S-glutathiolation of Ras mediates redox-sensitive signaling by angiotensin II in vascular smooth muscle cells.

Angiotensin II (AII) increases production of reactive oxygen species from NAD(P)H oxidase, a response that contributes to vascular hypertrophy. Here we show in cultured vascular smooth muscle cells that S-glutathiolation of the redox-sensitive Cys(118) on the small GTPase, Ras, plays a critical role in AII-induced hypertrophic signaling. AII simultaneously increased the Ras activity and the S-glutathiolation of Ras (GSS-Ras) detected by biotin-labeled GSH or mass spectrometry. Both the increase in activity and GSS-Ras was labile under reducing conditions, suggesting the essential nature of this thiol modification to Ras activation. Overexpression of catalase, a dominant-negative p47(phox), or glutaredoxin-1 decreased GSS-Ras, Ras activation, p38, and Akt phosphorylation and the induction of protein synthesis by AII. Furthermore, expression of a Cys(118) mutant Ras decreased AII-mediated p38 and Akt phosphorylation as well as protein synthesis. These results show that H(2)O(2) from NAD(P)H oxidase forms GSS-Ras on Cys(118) and increases its activity leading to p38 and Akt phosphorylation, which contributes to the induction of protein synthesis. This study suggests that GSS-Ras is a redox-sensitive signaling switch that participates in the cellular response to AII.

Considerable evidence implicates reactive oxygen species (ROS) 1 as mediators of cellular signaling by several kinds of stimuli, including cytokines, growth factors, hormones (1), or mechanical forces (2). However, the molecular mechanism by which ROS alter cellular signaling is not established. Reactive thiols on Cys residues of select proteins are among the most sensitive sites to be modified by ROS. For example the reactive Cys on phosphatases (3) or protein kinase C (4) can be modified by ROS and change their function. Oxidants react with these redox-sensitive thiols to form thiyl radicals that subsequently can react with other thiols to form mixed disulfide bonds (5)(6)(7). In mammalian cells GSH is the most abundant low molecular thiol, which as a result is most likely to bind to protein thiols to form mixed disulfides, a process termed S-glutathiolation. Many previous reports have demonstrated S-glutathiolation of (GSS-) isolated or purified proteins or in cells or tissues exposed to nonspecific thiol oxidants such as diamide or H 2 O 2 . A cellular regulatory role for GSS-protein adducts has only been rarely documented (8).
The small GTPase, Ras, modulates diverse signaling pathways, and a reactive thiol, Cys 118 , has been identified in the GTP-binding region of Ras. Previous work shows that Cys 118 can be S-nitrosated, leading to an increase in Ras activity and to downstream signaling (9 -11). However, Williams et al. (12) found that S-nitrosation itself may not directly change Ras conformation, thus questioning whether this modification could itself explain activation of Ras. S-Glutathiolation of the C-terminal Cys of Ras with diamide has been reported, but there have been no reports regarding S-glutathiolation of Cys 118 on Ras affecting its activity.
Angiotensin II (AII) induces vascular smooth muscle cell (VSMC) hypertrophy, proliferation, and migration via ROS formation by stimulation of NAD(P)H oxidase, a process that may contribute to a variety of vascular diseases, such as hypertension, atherosclerosis, and restenosis (13)(14)(15). Ushio-Fukai et al. (16,17) showed that AII-stimulated ROS from NAD(P)H oxidase plays a critical role in the activation of p38, Akt, but not ERK. They also showed that p38 and ERK pathways contribute synergistically to smooth muscle hypertrophy by showing that the MEK inhibitor, PD98059, and the p38 inhibitor, SB203580, each blocked ϳ40% of the hypertrophic response, whereas together they additively blocked the response. Therefore, AII-induced hypertrophy may be mediated in part by redox-sensitive activation of p38 and Akt and redoxinsensitive activation of ERK. In this study we employed AIIstimulated VSMC as a model for examining the relationship between GSS-Ras formation and redox-sensitive hypertrophic signaling with adenovirus-mediated genetic modulation of NAD(P)H oxidase, glutaredoxin-1 (GRX-1) and Cys 118 mutation of Ras (C118S Ras). Our studies indicate that oxidantinduced thiol modification of Ras mediates activation of p38 and Akt but not the ERK activation that contributes to the increase in protein synthesis.

EXPERIMENTAL PROCEDURES
Drugs and Antibodies-Sulfo-NHS biotin and an EZ-link Ras activation kit were obtained from Pierce. PD-10 desalting columns, protein A-Sepharose, and streptavidin-Sepharose were obtained from Amersham Biosciences. Bradford protein assay, Bio-spin 6 column, polyvinylidene difluoride membrane, and other reagents for immunoblotting were obtained from Bio-Rad. Polymer polyethylenimine was obtained from Aldrich. AG1478 was obtained from Calbiochem (San Diego, CA). Other chemicals were from Sigma. Epidermal growth factor (EGF) and anti-Ras antibody (clone Ras10) were obtained from Upstate Biochemical (Golden, CO). Rabbit anti-mouse IgG antibody was obtained from Jackson ImmunoResearch (West Grove, PA). Anti-phosphorylated and total Akt, p38, ERK, and p70 S6 kinase antibodies were obtained from Cell Signaling (Beverly, MA). Anti-p47 phox antibody was obtained from Transduction Laboratories (San Diego, CA). Anti-human catalase antibody was obtained from The Binding Site (Birmingham, UK). Anti-GRX antibody was obtained from American Diagnostica Inc. (Greenwich, CT). Ras cDNA and adenoviral vector were obtained from Clontech (Palo Alto, CA). Dulbecco's modified Eagle's medium, fetal bovine serum (FBS), pENTRD/Topo vector, and Gateway cassette were obtained from Invitrogen.
Cell Culture-Rat cultured VSMC were prepared as reported before (16,17). The cells were grown with Dulbecco's modified Eagle's medium with 10% FBS and penicillin/streptomycin and quiescent with 0.1% FBS for 36 -48 h before treating with AII.
Virus Preparation-Adenoviral constructs were developed as previously described. The catalase (ATCC) and dominant-negative p47 phox (DN-p47 phox , S303A/S304A double mutant, gift from Dr. B. M. Babior (18)) cDNAs were subcloned into the adenoviral shuttle vector (pShuttle), which contains the cytomegalovirus promoter. This shuttle vector was co-transfected with pADEasy into Escherichia coli, the resultant cosmid was transfected with calcium phosphate into HEK 293 cells, and adenoviral plaques were selected. The adenovirus was propagated in HEK 293 cells and purified via a double cesium chloride gradient. Using the TCID50 method, the viral titer was determined. Similar methods were used for the creation of ␤-galactosidase. The adenoviral vector to express GRX-1 was previously reported (19).
Adenovirus with C118S Ras was made as follows. Ras cDNA was subcloned from Ras expression vector to pENTRD/Topo vector by PCR. Site-directed mutagenesis of C118 was introduced by 18 cycles of PCR with forward primer 5Ј-TGCTGGTGGGGAACAAGTCTGACCTG-GCTGCACGC-3Ј and reverse primer 5Ј-GCGTGCAGCCAGGTCA-GACTTGTTCCCCACCAGCA-3Ј followed by digestion of the template with Dpn-1. The entire Ras sequence and mutation were confirmed by sequence analysis. Then adenoviral vector expressing C118S Ras was made by LR clonase reaction (Invitrogen) with an adenoviral vector in which the Gateway cassette had previously been inserted downstream of the cytomegalovirus promoter.
The cells at about 80% confluency were transfected with ϳ50 multiplicity of infection of adenovirus with 10 M polymer polyethylenimine in Dulbecco's modified Eagle's medium with 10% FBS for 36 h (20). After washing out the medium with adenovirus, the cells were quiescent in Dulbecco's modified Eagle's medium with 0.1% FBS for 36 h before the start of the experiments.
Ras Activity Assay-Ras GTP binding activity was assessed by the association with GST-Raf. Cultured VSMC were scraped into lysis buffer (50 mM Tris-HCl, pH 7.4, 0.25 M sucrose, 50 M diethylenetriaminepentaacetic acid, 2 mM phenylmethylsulfonyl fluoride) and centrifuged (10,000 ϫ g for 15 min). Protein concentration was measured by Bradford assay. Some cells were scraped into lysis buffer with 20 mM DTT to test the thiol dependence of activation.
After equilibration in lysis buffer, GST-Raf was passed over a Bio-6 spin column to eliminate DTT. In preliminary studies, by measuring the free thiols in bovine serum albumin by 5,5Ј-dithiobis(2-nitrobenzoic acid) fluorescence (412 nm), we confirmed that less than 1% of DTT remained after passing through the Bio-6 column. Approximately 700 -1000 g of total cell protein was incubated for 1 h with GST-Raf (75 g) coupled to 50 l of immobilized GSH-resin. The samples were washed four times with nonreducing lysis buffer and incubated with Laemmli buffer containing 5% ␤-mercaptoethanol and heated to 100°C for 5 min to dissociate Ras. Protein was run on SDS-PAGE, and Ras was quantified by immunoblotting with anti-Ras antibody (Upstate Biochemical, clone Ras10).
Biotinylated GSH Ester and Detection of S-Glutathiolated Ras-The methods followed those previously reported with some modifications (6). Biotinylated GSH ester was made by mixing 25 mM sulfo-NHSbiotin with 25 mM GSH ethyl ester in 50 mM NaHCO 3 at pH 8.5 for 2 h followed by the addition of 125 mM NH 4 HCO 3 at pH 8.5 for 1 h. Biotinylated GSH ester (250 M) was preincubated with VSMC in culture for 1 h. At each time point the cells were washed three times with cold phosphate buffer and lysed in buffer (Tris-HCl, pH 7.4, 1% Nonidet P-40, 150 mM NaCl, 50 M diethylenetriaminepentaacetic acid, 2 mM phenylmethylsulfonyl fluoride) containing 10 mM N-ethylmaleimide to block further thiol reactions. Approximately 1 mg of protein was passed through a PD-10 Sephadex-G25 column to eliminate the excess low molecular weight biotin products. The proteins were mixed with streptavidin-Sepharose beads for 1 h. The beads were washed five times with lysis buffer with 0.1% SDS, and the final precipitate was incubated for 10 min with 40 l of elution buffer (lysis buffer ϩ 20 mM DTT) to release S-glutathiolated proteins. After adding Laemmli buffer containing 5% ␤-mercaptoethanol, GSS-Ras was detected by immunoblotting with monoclonal anti-Ras antibody (UBI, clone Ras10).
Leucine Incorporation-The method followed previous reports (2,16,17). After quiescence, AII (0.1 M) was added to VSMC. After 16 h 1 Ci/ml of [ 3 H]leucine was added to each dish and incubated for 8 h. The cells were washed five times with phosphate-buffered saline and fixed with 10% trichloroacetic acid for 30 min at 4°C. After washing with ice-cold water once, protein was isolated with 0.5 M NaOH for 90 min, and radioactivity was counted with a scintillation counter. The values for cells treated with AII are normalized to the value obtained in untreated cells (16,17).
Ras Purification for MALDI-TOF Mass Spectrometry Analysis-The cells were scraped into lysis buffer containing iodoacetic acid (10 mM), and the proteins were separated with centrifugation (10,000 ϫ g for 15 min). After 1 h the proteins were passed through a PD-10 Sephadex-G25 column to eliminate excess iodoacetic acid. The lysate (2-3 mg of protein) was prewashed with protein A-Sepharose and incubated overnight with monoclonal anti-Ras antibody (25 g). Thereafter, protein A (Ϸ100 l) was added with rabbit anti-mouse Ig G antibody (30 g). The immunoprecipitate was released with nonreducing Laemmli buffer (50 l), and 40 l of sample was separated by nonreducing SDS-PAGE and stained with Coomassie Blue. Approximately 10 l of sample was used for immunoblotting Ras to identify the major band. The Ras protein band was cut from the gel, and protein was digested in the gel with trypsin (Sequencing grade; Sigma) and dried with a Speedvac (Savant, Albertville, MI). GSS-Cys containing peptides of Ras were detected by MALDI-TOF mass spectrometry and analyzed with the SWISS-PROT data base found on the Profound Web Site (prowl.rockefeller.edu/cgibin/ProFound). Reduced Cys-and GSS-Cys-containing peptides were identified on the basis of the predicted increase in mass (RSSG: ϩC 10 H 15 N 3 O 6 S). Accepted fit parameters for peptide mass were Ͻ100 ppm for monoisotopic mass, Ͻ200 ppm for average mass, and up to four tryptic miscleavages. The data fit well with the protein identified in the data base as gi 131873 sp P20171 RASH RAT TRANSFORMING PRO-TEIN P21/H-RAS-1 (C-H-RAS).
Data Analysis-For all of the experiments employing immunoblots, similar results were obtained in at least three or four separate experiments. The increase in GSS-Ras and Ras activity caused by AII was confirmed by densitometry (Molecular Analyzer, Hercules, CA), and differences in band densities were analyzed with Student's t test. p Ͻ 0.05 was considered to be statistically significant. Differences in the percentage of increase in [ 3 H]leucine incorporation between untreated and AII-treated cells were analyzed with Student's t test. p Ͻ 0.05 was considered to be statistically significant.

AII-induced S-glutathiolation and
Thiol-redox-sensitive Activation of Ras in VSMC-AII (0.1 M) increased ROS formation in VSMC as detected by 2Ј,7Ј-dichlorofluorscein fluorescence (data not shown) as shown previously (16). As detected by biotinylated GSH, AII (0.1 M) increased GSS-Ras formation (3.6 Ϯ 0.4-fold at 15 min p Ͻ 0.05, n ϭ 5; Fig. 1A). GSS-Ras was increased by AII in a concentration-dependent manner (Fig.  1B). H 2 O 2 (250 M) also increased GSS-Ras (Fig. 1B). Increased GSS-Ras was noted as early as 5 min, peaked at 30 min, and decreased at 60 min (Fig. 1C). Formation of GSS-Ras preceded the phosphorylation of p38 or Akt but not ERK (Supplemental Fig. 1). This suggests temporal correlation of GSS-Ras formation with activation of p38 and Akt, two downstream mediators previously shown to be involved in AII-induced protein synthe-sis in VSMC (16,17). To determine the importance of the thiol modification of Ras, Ras GTP binding activity was assessed with a GST-Raf pull-down assay. Ras activity was increased 2-fold at 15 min. Importantly, the increase in Ras activity was eliminated by DTT (Fig. 1D), suggesting that, like GSS-Ras formation, activation of Ras by AII is dependent on thiol-redox status.
AII-induced GSS-Ras and Ras Activation Were Mediated by ROS from NAD(P)H Oxidase-A NAD(P)H oxidase inhibitor, diphenyleneiodonium chloride (DPI; 10 M, 30 min) decreased GSS-Ras as well as the activation of Ras by AII at 15 min ( Fig.  2A). Previous reports showed that phosphorylation of the NAD(P)H oxidase subunit, p47 phox , by AII triggers release of ROS, including H 2 O 2 , which has been implicated in hypertrophic signaling (18,22). Therefore, either catalase (16,17) or DN-p47 phox (18) was overexpressed in VSMC to decrease the contribution of ROS to AII signaling. Either catalase or DN-p47 phox decreased GSS-Ras and Ras activation (Fig. 2B). We also assessed downstream signaling events affected by ROS from NAD(P)H oxidase. DPI, catalase, or DN-p47 phox decreased phosphorylation of p38 and Akt with lesser effects on ERK phosphorylation (Supplemental Fig. 2). In addition, compatible with previous results regarding redox-sensitive hypertrophic signaling by AII (16,17), catalase or DN-p47 phox inhibited the induction of protein synthesis by AII as indicated by the incorporation of [ 3 H]leucine (Fig. 2, C and D).
Overexpression of GRX-1 Inhibits GSS-Ras, Ras Activation, and Redox-sensitive Hypertrophic Signaling by AII-The time course (Fig. 1D) indicated that GSS-Ras formation is reversible, suggesting that an enzymatic process may be involved in the reduction of S-glutathiolated Ras. Thiol-disulfide exchange of GSS-protein mixed disulfides is specifically regulated by the enzyme GRX, which in turn is regulated by GSH and GSH reductase (5,19). To test the specificity of S-glutathiolation for the activation of Ras, GRX-1 was overexpressed in VSMC. Overexpression of GRX-1 inhibited the increase in Ras activity and GSS-Ras formation caused by AII, indicating the requirement of GSS-Ras formation for the increase in activity (Fig.  3A). Similar to the effects of inhibiting ROS production, over-expression of GRX-1 markedly decreased phosphorylation of p38 and Akt but was without effect on ERK (Fig. 3B). GRX-1 overexpression also inhibited the induction of protein synthesis (Fig. 3C).
The Identification of GSS-Cys on Ras by AII with MALDI-TOF Mass Spectrometry-To confirm that GSS-Ras is formed by endogenous GSH pools and to identify GSS-Cys sites, Ras was immunopurified from AII-treated VSMC, separated by SDS-PAGE, and trypsinized in gel. The peptides were analyzed by MALDI-TOF mass spectrometry, and reduced HS-Cys and GSS-Cys on Ras were analyzed (Table I). Approximately 60% coverage of the sequence of H-Ras was achieved, and three GSS-Cys sites were identified (Cys 80 , Cys 118 , and one of the C-terminal Cys 181 , Cys 184 , or Cys 186 ).
Effects of C118S Ras Expression on AII-induced Redox-sensitive Signaling-To test the functional importance of S-glutathiolation, Cys 118 on Ras was mutated to serine (C118S Ras), incorporated into an adenoviral vector, and overexpressed in VSMC. Overexpression of the C118S Ras mutant decreased phosphorylation of p38 and Akt but not ERK (Fig. 4A). p70 S6 kinase modulates protein synthesis in response to AII (23) and is a downstream target for both ERK and Akt. Overexpression of C118S Ras modestly decreased phosphorylation of p70 S6 kinase by AII (Fig. 4B). The C118S mutant also significantly inhibited the induction of protein synthesis by AII (Fig. 4C). These data indicate that the modification of Cys 118 , which was shown to be S-glutathiolated by mass spectrometry, plays a critical role in the redox-sensitive signaling, which leads to VSMC hypertrophy caused by AII.

GSS-Ras Is Not Downstream of EGF Receptor
Transactivation by AII-EGF receptor transactivation participates in the AII signaling cascade and is known to contribute to phosphorylation of ERK and Akt. Therefore, we tested whether EGF receptor transactivation is involved in GSS-Ras formation caused by AII. A selective inhibitor of EGF receptor activity, AG1478 (1 M, 30 min), completely blocked phosphorylation of ERK and Akt caused by EGF, and partially inhibited phosphorylation of ERK and Akt by AII (Fig. 5A). However, AG1478 had no significant effect on GSS-Ras formation by AII (Fig. 5B), indicating that GSS-Ras formation is not mediated through EGF receptor transactivation. DISCUSSION It is now well recognized that AII increases production of oxidants derived from NAD(P)H oxidase, which play an important role in vascular hypertrophy and the pathogenesis of vascular diseases (13)(14)(15). Previous reports indicate that the increased oxidants activate p38 and Akt, accounting for the redox-sensitive component of signaling leading to hypertrophy caused by AII (16,17); however, an upstream direct molecular target for ROS has not been identified. In this study the activation of Ras by S-glutathiolation at Cys 118 was identified to be a critical step in redox-sensitive signaling leading to AII-induced hypertrophy. Overexpression of GRX-1 confirmed the specificity of the activation of Ras by S-glutathiolation as well as its involvement in activating downstream signaling by AII. GSS-Cys 118 was identified with mass spectrometry and overexpression of C118S mutant Ras strikingly inhibited p38 and Akt phosphorylation and protein synthesis (Fig. 6). NAD(P)H oxidase is activated at the plasma membrane where we hypothesize it is well situated for ROS generated by it to attack Cys 118 and activate membrane-bound Ras. This hypothesis is supported by the fact that H 2 O 2 increased formation of GSS-Ras (Fig. 1B), and either DPI or overexpression of TABLE I MALDI-TOF mass spectrometry analysis of GSS-Cys modified peptides of Ras from AII-treated VSMC All-treated VSMC were lysed in buffer containing iodoacetic acid (IAA, 10 mM). After Ras was purified by immunoprecipitation and nonreducing SDS-PAGE, Ras protein was digested with trypsin, and MALDI-TOF mass spectrometry was obtained. Free Cys (SH) or Cys alkylated with IAA (IAA Cys: ϩC 2 H 2 O 2 Х 58 Da) during homogenization are listed as reduced Cys. GSS-Cys were identified with the SWISS-PROT database found at the ProFound Web Site (prowl.rockefeller.edu/cgi-bin/ProFound) by an increase in the molecular mass of H-Ras peptides containing Cys (RSSG: ϩC 10 H 15 N 3 O 6 S Х 305 Da). Reduced Cys and GSS-Cys containing peptides are listed with their measured mass, measurement error, start and end amino acid, and sequence. Errors are within 100 ppm for monoisotopic mass (Ͻ2000 Da) and 200 ppm for average mass (Ͼ2000 Da). The analysis took into account potential methionine oxidation, which is indicated in the column labeled MetO.  catalase inhibited GSS-Ras formation and Ras activation (Fig.  2). These agents also inhibited p38 and Akt phosphorylation and the increase in protein synthesis by AII as shown previously (16,17), and we further confirmed this by showing similar effects of overexpressing DN-p47 phox . Collectively, our data show that ROS from NAD(P)H oxidase are responsible for GSS-Ras formation and activation of Ras by AII, which is responsible for redox-sensitive signaling (p38, Akt) and induction of protein synthesis.
To identify the GSS-Cys site on Ras, we employed mass spectrometry, and three sites were identified (Cys 80 , Cys 118 , and one of the C-terminal Cys 181 , Cys 184 , or Cys 186 ). Among them, Cys 118 is located in the GTP-binding region, and Snitrosation of this Cys is associated with an increase in activity of purified Ras (9 -11). We did not detect S-nitrosated Cys 118 by mass spectrometry, nor is nitric oxide thought to be involved in AII signaling in VSMC. Therefore, it is reasonable to hypothesize that GSS-Ras formation on this Cys is responsible for the increase in Ras activity and for redox-sensitive signaling by AII. The specificity of the involvement of GSS-Ras for redoxsensitive hypertrophic signaling was tested with the overexpression of GRX-1, a cytosolic protein of the thioredoxin superfamily. GRX-1 inhibited Ras activation, p38 and Akt phosphorylation, and induction of protein synthesis by AII (Fig.  3), emphasizing the importance of the GSS-adduct of Ras in mediating redox-sensitive signaling.
It is controversial whether or not ERK activation by AII is redox-sensitive (16,25), and our data show that AII-induced ERK phosphorylation is much less sensitive to overexpression of catalase, DN-p47 phox , and GRX-1, as well as the C118S Ras mutant compared with their effects on phosphorylation of p38 or Akt. Although Ras/Raf/MEK1/ERK has been widely regarded as a major pathway, it is not clear that AII exclusively activates ERK via this pathway in VSMC. Takahashi et al. (26) reported that Ras may not be involved in ERK activation by AII in rat aortic smooth muscle cells. Others showed that Raf activation by AII is not mediated by Ras but rather by protein kinase C (27). These reports may help to explain our findings that GSS-Ras is not required for ERK activation but plays a role in phosphorylation of p38 and Akt. Recent studies showed that transactivation of the EGF receptor may mediate ERK activation by AII (23), and the EGF receptor/Grb/mSOS/Ras is the major pathway by which EGF activates ERK. Therefore, we tested the effects of AG1478, an EGF receptor tyrosine kinase inhibitor, on formation of GSS-Ras and phosphorylation of ERK and Akt. Compatible with previous reports (23,28), AG1478 partially inhibited ERK and Akt phosphorylation by AII but had no effect on GSS-Ras formation. Therefore, GSS-Ras formation by AII-stimulated ROS apparently is not a consequence of EGF receptor transactivation. Previously, the Ras C118S mutant was shown to have about 75% basal activity (9) and preserved ability to mediate growth factor-dependent mSOS-induced activation (10). However, the mutant prevented redox-dependent PI3 kinase activation. Therefore, GSS-Rasmediated Akt and p38 phosphorylation and mSOS-dependent activation of Ras via EGF receptor transactivation may be separate signaling pathways and be localized in different cellular compartments. AG 1478 also decreased AII-induced Akt phosphorylation, suggesting that both GSS-Ras and EGF receptor transactivation by AII may contribute to Akt activation.
In previous reports, ischemia-reperfusion increased S-thiolation labeling of Ras in heart (29), and diamide increased GSS-Cys 181 and Cys 184 in recombinant Ras (11). These reports did not assess the changes in Ras activity caused by GSS modification. A recent report suggests that S-nitrosation of Ras Cys 118 itself may not change the structure of Ras but that the increased activity of recombinant Ras was associated rather with the chemical process of S-nitrosation (12). This suggests that redox cycling of this critical thiol, possibly via either Snitrosation or sulfenilation could be an intermediate in Sglutathiolation, which would be favored because of the abundance of intracellular GSH. Because the glutamate residue of GSH has an extra carboxyl group, S-glutathiolation (but not S-nitrosation or sulfenilation) adds a negative charge to modified proteins that together with the steric effects of the tripeptide might directly change enzyme activity analogous to protein phosphorylation. Although many studies have shown that thiol antioxidants like N-acetylcysteine inhibit redox-sensitive signaling, they could do so by increasing intracellular GSH levels, enhancing the activity of GRX, and affecting S-glutathiolation of key signaling proteins including Ras. A recent report showed that fluid shear stress up-regulated GSH reductase activity in endothelial cells, inhibiting JNK activation by ROS (24). This result could be explained by the fact that GSH reductase activity couples with GRX for reducing the mixed disulfide. Thus, proteins in various signaling cascades could be reversibly regulated by S-glutathiolation, GRX, and GSH reductase to mediate redox-sensitive signaling in various cell types.
In summary, this study clearly shows, 1) that the ROSmediated modification of a critical thiol of Ras is important in mediating the actions of AII, 2) that intracellular GSS-Ras formation increases following AII, 3) that GSS-Ras is responsible for a portion of increased Raf-binding activity, 4) that p38 and Akt activation are most affected by GSS-Ras, and 5) that these events contribute to hypertrophic signaling induced by AII.