Reaction Kinetics for Nitrosation of Cysteine and Glutathione in

Jan 7, 1994 - Ingeborg Hanbauer/ George W. Cox/ Francoise Laval,II Jacques Laval,. John A. Cook/ Murali C. Krishna/ William G. DeGraff/ and James B...
0 downloads 0 Views 887KB Size
Chem. Res. Toxicol. 1994, 7, 519-525

519

Reaction Kinetics for Nitrosation of Cysteine and Glutathione in Aerobic Nitric Oxide Solutions at Neutral pH. Insights into the Fate and Physiological Effects of Intermediates Generated in the N0/02 Reaction David A. Wink,**tRaymond W. Nims,? John F. Darbyshire,t Danae Christodoulou,+ Ingeborg Hanbauer,l George W. Cox,$ Francoise Lava1,ll Jacques Laval, * John A. Cook,# Murali C. Krishna? William G. DeGraff? and James B. Mitchell# Chemistry Section, Laboratory of Comparative Carcinogenesis, National Cancer Institute, Frederick Cancer Research and Development Center, Frederick, Maryland 21 702, Laboratory of Chemical Pharmacology, National Heart, Lung, and Blood Institute, Bethesda, Maryland 20892, Laboratory of Experimental Immunology, Biological Response Modifiers Program, National Cancer Institute, Frederick Cancer Research and Development Center, Frederick, Maryland 21 702, Groupe "Radiochimie de I'ADN", U 147 INSERM, Institut Gustave-Roussy, 94805 Villejuif Cedex, France, Groupe "Reparation des lesions chimio- et radioinduites", UA 147 CNRS and U 140 INSERM, Institut Gustave-Roussy, 94805 Villejuif Cedex, France, and Radiobiology Section, Radiation Biology Branch, National Cancer Institute, Bethesda, Maryland 20892 Received January 7, 1994'

The critical regulatory function of nitric oxide (NO) in many physiologic processes is well established. However, in an aerobic aqueous environment NO is known to generate one or more reactive and potentially toxic nitrogen oxide (NO,) metabolites. This has led t o the speculation that mechanisms must exist in vivo by which these reactive intermediates are detoxified, although the nature of these mechanisms has yet to be elucidated. This report demonstrates that among the primary bioorganic products of the reaction of cellular constituents with the intermediates of the NO/O2 reaction are S-nitrosothiol (S-NO) adducts. Anaerobic solutions of NO are not capable of nitrosating cysteine or glutathione, while S-NO adducts of these amino acids are readily formed in the presence of 0 2 and NO. Investigation of the kinetics for the formation where of these S-NO adducts has revealed a rate equation of d[RSNOl/dt = ks~0[NOl~[023, k s ~= o (6 f 2) X lo6M-2 s-l, a value identical to that for the formation of reactive intermediates in the autoxidation of NO. Competition studies performed with a variety of amino acids, glutathione, and azide have shown that cysteine residues have an affinity for the NO, species that is 3 orders of magnitude greater than that of the nonsulfhydryl amino acids, and > l o 6 times greater than that of the exocyclic amino groups of DNA bases. The dipeptide alanyltyrosine reacts with the intermediates of the NO/O2 reaction with an affinity 150 times less than that of the sulfhydryl-containing compounds. Furthermore, Chinese hamster V79 lung fibroblasts depleted of glutathione display enhanced cytotoxicity on exposure to NO. Together, these results suggest that the S-NO adduct of glutathione may represent a physiological scavenger of NO, species and that enzymes containing cysteine residues critical to their function may be subject t o inhibition by reactive intermediates generated in the NO/O2 reaction.

Introduction The role of nitric oxide (NO) as a mediator of a variety of physiological functions is surprising, due to the toxicity normally associated with this molecule in aerobic environments ( 1 , 2). Although NO is a major participant in a number of physiological functions, such as blood pressure regulation and neurotransmission,it has also been shown to promote mutations by deamination of DNA bases (37). The latter effect was shown to be caused by reactive species generated in the NO/O2 reaction (3-7). Furthermore, such reactive nitrogen oxide (NO,) species can inhibit

* To whom correspondenceshould be addressed at Building 538,Room 205E, Frederick Cancer Research and Development Center, Frederick, MD 21702. + Laboratory of Comparative Carcinogenesis, NCI-FCRDC. t National Heart, Lung, and Blood Institute. 8 Laboratory of Experimental Immunology, NCI-FCRDC. I U 147 INSERM, Institut Gustave-Roussy. * UA 147 CNRS and U 140 INSERM, Institut Gustave-Roussy. # Radiation Biology Branch, NCI. Abstract published in Adoance ACS Abstracts, June 1, 1994.

certain enzymes ( 4 9 ) . The irreversible inhibition of the catalytic activity of cytochrome P450 by NO has been proposed to be mediated by NO, (8). This proposition is based on the observation that the inhibitory effects on the catalytic activity could be prevented by the inclusion of bovine serum albumin in the reaction mixture, presumably as a result of scavenging by albumin of the NO, species (8). The cytotoxic effects of NO have been suggested to result, in part, from S-nitrosation of glyceraldehyde-3phosphate dehydrogenase and its subsequent self ADPribosylation and inhibition (9). Stamler et al. (10)' have demonstrated that S-nitrosothiol (S-NO) adducts of human serum albumin are formed from endogenously generated NO and have speculated that such adducts may Abbreviations: ABTS,2 , 2 ' - a z i n o b i s ( 3 - e t h y l ~ ~ ~ ~ ~ ~ s ~ o n i c acid);DMF', dimethylformamide;lc,linear correlationcoefficient(Pearson product moment); SCE, saturated calomel electrode; DENNO, 1,ldiethyl-2-hydroxy-2-nitrosohydrazine, sodium salt;S-NO, S-nitrosothiol; (TMMP)Cl, tetra(N-methyl-4-pyridy1)porphyrin chloride; BSO, Lbuthionine sulfoximine.

This article not subject to US.Copyright. Published 1994 by the American Chemical Society

520 Chem. Res. Toxicol., Vol. 7, No. 4, 1994

serve as a source of circulating NO for the regulation of vascular tone. Such findings suggest that the formation of S-NO complexes may have several roles in vivo. The elucidation of the chemical mechanism of the formation of these S-nitrosocysteineadducts may yield more insight into the physiologic functions of NO and NO,. The nitrosation of thiols by various nitrosating agents under a variety of conditions has been studied (11). In neutral anaerobic solutions,NO does not react withvarious thiol-containing compounds to form S-nitroso adducts; under these conditions, however, the thiol is slowly oxidized to the corresponding disulfide (12). Yet in the presence of acidic nitrite solutions, NO2, or other nitrogen oxide intermediates generated from the NO/Oz reaction in the gas phase, rapid nitrosation of thiols and amines occurs (11,12). However, the nitrogen oxide species generated in the autoxidation of NO in aqueous solution displays fundamentally different chemistry than the nitrosating agents previously studied (3,131. In particular, competitive kinetics studies would appear to rule out the involvement of NO2, NO+, or peroxynitrite anion in the chemistry mediated by intermediates generated during the autoxidation of NO in neutral aqueous solution (3). Comparison of the kinetics results (3) with published rate constants for N203 demonstrates that the latter intermediate, formed in weakly acidic nitrite solutions, displays different selectivity toward various substrates than the intermediates generated during NO autoxidation under neutral conditions. Examination of the chemical reactivity of these nitrosating species toward other biological compounds may, therefore, provide insights into the biological targets of these intermediates. This report demonstrates that, under aerobic conditions, the formation of S-nitrosocysteine results from reaction of the amino acid with the intermediates generated in the NO/O2 reaction and that cysteine-containing peptides have a higher affinity for these reactive intermediates than other biologic components.

Materials and Methods Chemicals. Sodium azide, potassium superoxide, and nickel acetate were obtained from Aldrich Chemical Co. (Milwaukee, WI). Tetra(N-methyl-4-pyridy1)porphyrinchloride [(TMPP)Cl] was purchased from Midcentury (Posen, IL). Potassium ferrocyanidewas purchased from Fisher ScientificCo. (Fairlawn, NJ), and 2,2'-azinobis(3-ethylbenzothiazoline-6-sulfonicacid) (ABTS) and the various amino acids examined were obtained from Sigma Chemical Co. (St. Louis, MO). The above-named reagents were used without further purification. NO gas was purchased from PotomacAirGas (Frederick,MD). Peroxynitrite anion (OONO-) was prepared as previously described (13). NO Preparation and Assay. NO solutions were prepared as previously described (13). In brief, buffer solutions were degassed by vacuum for 1 min per milliliter of buffer. The headspacewas then replaced with argon. NO, purified by passage through a 1 M sodium hydroxide solution,was bubbled through the degassed buffer. The resulting solutions were assayed for NO content by noting the absorbance changes of an air-saturated solution of ABTS,monitored at 660 nm (ACNO= 12 000M-l cm-l), and 750 nm ( A ~ N o= 15000 M-1 cm-1). There was a 1:l ratio between ABTS+and NO (13). This assay has been validated by comparison with the more commonly employed oxymyoglobin assay (14). SpectrophotometricTechniques. Absorbancechangeswere measured with a Hewlett-Packard Model 8451 diode-array spectrophotometer. Absorbancechangesat 338nm were followed in cuvettesreferenced againsta solution exposed to the identical

Wink et al.

concentration of NO, to prevent interference from nitrite. Stopped-flow experiments were carried out with a Hi-Tech ScientificMulti-Mixingstopped-flowspectrophotometer,Model SF-BlMX, with an IS 1.0 Rapid Kinetics Software Suite (HiTech Scientific Limited,Salisbury, Wiltshire, England) (13).All experiments were performed at room temperature (22 f 1 O C ) in 10mM phosphate buffer (pH7.4)with concentrationsas noted. The pH was invariant thoroughout the reaction course. NO Electrode Technique. The method for the detection of NO with a microsensor was a modification of that previously described (15,16). Briefly: One gram of (TMPP)Cl and 0.5 g of nickel acetate were placed in 50 mL of dimethylformamide (DMF) and allowed to reflux overnight. The solid was filtered and dried with diethyl ether (Chempure, Houston, TX). The resultant solid (200 mg) was then dissolved in 0.1 M sodium hydroxide. The electrochemical experiments were performed with a PAR 273 potentiostat/galvanostat(EG&G,Princeton,NJ) using the Model 270 software. The porphyrin was coated onto a glassy carbon fiber, 35 pm in diameter and 100-190 pm long, purchased from Medical Services (Greenville,NY). Controlled potential coulometryat +0.7 V for 4 min [total coulombs = (2-7) X 1od Cl was used to coat the porphyrin to the electrode surface. The electrode was dried at 85 "C and dipped in a 4 % Ndion solution (Aldrich)for about 3 s. This dry-dip cycle was repeated 5 times. Various aliquotsof NO were added to a 0.1 M phosphate buffer solution (pH 7.4)to yield final concentrationsof 0.5-50 pM, and the electrode response was monitored at either 0.7 or 0.85 V versus the saturated calomel electrode (SCE) using amperometry techniques. The current for the electrode was directlyproportionalto NO concentration,with 1 nA elicited by 1.7 pM NO. The disappearance of NO was monitored using chronoamperometry. The decay curve at 1 and 5 pM NO indicated half-lives for NO as predicted (13,17,18). The rate of disappearance of the NO signal was not affected by addition of 1 mM cysteine or glutathione. Cell Survival Experiments. Chinese hamster V79 lung fibroblasts were cultured in F-12 medium supplemented with 10% (v/v) fetal calf serum and antibiotics. Cell survival was determined by clonogenicassay, with platingefficienciesranging from 85% to 95%. Stock cultures containing exponentially growing cells were trypsinized,rinsed, and plated as previously described (19, 20). The NONOate complex, DEA/NO (1,ldiethyl-2-hydroxy-2-nitrosohydrazine, sodiumsalt),was prepared as previously reported (20). Stock solutions were prepared by dissolving the solid DEA/NO in basic aqueous solution (pH = 11.5)to yield a final concentration of 10 mM (20).Concentrations were checked via the characteristicchromophoreat 248 nm (e = 6500 M-1 cm-1). For cellular GSH depletion, V79 cells were treated with 0.5 mM L-buthionine sulfoximine (BSO)for 15 h. The glutathione levels were measured as previously described (19). Both BSOtreated and control cells were rinsed and then exposed to various concentrationsof DEA/NO for 1 h as previously described (20). Cells were rinsed twice, trypsinized, counted, and plated in triplicate for clonogenic survival. Plates were incubated for 7 days at 37 O C in a humidified atmosphere containing 5 % COz, after which the colonieswere fixed with methanol/aceticacid 3:l (v/v), stained with crystal violet, and counted.

Results and Discussion Kinetics for S-NitrosothiolFormation. When aerated solutions of 1 mM glutathione or cysteine were exposed to 0.06 mM NO and 0.27 mM 02 (211, the rapid formation of a species with an absorbance with ,A, = 338 nm was observed (Figure 1). This absorbance maximum corresponded to that reported for the S-NO adduct (10, 11). From a plot of absorbance a t 338 nm versus NO concentration an extinction coefficient (per molar NO) of 1400 f 200 MNO-1cm-1 was calculated for the glutathione adduct (Figure l A , inset) and 1300 f 200 M ~ 0 - cm-l l for

Chem. Res. Toxicol., Vol. 7,No. 4, 1994 521

Mechanism of S-NO Formation by NO

, 0.08 ,

,

0.08

I

0.04 (0

0.02 0.04

\

300 0

50 [NO] W) loo

0.02

150 I

15

o

20

5

io

15

za

time (sec) 350

400

450

500

3.5

P

,

I

-

0.10

I

VI v

0.08

111

n

0.06

Y

0.04

0.02 0.00 300

350

400

450

500

A (nm) Figure 1. (A) Spectral changes that occurred when various aliquots of an NO stock solution were added via air-tight syringe to a stirred air-saturated 0.01 M phosphate buffer solution (pH 7.4)containing1 mM glutathioneto yield final NO concentrations of 0.05, 0.1, or 0.15 mM NO. Spectra were taken 3 min after addition of NO. The inset is a plot of absorbance at 338 nm versus NO concentration (slope = 1400 f 200 M-l cm-1). (B) Spectral changes observed when NO was introduced as described above to yield a final concentration of 0.05, 0.1, or 0.15 mM to an air-saturated 0.01 M phosphate buffer solution (pH 7.4) containing 1mM cysteine. The inset is a plot of absorbance at 338 nm versus NO concentration (slope = 1300 200 M-l cm-9. Each spectrum was referenced against a buffer solution treated with the same amount of NO to avoid any interference of nitrite.

*

the cysteine adduct (Figure lB, inset). In the absence of oxygen, no formation of this absorbance band was observed, which is consistent with previous findings (12). When the formation of S-NO complexes was studied with stopped-flow techniques performed under limiting [ 0 2 1 conditions, a first-order absorbance increase was observed in the presence of 1 mM cysteine (Figure 2A) or 1 mM glutathione (Figure 2B). A plot of kobs versus the square of the NO concentration was linear, with a slope of (7 f 1)X 106M-2 s-l (Figure 2C). Therefore, the kinetics of the reaction are identical to those for nitrite formation, as well as those for the nitrosation and oxidation reactions that occur during the autoxidation of NO (13,17,18). It should be noted that these reaction rates were independent of thiol concentration within the range of 1-10 mM. Furthermore, when NO was limiting ([NO] = 0.09 mM), 1 mM solutions of glutathione or cysteine containing 1 mM 02 also showed an increase in absorbance a t 338 nm (data not shown). These temporal absorbance changes were second-order, yielding traces similar to those previously reported for nitrite formation and ferrocyanide oxidation under analogous conditions (13). The hob was (5 f 1) X lo3 M-' s-l as derived from the extinction coefficient from Figure 1, which, when divided by the oxygen concentration, indicated a rate constant of (5 f 1) X 106 M-2 5-1. This is consistent with the rate constant determined from Figure 2. These values for the rate constants of S-nitrosothiol formation are consistent with those for the autoxidation of NO in aqueous media (Figure 1; 13, 17, 18).

00

02

01

[NO]'

03

04

(mM)'

Figure 2. (A) Temporal absorbance changes monitored at 338 nm via stopped-flow techniques by mixing an aerobic solution

containingthiol with a specific volume of saturated NO solution to yield final concentrationsof 670pM NO, 40 pM 0 2 , and 1 mM cysteine. Using the Hi-Tech software, first-order line fits were obtained. (B)Temporal absorbance changes, monitored at 338 nm with stopped-flowtechniques, of aerobic thiol solutionswith NO-saturated solutions to yield a final concentration of 670pM NO, 40 pM 02,and 1 mM glutathione. (C)A plot of kat. versus [N0I2for the formation of the S-NO adducts of cysteine and glutathione when 02 concentration was limiting (40pM) in the presence of either 1mM cysteine ( 0 )(slope = 7.3 X 108 M-12 8-1, IC = 0.986)or 1 mM glutathione (0) (slope = 8.1 X 108 M-12 8-1, IC = 0.998). The kinetics for the formation of the S-NO adducts indicate that the nitrosation of these amino acids proceeds through the intermediates generated during the NO/Oz reaction. The lack of appreciable formation of S-NO adducts under anaerobic conditions (12) suggests that a primary mechanism for the formation of these S-NO adducts in vivo might be the direct nitrosation by intermediates in the NO/Oz reaction. One possible reaction pathway at low NO concentrations is nitrosation of the sulfhydryl group mediated by an intermediate such as the peroxynitrite radical (3,13). If such were the case, however, the reaction kinetics should be first-order, not second-order, with respect to NO. Therefore, the disappearance of NO would be expected to be accelerated in the presence of cysteine. To test this possibility, the kinetics for the autoxidation of NO were studied at low NO concentrations (1-5 pM). At low NO concentrations the kinetics of the S-nitrosation reaction could not be evaluated by UV-visible techniques because the molar extinction coefficient of the S-nitroso adduct is not sufficiently great to allow reproducible measurement. Therefore, it was necessary to use a porphyrin-coated electrode with a Nafion coating similar to that previously described (15,16). The response of this electrode to NO was not affected by nitrite, glutathione, or cysteine (1mM). The disappearance of NO in buffered solutions (initial concentrations of 1-5 pM NO) indicated a half-life consistent with the kinetics describing the NO/Oz reaction (13, 17, 28). In the presence of 1 mM cysteine or glutathione, there was no effect on the rate of disappearance of NO. These dataindicate that the sulfhydryl groups do not intercept any intermediates in the rate-limiting

Wink et al.

522 Chem. Res. Toxicol., Vol. 7,No. 4, 1994

steps of the NO autoxidation reaction, such as the peroxynitrite radical or the NO dimer, under these conditions. The kinetics clearly show that the reaction of two NO molecules with one of 02 must occur before the formation of an S-NO adduct.

2N0 + 0, NO,

+ RSH

-

-

a,

NO,

RS-NO

(1)

+ H+

0

(2)

+ H,O

-

NO;

(3)

A comparison of the selectivities of NO, for S-NO formation (eq 2) versus hydrolysis (eq 3) [which proceeds through a unimolecular process (13) much like the hydrolysis of N2O3 and N204 to form nitritehitrate, for which the reported rate constant is 1 X lo3 s-l (22)l indicates a selectivity ratio of 10 000 M-':5000 M-l:l for g1utathione:cysteine:water.These ratios allow the determination of the rate of hydrolysis of NO, compared to 2 The product ratio between reactions leading to RSNO (eq 2) or to NO2- (eq 3) may be expressed in terms of their individual rate equations in analogy with a Lineweaver-Burk plot (33), as follows:

l/[RSNOI = l/[RSNO],

+ (kd(k,[RSHl))(l/[RSNOl,)

(4)

where [RSNO] is the concentration of S-nitrosothiol formed from a given [RSH] and [RSNOlb is the concentration of S-nitrosothiol formed at "infinite" [RSH]. [RSNO! = Abs/cm and [RSNO], = Absde-. Substituting these expressions into eq 4 results in the following: 1/Abs = l/Abs,

+ (kd(k,[RSH]))(l/Absi)

(5)

A plot of l/[RSNOl vs 1/[RSH] ,such as that displayed in Figure 3A, has y-intercept = l/Absband slope = (ks/kz)(l/Absb).Smce thex-intercept = -(y-intercept/slope), the x-intercept may also be expressed as -k2/k3. Equation 5 may be rearranged to: Absb/Abs = 1+ (ks/(k2[RSHl))

(6)

At a thiol concentration ([RSHIw) at which half the NO. intermediate is intercepted by RSH to form RSNO [i.e., Abs = 0.5(Absi)Itthe following is true: [RSH], = (k2/k,) and

n

6

v

These results are consistent with our previousresults which demonstrated that the nitrosation of amines proceeds by the same rate equation as other oxidative reactions (formation of nitrite, oxidation of ferrocyanide) involving the reactive intermediates formed in the NO/O2 reaction (13). Selectivity of NO, for Cysteine and Glutathione. In a previous study of the autoxidation of NO, competitive kinetics studies were conducted to determine the relative reactivity for oxidation and nitrosation reactions by these nitrogen oxide intermediates (13). To determine the relative rates of reaction for sulfhydryl-containing compounds toward NO,, a series of competitive kinetics experiments was performed. Aliquots of NO solutions were added to stirring aqueous solutions containing the thiol. As the concentration of glutathione or cysteine increased in the presence of 0.06 mM NO, the intensity of the characteristic 338-nm absorbance band also increased. A plot of l/(Absad versus l/[glutathionel or l/[cysteinel was linear (Figure 3A). For the plots inFigure 3A, the -(x-intercept)-' (X,)represents the concentration of glutathione or cysteine required for the formation of half the maximal S-NO adduct concentration.2 At this substrate concentration, 50 % of the reactive intermediate forms S-NO (eq 2), while the other 50% is hydrolyzed to nitrite by a competing pathway: NO,

C O

n L

:. = -l/x-intercept

(7)

-12 -10

-8

-6

-4

-2

2

0

4

6

8

10

12

(mu-')

[RSH]-l 300

-7

250

0 -0.1 -2

4

150

3

100

-0.1

0.0

0.1

[Azide] (

0.0

0.1

[Azide]

r

n

0.2

/ M

0.3

y

0.4

4

0.5

(mM)

Figure 3. (A) Double-reciprocalplot of the absorbance changes at 338 nm with various glutathione (0)and cysteine ( 0 ) concentrationswhen [NO] = 65 pM in an air-saturated phosphate buffer (pH 7.4)(glutathione, slope = 1.9 mM, IC= 0.97;cysteine, slope = 3.4 mM, IC = 0.95).The inset is a double-reciprocalplot of the absorbance changes at 474 nm with various alanyltyrosine concentrations ([NO] = 65 pM) in an air-saturated phosphate buffer, pH 7.4 (slope = 411 M, IC = 0.97). (B)A plot of the quenching of azide versus the reciprocal of the resulting absorbance at 338 nm when [NO] = 65 pM in an air-saturated phosphate buffer (pH7.4)wherethe concentrationsof glutathione (0) or cysteine (0) were 1 mM (glutathione, slope = 250 mM-l, IC = 0.99;cysteine, slope = 500 mM-l, IC = 0.99).The inset is a plot of the quenchingof azideversusthe reciprocal of the resulting absorbance at 338 nm when [NO] = 65 pM in an air-saturated phosphate buffer (pH7.4)and the concentrationof alanyltyrosine was 65 mM (slope = 590 mM-l, IC = 0.98). S-NO formation. The percent of total product formation represented by thiol nitrosation is determined simply by multiplying the ratio by the concentration of the thiol present. For example, in the presence of 2 mM cysteine or 2 mM glutathione I(0.002 M) X (5000 M-l):l is 10:11, >90% of the NO, species available will react with the sulfhydryl group of the amino acid to form the S-NO adduct. Azide was used to probe the thiol nitrosation reaction in order to further investigate the reactivity of the NO, intermediate. Azide has previously been observed to compete for the intermediates in the NO/O2 reaction (13) as well as other nitrogen oxide intermediates (10). When increasing amounts of sodium azide were added to an airsaturated solution containing either 1mM glutathione or 1mMcysteine and0.08 mM NO, the increase in absorbance at 338nm was markedlyreduced. Aplotof l/Abs3%versus [azide] was linear, yielding a value for Xi (product of the slope X the absorbance at 338 nm in the absence of azide) of 0.1 mM for cysteine and 0.2 mM for glutathione. These

Mechanism of 5'-NO Formation by NO

values represent the concentration at which the S-NO formation is competitively quenched to 50% of that observed in the absence of azide (the S-NO adduct itself does not react with azide concentrations as high as 10 mM). On the basis of these results, the relative selectivity ratio for azide:glutathione:cysteine:Fe(CN)&:water = lo6 M-l:104 M-% X lo3 M-l:500 M-1:l.3 Another potential pathway that might be relevant to a biological system involving NO is the formation of S-NO by peroxynitrite anion (OONO-) formed from superoxide and NO. Exposure of solutions of 1 mM glutathione or 1mM cysteine to 150pM OONO-resulted in no significant increase in absorbance at 338 nm, indicative of S-NO; thus the formation of the S-NO adduct involves the interaction of NO with 0 2 , not superoxide. Similarly, cysteine did not react with OONO- to form an S-NO adduct as monitored by UV-visible spectrophotometry. Such results are consistent with those reported by Radi et al. (23)when characterizing the oxidation of sulfhydryl groups by peroxynitrite anion. Together, these observationssuggest that a likely mechanism for the formation of S-NO adducts in vivo is through intermediates generated by the NO/O2 reaction at neutral pH. To determine if other amino acids appreciably scavenge the NO/Oz intermediate, competitive kinetics experiments were performed using as a probe the oxidation of Fe(CN)& during the NO/O2 reaction (13). Solutions of 2 mM Fe(CN)& containing various concentrations of different amino acids were exposed to 0.1 mM NO. A survey of the amino acids serine, valine, tryptophan, glycine, methionine, histidine, phenylalanine, glutamate, arginine, and lysine (final concentrations of 200 mM) a t pH 7.4 revealed no appreciable decrease in oxidation of Fe(CN)&. Since the ratio of ferrocyanide oxidation to nitrite formation from the intermediates of the NO/Oz reaction is 500 M-l:1(13), the selectivity for these amino acids as compared to nitrite formation is