Nitric Oxide and N -Acetylcysteine Inhibit the Activation of Mitogen-activated Protein Kinases by Angiotensin II in Rat Cardiac Fibroblasts*

Angiotensin II acts on the cardiac fibroblast to produce a mitogenic response. Nitric oxide and N -acetylcys-teine have been used to determine if oxidative stress influenced the effects of angiotensin II on the cardiac fibroblast. Angiotensin II activated the mitogen-acti-vated protein kinases designated extracellular signal-regulated kinases within 5 min by interacting with the AT 1 receptor. This activation was completely inde- pendent of protein kinase C and was inhibited when farnesylation was blocked, implicating Ras involvement. Pretreatment of cardiac fibroblasts with either N -acetylcysteine for 8 h or nitric oxide for 10 min suppressed this activation by angiotensin II in a dose-de-pendent manner. However, when both agents were added, inhibition was essentially complete. This combined effect of N -acetylcysteine and nitric oxide to block ERKs activation also was found if the activity was stim-ulated by either another growth factor (platelet-derived growth factor) or by the addition of phorbol ester, suggesting the effect was not limited to the receptor site alone. The results are consistent with the hypothesis that hormonal activation of mitogenic steps such as ERKs is influenced by increased oxidative stress, which is reduced by the combined effects of N -acetylcysteine and nitric oxide. medium/F-12 blue activity cells Glutathione minor

Angiotensin II (Ang II) 1 has pleiotrophic effects on several cell types, leading to diverse responses including the regulation of cell growth, programmed cell death, cell migration, and modification of the extracellular matrix (1)(2)(3). Ang II has been shown to activate many signaling pathways through its G qlinked AT 1 receptor. In cultured cardiac cells, both the activation of the AT 1 receptor, leading to the activation of phospholipase C and protein kinase C (4,5), and cross-talk between AT 1 receptor with other growth factor receptors having intrinsic tyrosine kinase activity leading to activation of mitogen-acti-vated protein (MAP) kinases have been documented (6,7).
The MAP kinases are a family of serine-threonine kinases, which include the extracellular signal-regulated kinases (ERKs), the c-Jun N-terminal kinases (JNKs/stress-activated protein kinases), and p38 MAPK. The ERKs are well studied and generally activated by growth factors and mitogenic stimuli via a Ras/Raf1/MEK1 signaling cascade leading to activation of transcription factors such as Elk1 and c-Fos, whereas JNKs and p38 MAPK exhibit substantial sequence homology and respond primarily to various cellular stress conditions such as proinflammatory cytokines and UV irradiation, although it is not clear whether this occurs via common or parallel upstream kinases (8 -10).
The signaling system leading to the activation of MAP kinases is subject to diverse and complex regulation. Several recent studies have suggested that the balance of the oxidative and reductive potentials within the cell (cellular redox state) may substantially influence this pathway (11,12). In the presence of reactive oxygen species, such as superoxide anion or hydroxyl radical, or by the addition of N-acetylcysteine (NAC), an agent that is thought to influence intracellular glutathione and the redox state of the cell, the response of cells to mitogens or cytokines can be impaired, presumably by changing the conditions for thiol oxidation or reduction (13)(14)(15). Nitric oxide (NO), a redox-active molecule, has been identified as an important potential regulator of certain signaling events. NO acts by stimulating soluble guanylate cyclase, leading to enhanced production of intracellular cyclic GMP, an intracellular second messenger that can activate cyclic GMP-dependent protein kinases (16). NO also is capable of reacting with oxygen radicals such as superoxide anion (17) as well as directly modulating the activity of signaling molecules (18,19). The interaction with superoxide anion was suggested to be important in mechanisms where NO was implicated in modulating cytotoxic mechanisms, presumably by influencing oxidative stress (19).
Oxidative stress, the term used to encompass changes in the cellular redox state, has been implicated in inflammatory processes such as fibrosis. The cardiac fibroblast, which is the cell type known to proliferate during cardiac fibrosis and produce the excess matrix proteins characteristic of that condition, is a target cell for Ang II. Ang II has been shown to cause proliferation of cardiac fibroblasts in culture (20). We have recently shown that nitric oxide can modulate this proliferative effect (21), consistent with the known ability of NO to antagonize the actions of Ang II in many cell types. To better understand the role that NO might have in influencing Ang II action in cardiac fibroblasts, we have characterized the activation of ERKs in these cells by Ang II and have examined the effects of Nacetylcysteine and NO on this pathway. We have found novel effects of both NAC and NO on Ang II-induced ERKs activation, suggesting that both NO and oxidative stress, which ac-* This work was supported by National Institutes of Health Grants HL53471 and HL55001. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Cell Culture-Cardiac fibroblasts were obtained from 8-day-old rats following an isolation procedure described previously by us (22). Cells in the 3rd to 4th passage were grown to 80% confluence in either 60-or 100-mm culture dishes and then maintained for 24 h in 0.4% fetal calf serum/Dulbecco's modified Eagle's medium/F-12. Fresh medium was routinely added 2 h before the experiment.
Cell Treatments-Ang II was routinely added to the cells for 5 min at a concentration of 0.1 M. SNAP was routinely added 10 min prior to the addition of agonists. NAC pretreatment was typically for 8 h, and the medium containing NAC was removed and replaced with Dulbecco's modified Eagle's medium/F-12 lacking NAC for 2 h prior to adding other agonists. Care was taken to adjust the pH of medium containing NAC prior to adding it to the cells. Cell viability was monitored routinely using either trypan blue exclusion or by a measurement of lactate dehydrogenase activity into the culture medium using a commercially available kit.
Radioassay for ERKs-This radioassay was essentially that provided in a kit purchased from Amersham Pharmacia Biotech (catalog no. RPN 84) in which a cell lysate is used to generate a radiolabeled product using a synthetic peptide substrate specific for ERKs. The instructions provided with the kit were followed except for the initial procedures used to obtain a cell lysate. Following treatment of the cells with hormone or drugs, the cells were washed twice with ice-cold PBS, and then cell lysis was accomplished by adding a lysis buffer containing 0.1% Triton X-100, 10 mM Tris, pH 7.4, 50 mM NaCl, 2 mM EGTA, 1 mM dithiothreitol, 1 mM orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, and 10 g/ml aprotinin. The cells were briefly frozen at Ϫ70°C for 10 min and allowed to thaw at 4°C, and then the lysed cells were transferred to a 1.5-ml Eppendorf tube for subsequent centrifugation at 10,000 ϫ g for 10 min. Fifteen-l aliquots were then used for the radioassay, following the instructions provided in the kit. Following incubation of the extract with labeled ATP and the synthetic substrate, the radiolabeled product was adsorbed to phosphocellulose paper and washed extensively with 1% acetic acid, and then the radioactivity was measured in a scintillation counter. Protein was determined by using the Bio-Rad protein assay system and bovine serum albumin as a standard. All data were expressed relative to the control values.
Immunoassay for ERKs, JNKs, and p38 MAPK-The procedures followed were essentially those described in the instructions for kits provided by New England Biolabs (ERKs, catalog no. 9800; JNKs, catalog no. 9810; p38 MAPK, catalog no. 9820). Cell lysates were obtained as described above for the radioimmunoassay. In the ERK assays, 200 g of protein was immunoprecipitated with 1 g of a monoclonal antibody directed against phosphorylated ERKs. Following the addition of protein A-Sepharose beads (Amersham Pharmacia Biotech), the immunoprecipitated enzymes were subsequently incubated with 1 g of GST-Elk1 fusion protein for 30 min at 30°C in the presence of 100 M ATP. The phosphorylated products were solubilized, resolved by 10% SDS-PAGE, and subjected to immunoblotting using an antibody against phosphorylated Elk1 as described in the kit. In the p38 MAPK assay, 200 g of protein was immunoprecipitated with 1 g of a polyclonal antibody against phosphorylated p38 MAPK. The immune complexes, adsorbed to protein A-Sepharose beads, were incubated with 1 g of GST-ATF2 fusion protein and 100 M ATP for 15 min at 30°C. The phosphorylated products were determined by Western blot analysis using an antibody against phosphorylated ATF2 as described in the kit. In the JNKs assay, 300 g of protein was incubated with 2 g of GST-c-Jun-(1-89) fusion protein bound to glutathione-Sepharose beads to selectively precipitate JNKs. The kinase reaction was performed by the addition of 100 M ATP for 30 min at 30°C. c-Jun phosphorylation was selectively measured using a phosphospecific c-Jun antibody as described in the kit. Immune complexes on nitrocellulose membrane were treated with an appropriate secondary antibody conjugated with horseradish peroxidase and visualized with ECL (Amersham Pharmacia Biotech). Densitometric analysis of immunoblots from these assays and all other immunoassays used in this study were performed using a PDI scanner (model 420oe), and the data were reported as a -fold increase over unstimulated cells (control), which arbitrarily were set at 1.
Direct Immunoassay for ERKs Using Whole Cell Lysates-Following treatment with Ang II for 5 min, cells were washed twice with ice-cold PBS and then lysed with a concentrated sample buffer containing 250 mM Tris, pH 6.8, 8% SDS, 40% glycerol, 200 mM dithiothreitol, and 0.04% bromphenol blue. Following boiling for 5 min, the suspension was centrifuged at 10,000 ϫ g for 10 min at 4°C, and the supernatant was used for direct application onto 10% SDS-PAGE. Following transfer to nitrocellulose membrane and blockage with 5% nonfat milk in a Trisbuffered saline solution containing 0.1% Tween 20, the blot was incubated with antibody (1:1000) specific for phospho-ERK44 and -42 (New England Biolabs). After extensive washing, the blot was incubated with a second antibody conjugated with horseradish peroxidase and visualized with ECL (Amersham Pharmacia Biotech).
Radioimmunoassay for MEK1-500 g of protein in cell lysates obtained as described above for the ERKs radioimmunoassay were incubated 3 h at 4°C with 5 g of a polyclonal antibody against MEK-1 (Santa Cruz Biotechnology, catalog no. sc-436). Protein A-Sepharose beads were added, and the immunoprecipitated enzyme was incubated in 30 l of kinase buffer (25 mM Tris, pH 7.5, 2 mM dithiothreitol, 0.1 mM orthovanadate, 10 mM MgCl 2 , 5 M ATP, 10 Ci of [␥-32 P]ATP) with a recombinant GST-[K71A]Erk44 lacking enzymatic activity as a substrate (Upstate Biotechnology, Inc., catalog no. 14-135). The labeled substrate was detected by autoradiography following 10% SDS-PAGE.
Glutathione Assay-Intracellular glutathione was measured spectrophotometrically following a minor modification of the method of Tietze (23). Cells were lysed with 0.2 ml of 0.35 N perchloric acid. Following centrifugation at 10,000 ϫ g for 10 min at 4°C, different aliquots of the lysate were incubated with 0.1 M sodium phosphate, pH 7.5, 5 mM EDTA, 0.21 mM NADPH, 1 unit of glutathione reductase (Boehringer Mannheim), and 0.6 mM 5,5Ј-dithiobis(nitrobenzioc acid) in a total volume of 1 ml. The rate of change in absorbency at 412 nm was measured over a 5-min period to reflect the formation of reaction product and correlated with a standard curve using reduced glutathione. Data are expressed as nmol of glutathione/mg of protein in the lysate.
Statistics-Data are presented as the mean Ϯ S.E. of at least three experiments unless designated otherwise. Statistical analysis was performed using analysis of variance and Student's t test as appropriate. A value of p Ͻ 0.05 was considered to be statistically significant. Fig. 1A shows the rapid and transient increase in the activity of ERKs following the addition of 0.1 M Ang II in quiescent rat cardiac fibroblasts, where maximal activation occurred 2-5 min after hormone addition and then gradually decreased to basal levels. Fig. 1B shows that the 5-min response to Ang II was dose-dependent with peak activity occurring at concentrations between 0.01-0.1 M Ang II and a clear increase even at concentrations less than 1 nM. In 35 separate experiments using five different preparations of cardiac fibroblasts in the 3rd or 4th passage, 0.1 M Ang II increased ERK activity an average of 8-fold over control levels, with a range of 5-15-fold. Fig. 1C shows that the response to Ang II was mediated through AT 1 but not AT 2 receptors, based on the almost complete blockade of activation if cells were pretreated with losartan (an AT 1 receptor antagonist), whereas pretreatment with PD123319 (an AT 2 receptor antagonist) was without effect. The experiments shown in Fig. 1, A-C, were performed using the radioassay for ERKs. To further establish the response to Ang II, experiments using an immunoprecipitation assay for ERKs were performed (Fig. 2, A-C). This assay detects immunoreactive phosphorylated Elk1, a specific substrate for the immunoprecipitated ERKs. In Fig. 2, A-C, the effects of genistein, farnesyltransferase inhibitor-3, and PD98059, respectively, were studied. In each case, pretreatment with the respective inhibitor almost completely abolished the response to Ang II, suggesting that the activation of ERKs by Ang II involved tyrosine kinase activity, the presence of an activated form of Ras, and the subsequent activation of MEK1. It is noted that in the immunoassay procedure we used, the lower and upper bands for phosphorylated Elk1 reflect different degrees of phosphorylation. When activity is relatively low in the immunoprecipitated sample, the lower, less phosphorylated band predominates, whereas when activity is relatively high, the hyperphosphorylated, upper band is most obvious.

Characteristics of ERK Activation by Ang II-
Role of Protein Kinase C in ERK Activation- Fig. 3 shows experiments examining the role of protein kinase C in the Ang II-induced activation of ERKs. In Fig. 3A, using the radioassay for ERKs, cells were pretreated for only 15 min with protein kinase C inhibitors GF109203X (1 M) or Gö6983 (1 M), and the subsequent activation by PMA was abolished, whereas no effect on Ang II activation was found. Similar results were shown in Fig. 3B by using the immunoassay for ERKs. Cells were preincubated with GF109203X. Responses to PMA were blocked with no effect on Ang II activation. In Fig. 3C, cells were pretreated for 24 h with 1 M PMA to down-regulate protein kinase C, and then either Ang II (0.1 M) or PMA (1 M) was added. After 5 min, the cells were assayed for ERK activity. Again, pretreatment with PMA for 24 h completely prevented subsequent activation of ERKs by additional PMA but had almost no effect on the increased activity in response to Ang II. Thus, it appears that Ang II activates ERKs through a mechanism that is predominantly independent of protein kinase C.
Role of Glutathione in ERK Activation-To assess the role of intracellular GSH, a major determinant of the redox state of the cell, on the activation of ERKs by Ang II, cells were pretreated with BSO, an inhibitor of glutathione biosynthesis that decreases intracellular GSH levels. Cells also were treated with NAC, an agent known to increase GSH levels, and additionally with a combination of both drugs. Fig. 4A shows the effects of these drugs on GSH levels in the cardiac fibroblasts. BSO (50 M) dramatically decreased cellular GSH levels when pretreatment of the cells was performed for 24 h. NAC (8 mM) increased GSH approximately 3-fold if pretreatment was for at least 8 h. Higher concentrations of NAC (20 -30 mM) caused cell damage within 30 min of treatment as measured by the release of lactate dehydrogenase into the culture medium. The combination of BSO and NAC pretreatment blocked the increase in GSH levels generated by exposure to NAC, indicating that the effects of BSO override those of NAC. However, as shown in Fig. 4B, activation of ERKs by Ang II was essentially unaffected by BSO treatment, whereas NAC pretreatment caused a decrease in ERK activity. This effect of NAC was clearly independent of GSH levels, since ERK activity was lowered even when BSO was added in the presence of NAC. Fig. 4C shows that the inhibition of ERK activation by NAC was dose-dependent, but NAC did not completely inhibit the response. Since a low GSH concentration presumably makes the cell more sensitive to oxidative stress, and Ang II-mediated effects may be mediated by reactive oxygen species (24, 25), we performed a series of experiments with BSO-pretreated cells to determine if the response to suboptimal levels of hormone would be influ- All experiments were performed using the immunoassay for ERKs as described under "Experimental Procedures." A, following pretreatment with 50 M genistein for 15 min, the cells were incubated with 0.1 M Ang II or 20 ng/ml PDGF-BB for 5 min, and activated ERKs were immunoprecipitated using a phospho-ERK-specific antibody and then used in kinase reactions to phosphorylate GST-Elk1 fusion protein as substrate. Elk1 phosphorylation was then detected by immunoblotting using a phospho-Elk1-specific antibody. B, following pretreatment with 10 M farnesyltransferase inhibitor-3 (FT-3), a commercially available enzyme inhibitor (Calbiochem) for 1 h, the cells were incubated with either 0.1 M Ang II or 20 ng/ml PDGF-BB for 5 min, and then extracts were prepared for immunoprecipitation. C, following pretreatment with 50 M PD98059 for 45 min, the cells were incubated for 5 min with 0.1 M Ang II, 1 M PMA, or 20 ng/ml PDGF-BB, and then extracts were prepared for immunoprecipitation. Each immunoblot is representative of at least two separate experiments. pElk1, phosphorylated Elk1. enced in the presence of lower GSH levels. Fig. 4D shows that dose-response curves to Ang II activation were similar in the absence or presence of BSO, indicating that intracellular GSH apparently does not affect the signaling pathways used by Ang II. In separate experiments, not shown, we also found that BSO pretreatment had no effect on PMA activation of ERKs throughout a concentration range of 10 Ϫ10 to 10 Ϫ6 M. In additional experiments, pretreatment with glutathione monoethylester (0.5 mM), which increased intracellular GSH levels about 2-fold, did not affect Ang II-activated ERK activity, whereas relatively brief pretreatment (60 min) with either ␤-mercaptoethanol (1 mM) or dithiothreitol (0.5 mM) reduced Ang II activation of ERKs by 50 -75%. These latter findings suggested that it was the efficacy of NAC as a reducing agent that might account for its inhibitory effect. In all experiments with BSO and NAC pretreatments, cell viability was monitored visually by trypan blue exclusion and in selected experiments by measuring the release of lactate dehydrogenase into the culture medium. No evidence of cell damage to the cardiac fibroblasts was found during the pretreatment protocols.
The Effect of NO on ERK Activation-To assess the role of NO, we pretreated the cells with varying concentrations of the NO donor SNAP. Fig. 5A shows that pretreatment for 10 min with varying doses of SNAP resulted in a dose-dependent inhibition of ERK activation, but complete inhibition was not obtained at SNAP concentrations that maintained cell viability. In the above experiments, equimolar amounts of cysteine were added with SNAP to increase the efficacy of NO release. Fig. 5B shows that in the absence of cysteine, 50 M SNAP was only slightly effective in inhibiting ERK activation, whereas a statistically significant reduction in activity was found when cysteine was included. Using another NO donor, S-nitrosoglu-tathione, a similar effect of cysteine addition was observed. We routinely checked to see if the SNAP addition, either in the presence or absence of cysteine, influenced the basal levels of ERK activity and found no change whatsoever in the control levels throughout the concentration range of SNAP used. To confirm that the inhibitory action of SNAP was due to the presence of NO in the medium, we pretreated cells with 10 M ODQ, an inhibitor of soluble guanylate cyclase, or 20 M oxymyoglobin, which sequesters NO; the inhibition of ERK activation by SNAP was prevented (Fig. 5C). In an additional experiment (not shown), we added 10 M LY83583, another inhibitor of soluble guanylate cyclase, and found a similar reversal of the inhibition produced by SNAP. With each new agent added to the cells, viability was checked both visually and by lactate dehydrogenase release into the medium, and no adverse effects were noted due to SNAP, cysteine, or the other agents added.
Combined Effect of NAC Pretreatment and NO Addition on ERK Activation-In Fig. 6, A-C, we compared the effects of Each immunoblot is representative of at least two separate experiments. pElk1, phosphorylated Elk1.

FIG. 4. Effect of NAC and BSO pretreatment on intracellular glutathione levels and ERK activation by Ang II. A, cells were pretreated with NAC (8 mM) for 8 h and/or BSO (50 M) for 24 h.
Glutathione was then measured in perchloric acid extracts prepared as described under "Experimental Procedures." In the designated experiment, cells were pretreated with 0.1 M Ang II for 5 min prior to the addition of perchloric acid. The data shown represent five separate determinations. *, p Ͻ 0.05 versus control. B, cells were pretreated with NAC and/or BSO as described above and then treated for 5 min with 0.1 M Ang II. ERK activity was measured by the immunoassay. The upper part is a representative immunoblot of Ang II-induced ERK activation. pElk1, phosphorylated Elk1. The lower part represents densitometric analysis of the immunoblot. C, cells were pretreated with NAC for 8 h at the designated concentration and then treated with 0.1 M Ang II for 5 min prior to measuring ERK activity by a radioassay. Values are the mean Ϯ S.E. for three independent experiments. *, p Ͻ 0.05 versus Ang II alone. D, cells were pretreated with 50 M BSO for 24 h, and then both treated and untreated cells were incubated with the designated concentration of Ang II for 5 min prior to measuring ERK activity. Data are presented as the mean Ϯ S.E. for three independent experiments using a radioassay. No significant difference was found between the groups with or without BSO pretreatments (two-way analysis of variance). NAC and SNAP treatment on ERK activation using Ang II, PDGF-BB, or PMA as the agonist. The most striking finding with all agonists used was that in combination, SNAP and NAC produced greater inhibition than when either drug was added alone.
Direct Measurement of ERK Phosphorylation-Since both a radioassay and an immunoassay have been used in experiments described above to document the effects of NAC and NO either alone or in combination, and some quantitative discrepancies sometimes were found, we performed an additional assay for ERK activation, which directly measures the phosphorylated ERK44 and ERK42 in a whole cell extract. Fig. 7 shows representative results and densitometric analysis from such assays using either SNAP or NAC, alone or in combination, prior to the Ang II addition. Consistent with the other assays, there was a greater inhibition of activity when both agents were added together. In this procedure, we found it necessary to obtain concentrated extracts of the cells prior to SDS-PAGE in order to have enough activated enzyme to detect by Western blot analysis.
Measurement of MEK1 Activity-To obtain insight into the possible upstream sites involved in the above effects of NAC and NO, we measured the effects of these agents on the activation of MEK1, an upstream kinase of ERKs, by Ang II. Fig.  8 shows that MEK1 activation by Ang II is marked and that pretreatment with either NAC or SNAP caused significant inhibition. However, combined treatment with both NAC and SNAP inhibited MEK1 activation by Ang II completely. In an additional set of experiments, we directly added SNAP to the assay tubes containing the immunoprecipitated, activated MEK1. There was no effect on MEK1 activity even when SNAP was added at a final concentration of 100 M, suggesting that there was no in vitro nitrosation and subsequent inactivation of MEK1. Measurement of JNKs and p38 MAPK Activity-Ang II also activated both JNKs and p38 MAPK in the cardiac fibroblasts (Fig. 9, A and B). The activation of both these MAP kinases ranged between 8-and 15-fold using 10 Ϫ7 M Ang II by immunoprecipitation assays. Following Ang II addition, maximal activation of JNKs routinely occurred after 15 min, whereas for p38 MAPK, activation was maximal after 5 min. Of particular interest was the observation that pretreatment with either NAC or SNAP also inhibited the activation of these kinases by Ang II in a manner analogous to the inhibition observed with ERK activation. DISCUSSION This study shows that NO inhibited the activation of ERKs by Ang II in rat cardiac fibroblasts, providing a molecular model for the known counterregulatory roles that Ang II and NO have in diverse tissues. The inhibitory effect of NO was enhanced when cells were pretreated with NAC, which by itself also inhibited the response, albeit to a lesser extent than when combined with NO treatment. The combination of NAC pretreatment and the addition of SNAP almost completely inhibited ERK activation by pathways both dependent on and independent of protein kinase C and also inhibited ERK activation initiated by PDGF and PMA, suggesting the effect was not localized to the hormone receptor alone.
We characterized the response to Ang II using agents that implicated the AT 1 receptor (losartan), tyrosine kinase activity (genistein), Ras activation (farnesyltransferase inhibitor-3), and the requirement for MEK1 activation (PD98059). The activation of ERKs through such a signaling pathway has been implicated in other studies using cardiac fibroblasts and is consistent with recent studies showing that divergent pathways were involved in Ang II action, which conventionally had been thought to work solely through G protein-mediated activation of phospholipase C (4). The absence of a role for protein kinase C in our study is in agreement with recent findings showing that protein kinase C was not involved in Ang II activation of ERKs using cultured neonatal cardiac fibroblasts and vascular smooth muscle cells (26,27), but differs from other studies where protein kinase C was involved in Ang II action both in cardiac myocytes and aortic smooth muscle cells (28,29). We inactivated protein kinase C using both downregulation by PMA pretreatment for 24 h and by using specific inhibitors (GF109203X and Gö6983) that only required brief pretreatment. Since it is known that many isoforms of protein kinase C exist, some being independent of PMA down-regulation, it is difficult to absolutely rule out any involvement of protein kinase C, although the inhibitors we used are thought to also act on the PMA-independent forms of protein kinase C (30). A recent study (31) has shown that Ang II activates ERKs in vascular smooth muscle cells through a mechanism involving protein kinase C-. This kinase is calcium-independent and not activated by diacylglycerol or phorbol esters.
Studies with the agent that inhibited farnesyltransferase implicated Ras activation in the signaling pathway. Ras is known to be involved in mediating the effects of Ang II in several cell types, although a recent study has reported that Ras was not activated by Ang II in cultured cardiac myocytes (28). Of particular interest to our studies is the report showing that Ras can be directly nitrosated and activated by NO (19) and studies showing that basal activity of ERKs and other MAP kinases were increased following the addition of NO (32). Those studies, conducted in the Jurkat human T-cell line, clearly differ from our own, since we repeatedly found no effect of SNAP addition on basal activity, either under our standard assay conditions or if assays were performed between 30 s and 30 min following the SNAP addition (data not shown).
Cysteine increased the ability of the NO donor SNAP to inhibit ERK activity. Cysteine was shown to enhance the effect of the S-nitrosothiol SNAP, which commonly is used as an NO donor in in vitro experiments with intact cultured cells. It was suggested (33) that cysteine could act by promoting transnitrosation from SNAP to S-nitrosocysteine, which would be more susceptible to intracellular decomposition, leading to more effective storage or transport of NO within the cell. Reduced GSH would presumably be the major intracellular mediator of NO storage and transport, implicating a transnitrosation from the original NO donor (SNAP) to a labile intermediate carrier S-nitrosocysteine and ultimately to intracellular GSH.
The reported effects of NAC pretreatment on several cell types include increased levels of reduced glutathione (34) and action as a reducing agent (35), a potential contributory factor in the intracellular storage and transport of NO (36), and a free radical scavenger (37). NAC has been shown to both increase survival and inhibit the proliferation of PC12 cells by mechanisms that are independent of intracellular glutathione levels and do not appear to be dependent on its antioxidant/radical scavenging properties (15,38). In one study (38), the effect of NAC was suggested to be dependent on its ability to act as a reducing agent or on the transcription activation of as yet unidentified genes. Our findings with NAC required prolonged pretreatment (about 8 h) of the cardiac fibroblasts to partially inhibit the acute activation of ERKs in response to Ang II or to further enhance the inhibitory effect of SNAP. We also found that glutathione monoethylester, an agent that also increases intracellular level of glutathione, does not share with NAC the capability to inhibit Ang II-induced ERK activation. Although we have shown that changes in the absolute amount of glutathione do not account solely for the inhibitory effects of NAC on the activation of ERKs, it is possible that changes in the concentration of reduced glutathione combined with an increased ratio of reduced/oxidized glutathione (GSH/GSSG) might account for the effect of NAC. It has been proposed that the GSH/GSSG ratio, by controlling the reduction state of critical sulfhydryl groups, may reflect the redox state of the cell and regulate enzyme activity (39). Recently, the effect of NAC on cell redox environment has been studied in endothelial cells (40), where NAC caused a dramatic increase in GSH/GSSG ratio from 200 to more than 400, which has been suggested as one possible mechanism through which NAC exerts its inhibitory effect.
A role for reactive oxygen species influencing the MAP kinase pathway has been suggested, with the data available implicating superoxide anion as a positive effector of MAP kinase activation (41). Ang II, by virtue of activating the NADPH oxidase system, produces the superoxide anion, which triggers the kinase cascade and promotes cell proliferation (25). If true, scavengers of free radicals such as the superoxide anion would inhibit activation. NAC has been suggested as a direct scavenger of free radicals in studies where it reduced the activation of MAP kinase phosphorylation in response to PDGF (42) and where it lowered MAP kinase activity and Ras activation in response to lactosylceramide (43). Data have shown an interaction between NAC and reactive oxygen species (37), although the reaction may not be rapid at physiological conditions. Nonetheless, the inhibition of ERK activation by Ang II (and by PDGF and PMA) could be mediated by the ability of NAC to scavenge requisite free radicals.
One possible mechanism for the NO action on the ERKs could be the redox reaction of NO with critical thiol group(s) of the enzymes, leading to its S-nitrosation. Nitrosation of thiol groups has been shown to modulate the activity of certain enzymes (16). Direct evidence of the redox regulation of Ras has been presented (19), where NO binds to a cysteine residue on the surface of the activated Ras molecule, leading to activation of the downstream MAP kinase cascade. This apparently does not happen in our system, since no activation of ERKs in the basal state was observed when SNAP was added to the cardiac fibroblasts. However, NO could react with superoxide anion, giving rise to the formation of peroxynitrite, a reactive substance that could lead to modification of cellular thiols (17).
We found that MEK1 was activated by Ang II and inhibited by both NAC and SNAP, suggesting that it could be a potential site of action for these drugs, although any site upstream of MEK1 also could be implicated. Evidence suggesting that MEK1 is not the sole site of action for these drugs comes from our observations that both JNKs and p38 MAPK activation also were inhibited by NAC and SNAP, and these pathways are generally believed to be independent of MEK1.
In our studies, NAC facilitated the effects of NO on the inhibition of an essential signaling system. This is a novel finding that contrasts with recent studies in other cell types where NAC pretreatment reduced heme oxygenase activity induced by NO (44) and in another study where the inhibitory effect of NO on the respiration of digitonin-treated ascites hepatoma cells was suppressed by NAC (45). The interrelationship between NO and the superoxide anion has been intensely studied in recent years, and an emerging hypothesis is that such interactions have an important physiological role in modulating the biological activities of both NO and the superoxide anion (46). It is plausible that the Ang II signaling pathway consists of redox-sensitive steps and that NAC pretreatment could modulate the redox state of the cell in an as yet undetermined manner that would lead to NO being either stabilized or transported intracellularly in a more effective manner so that it would either interact more readily with superoxide anion or could act directly on redox-sensitive target proteins by nitrosation reactions.

Additions and Corrections
Vol. 273 (1998) 33027-33034 Nitric oxide and N-acetylcysteine inhibit the activation of mitogen-activated protein kinases by angiotensin II in rat cardiac fibroblasts.

Deming Wang, Xin Yu, and Peter Brecher
Page 33031, Fig. 5A: The plus (ϩ) designation for AII in the second bar from the left is incorrect and should be a minus sign, so that the bar represents an experiment lacking AII(Ϫ), containing Cys(ϩ), and lacking SNAP(Ϫ). A corrected version of V. Srinivas, X. Zhu, Susana Salceda, R. Nakamura, and Jaime Caro We have been unable to reproduce the studies on the iron content of the recombinantly expressed HIF-1␣ protein, as represented in Table I of the above cited article. It appears that the iron was due to a contaminated reagent. Consequently, we wish to retract those data. The results concerning the effect of CO and heme synthesis inhibitors on the hypoxia response of B-1 cells are not in doubt. We apologize for the inconvenience caused by our mistake. We suggest that subscribers photocopy these corrections and insert the photocopies at the appropriate places where the article to be corrected originally appeared. Authors are urged to introduce these corrections into any reprints they distribute. Secondary (abstract) services are urged to carry notice of these corrections as prominently as they carried the original abstracts.