,† and Bernhard Bru¨ne2*

The exposure of rat mesangial cells to cytokines promoted activation of the p42/p44 mitogen-activated protein kinase (MAPK). We identiﬁed a rapid and delayed phase of MAPK activation with distinctive activity increases at 5 to 15 min and 15 to 24 h. Rapid and late MAPK activation were attenuated by the redox-modulating agent N-acetylcysteine. Speciﬁcally, late-phase activation coincided with endogenous nitric oxide (NO) generation and in turn was suppressed by the NO synthase-blocking compounds diphenyliodonium or nitroarginine methyl ester. By using NO-liberating agents such as S-nitrosoglutathione and 3-morpholinosydnonimine, we investigated intermediary signaling elements of NO in promoting MAPK activation. Early and transient activation at 5 min was suppressed by the soluble guanylyl cyclase-blocking agent 1H-(1,2,4)-oxdiazolo-(4,3-␣)-6-bromoquinoxazin- 1-one (NS 2028) and, moreover, was mimicked by the lipophilic cyclic GMP (cGMP) analogue 8-bromo-cGMP. In contrast, NO-mediated activation achieved within hours was unrelated to cGMP signaling. Late and persistent MAPK activation, induced by NO donors or endogenously generated NO, was found in association with inhibition of phosphatase activity. In vitro dephosphorylation of activated and immunoprecipitated p42/p44 by cytosolic phosphatases was sensitive to the readdition of NO and was found to be inhibited in cytosol of S-nitrosoglutathione-stimulated cells. Also, cells that had been exposed to cytokines for 24 h revealed a blocked phosphatase activity, which was successfully attenuated by the NO synthase inhibitor nitroarginine methyl ester and, therefore, was NO mediated. Conclusively, NO affects p42/p44 MAPK in rat mesangial cells twofold: rapid activation is cGMP mediated, whereas late activation is transmitted via inhibition of tyrosine dephosphorylation. The Journal of Immu- nology, 1998, 161: 4852–4858.

esangial cells (MC)3 are specialized smooth muscle characterized as an antiproliferative agent in some systems (8, 9), cells located in the glomeruli of the kidney (1). MC but emerged as an upstream signal of cell proliferation in endo- M take part in the pathogenesis of severe inflammatory thelial cells, keratinocytes, and fibroblasts (10–12). conditions, i.e., glomerulonephritis (2, 3). Cellular proliferation of Cell proliferation during glomerulonephritis may be augmented MC is thought to be pivotal for the development of crescent for- by endogenously produced NO by activating p42/p44 mitogen- mation and progression of the proliferative glomerulonephritis (4). activated protein kinases (MAPK), which are established prolifer- Following cell activation, in part as a result of cytokine release by ative signal components (4, 13). p42 MAPK (also known as ex- infiltrating macrophages, MC respond with proinflammatory me- tracellular-regulated kinase-2 or ERK-2) and the p44 MAPK diator secretion and inducible nitric oxide synthase (iNOS) up- (ERK-1) are activated as a result of a cascade of different upstream regulation (5, 6). Nitric oxide (NO) is recognized for its partici- kinases (14, 15). Mitogen-activated serine/threonine kinases are pation in diverse biologic processes, ranging from physiologic to unusual in requiring phosphorylation on both tyrosine and threo- pathophysiologic/pathologic biologic processes, including inflam- nine residues within the signature sequence T-E-Y (threonine-183 matory and proliferative/antiproliferative responses (2, 3, 7). The and tyrosine-185) (16). Phosphorylation of the T-E-Y motif in role of NO in promoting cell proliferation is controversial. NO was MAPK is catalyzed by a dual-specificity kinase termed MAPK kinase (MAPKK, also known as MEK). MEK is itself activated by phosphorylation on either of two serine residues by a MEK kinase *University of Erlangen-Nu¨rnberg, Faculty of Medicine, Department of Medicine IV, (MEKK) (17, 18). Experimental Division, Erlangen, Germany; and †Zentrum der Pharmakologie, Klini- Activation of MAPK is transient and highly regulated by the kum der Johann Wolfgang Goethe-Universita¨t, Frankfurt am Main, Germany antagonizing or cooperating action of upstream kinases and phos- Received for publication January 20, 1998. Accepted for publication June 24, 1998. phatases (19, 20). MAPK phosphatases are classified into two sub- The costs of publication of this article were defrayed in part by the payment of page families according to their substrate specificity. There are serine/ charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. threonine phosphatases and tyrosine phosphatases, which are 1 This work was supported by the Deutsche Forschungsgemeinschaft and in part by further subgrouped into VH-1-like dual-specificity phosphatases the European Community. and tyrosine-specific phosphatases (20, 21). Dephosphorylation of 2 Address correspondence and reprint requests to Dr. Bernhard Bru¨ne, University of p42/p44 MAPK in vitro and in vivo is achieved by the cytosolic Erlangen-Nu¨rnberg, Faculty of Medicine, Loschgestrasse 8, 91054 Erlangen, Ger- serine/threonine phosphatase PP2A, various inducible dual-spe- many. E-mail address: [email protected] cific phosphatases such as MKP-1/2 or PAC, and thus far unchar- 3 Abbreviations used in this paper: MC, mesangial cells; cGMP, cyclic GMP; NO, acterized tyrosine phosphatases (21–24). Common to all dual-spe- nitric oxide; IP, immunoprecipitate; MBP, myelin basic protein; CM, cytokine mix; G NAME, N -nitro-L-arginine methyl ester; NAC, N-acetylcysteine; GSNO, S-nitroso- cific and tyrosine-specific phosphatases is an active site motif glutathione; SIN-1, 3-morpholinosydnonimine; NS 2028, 1H-(1,2,4)-oxdiazolo-(4,3- consisting of a cysteine and an arginine separated by five residues ␣)-6-bromoquinoxazin-1-one; Br-cGMP, 8-bromo-cyclic GMP; DPI, diphenyliodonium; MAPK, mitogen-activated protein kinase; MEK, MAPK kinase; iNOS, (commonly known as the Cx5R motif) (25). Mutation of the cat- inducible nitric oxide synthase. alytically active cysteine abrogated all phosphatase activity, which

can be rationalized by the requirement of a phosphocysteine in- SDS, 10% glycerin, 1 mM DTT, 0.002% bromphenol blue, pH 6.9) and termediate during catalysis (26, 27). boiled for 5 min. Proteins were resolved on 10% SDS-polyacrylamide gels With the notion that histidine-activated thiols are preferentially and blotted onto nitrocellulose sheets. The m.w. of corresponding proteins was determined in relation to m.w. rainbow markers. Transblots were targeted by NO redox species (28), we proposed that tyrosine washed twice with TBS (140 mM NaCl, 50 mM Tris-HCl, pH 7.2) con- phosphatases, which are involved in the regulation of the p42/p44 taining 0.06% Tween 20 before blocking unspecific binding with TBS/5% MAPK cascade, are NO sensitive. Besides up-regulation of cyto- skim milk. Phosphospecific p44/42 MAPK Abs, dissolved in TBS/5% skim ␮ kine-inducible iNOS, which generates NO by converting L-argi- milk at a final concentration of 0.1 g/ml, were incubated overnight at 4°C. Blots were washed five times before the anti-rabbit peroxidase-labeled sec- nine to NO and citrulline (29, 30), a structurally diverse class of ondary polyclonal Ab was added (1 h, diluted 1:10,000 in TBS/0.5% milk). NO-releasing compounds such as S-nitrosoglutathione (GSNO) or Blots were washed five times and followed by enhanced chemilumines- 3-morpholinosydnonimine (SIN-1) (31) was used to study NO- cence detection. Afterwards, blots were stripped (62.5 mM Tris, 2% SDS, signaling pathways in rat MC. 95 mM ␤-mercaptoethanol, pH 6.7) for 30 min at 55°C and then blocked In the present study, we demonstrate that the inflammatory me- for 1 h with TBS/5% skim milk. In turn, the anti-MAPK rabbit polyclonal Ab, dissolved in TBS/5% skim milk at a final concentration of 0.1 ␮g/ml, diator NO provoked rapid and delayed p42/p44 MAPK activation. was incubated for 1 h before enhanced chemiluminescence detection. Whereas rapid activation was cyclic GMP (cGMP)-mediated, delayed activation resulted from NO-evoked inhibition of tyrosine MAPK assay phosphatases. Dishes (10 cm) of confluent, quiescent cells were incubated for the times indicated. After removing the medium, cells were washed once with ice- Materials and Methods cold PBS (supplemented with 1 mM sodium vanadate) and scraped into 400 ␮l of lysis buffer (20 mM Tris-HCl, pH 7.5, 1% Triton X-100, 10% Materials glycerol, 137 mM NaCl, 2 mM EDTA, 25 mM ␤-glycerophosphate, 2 mM RPMI 1640 and medium supplements were ordered from Biochrom (Ber- sodium pyrophosphate, and 1 mM sodium vanadate). Cells were lysed for lin, Germany). FCS was purchased from Life Technologies (Karlsruhe, 5 min on ice, vortexed, and centrifuged (10,000 ϫ g, 20 min, 2°C). Protein Germany). Insulin, ␤-glycerophosphate, 1,1,2-trichlorotrifluoroethane, tri- content of the supernatant was measured and 1 mg of protein in an equal- N-octylamine, trichloroacetic acid, sodium pyrophosphate, sulfanilamide, ized sample volume was used for immunoprecipitation. Anti-p42/p44 myelin basic protein (MBP), protein A-Sepharose, N-naphthylethylenedia- MAPK Abs (35) were added to the lysate, followed by gentle rotation at mine, reduced glutathione, and LPS came from Sigma (Deisenhofen, Ger- 4°C. After 2 h, protein A-Sepharose was added, and the immunocomplexes many). IL-1␤, IFN-␥, sodium vanadate, and sodium fluoride were bought were incubated for another 60 min. p42/p44 MAPK/protein A-Sepharose from Boehringer Mannheim (Mannheim, Germany). PMSF was ordered complexes were washed twice with lysis buffer and once with kinase buffer G ␤ from Fluka (Deisenhofen, Germany). N -nitro-L-arginine methyl ester (25 mM HEPES, pH 7.4, 25 mM -glycerophosphate, 25 mM MgCl2,1 (NAME) was delivered by Alexis (Gru¨nberg, Germany). Anti-MAPK rab- mM DTT, and 0.1 mM sodium vanadate). Immunocomplexes were spun bit polyclonal Abs for Western blot analysis were purchased from IC down (10,000 ϫ g, 2 min, 4°C) and resuspended in 25 ␮l kinase buffer with 32 Chemikalien (Ismaning, Germany). The phosphospecific p44/42 MAPK the addition of 20 ␮g MBP and 5 ␮Ci [␥- P]ATP. Phosphorylation was Ab came from New England Biolabs (Schwalbach/Taunus, Germany), performed for 20 min at 37°C. The reaction was stopped by adding 25 ␮l while the secondary Ab was purchased from Promega/Serva (Heidelberg, 2ϫ SDS-sample buffer. Samples were heated for 5 min at 95°C and sep- Germany). Enhanced chemiluminescence detection reagents and arated on 15% SDS-polyacrylamide gels. Gels were fixed, dried, and sub- [␥-32P]ATP were from Amersham (Braunschweig, Germany). SIN-1 and jected to autoradiography. ␣ TNF- were provided by Hoechst Marion Roussel (Frankfurt, Germany) Dephosphorylation assay and Bayer (Leverkusen, Germany). PD 98059 came from Calbiochem (Bad Soden, Germany). 1H-(1,2,4)-oxdiazolo-(4,3-␣)-6-bromoquinoxazin-1- Dishes (10 cm) of confluent, quiescent cells were treated with 20% FCS for one (NS 2028) was graciously supplied by Prof. Dr. R. Busse (University 5 min. After removal of the medium, cells were washed with ice-cold PBS of Frankfurt, Frankfurt, Germany). All other chemicals were of the highest (supplemented with 1 mM sodium vanadate, 1 mM sodium fluoride, and 1 grade of purity and commercially available. mM PMSF), scraped into lysis buffer (as outlined for the MAPK assay), kept for 10 min on ice, vortexed, and centrifuged (10,000 ϫ g, 20 min, MC culture 4°C). The supernatant was used for immunoprecipitation of p42/p44 Rat MC were cultured, cloned, and characterized as described previously MAPK. Pelleted p42/p44 MAPK immunocomplexes were washed twice (32). Cells were grown in RPMI 1640 supplemented with 10% FCS, pen- with lysis buffer and twice with hypotonic buffer (20 mM Tris, 10 mM icillin (100 U/ml), streptomycin (100 ␮g/ml), and bovine insulin at 5 ␮g/ NaCl, 1 mM PMSF, pH 7.5). Immunocomplexes were centrifuged ϫ ml. One day before and during the experiments, cells were kept in medium (10,000 g, 2 min, 4°C) and resuspended in hypotonic buffer. For indi- without FCS. For experiments, passages 10 to 25 of MC were used. vidual dephosphorylation experiments, the resuspended p42/p44 MAPK immunocomplexes (immunoprecipitate (IP)) were divided into equal ali- GSNO synthesis quots, which then were incubated with 500 ␮g of cytosolic cell extracts of untreated cells (controls) with or without the further addition of 1 mM GSNO was synthesized and characterized as described previously (33). GSNO and/or 10 ␮M PD 98059 during an incubation period of 30 min at Nitrite determination 37°C. Alternatively, equal aliquots of IP were incubated with 500 ␮gof cytosolic protein derived from untreated, GSNO (1 mM, 8 h)-, or cytokine Nitrite, a stable NO oxidation product, was determined by the Griess mix (CM) (24-h exposure period with 25 U/ml IL-1␤, 100 U/ml IFN-␥,10 reaction. Cell-free culture supernatants were collected (200 ␮l) and mixed ng/ml TNF-␣, and 25 ng/ml LPS, in the presence or absence of 1 mM with 20 ␮l of sulfanilamide (dissolved in 1.2 M HCl)/20 ␮lofN- NAME)-exposed cells. For the preparation of cytosol, cells were scraped naphthylethylenediamine dihydrochloride. After 5 min at room tempera- into ice-cold PBS (supplemented with 1 mM PMSF), centrifuged ture, the absorbance was measured at 560 nm with a reference wavelength (10,000 ϫ g, 10 min, 4°C), resuspended in hypotonic buffer (supplemented at 690 nm. Nitrite concentrations were calculated using a NaNO2 standard. with 1 mM PMSF), sonicated, and centrifuged (10,000 ϫ g, 15 min, 4°C). For details, see Ref. 34. Dephosphorylation was terminated by centrifugation (1000 ϫ g, 5 min, 4°C). IP were washed once with hypotonic buffer (supplemented with 5 Western blot analysis mM sodium vanadate, 2 mM sodium fluoride, and 1 mM PMSF) and with Confluent and quiescent cells were incubated for the times indicated. After kinase buffer. MAPK activation was determined as described for the removing the medium, cells were scraped into ice-cold PBS (supplemented MAPK assay. with 1 mM sodium vanadate, 1 mM sodium fluoride, and 1 mM PMSF), Determination of cGMP centrifuged (700 ϫ g, 10 min, 4°C), and resuspended in 200 ␮l of ice-cold lysis buffer (50 mM Tris, 5 mM EDTA, 150 mM NaCl, 0.5% Nonidet P-40, Confluent, quiescent cells were incubated for the times indicated. After 1 mM PMSF, 1 mM sodium vanadate, and 1 mM sodium fluoride, pH 8.0). removal of the medium, cells were lysed by the addition of 1 ml ice-cold Cell debris was sonicated with a Branson sonifier (10 s, duty cycle 100%, 10% TCA. Lysates were kept for 10 min at 4°C and centrifuged (10,000 ϫ output control 1), centrifuged (14,000 ϫ g, 15 min), and measured for g, 15 min, 4°C). The supernatant was neutralized with 750 ␮l of a 1:1 (v/v) protein content. For equal protein loading, 150 ␮g of protein was mixed mixture of tri-N-octylamine and 1,1,2-trichlorotrifluoroethane while vor- with the same volume of 2ϫ SDS-sample buffer (125 mM Tris-HCl, 2% texing. Centrifugation for 1 min at 10,000 ϫ g generated three separate 4854 NO ACTIVATES p42/p44 MAPK

FIGURE 2. NAC, DPI, and NAME suppressed p42/p44 MAPK activation. Rat MC were activated by CM for 24 h in the presence or absence of 10 ␮M DPI and/or 1 mM NAME. 30 mM NAC was added for the last 30 of the incubation period. Phosphorylation and expression of the (ء) min p42/p44 MAPK were detected by Western blot analysis. Nitrite accumulation in the cell supernatant was determined by the Griess reaction as outlined in Materials and Methods. Blots are representative of at least three similar experiments. Further details are as in Figure 1.

stable NO oxidation product, significantly accumulated after 15 to FIGURE 1. Cytokine-elicited biphasic p42/p44 MAPK activation. For 24 h following CM addition (Fig. 1, A and B). In extending ex- the times indicated, confluent and quiescent rat MC were stimulated with periments, we reestablished the concurrence between MAPK phos- CM (25 U/ml IL-1␤, 100 U/ml IFN-␥, 10 ng/ml TNF-␣, 25 ng/ml LPS). phorylation and its kinase activity. Activity determination is based 32 Phosphorylation/activation of the p42/p44 MAPK was detected by Western on phosphate incorporation from radiolabeled [␥- P]ATP into blot analysis using a phosphospecific p44/42 MAPK Ab (A)orbyara- MBP. MC were time-dependently exposed to the CM followed by dioactive MAPK phosphorylation assay with the substrate MBP as de- radioactive activity determination. A first and transient activation scribed in Materials and Methods (B). The amount of p42/p44 MAPK was appeared 5 to 15 min after CM application. A second activation determined by Western blot analysis using an anti-MAPK Ab (A). Nitrite phase was noticed after 24 h in accordance with iNOS up-regula- accumulation in the cell supernatant was measured by the Griess reaction tion as measured by nitrite accumulation in the cell supernatant as described in Materials and Methods (A and B). Data are representative (Fig. 1B). As outlined in the manufacturer description, we ob- of three similar experiments. served MBP to be separated on SDS-gels into two distinct m.w. fractions. phases. Water-soluble components were recovered from the upper, neu- Delayed MAPK activation, seen at 24 h after CM exposure, was tralized phase. Determination of cGMP was by RIA as described (36). attenuated by the addition of 10 ␮M diphenyliodonium (DPI), an inhibitor of O Ϫ-producing NAD(P)H-like oxidases (38), or by Statistical analysis 2 including N-acetylcysteine (NAC) for the last 30 min of CM ex- Each experiment was performed at least three times. posure (Fig. 2). Further, the second but not the first phase (data not shown for Results the first phase) of MAPK activation was compromised by NAME, Rapid and delayed p42/p44 MAPK activation by cytokine a specific NO synthase inhibitor. As expected, nitrite formation stimulation was additionally attenuated by NAME treatment. Conclusively, Incubating rat MC with CM (25 U/ml IL-1␤, 100 U/ml IFN-␥,10 delayed activation of MAPK in MC following CM addition coin- ng/ml TNF-␣, and 25 ng/ml LPS) resulted in p42/p44 MAPK ac- cided in time with endogenous NO formation and was compro- tivation as determined by Western blot analysis. MAPK activation mised by NO synthase inhibitors. The redox-modulating agent was detected by visualizing the tyrosine phosphorylated form of NAC reduced rapid as well as delayed MAPK activation. Albeit p42/p44 MAPK (Fig. 1A). The phosphospecific Ab mainly de- modulation of MAPK activity, the amount of MAPK remained tected p42 and to a lesser extent p44. Cytokines elicited a biphasic invariable (Fig. 2). response, showing rapid and delayed kinase activation. Immediate MAPK phosphorylation occurred within 5 to 15 min and vanished Rapid and delayed activation of p42/p44 MAPK by NO donors at 30 to 60 min following CM addition. Delayed kinase activation NO-releasing compounds such as GSNO or SIN-1 were employed was observed after 15 to 24 h. Active MAPK was absent in con- to study MAPK activation in rat MC. GSNO elicited a biphasic trols and kinase activation was blocked by the MEK-1-specific MAPK activation (Fig. 3A). Following GSNO addition, we ob- inhibitor PD 98059 (data not shown) (37). served a rapid and transient MAPK phosphorylation, with a max- To rule out variations in the amount of MAPK to account for imum at 5 min, that was suppressed by the addition of the specific activity alterations, we probed for the protein amount of p42/p44 inhibitor PD 98059 (data not shown). After 5 min, phosphorylation following CM addition. Within the time frame of the experiments disappeared to basal values and finally reappeared after 4 h. Fol- and especially at longer exposure periods, the amount of MAPK lowing that time point, MAPK stayed in its activated, i.e., phos- remained constant (Fig. 1A). Further, we correlated MAPK and phorylated, form for up to 15 h. iNOS activation in rat MC. Up-regulation of iNOS in response to Direct kinase activity of p42/p44 MAPK as determined by MBP cytokines is known and was corroborated in these experiments by phosphorylation delivered identical results in response to GSNO measuring nitrite accumulation in the cell supernatant. Nitrite, a (Fig. 3A) and assured the interchange of data obtained with either The Journal of Immunology 4855

FIGURE 3. NO-donor-induced biphasic p42/p44 MAPK activation. Rat MC were stimulated with 1 mM GSNO (A) or 1 mM SIN-1 (B) for times indicated. Phosphorylation/activation of p42/p44 MAPK was detected by Western blot analysis with a phosphospeciﬁc p44/42 MAPK Ab (A and B) or by the radioactive MAPK phosphorylation assay (A) with the substrate MBP as described in Materials and Methods. Protein amount of p42/p44 MAPK was assessed by Western blot analysis using an anti-MAPK Ab (B). Results are representative of three similar experiments.

of the assay systems. Further controls revealed that NAME left GSNO-mediated MAPK activation unaltered (data not shown). To establish MAPK activation by NO donors irrespective of the chemical structure of NO-delivering agents, we tested kinase activation by SIN-1 (Fig. 3B). SIN-1 shared with GSNO the ability to evoke a biphasic MAPK activity modulation. Rapid activation was established between 5 to 15 min, whereas delayed activation occurred between 4 and 24 h. Intermediate time points revealed no kinase activity. For control reasons, we again probed for variations in the amount of MAPK during NO exposure (Fig. 3B). Evidently, neither rapid nor delayed MAPK activation was associated with FIGURE 4. MAPK phosphorylation in relation to the cGMP-signaling fluctuations in the amount of protein. Importantly, increased en- cascade. Rat MC were exposed to 1 mM Br-cGMP for indicated times (A). zyme activity at 4 to 24 h was unrelated to de novo kinase protein Phosphorylation/activation of p42/p44 MAPK were determined by a phos- synthesis. phospecific Ab and by a MAPK phosphorylation assay as outlined in Ma- terials and Methods (A). NAC (30 mM) was incubated for 30 min before Some additional experiments again addressed the relation be- B). NS 2028 (5 ␮M) was)(ء) mM Br-cGMP was added for the last 5 min 1 tween p42/p44 phosphorylation and redox alterations. As a result, incubated for 30 min before 1 mM GSNO was added for the last 5 min (C) it became clear that the rapid and delayed GSNO-mediated MAPK or before CM (specified in Fig. 1) was added for the last 15 min (D). NS activation was largely abrogated by NAC (data not shown). NAC 2028 (5 ␮M) was continuously present together with 1 mM GSNO during by itself showed no kinase activation. Basal MAPK activation, an 8-h incubation period (C). C and D mark unstimulated and DMSO- which sometimes appeared, was also NAC sensitive. treated controls. The MEK-1 inhibitor PD 98059 (20 ␮M) was continuously present with 1 mM GSNO for an 8-h incubation period or was in- Transducing pathways during NO-mediated MAPK activation cubated simultaneously with the CM for 24h (E). Phosphorylation of Experiments were designed to identify NO-associated signaling MAPK was measured by using a phosphospecific Ab as described in Ma- terials and Methods. Results are representative of four similar experiments. pathways that transmitted rapid and delayed MAPK activation. Based on the knowledge that soluble guanylyl cyclase is a primer target for NO, we intended to mimic MAPK activation by expos- activation within 5 to 15 min. This was assured by Western blot ing cells to lipophilic cGMP analogues. Exposing rat MC for vari- analysis and the radioactive MAPK assay (Fig. 4A). With pro- able times to 1 mM 8-bromo-cGMP (Br-cGMP) revealed MAPK longed incubation, MAPK activation declined to basal values with 4856 NO ACTIVATES p42/p44 MAPK no indication of a second activation phase. Rapid MAPK phosphorylation, initiated by lipophilic cGMP derivatives within 5 min, was largely compromised by preincubating cells with NAC (Fig. 4B). We noticed a strongly decreased Br-cGMP-mediated MAPK phosphorylation response when cells were incubated beforehand for 30 min with 30 mM NAC. In extending experiments, we employed NS 2028, which is a recognized soluble guanylyl cyclase inhibitor (39). NS 2028 by itself did not affect MAPK, but it largely attenuated rapid MAPK activation in response to GSNO (Fig. 4C). In contrast, late GSNO- mediated MAPK activation observed after 8 h (Fig. 4C) was unaffected by NS 2028. Our results imply that rapid and delayed NO-mediated MAPK activation are differently regulated. Whereas the first activation period occurred in close association with soluble guanylyl cyclase activation and cGMP formation, the second activation phase seemed unrelated to cGMP signaling. To further support a role of cGMP during the rapid phase of NO-mediated MAPK activation, we determined cGMP levels by RIA in MC in response to GSNO. Stimulation with 1 mM GSNO for 2.5 min promoted a cGMP increase from a basal intracellular cGMP level of 10 Ϯ 3 pmol/ml to stimulated values of 1047 Ϯ 148 pmol/ml. At a 5-min incubation period with 1 mM GSNO, cGMP FIGURE 5. Inhibition of MAPK dephosphorylation by NO. Confluent, Ϯ further increased to 1588 32 pmol/ml. Neither basal nor stim- quiescent rat MC were treated with FCS (20%) for 5 min. Activated p42/ ulated cGMP responses were affected by the redox-modulating p44 MAPK was immunoprecipitated (IP p42/p44) as described in Mate- agent NAC (data not shown). rials and Methods. IP aliquots were divided into equal aliquots that were Further analysis investigated the mechanism of the CM-medi- then incubated for 30 min at 37°C with 500 ␮g of cytosolic protein from ated rapid MAPK activation. Compared with the rapid NO-in- control cells (cytosol) with or without the further addition of 1 mM GSNO duced MAPK activation, CM-stimulated MAPK activation was and/or 10 ␮M PD 98059 (A). IP of FCS (20%, 5 min)-treated quiescent MC not affected by the soluble guanylyl cyclase inhibitor NS 2028 (IP p42/p44) were divided into equal aliquots and then incubated under (Fig. 4D) and so is unrelated to cGMP signaling pathways. NO- conditions as outlined (A) without any addition or with cytosol of untreated induced as well as CM-mediated MAPK activation was abolished (controls), GSNO (1 mM, 8 h)-, or CM-stimulated cells. Cytokines were used as indicated in Figure 1, in the absence (CM) or presence (CM/ by the MEK-1 inhibitor PD 98059 (Fig. 4E). NAME) of 1 mM NAME, exposed for 24 h. Phosphorylation/dephosphor- To gain mechanistic insights into the second delayed MAPK ylation of MAPK was then determined in a radioactive MAPK assay using activation period, we performed in vitro MAPK dephosphorylation the substrate MBP as outlined in Materials and Methods. Data are repre- assays. For these experiments, phosphorylation of p42/p44 MAPK sentative of three similar experiments. was initiated in serum-starved rat MC by a 5-min readdition of serum, which is known to activate p42/p44 MAPK (40). Phos- phorylated, i.e., activated, p42 and p44 MAPK was immunopre- Discussion cipitated (p42/p44 IP) and served as the substrate in the following dephosphorylation assays (Fig. 5). Following IP dephosphoryla- Severe inflammatory conditions in the kidney mesangium, i.e., tion, the remaining MAPK activity was then measured by its abil- glomerulonephritis, are often associated with NO formation and ity to cause MBP phosphorylation. Dephosphorylation of the sub- increases in number of MC (2, 3). Consequently, a major focus strate, i.e., phosphorylated p42/p44, was achieved during underlying the pathogenesis of glomerulonephritis has been to de- incubations with a cytosolic fraction of untreated, GSNO-, CM-, fine mechanisms of MC proliferation. Presently, the role of NO in and/or NAME-exposed cells that contained active or blocked ty- modulating cell proliferation is discussed controversially. Al- rosine phosphatases. though NO acts as an antiproliferative agent under some condi- Cytosol of unstimulated cells promoted a rapid and near com- tions (8, 9), NO has been established as a downstream signal in plete dephosphorylation (deactivation) of the p42/p44 MAPK sub- vascular endothelial growth factor-induced endothelial cell prolif- strate (Fig. 5A). Readdition of the NO donor GSNO to the cytosol eration (10). The ability of NO to activate the p42/p44 MAPK prevented dephosphorylation of the substrate, thus pointing to in- pathway in cytokine-activated rat MC may serve as an explanation hibition of tyrosine phosphatases by NO. MAPK dephosphoryla- for cell proliferation during glomerulonephritis and is achieved tion and/or inhibition of dephosphorylation was unaffected by the through distinct signaling pathways. MEK-1 inhibitor PD 98059. This ruled out the involvement of MAPK phosphorylation under the specified experimental condi- Cytokine-mediated MAPK activation tions. Cytosol derived from cells that were pretreated with 1 mM The present study attributed special attention to cytokine-mediated GSNO for8horcytosol from cells that had been stimulated with NO generation and its signal-conveying properties leading to al- CM for 24 h revealed no dephosphorylation activity (Fig. 5B). terations in MAPK activity. Although primer effects of cytokines However, phosphatase activitiy was apparent under conditions of such as IL-1␤, IFN-␥, and TNF-␣ are receptor-mediated (41, 42), costimulation with CM and the iNOS-blocking agent NAME for cellular responses integrate activating and antagonistic signals and 24 h. MAPK dephosphorylation and/or inhibition of dephosphor- combine rapid as well as delayed alterations in the formation of ylation in this asssay was unaffected by the MEK-1 inhibitor PD diffusible second messengers and/or modulations in gene expres- 98059 (data not shown). Evidently, exogenously as well as endo- sion (6, 43). A major signal component in rat MC that is produced genously generated NO in rat MC caused phosphatase inhibition, in response to cytokines is NO, which is found in association with thus promoting p42/p44 activation. iNOS up-regulation (3). The Journal of Immunology 4857

In corroboration with earlier reports (44, 45), we noticed a fast rosine phosphatases may result in increased phosphorylation, we MAPK activation following cytokine addition, which successfully addressed the possibility that NO interfered with the MAPK phos- was attenuated by NAC addition. Transient MAPK activation is phorylation/dephosphorylation equilibrium. By using FCS-acti- compatible with the notion that protein tyrosine phosphatases be- vated p42/p44, we established dephosphorylation of MAPK by come active shortly following kinase activation (46). The resultant cytosolic tyrosine phosphatases. Dephosphorylation was sensitive domination of dephosporylation abrogated MAPK activity. to the readdition of NO and further was down-regulated in cytosol Surprisingly, a second phase of MAPK activation occurred 15 to of NO- or cytokine-stimulated cells. An active role of NO in block- 24 h after cytokine stimulation. Delayed and pronounced MAPK ing phosphatase activity was proven in cells that were treated with activation coincided with iNOS activation. A cause/effect relation- cytokines in the presence of iNOS inhibitors. Inhibition of phos- ship between NO generation and MAPK activation was estab- phatase activity by NO is rationalized when we consider the active lished by the successful use of iNOS inhibitors such as NAME and and catalytically necessary thiol group in the Cx5R motif of ty- DPI, which attenuated kinase activation. As a result of these ex- rosine-specific phosphatases (25). Taking into account that the ac- periments, we conclude that endogenously derived NO, stemming tive cysteine residue of tyrosine phosphatases exists as a thiolate from an active iNOS, caused late-phase MAPK activation. anion (27, 51), inhibition by S-nitrosation or oxidation is predict- able. Knowing that As3ϩ abrogated tyrosine phosphatase activity, NO-derived signaling pathways that promote rapid MAPK and thereby promoted tumor growth, and binds with high affinity activation to vicinal thiols (52), and in view of the circumstances that some With the use of NO donors, we underscored an active role of NO tyrosine phosphatases contain a second cysteine five residues away in promoting rapid and delayed MAPK activation. Activation of from the invariant thiol common to all tyrosine phosphatases, mod- MAPK by authentic NO and NO donors is in accordance with ifications and inhibition via intramolecular disulfide formation reports stating that activation of p21ras promoted rapid MAPK may attenuate protein activity. Although inhibition of tyrosine activation in Jurkat cells (47, 48). However, delayed MAPK acti- phosphatases by NO donors has been performed, in part by mea- vation by NO donors and more importantly by endogenous NO suring hydrolysis of p-nitrophenyl phosphate (53–55), cellular in- generation is unknown. vestigations with NO donors, and more importantly by iNOS in- In MC, rapid MAPK activation is compatible with NO-mediated duction in relation to MAPK activation, are elusive. Conclusively, soluble guanylyl cyclase activation and cGMP formation. Exper- late-phase activation of MAPK by NO resulted from inhibition of imental evidence supporting this assumption is threefold. First, tyrosine phosphatases rather than activation of tyrosine kinases. MAPK activation is mimicked by membrane-permeable cGMP an- NAC-evoked redox modulation of tyrosine phosphorylation may alogues. Second, inhibition of soluble guanylyl cyclase by NS be in line with a regulatory role of the phosphatase reaction as 2028, a recognized enzyme-blocking agent, abrogated immediate well. Significantly, the use of NAC does not automatically imply MAPK activation in response to NO donors (39). Third, cGMP activation of MAPK through an oxidation-dependent mechanism; formation in response to GSNO coincided with MAPK activation. rather, it may point to the activation of oxidation-sensitive phos- As a further working hypothesis, we assume that cGMP-mediated phatases, thereby shifting the phosphorylation equilibrium toward phosphorylation initiates upstream signaling in the cascade leading dephosphorylation, i.e., MAPK deactivation. to MAPK activation because the MEK-1 inhibitor PD 98059 hin- NO-mediated rapid and delayed MAPK activation was sepa- dered the cGMP response. Our results are in line with a recent rated on the basis of cGMP signaling vs inhibition of protein ty- study showing that NO mediates vascular endothelial growth fac- rosine phosphatases and may contribute to MC hypercellularity tor-dependent p42/p44 MAPK activation via a cGMP-elicited re- under inflammatory conditions. Our results emphasize the impor- sponse (10). The cGMP response may be in some analogy to the tance of NO as a balancing element in the molecular events af- action of cAMP, which is known to activate p42/p44 MAPK fecting diverse cell responses such as cell proliferation and death. through a B-Raf-dependent pathway (49). Transient activation is explained on the basis that cGMP-elevating agents, such as atrial natriuretic peptide or NO donors, contribute to MAPK inactivation References through the induction of dual-protein tyrosine/threonine-specific 1. Michael, A. F., W. F. Keane, L. Raij, R. L. Vernier, and S. M. Mauer. 1980. The phosphatases that show some selectivity toward MAPK (50). Con- glomerular mesangium. Kidney Int. 17:141. 2. Pfeilschifter, J. 1995. Does nitric oxide, an inflammatory mediator of glomerular trol experiments in the presence of NS 2028 rule out a cGMP mesangial cells, have a role in diabetic nephropathy? Kidney Int. Suppl. 51:S50. action in cytokine-stimulated MC. However, cytokine and cGMP 3. Raij, L., and C. Baylis. 1995. Glomerular actions of nitric oxide [editorial]. Kid- signals that led to a rapid MAPK activation were both PD 98059 ney Int. 48:20. sensitive. The integration of NO formation and MAPK activation 4. Bokemeyer, D., K. E. Guglielmi, A. McGinty, A. Sorokin, E. A. Lianos, and M. J. Dunn. 1997. Activation of extracellular signal-regulated kinase in prolif- may represent a so far underestimated target system that contrib- erative glomerulonephritis in rats. J. Clin. Invest. 100:582. utes to cGMP-evoked signal transduction. 5. Cattell, V., E. Lianos, P. Largen, and T. Cook. 1993. Glomerular NO synthase activity in mesangial cell immune injury. Exp. Nephrol. 1:36. NO-derived signaling pathways that promote delayed MAPK 6. Pfeilschifter, J., P. Rob, A. Mu¨lsch, J. Fandrey, K. Vosbeck, and R. Busse. 1992. activation Interleukin 1 ␤ and tumour necrosis factor ␣ induce a macrophage-type of nitric oxide synthase in rat renal mesangial cells. Eur. J. Biochem. 203:251. Strikingly, NO evoked a second, delayed phase of activation, ir- 7. Narita, I., W. A. Border, M. Ketteler, and N. A. Noble. 1995. Nitric oxide me- respective of whether NO was endogenously produced or deliv- diates immunologic injury to kidney mesangium in experimental glomerulonephritis. Lab. Invest. 72:17. ered by NO donors. Delayed MAPK activation was not mimicked 8. Pipili-Synetos, E., A. Papageorgiou, E. Sakkoula, G. Sotiropuolou, T. Fotsis, by lipophilic cGMP derivatives, was insensitive to the soluble gua- G. Karakiulakis, and M. E. Maragoudakis. 1995. Inhibition of angiogenesis, tu- nylyl cyclase blocking agent NS 2028, and thus is unrelated to mour growth and metastasis by the NO-releasing vasodilators, isosorbide mono- nitrate and dinitrate. Br. J. Pharmacol. 116:1829. cGMP signaling. 9. Kuzin, B., I. Roberts, N. Peunova, and G. Enikolopov. 1996. Nitric oxide regu- Physiologic inactivation of MAPK is achieved by at least three lates cell proliferation during Drosophila development. Cell 87:639. distinctive MAPK phosphatase subfamilies with more than 100 10. Parenti, A., L. Morbidelli, X. L. Cui, J. G. Douglas, J. D. Hood, H. J. Granger, F. Ledda, and M. Ziche. 1998. Nitric oxide is an upstream signal of vascular isoenzymes that differ in structure and mode of function (25). Fol- endothelial growth factor-induced extracellular signal-regulated kinase 1/2 acti- lowing the assumption that blocking thiol-sensitive protein ty- vation in postcapillary endothelium. J. Biol. Chem. 273:4220. 4858 NO ACTIVATES p42/p44 MAPK

11. Clark, J. E., C. J. Green, and R. Motterlini. 1997. Involvement of the heme cascade in renal mesangial cells: comparison of hypertrophic and hyperplastic oxygenase-carbon monoxide pathway in keratinocyte proliferation. Biochem. agonists. Biochem. J. 305:777. Biophys. Res. Commun. 241:215. 36. Wu, X. B., B. Bru¨ne, F. von Appen, and V. Ullrich. 1993. Efflux of cyclic GMP 12. Du, M., M. M. Islam, L. Lin, Y. Ohmura, Y. Moriyama, and S. Fujimura. 1997. from activated human platelets. Mol. Pharmacol. 43:564. Promotion of proliferation of murine BALB/C3T3 fibroblasts mediated by nitric 37. Alessi, D. R., A. Cuenda, P. Cohen, D. T. Dudley, and A. R. Saltiel. 1995. PD oxide at lower concentrations. Biochem. Mol. Biol. Int. 41:625. 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase 13. Pages, G., P. Lenormand, G. L’Allemain, J. C. Chambard, S. Meloche, and kinase in vitro and in vivo. J. Biol. Chem. 270:27489. J. Pouyssegur. 1993. Mitogen-activated protein kinases p42mapk and p44mapk 38. Cross, A. R., and O. T. Jones. 1986. The effect of the inhibitor diphenylene are required for fibroblast proliferation. Proc. Natl. Acad. Sci. USA 90:8319. iodonium on the superoxide-generating system of neutrophils: specific labelling 14. Davis, R. J. 1993. The mitogen-activated protein kinase signal transduction path- of a component polypeptide of the oxidase. Biochem. J. 273:111. way. J. Biol. Chem. 268:14553. 39. Mu¨lsch, A., J. Bauersachs, A. Schafer, J. P. Stasch, R. Kast, and R. Busse. 1997. 15. Guan, K. L. 1994. The mitogen activated protein kinase signal transduction path- Effect of YC-1, an NO-independent, superoxide-sensitive stimulator of soluble way: from the cell surface to the nucleus. Cell. Signal. 6:581. guanylyl cyclase, on smooth muscle responsiveness to nitrovasodilators. 16. Anderson, N. G., J. L. Maller, N. K. Tonks, and T. W. Sturgill. 1990. Require- Br. J. Pharmacol. 120:681. ment for integration of signals from two distinct phosphorylation pathways for 40. Ottlinger, M. E., L. A. Pukac, and M. J. Karnovsky. 1993. Heparin inhibits activation of MAP kinase. Nature 343:651. mitogen-activated protein kinase activation in intact rat vascular smooth muscle 17. Zheng, C. F., and K. L. Guan. 1994. Activation of MEK family kinases requires cells. J. Biol. Chem. 268:19173. phosphorylation of two conserved Ser/Thr residues. EMBO J. 13:1123. 41. Dinarello, C. A. 1991. Interleukin-1 and interleukin-1 antagonism. Blood 77: 18. Seger, R., and E. G. Krebs. 1995. The MAPK signaling cascade. FASEB J. 9:726. 1627. 19. Waskiewicz, A. J., and J. A. Cooper. 1995. Mitogen and stress response path- 42. Coyne, D. W., and A. R. Morrison. 1990. Effect of the tyrosine kinase inhibitor, ways: MAP kinase cascades and phosphatase regulation in mammals and yeast. genistein, on interleukin-1 stimulated PGE2 production in mesangial cells. Bio- Curr. Opin. Cell Biol. 7:798. chem. Biophys. Res. Commun. 173:718. 20. Hunter, T. 1995. Protein kinases and phosphatases: the yin and yang of protein 43. Pfeilschifter, J., and H. Schwarzenbach. 1990. Interleukin 1 and tumor necrosis phosphorylation and signaling. Cell 80:225. factor stimulate cGMP formation in rat renal mesangial cells. FEBS Lett. 273: 21. Keyse, S. M. 1995. An emerging family of dual specificity MAP kinase phos- 185. phatases. Biochim. Biophys. Acta 1265:152. 44. Huwiler, A., and J. Pfeilschifter. 1994. Interleukin-1 stimulates de novo synthesis 22. Wu, J., L. F. Lau, and T. W. Sturgill. 1994. Rapid deactivation of MAP kinase of mitogen-activated protein kinase in glomerular mesangial cells. FEBS Lett. in PC12 cells occurs independently of induction of phosphatase MKP-1. FEBS 350:135. Lett. 353:9. 45. Wilmer, W. A., L. C. Tan, J. A. Dickerson, M. Danne, and B. H. Rovin. 1997. 23. Sun, H., C. H. Charles, L. F. Lau, and N. K. Tonks. 1993. MKP-1 (3CH134), an Interleukin-1␤ induction of mitogen-activated protein kinases in human mesan- immediate early gene product, is a dual specificity phosphatase that dephospho- gial cells: role of oxidation. J. Biol. Chem. 272:10877. rylates MAP kinase in vivo. Cell 75:487. 46. Brondello, J. M., A. Brunet, J. Pouyssegur, and F. R. McKenzie. 1997. The dual 24. Alessi, D. R., N. Gomez, G. Moorhead, T. Lewis, S. M. Keyse, and P. Cohen. specificity mitogen-activated protein kinase phosphatase-1 and -2 are induced by 1995. Inactivation of p42 MAP kinase by protein phosphatase 2A and a protein the p42/p44 MAPK cascade. J. Biol. Chem. 272:1368. tyrosine phosphatase, but not CL100, in various cell lines. Curr. Biol. 5:283. 25. Fauman, E. B., and M. A. Saper. 1996. Structure and function of the protein 47. Lander, H. M., A. T. Jacovina, R. J. Davis, and J. M. Tauras. 1996. Differential tyrosine phosphatases. Trends Biochem. Sci. 21:413. activation of mitogen-activated protein kinases by nitric oxide-related species. 26. Guan, K. L., and J. E. Dixon. 1991. Evidence for protein-tyrosine-phosphatase J. Biol. Chem. 271:19705. catalysis proceeding via a cysteine-phosphate intermediate. J. Biol. Chem. 266: 48. Lander, H. M., D. P. Hajjar, B. L. Hempstead, U. A. Mirza, B. T. Chait, 17026. S. Campbell, and L. A. Quilliam. 1997. A molecular redox switch on p21(ras): 27. Stone, R. L., and J. E. Dixon. 1994. Protein-tyrosine phosphatases. J. Biol. Chem. structural basis for the nitric oxide-p21(ras) interaction. J. Biol. Chem. 272:4323. 269:31323. 49. Vossler, M. R., H. Yao, R. D. York, M. G. Pan, C. S. Rim, and P. J. Stork. 1997. 28. Stamler, J. S. 1994. Redox signaling: nitrosylation and related target interactions cAMP activates MAP kinase and Elk-1 through a B-Raf- and Rap1-dependent of nitric oxide. Cell 78:931. pathway. Cell 89:73. 29. Nathan, C., and Q. W. Xie. 1994. Nitric oxide synthases: roles, tolls, and controls. 50. Sugimoto, T., M. Haneda, M. Togawa, M. Isono, T. Shikano, S. Araki, Cell 78:915. T. Nakagawa, A. Kashiwagi, K. L. Guan, and R. Kikkawa. 1996. Atrial natri- 30. Nathan, C. 1992. Nitric oxide as a secretory product of mammalian cells. FASEB uretic peptide induces the expression of MKP-1, a mitogen-activated protein J. 6:3051. kinase phosphatase, in glomerular mesangial cells. J. Biol. Chem. 271:544. 31. Feelisch, M., and J. S. Stamler. 1996. Donors of nitrogen oxides. In Methods in 51. Barford, D. 1995. Protein phosphatases. Curr. Opin. Struct. Biol. 5:728. Nitric Oxide Research. M. A. Feelisch and J. S. Stamler, eds. John Wiley & Sons, 52. Cavigelli, M., W. W. Li, A. Lin, B. Su, K. Yoshioka, and M. Karin. 1996. The New York, p. 71. tumor promoter arsenite stimulates AP-1 activity by inhibiting a JNK phospha- 32. Pfeilschifter, J., and K. Vosbeck. 1991. Transforming growth factor ␤ 2 inhibits tase. EMBO J. 15:6269. interleukin 1 ␤- and tumour necrosis factor ␣-induction of nitric oxide synthase 53. Caselli, A., P. Chiarugi, G. Camici, G. Manao, and G. Ramponi. 1995. In vivo in rat renal mesangial cells. Biochem. Biophys. Res. Commun. 175:372. inactivation of phosphotyrosine protein phosphatases by nitric oxide. FEBS Lett. 33. Hart, T. W. 1985. Some observations concerning the S-nitroso and S-phenylsul- 374:249. phonyl derivates of L-cysteine and glutathione. Tetrahedron Lett. 26:2013. 54. Caselli, A., G. Camici, G. Manao, G. Moneti, L. Pazzagli, G. Cappugi, and 34. Messmer, U. K., E. G. Lapetina, and B. Bru¨ne. 1995. Nitric oxide-induced ap- G. Ramponi. 1994. Nitric oxide causes inactivation of the low molecular weight optosis in RAW 264.7 macrophages is antagonized by protein kinase C- and phosphotyrosine protein phosphatase. J. Biol. Chem. 269:24878. protein kinase A-activating compounds. Mol. Pharmacol. 47:757. 55. Peranovich, T. M., A. M. da Silva, D. M. Fries, A. Stern, and H. P. Moneiro. 35. Huwiler, A., S. Stabel, D. Fabbro, and J. Pfeilschifter. 1995. Platelet-derived 1995. Nitric oxide stimulates tyrosine phosphorylation in murine fibroblasts in growth factor and angiotensin II stimulate the mitogen-activated protein kinase the absence and presence of epidermal growth factor. Biochem. J. 305:613.