| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,2
* Department of Biochemistry and Molecular Biology and
Institute of Molecular Cardiology, University of Louisville, Louisville, Kentucky, USA
2Correspondence: Division of Cardiology, Department of Medicine, Delia Baxter Bldg., 580 S. Preston St., Rm. 421F, University of Louisville, Louisville, KY 40202, USA. E-mail: aruni{at}louisville.edu
ABSTRACT
This study was designed to examine whether NO regulates protein glutathiolation. Exposure to NO donors increased protein glutathiolation in COS-7 or rat aortic smooth muscle cells as detected by anti-protein glutathione (GSH) antibodies. This process was reversible and saturable. Stimulation with acetylcholine (ACh) increased protein glutathiolation in isolated rat aortic rings. This was prevented by inhibiting endothelial NO synthase (eNOS). In ACh-treated rings, proteins showing positive immunoreactivity with the anti-PSSG antibody (Ab) were identified by matrix assisted laser desorption-time-of-flight mass spectrometry to be actin, vimentin, and heat shock protein 70. Purified actin was more readily glutathiolated by S-nitrosoglutathione than by oxidized GSH as determined by electrospray-ionization mass spectrometry, and nitrosylated actin was glutathiolated by reduced GSH. Relative to wild-type (WT) mice, increased protein glutathiolation was observed in hearts of mice with cardiac-specific expression of inducible NO synthase (iNOS). Proteins immunoprecipitated from transgenic hearts revealed GSH-adducted peptides corresponding to adenine nucleotide translocator and the 
-subunit of F1F0ATPase. These data suggest that exogenous NO or NO generated by eNOS or iNOS regulates protein adduction with GSH. This could be due to a direct reaction of proteins with S-nitrosoglutathione or denitrosylation of S-nitrosylated proteins by reduced GSH. Glutathiolation of cytoskeletal and mitochondrial proteins may be a significant feature of NO bioreactivity.—West, M. B., Hill, B. G., Xuan, Y.-T., Bhatnagar, A. Protein glutathiolation by nitric oxide: An intracellular mechanism regulating redox protein modification
Key Words: iNOS-TG mice • aortic rings • endothelium-dependent relaxation • posttranslational modification • rat aortic smooth muscle cells
NO IS A PLEIOTROPIC SIGNALING molecule involved in multiple processes that regulate smooth muscle cell relaxation, neurotransmission, platelet aggregation, cell growth, survival, differentiation, and apoptosis (1
2
3
4)
. The primary targets of NO signaling are heme-containing proteins. NO nitrosates heme with high affinity, and this reaction accounts for the activation of soluble guanylate cyclase leading to smooth muscle cell relaxation and inhibition of platelet aggregation (5)
. Binding of NO to other heme proteins, such as cytochrome P450 (6
, 7)
, NO synthases (NOS; refs 8
, 9
), and cytochrome c oxidase (10
, 11)
, competes with oxygen and regulates drug metabolism, NO synthesis, and oxygen consumption by the mitochondria. Due to its radical nature, NO also reacts with other radicals, particularly molecular oxygen (12
, 13)
, to yield N2O3, which is a potent nitrosating agent with high affinity for nucleophiles such as thiols (12
, 14)
. Although NO could induce a range of thiol modifications, S-nitrosylation has emerged as the most well-characterized effector mechanism by which NO reversibly regulates protein structure and function (15
, 16)
. S-nitrosylation has been shown to regulate ion channel activities, enzyme catalysis, cofactor binding, and the protein-protein interactions of a wide range of proteins (17)
. In most cases, however, S-nitrosylation is an inferred protein modification modality and the mechanisms and the implications of other sulfur oxidation products generated by NO remain to be fully assessed.
In addition to S-nitrosylation, NO could induce a variety of other modifications including S-thiolation, disulfide bond formation, and the formation of cysteine oxy-sulfur acids, i.e., sulfenic, sulfinic, or sulfonic acid derivatives (18
19
20)
. The formation of intramolecular or intermolecular disulfide bonds (21)
by NO could affect not only the structure of individual proteins but also the reactivity between protein and nonprotein thiols. In agreement with this view, it has been shown that reactions between nitrosothiols and reduced thiols yield mixed disulfides in vitro (20)
, suggesting that induction of protein-thiol mixed disulfides may be an important facet of NO biology. Indeed, several proteins such as aldose reductase, glyceraldehyde-3-phosphate dehydrogenase, cathepsin K, c-Jun, and calcium (Ca2+) ATPase form GSH adducts when incubated with S-nitrosoglutathione (GSNO; ref 18
). In several instances, e.g., aldose reductase (22)
, c-Jun (23)
, and cathespin K (24)
, glutathiolation by GSNO is nearly stoichiometric, localized to a specific cysteine, and results in changes in protein function, indicating that NO-mediated protein glutathiolation may be an important mechanism regulating protein activity. Increased glutathiolation of sarcoplasmic calcium ATPase in response to physiological NO generation has also been demonstrated (25)
.
Despite this evidence, it remains unclear whether glutathiolation is a general signaling modality engaged by NO and how NO induces glutathiolation of specific proteins. Moreover, the identities of the proteins most abundantly glutathiolated by physiological increases in NO are largely unknown. Hence, the aim of the present study was to determine whether exogenous or endogenous NO generated by eNOS/inducible NOS regulates protein glutathiolation and to identify the proteins that are glutathiolated in NO-exposed cardiovascular tissue. Our results indicate that increasing NO synthesis promotes glutathiolation of cytoskeletal and mitochondrial proteins. Additionally, our mechanistic studies with actin suggest that protein-glutathione adducts could arise either as a result of a direct reaction between reduced proteins and GSNO or as a result of denitrosylation of S-nitrosylated proteins by reduced GSH. These observations suggest new concepts regarding the cellular origins of protein-glutathione mixed disulfides and their role in the spectrum of thiol-based protein modifications induced by NO. A preliminary report of these findings has been previously published (26)
.
MATERIALS AND METHODS
Animal models and care
These studies were performed in accordance with the guidelines of the Animal Care and Use Committee of the University of Louisville (KY) School of Medicine. The iNOS transgenic mice expressing iNOS under the control of the -myosin heavy chain (MHC) promoter were developed and characterized as described previously (27)
.
Cell culture experiments
Primary rat aortic smooth muscle cells (RASM) or COS-7 cells were grown to 80% confluency in Dulbecco’s modified Eagle’s medium (Life Technologies-Invitrogen) supplemented with 10% FBS and 0.1% streptomycin/penicillin. At 80% confluency, the growth medium was removed and the cells were washed twice with Hanks’ balanced salt solution (HBSS). S-nitroso-N-acetyl-D,L-penicillamine (SNAP; Molecular Probes) was added to each flask in varying concentrations using equal volumes of DMSO as vehicle control. After treatment, cells were washed once in fresh HBSS and then scraped with a rubber policeman in homogenization buffer. For immunocytochemistry, cells were grown on multiwell glass slides and after treatment were fixed in ice-cold 100% acetone and immunostained as described below. For biochemical analyses, the cells were harvested in homogenization buffer containing (in mM) 25 HEPES (pH 7.0), 150 NaCl, 5 EDTA, 5 EGTA, 5 NaF, 2 NaVO3, 50 N-ethylmaleimide (NEM), and 1:100 protease inhibitor cocktail. Cell suspensions were lysed by sonication, and homogenates were centrifuged at 2000 g for 15 min at 4°C. Total protein concentration was measured by a modified Lowry method (28)
.
Vascular reactivity studies
Male Sprague-Dawley rats (10–12 wk, 250–350 g, Harlan; Indianapolis, IN) were anesthetized with sodium pentobarbital (60 mg/kg ip), and the descending thoracic aorta was excised and placed into aerated buffer containing: (in mM) 118 NaCl, 4.7 KCl, 2.5 CaCl2, 1.2 KH2PO4, 1.2 MgSO4, 12.5 NaHCO3, and 11.1 glucose (Glc). The aorta was cleaned of its connective and fat tissues and cut into 4 mm thick rings. The rings were mounted on triangular steel wire holders, placed into jacketed glass organ chambers at 37°C under a passive tension of 1 g, and constantly aerated with a 95% O2–5% CO2 mixture. The contraction and relaxation of the rings were measured with FT03 force displacement transducers and recorded using Chart software (ADInstruments). NO-mediated vascular tone was assessed by recording relaxation induced by ACh (1 µM) on aortic rings pre-equilibrated in the presence or absence of 100 µM N
-nitro-L-arginine methyl ester (L-NAME) hydrochloride and precontracted with phenylephrine (1 µM). After contraction/relaxation, the rings were either snap-frozen in liquid N2 or fixed in formalin. Aortic rings were homogenized in a glass homogenizer in the same buffer as cultured cells with the addition of 0.1% Triton X-100.
Western blot, immunocytochemistry, immunohistochemistry, and immunoprecipitation
Monoclonal antibodies recognizing glutathiolated proteins (anti-PSSG Ab) were obtained from Virogen (Watertown, MA) and diluted to 1:500 for immunoblotting. Immunoreactivity was visualized by chemiluminescence using a Typhoon 9400 detector (Amersham Biosciences), and the intensity of resulting bands was analyzed with ImageQuantTL software (Amersham Biosciences). For immunocytochemistry, cells were fixed in 100% acetone, washed with PBS, and then incubated in PBS containing 10% goat serum for 30 min, primary anti-PSSG antibodies overnight, and fluorescent secondary antibodies (Alexa-488, Molecular Probes) for 2 h at room temperature. The slides were then rinsed with PBS and mounted with a coverslip using Fluorsave reagent (Calbiochem). For immunohistochemical analysis, rat aortic rings were fixed in formalin and stored in 70% EtOH. The rings were immunostained with a 1:200 dilution of anti-PSSG antibodies. Fluorescent and phase contrast pictures of cells were acquired using a Zeiss LSM 510 microscope using a x63 water immersion objective. Images of aortic rings were acquired on a Nikon Eclipse E600 microscope with a x20 objective using a Spot Insight QE camera. Immunostaining in aortic rings was quantified using Metamorph software. Glutathiolated proteins were immunoprecipitated with anti-PSSG antibodies using a protein immunoprecipitation assay kit (Sigma) from cells homogenized in the nonreducing buffer described above. Mouse hearts from WT and iNOS-TG mice were glass homogenized in 50 mM Tris, pH 7.4, containing 250 mM sucrose, 10 mM iodoacetic acid (IAA), and 1% protease inhibitor cocktail. The homogenates were centrifuged at 14,000 g for 15 min at 4°C, and the supernatant was incubated in the dark for 1 h at room temperature. The homogenate was then passed through a Sephadex G25 (PD-10) column to remove excess IAA, and the glutathiolated proteins were immunoprecipitated with the anti-PSSG Ab. Nonspecific mouse IgG served as a control for the immunoprecipitations. The proteins were eluted by boiling the agarose beads in Laemmli buffer containing 50 mM NEM. The proteins were separated by nonreducing SDS-PAGE and visualized by silver staining.
Identification of glutathiolated proteins
Trichloroacetic acid (10% v/v) was added to aortic ring homogenates, and protein pellets were resuspended in 2D sample buffer, pH 6.8, containing 20 mM Tris, 8 M urea, and 1 mM EDTA. Proteins (40 µg) were focused on pH 5–8 IPG strips (Bio-Rad) and separated by SDS-PAGE. Parallel gels were developed for either immunoblot analysis or Sypro Ruby staining. To obtain peptides for matrix assisted laser desorption ionization-time-of-flight mass spectrometry (MALDI-TOF/MS), protein spots that were immunoreactive with anti-PSSG antibodies were excised from parallel Sypro Ruby-stained gels and digested with trypsin using a modified version of the method described by Jensen et al. (29)
. Additionally, protein bands from anti-PSSG immunoprecipitates were excised, digested, and analyzed by MALDI-TOF/MS. Peptide masses obtained by MALDI-TOF/MS analysis were used to search the National Center for Biotechnology Information (NCBI) database to identify the parent protein (30)
.
Electrospray-ionization mass spectrometry analysis
Rabbit skeletal muscle actin (Sigma) was reduced with 100 mM DTT and then excess DTT was removed by gel filtration on a Sephadex G-25 PD-10 column in 20 mM Tris, pH 7.5. The protein was then incubated with varying concentrations of GSNO or oxidized GSH (GSSG) for 1 h at 25°C. The protein was desalted again by gel filtration in 10 mM ammonium acetate and diluted in acetonitrile-water-acetic acid (50:50:0.5; v/v/v) before direct injection into the spectrometer (LCZ; Micromass, Manchester, UK) with a syringe pump (Harvard Apparatus, Holliston, MA) at a rate of 10 µl/min. The operating conditions were as follows: capillary voltage, 3.0 kV; cone voltage, 25 V; extractor voltage, 4 V; source block temperature, 80°C; and dissolvation temperature, 200°C. Spectra were acquired at the rate of 200 AMU/s over the range of 20–2000 amu. The instrument was calibrated with myoglobin. The spectra from each ion were summed and deconvoluted with MaxEnt software (MaxEnt Solutions, Suffolk, UK).
Myocardial NOx and iNOS activity measurements
NOS activity was determined by measuring the conversion of L-[14C]arginine to L-[14C]citrulline as described by Bredt and Snyder (31)
. For measuring NOx, tissue samples were homogenized in a buffer containing 25 mM Tris·HCl (pH 7.5), 0.5 µM EDTA, and 0.5 µM EGTA and centrifuged at 14,000 g for 15 min, and the resulting supernatants were collected as cytosolic fractions. The supernatants were loaded onto a Centricon-30 filter and centrifuged to remove molecules <30 kDa. Nitrite was assayed using the Griess reaction. Nitrate content was determined after conversion of nitrate to nitrite with Aspergillus nitrate reductase.
Statistical analysis
Data are reported as mean ± SE. Comparisons between two groups were performed by unpaired Student’s t-tests. A P value of < 0.05 was considered statistically significant.
RESULTS
Exogenous NO induces saturable and reversible glutathiolation in specific proteins
To examine NO-mediated changes in protein glutathiolation, COS-7 cells were incubated with 0.5–2.0 mM SNAP (corresponding to a cumulative concentration of NO of 12–48 µM) for 30 min. The cells were lysed, and total protein-glutathione formation was assessed by immunoblotting using monoclonal anti-PSSG antibodies that have been previously shown to selectively recognize glutathiolated proteins (32
33
34
35)
. In our studies, these antibodies recognized proteins (aldose reductase, creatine kinase, and glyceraldehyde-3-phosphate dehydrogenase) glutathiolated with 5 mM GSSG but not the same proteins treated with cystine. Treatment with DTT abolished the reactivity of the GSSG-modified proteins (data not shown), indicating that the antibodies recognized only the glutathiolated form of proteins. As shown in Fig. 1
A, incubation with SNAP led to a dose-dependent increase in immunoreactivity over the vehicle control. In unstimulated cells only basal levels of glutathiolated proteins were observed (Fig. 1Ba
), whereas treatment with SNAP led to the significant increase in Ab reactivity of multiple protein bands corresponding to 37, 50, and 75 kDa (Fig. 1Ba
). This increase depended on SNAP concentration (Fig. 1B, b
). However, prolonged incubation of the cells led to a time-dependent decrease in immunoreactivity, and after 60–120 min of incubation in the presence of SNAP, glutathiolation was at or below the levels observed before treatment (Fig. 1C, a
). Furthermore, no change in cell viability was observed (data not shown). Removal of SNAP after 30 min of exposure accelerated the loss of glutathiolated proteins. For instance, the 37 kDa protein was completely deglutathiolated 30 min after SNAP removal (Fig. 1C, b
). To rule out the possibility that GSH is adducted to proteins during cell lysis, the SNAP-treated cells were fixed immediately after treatment and stained with anti-PSSG antibodies. Cells treated with diamide, a thiol reagent that increases PSSG formation (36)
, were used as positive controls. As shown in Fig. 1D
, the SNAP-treated cells were stained more intensely than the vehicle-treated cells, although significantly less than the diamide-treated cells.
|
In addition to COS-7 cells, NO also induced a dose-dependent increase in protein glutathiolation in primary rat aortic smooth muscle (RASM) cells. Incubation of these cells with SNAP led to a dose-dependent increase in PSSG formation (Fig. 1E
, top inset). Also, incubation of these cells with the nonthiol NO donor spermine NONOate resulted in similar levels of glutathiolation (data not shown). Pooled data from three sets of gels could be well described by a simple binding isotherm with a K1/2 of 3 mM SNAP (Fig. 1E
). Pretreatment of the cells with the guanylate cyclase inhibitor 1H-[1,2,4]oxadiazolo-[4,3-a]quinoxalin-1-one did not significantly change NO-induced glutathiolation (data not shown) indicating that it is a cGMP-independent process. To identify the specific proteins that are glutathiolated in the presence of NO, immunoprecipitates of lysates prepared from SNAP-treated RASM cells were separated by SDS-PAGE. After silver staining, the most intense bands of 42 and 75 kDa (Fig 1E
, bottom inset) were excised, digested by trypsin, reduced by DTT, and identified by MALDI-TOF/MS. As listed in Table 1
, the two most glutathiolated proteins in SNAP-treated cells were identified to be heat shock protein 70 (HSP70) and ß-actin. Taken together, these data indicate that exposure to exogenous NO increases glutathiolation of a specific set of proteins in cells in culture. This modification was induced in intact cells and was saturable and reversible even in the continued presence of the NO donor.
|
NO-dependent vascular relaxation induces protein glutathiolation
To examine the role of eNOS, we examined protein glutathiolation in arterial rings during ACh-induced relaxation. For this, rat aortic rings were preincubated in the presence or absence of the NOS inhibitor L-NAME and then precontracted with phenylephrine. The rings were then treated with ACh to stimulate eNOS. Representative tracings are shown in Fig. 2
A. After addition of ACh, the aortas were allowed to relax for 10 min (rings with irregular relaxation or <40% relaxation were not used) and then removed from the organ baths and either frozen in liquid N2 or fixed in formalin. As shown in Fig. 2B
, sections from untreated vessels displayed relatively low levels of glutathiolated proteins as assessed by immunostaining with anti-PSSG antibodies. However, the ACh-treated vessels displayed a significant increase in PSSG staining that was abolished on pretreatment with L-NAME (Fig. 2C
). The ACh-induced increase in PSSG formation was also evident by immunoblot analysis, which revealed an increase in immunoreactivity of the total protein extract from ACh-treated rings (Fig. 2D
). The increase in immunoreactivity by ACh was prevented in rings pretreated with L-NAME and was abolished when the homogenates were treated with ß-mercaptoethanol (Fig. 2D
, inset), indicating increased eNOS-dependent glutathiolation in aortic rings.
|
To identify arterial proteins glutathiolated upon stimulation of eNOS, extracts of aorta treated with ACh in the absence and the presence of L-NAME were separated by nonreducing 2D SDS-PAGE followed by immunoblot using anti-PSSG antibodies. Immunopositive spots were excised from parallel Sypro Ruby-stained gels, and proteins were identified by MALDI-TOF/MS analysis. As shown in Fig. 2E, a
, multiple immunopositive spots were observed on 2D Western blots, the intensity of which was much less in L-NAME pretreated rings (Fig. 2E, b
), indicating NO-specific protein modification. Excision and MALDI-TOF/MS analysis of the L-NAME-responsive spots led to the detection of six proteins, four of which were identified as -actin. Multiple spots, corresponding to monomers and dimers of actin, were observed, and extensive sequence coverage and significant MOWSE scores suggest unambiguous protein identification (Table 2
). Spots 1 through 4 were confirmed to be actin by transblotting the proteins with anti-actin antibodies (Fig. 2E, a
, inset). In addition to actin, vimentin and HSP70 were also identified as proteins that displayed positive immunoreactivity with anti-PSSG antibodies (Table 2)
.
|
Glutathiolation results from direct modification of proteins by GSNO or denitrosylation
To delineate the mechanisms by which NO promotes protein glutathiolation, we examined the reactivity of actin, which was glutathiolated by NO in cells in culture (Table 1)
and isolated aortic rings (Table 2)
. To determine whether actin is modified by nitrosothiols, we incubated purified skeletal muscle actin with 1 mM GSNO for 1 h at 25°C. The charge states of actin obtained on electrospray-ionization mass spectrometry of the modified protein converged to molecular masses of 41,887 and 42,193 Da (Fig. 3
A, a). The 41,887 peak was ascribed to native actin, while the peak with a high mass (+305 Da) was assigned to glutathiolated actin. Thus under the experimental conditions used, modification of actin by GSNO results in the adduction of a single GSH molecule to the protein. In contrast, incubation with 1 mM GSSG did not increase the molecular mass of actin, indicating that actin was not glutathiolated by GSSG. Glutathiolation of actin was confirmed by Western analysis, which showed a concentration-dependent increase in actin glutathiolation by GSNO. In contrast, GSSG-treated actin showed only weak immunoreactivity at 1 mM GSSG concentration (Fig. 3B
). Interestingly, no nitrosylated actin was observed with the GSNO-treated protein. To ensure that this was not due to the breakdown of the actin-NO bond during mass spectrometry, reduced actin was incubated with a nonthiol NO-donor, spermine NONOate (NOC-22). As shown in Fig. 3A, c
, NOC-22 induced extensive nitrosylation of actin which is evident from the multiple +29 Da molecular mass peaks. Collectively, these data indicate that actin is not only more avidly glutathiolated by GSNO than GSSG but that it is not S-nitrosylated by GSNO.
|
To determine whether glutathiolated proteins could be generated during reduction of S-nitrosylated proteins by GSH, actin was first nitrosylated by incubating the protein with NOC-22, desalted, and then incubated with GSH. The glutathiolated form of the protein was identified by immunoblot analysis. As shown in Fig. 3C
, incubation of reduced actin with GSH did not induce protein glutathiolation; however, nitrosylated actin was readily glutathiolated by GSH, suggesting that in addition to direct glutathiolation by GSNO, actin could also be glutathiolated during GSH-assisted protein denitrosylation.
Overexpression of iNOS increases protein glutathiolation
To determine whether an increase in iNOS activity increases protein glutathiolation in situ, we used transgenic mice that express the iNOS gene specifically in the heart under control of the -myosin heavy chain promoter. These mice are healthy, and their cardiovascular function as measured by echocardiography is normal (data not shown). These animals display a significant increase in myocardial NO production compared with wild-type animals as evinced by NOx levels (Fig. 4
A, a) in the heart, due to a significant increase in iNOS activity compared with WT animals (Fig. 4A, b
). Western analysis of whole heart homogenates of iNOS-TG mice showed a much higher level of positive immunoreactivity than their WT littermates. Particularly intense staining was observed with protein bands of 
30, 50, and 120 kDa, although high reactivity was also observed with high MW proteins (>250 kDa; Fig. 4B
). MALDI-TOF/MS analysis led to the identification of the proteins immunoprecipitated from the iNOS-TG hearts (Table 3
). The intense immunopositive bands at 33 and 51 kDa were identified to be the adenine nt translocator (ANT; Fig 4C, a
) and the -subunit of the F1F0 ATPase. Peptides from these proteins displayed an increase in their expected mass by 305 Da, indicating that they were adducted to a single GSH molecule (Fig 4C, b
; Table 3
). These observations further validate the efficacy of the anti-PSSG Ab in identifying glutathiolated proteins and additionally provide direct evidence for protein glutathiolation in situ in hearts overexpressing iNOS.
|
|
DISCUSSION
The results of this study show that NO regulates the formation of protein-glutathione mixed disulfides. Our data demonstrate that exposure to exogenous NO increases glutathiolation in several cell types and that glutathiolation could also be induced by endogenous NO generated by either eNOS or iNOS, suggesting that NO could regulate protein glutathiolation via both autocrine and paracrine mechanisms. In our analysis, some proteins seem to be preferentially glutathiolated by NO; therefore, glutathiolation does not appear to be an inevitable consequence of NO reacting indiscriminately with protein thiols. NO-induced protein glutathiolation was saturable and reversible, indicating that glutathiolation is a regulated mechanism by which NO induces posttranslational modification of proteins with reactive cysteine residues.
Glutathiolation of proteins by NO is likely to be of functional significance. Previous studies have shown that NO-dependent glutathiolation of aldose reductase decreases the reduction of Glc to sorbitol (37)
, and glutathiolation of the calcium release channel upon stimulation of eNOS inhibits its activity (25)
. In the present study, increased glutathiolation was also observed in iNOS-TG hearts demonstrating that NO stimulates protein glutathiolation in situ and under basal oxygen concentrations. These observations suggest that NO-induced glutathiolation is not dependent on the high oxygen concentrations that prevail in isolated tissue experiments or cells in culture. Collectively, these data are in accordance with the view that NO is a physiological regulator of protein glutathiolation.
Glutathiolated proteins have been detected under a variety of physiological and pathological conditions, and it has been estimated that 3% of all proteins remain bound to GSH under basal conditions (18
, 38
, 39)
. Protein-glutathione adducts accumulate during oxidative stress (38
, 39)
, and an increased abundance of glutathiolated proteins has been reported in ischemic hearts (40)
, stimulated neutrophils (41)
, and cataractous lenses (42)
. Proteins are also glutathiolated by specific signaling events such as epidermal growth factor (EGF)-induced glutathiolation of protein-tyrosine phosphatase 1B (43)
, TNF-mediated glutathiolation of thioredoxin peroxidase II and annexin II (44)
, and the glutathiolation of Ras by angiotensin II (45)
. In addition, several enzymes, such as glutaredoxin, thioredoxin, and protein disulfide isomerase, have been shown to catalyze deglutathiolation of proteins (18
, 46)
, suggesting that protein glutathiolation is a highly regulated and reversible posttranslational mechanism. Under basal conditions, addition or removal of GSH could regulate protein function, whereas under oxidative stress this modification could protect protein thiols from irreversible oxidation by providing a "glutathione cap." However, the physiological and pathological significance of this form of post-translational modification remains to be fully assessed. In particular, it remains unclear how glutathiolated proteins are generated and to what extent they contribute to cell signaling and cytoprotection.
The most straightforward mechanism for the generation of S-glutathiolated proteins is a direct thiol-disulfide exchange reaction between protein thiols and GSSG. Indeed, for many years this was considered to be the sole mechanism for the generation of glutathiolated proteins (47)
. However, in most cells the concentration of GSSG is 10- to 100-fold lower than that of GSH (48)
, thereby making glutathiolation of proteins under basal conditions thermodynamically and kinetically unlikely (47)
. Moreover, during oxidative stress, most of the GSSG formed from oxidation of GSH is actively extruded (48)
, and therefore it does not readily accumulate (to levels higher than the Km of the GSSG transporter) in metabolically active cells. Additionally, GSSG is a poor glutathiolating agent and does not readily glutathiolate proteins even when incubated with proteins with highly reactive cysteines such as calbindin and CuZnSOD (49)
and, as shown here (Fig. 3)
, actin. Thus, even if GSSG could accumulate to levels exceeding those of GSH, it is unlikely to glutathiolate these proteins.
If direct thiol-disulfide exchange with GSSG is unlikely, what could account for the abundance of glutathiolated proteins in most tissues? It has been suggested that during oxidative stress GSSG could be oxidized to GSH thiosulfinate [GS(O)GS)], which is a more potent glutathiolating agent than GSSG (49)
. However, in a manner similar to GSSG, GS(O)SG is also likely to be readily extruded from cells. Moreover, it is currently unclear whether this species is generated in vivo and whether it accumulates in cells in concentrations high enough to glutathiolate susceptible proteins. As an alternative or additional mechanism, we propose that NO could regulate protein glutathiolation at least in tissues that are exposed to NO in vivo (Fig. 5
).
|
One possible mechanism by which NO induces protein glutathiolation may relate to direct reactivity of GSNO with protein thiols. GSNO is the most abundant nonprotein thiol in cells (50)
, and it could glutathiolate proteins (PSH) by undergoing a group transfer reaction involving an N-hydroxysulfenamide-like intermediate (20)
:
 | (1) |
, and the N-hydroxysulfenamide-like complex between aldose reductase and nitrosothiols has been observed by mass spectrometry (51)
. Several other proteins including cathespin K (24)
, glyceraldehyde-3-phosphate dehydrogenase (52)
, papain (53)
, and caspase-3 (54)
have also been shown to be directly glutathiolated by GSNO. Supporting a role of GSNO in glutathiolating proteins, our current studies show direct glutathiolation of actin by GSNO at concentrations much lower than GSSG (Fig. 3)
. Other proteins have also been shown to be preferentially glutathiolated by GSNO rather than GSSG (18
, 22
, 49)
. Taken together, this evidence supports the notion that under some conditions thiol transfer by nitrosothiols may be kinetically more efficient than direct thiol-disulfide exchange. Nevertheless, whether direct addition of GSNO accounts for NO-induced protein glutathiolation remains to be established. In our in vitro studies, actin was glutathiolated by GSNO only at concentrations exceeding 100 µM. In other studies, a concentration range of 50–400 µM GSNO was used to glutathiolate proteins (22
, 24
, 52
53
54)
. Although this is higher than the bulk concentration of GSNO expected to be present in most cells, it is possible that endogenous protein glutathiolation is facilitated by localized generation of GSNO at privileged sites close to the protein or catalyzed by neighboring metal complexes and proteins such as ceruloplasmin (55)
, hemoglobin (56)
, and possibly myoglobin.
Intracellular proteins could also be glutathiolated by GSNO generated from the reaction between GSH and peroxynitrite. Indeed, the work of Adachi et al. (25)
, showing that the glutathiolation of SERCA is prevented by superoxide scavengers demonstrates that peroxynitrite may be a critical mediator of NO-induced protein glutathiolation. Although peroxynitrite is a poor nitrosylating agent, it could oxidize cysteine residues of proteins or GSH to oxy-sulfenic acids (57)
, which readily form mixed disulfides (58)
. As shown here, however, proteins could also be glutathiolated directly by GSNO or during denitrosylation by GSH, without the involvement of peroxynitrite, suggesting that there may be peroxynitrite-dependent and -independent pathways of protein glutathiolation. Additional work is required to further delineate the contribution of these pathways under specific physiological and pathological conditions.
Although nitrosylation is an established posttranslational modification, less is known about denitrosylation. Previous studies suggest that proteins could be denitrosylated by thioredoxin and protein-disulfide isomerase (17)
, and our current data indicate that glutathiolation may be another fate of nitrosylated proteins. Reduced GSH could react with nitrosylated proteins to form a protein-glutathione mixed disulfide:
 | (2) |
. However, in vivo this reaction will compete with transnitrosylation:
 | (3) |
. We propose that glutathiolation may represent one mechanism of protein-denitrosylation that regulates the equilibrium between nitrosylated proteins and GSNO, where the free GSNO formed could feed into protein glutathiolation (Eq. 1)
. In any case, GSH-assisted denitrosylation would be advantageous in regulating the signaling functions of nitrosylation, since it builds in a refractory period that could be variably regulated (for the same reason inactivation and recovery from inactivation are the most regulated steps in ion conductance; ref 60
). Similarly, recovery of nitrosylated proteins could be controlled by the activity of dethiolating enzymes, GSH levels, and the prevailing redox state, thus linking cell signaling to intermediary metabolism and the generation of reactive oxygen species. Moreover, removing NO and restoring basal protein function are consistent with the cytoprotective and homeostasis-maintaining role of reduced GSH (48)
. Finally, the subset of proteins identified in our study (heat shock protein, vimentin, and actin) is similar to the recently identified set of nitrosylated proteins in endothelial cells treated with NO donors (61)
, suggesting that glutathiolation is a significant fate of nitrosylated proteins.
The role of protein-glutathiolation (and subsequent dethiolation) in NO signaling is also consistent with the predominant modification of heat shock protein and cytoskeletal proteins such as vimentin and actin. Glutathiolated actin has been detected in many cells including neutrophils on stimulation of the respiratory burst (62)
and gastric mucosal cells (63)
, T cells (36)
, hepatocytes, HepG2 cells (64)
, and erythrocytes (65)
exposed to peroxide. The appearance of glutathiolated actin is also associated with many pathological conditions that are associated with oxidative stress such as myocardial ischemia-reperfusion (40)
, sickle cell disease (66)
, and Friedreich’s ataxia (33)
. Hence, glutathiolation of actin could be one of the mechanisms by which NO disrupts the cytoskeleton and induces actin disassembly (67
, 68)
.
NO-induced glutathiolation of proteins other than actin could also be functionally important. HSP70 for instance is glutathiolated in vitro (69)
and during oxidative stress (36)
, and glutathiolation has been shown to convert the protein into an active chaperone (69)
. Similarly, glutathiolation of ANT at Cys-57 in iNOS-TG hearts may be related to the mechanism by which NO regulates mitochondrial permeability transition (70)
, which is currently believed to be a key step in both apoptotic and necrotic cell death pathways (71)
. Mitochondrial permeability transition is preceded by the formation of a permeability pore via ANT and cyclophilin D interaction (72)
. Cysteine-57 of ANT is located on the matrix face of the protein at the cyclophilin binding site (73)
. Hence, glutathiolation of ANT at Cys-57 could prevent MPT and therefore contribute to the antiapoptotic effects of NO.
The proteins that we found to be glutathiolated in both cells and tissues are likely to be the most abundant of a set of modified proteins. There may be additional proteins that were modified but not identified in our analysis. Importantly, we found that proteins glutathiolated in the rat aorta were different from those glutathiolated in iNOS-TG mouse hearts, suggesting that the specific protein modified may depend on the tissue or cell type or the proximity of the protein to the source of NO generation. Clearly, much additional work is required to identify other proteins that are modified in cells exposed to NO and how iNOS and eNOS cause differential protein glutathiolation. Nonetheless, data presented here support the view that increased NO generation increases glutathiolation of several proteins, which could result in multiple, tissue-specific changes.
In summary, our results with multiple experimental systems suggest that NO-mediated protein glutathiolation may be a sensitive, reversible, and regulated posttranslational mechanism. Additionally, we speculate that NO-mediated glutathiolation may be a regulatable step in denitrosylation of proteins, a protective strategy to prevent irreversible thiol oxidation, or a distinct signaling event in its own right. Additional investigations are required to ascertain fully the physiological significance of protein glutathiolation by NO.
ACKNOWLEDGMENTS
This work was supported in part by National Institutes of Health Grants HL-59378 and ES-11860 (to A. Bhatnagar), American Heart Association predoctoral fellowships (to M. B. West and B. G. Hill), and HL-65660 (to Y. T. Xuan). We thank Dr. Roberto Bolli for providing the iNOS transgenic mice. We also thank Jian Cai and Bill Pierce in the University of Louisville Mass Spectrometry Core Facility for assistance in identification of proteins and Joseph D. Hoetker for help in electrospray mass spectrometry.
FOOTNOTES
1 These authors contributed equally to this work. 
Received for publication February 3, 2006. Accepted for publication March 20, 2006.
REFERENCES
-activated prodrug causes kinase activation concurrent with S-glutathionylation of proteins. Mol. Pharmacol. 69,501-508
S-crystallin in human lens. Biochem. J. 379,595-600[CrossRef][Medline] induces protein glutathiolation. Biochemistry 39,11121-11128[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  X.-H. Gao, M. Bedhomme, D. Veyel, M. Zaffagnini, and S. D. Lemaire Methods for Analysis of Protein Glutathionylation and their Application to Photosynthetic Organisms Mol Plant, March 1, 2009; 2(2): 218 - 235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Y. Lim, M. Raftery, H. Cai, K. Hsu, W. X. Yan, H.-L. Hseih, R. N. Watts, D. Richardson, S. Thomas, M. Perry, et al. S-Nitrosylated S100A8: Novel Anti-Inflammatory Properties J. Immunol., October 15, 2008; 181(8): 5627 - 5636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Reinartz, Z. Ding, U. Flogel, A. Godecke, and J. Schrader Nitrosative Stress Leads to Protein Glutathiolation, Increased S-Nitrosation, and Up-regulation of Peroxiredoxins in the Heart J. Biol. Chem., June 20, 2008; 283(25): 17440 - 17449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Lee, C. Wie, B. O. Fernandez, M. Feelisch, and E. Vierling Modulation of Nitrosative Stress by S-Nitrosoglutathione Reductase Is Critical for Thermotolerance and Plant Growth in Arabidopsis PLANT CELL, March 1, 2008; 20(3): 786 - 802. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Into, M. Inomata, M. Nakashima, K.-i. Shibata, H. Hacker, and K. Matsushita Regulation of MyD88-Dependent Signaling Events by S Nitrosylation Retards Toll-Like Receptor Signal Transduction and Initiation of Acute-Phase Immune Responses Mol. Cell. Biol., February 15, 2008; 28(4): 1338 - 1347. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Diet, K. Abbas, C. Bouton, B. Guillon, F. Tomasello, S. Fourquet, M. B. Toledano, and J.-C. Drapier Regulation of Peroxiredoxins by Nitric Oxide in Immunostimulated Macrophages J. Biol. Chem., December 14, 2007; 282(50): 36199 - 36205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sun, M. Morgan, R.-F. Shen, C. Steenbergen, and E. Murphy Preconditioning Results in S-Nitrosylation of Proteins Involved in Regulation of Mitochondrial Energetics and Calcium Transport Circ. Res., November 26, 2007; 101(11): 1155 - 1163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Martinez-Ruiz and S. Lamas Signalling by NO-induced protein S-nitrosylation and S-glutathionylation: Convergences and divergences Cardiovasc Res, July 15, 2007; 75(2): 220 - 228. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Spurny, K. Abdoulrahman, L. Janda, D. Runzler, G. Kohler, M. J. Castanon, and G. Wiche Oxidation and Nitrosylation of Cysteines Proximal to the Intermediate Filament (IF)-binding Site of Plectin: EFFECTS ON STRUCTURE AND VIMENTIN BINDING AND INVOLVEMENT IN IF COLLAPSE J. Biol. Chem., March 16, 2007; 282(11): 8175 - 8187. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Clavreul, M. M. Bachschmid, X. Hou, C. Shi, A. Idrizovic, Y. Ido, D. Pimentel, and R. A. Cohen S-Glutathiolation of p21ras by Peroxynitrite Mediates Endothelial Insulin Resistance Caused by Oxidized Low-Density Lipoprotein Arterioscler Thromb Vasc Biol, November 1, 2006; 26(11): 2454 - 2461. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Zimmet and J. M. Hare Nitroso-Redox Interactions in the Cardiovascular System Circulation, October 3, 2006; 114(14): 1531 - 1544. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
