THEJOURNAL BIOLO(i1CALCHEMISTRY OF 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269, No. 8, Issue of February 25, pp. 5881-5883, 1994 Printed in U.S.A.

Kinetics of Nitric Oxide Autoxidation Aqueous Solution* in

(Received for publication, July 19, 1993, and in revised form, September 13, 1993)

Vladimir G. Kharitonov, AlfredR. Sundquist, and Viiay S Sharma .

From the Department of Medicine, University of California at Sun Diego, La Jolla, California 92093-0652

A kinetic study of the reaction of NO with O2 in aqueous solution has been carried out using a colorimetric methodbasedona pH indicatorandastopped-flow spectrophotometer. The reaction is second order inNO concentration and first order in O2 concentration; the overall third order rate constant (e 0.4) x lo6 ( - ) is 6.3 M s-l. Model calculations forthereaction at estimated physiological concentrations NO and O2 indicate that of its the simple autoxidation NO will not limit diffusion of from the site of production in endothelial cells to a spatially removed target molecule such as guanylate cyclase in myocytes and platelets.

red and a stopped-flow spectrophotometer. Phenolred was chosen because of its pK- of 7.6 and because of the large extinction change that accompanies its protonation , A ( of deprotonated form, 558 nm). SOlutions of NO and 0, in buffer (2.0 m sodium phosphate, pH 7.75) M containing phenol red were prepared in gas-tight syringes by first degassing the buffer for at least l h with argon and then saturating with the respective gas; lower concentrations were achieved by dilution into degassed buffer (Cassoly Gibson, 1975). Solutions of very high[NO] and or low [O,] were used immediately after preparation to minimize the effect of atmospheric 0, on reactant concentration.All kinetic measurements were made at 20 C with a Durrum (Palo Alto, CA) instrument.

RESULTS The reactionof N O with 0 2 in aqueous solution was followed as a time-dependent acidification of the solution using a pH indicator and stopped-flow spectrophotometer. Fig. 1 shows a While a host of functions and effectsof the bioregulator N O double In plot of initial rate of H+ production as a function of is 2, have been described (Moncada et al., 1991; McCall a n d Val- NO concentration. The slope of the least squares line lance, 1992; Traylor and Sharma, 1992; Ribeiro et al., 1993), which demonstrates a second order concentration dependence some basic questions concerning the accessibilityof NO to its on NO. A similar plot for O2 concentration is given in Fig. 2, target enzyme guanylate cyclase remain open. is produced showing a first order dependence. NO A calculated reaction time course was thenfit to experimenin endothelial cells by a specific synthase, and, from there, it tal data as described below. (For further details of numerical can diffuse to neighboring arterial and cardiac myocytes and the heme of guanylate analysis of kinetic data, see Olson (19811.) The following equaplatelets, where it then reacts with of N cyclase to stimulate enzyme activity up200-fold (Moncada et tions were used to calculate concentrations O or O2 at time to t (Le. or al., 1991). However,in addition to being an exceptionally strong [NO], [021t): ligand of ferroheme, N O is redox active andso may also react d[NO] d [H] with O2 to a significant extent, a condition t h a t could limit the (Eq. 2) [Ozl dt dt - k diffusion of NO through an aerobic environment. In aqueous, aerobic solutions,NO autoxidizes toNO2, which rapidly hydrod [O,] I d [Hl k (Eq. 3) - [NO] [O,I lyzes to yield nitrite (Kelm and Schrader, 1990). The overall dt 4 dt 4 reaction in buffer is: A.[Ozlt = 4 A [NO], (Eq. 4) 4N0 + 0, + 2H,O 4H++4NO; (Eq. 1) For reactions in which NO concentration was limiting, EquaTo assess the effect the reaction of NO with O2 on the avail- tions 2 a n d 4 were used; when was limiting, Equations a n d of O2 3 ability of NO at the cellular level, the kinetic study described were used. The calculated value 4 in of the fraction the reaction of this report was undertaken. the following completed at time t , Fcalo was obtained from Although the gas-phase reaction of N O with O2 has been equation: investigated extensively (Olbregts, 19851, studies in aqueous solutions are only few and they disagree with respect to reac(Eq. 5) tion order. Eguchi et al. (1989) and Winket al. (1993) reported that the reaction is second order in O concentration and first N whereas Tahaet al. (1992) reported zero order andwhere the subscript tot indicatesthe initial concentration of order in02, The experimentally observed first order concentration dependence for NO a n d 02, respec- the limiting reactant,N O or 02. value of the fraction of the reaction completed at time t , Fobs, tively. was obtained from the reaction time course: EXPERIMENTALPROCEDURES (Eq. 6) The autoxidation of NO in aqueous solution was studied by rapidly mixing weakly buffered solutions of NO and 0, and following the reacof tion as a change in [H] (see Equation 1) using the pH indicator phenol with A representing absorbanceat 558 nm. The sum square the residuals (Equation was then minimized to obtain 7) a least squares fit and third order rate constant, k ( i is the ith data * This work was supported by National Institutes of Health Grants of HL48014 and HL1358. The costs of publication of this article were point and n is the total number data points, which in every defrayed in part by the payment of page charges. This article must case was >loo). therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact. (Eq. 7) This article is dedicated to the memory of Prof. T. G. Traylor (19251993).
+

5881

5882
"3 0 .

Solution Autoxidation Aqueous in NO

-18.0 I

8
-11.3

180

-12.5

- 10.0

SEC.

200

380

In [NO]
FIG.1. Second order dependence reaction rate on NO conof centration.The initial [021 200 p for each was run; all runs were done in 2 m~ sodium phosphate, pH 7.75, containing phenol red. Concentrations are aRer mixing in the stopped-flow spectrophotometer (see "Experimental Procedures").
-11.07

FIG.3. Example of a third order of the reaction time course fit of NO autoxidation.[NO1 = 25 p ~ LO,] = 200 p ~ all other conditions , ; as in Fig. 1. To improve clarity, only everyfourth data point has been plotted. F = fraction of reaction completed.
1.2 T

E
C

0.7

00
Lo

Lo 0.2

c -0.3

-0.8

L
0

L
4 5

time (sec)
.
- 1 1.5

-1 0.5

-9.5

In

Lo21

FIG.4. The autoxidation of NO under pseudo-first order conditions. The initial concentrations of NO and O2were 500 and 20 p ~ respectively. Measurements were made as described in Fig. 1.

FIG.2. First order dependence of reaction rateon 0% concentration. The initial [NO] was 500 p~ in each case; all other conditions are as in Fig. 1.

An initial estimate of k was obtained from the y-intercept in Fig. 1. An example of reaction time course data and least squares fit is given in Fig. 3. Analyses of over 20 reaction time courses were carried out for the following conditions.1)O2 concentration-limiting (i.e. [02] << [NO]):(a) initial O2concentration constant a t 15 1.1~ NO and ( concentration varied from 100 to 1000 1.1~; b ) initial NO conand centration constant at 500 1.1~ O2 concentration varied from 15to 50 p ~ 2) NO concentration-limiting(i.e. [NO] << [021):(a) . initial NO concentration constant at 25 1.1~ and 0 2 concentrab tion vaned from 150 to 600 p ~ ( ; ) initial O2 concentration constant at 200 1.1~ and NO concentration varied from 5 to 40 VM. The rate constant calculated from the data was 6.3 (2 0.4) X lo6 ("'I2 S ' . In Fig. 4,kinetic data for an experiment is analyzed graphi<< cally. Since [021 [NO], the reaction becomes pseudo-first order with respect to O2 concentration, and theslope of the plot In hA versus time is the pseudo-first order rate constant k', which is related to the third order rate constant, k, as follows ) (see Equation 3 :
k k' = - [NO]'
4

autoxidation of NO wasalso studied by a noncolorimetric method. At relatively high initial NO and O2 concentrations, the reaction could be followed as an increase in absorbance at 354 nm resulting from the formation of nitrite, and this data s-l, consistent with that yielded an average k of 6.9 x lo6 ("'I2 determined by the pH indicator method.
DISCUSSION

(Eq. 8)

The value of k calculated from these data was 6 x lo6 (M-')~ s-', which is within the range of values derived from the numerical analysis (see above). the As validation of the colorimetric method described above,

The kinetic method presented here is simple and allows one to obtain time courses for the reaction ofNO with O2 over a wide range of reactant concentrations. Inall experiments, whether NO or O2 concentration was limiting, the reaction was second order in NO concentration and firstorder in O2 concentration. These results disagree with those of an earlier study which suggested a zero order dependence of the reaction on NO concentrations (Taha et al., 1992). In addition, the rate constant for the reaction of NO with O2 in aqueous solutions, 6.3(? 0.4) lo6 ("'I2 x s-', is -1000-fold larger than that for the gas phase reaction (Olbregts, 1985). Using the rate law (Equation 2) and the value for k determined here, together with estimates of cellular NO and O2 concentrations taken from the literature, a reasonable prediction of the time course of NO autoxidation under physiological conditions can be made. In situ NO levels have been measured " in cerebellar slices, and increases in NO of up to 75 I accompanied electrical stimulation of the tissue(Shibuki and Okada, 1991).Cellular and peripheral O2 concentrations will depend on proximity to blood vessels, but levels greater than that in air-saturated waterat 37 "C are not expected. Withinitial NO and O2 concentrations of 100 n and 230 p ~ respectively, the M ,

NO Autoxidation Solution in Aqueous

first half-life ( i e . the time required for a halving of the NO concentration) for the simple autoxidation of NO is calculated to be 2 h. This value was obtained from both a simulated reaction time course and from the integrated rate equation for a pseudo-second order reaction. The highest reported level for cellular NO that we found in the literature was 4 p~ (Malinski et al., 1993), which, when substituted into the above calculations, yielded a half-life of 3 min for the autoxidation reaction. In the study with cerebellar tissue mentioned above, NO levels dropped within seconds following cessation of the electrical stimulus. Other studies have also indicated that thelifetime of NO in the presence of cells is on the orderof seconds (Palmer et al., 1987; Kelm and Schrader, 1990; Taha et al., 1992). From these considerations, it is clear that autoxidation alone will not significantly impede diffusion of NO through the cellular and extracellular milieu, and that other reactions of NO account for its short lifetime in biological systems. Nitric oxide produced in endothelialcells has several possible fates. Itcan diffuse into the lumen blood vessels where most of will be trapped by oxy- and deoxyhemoglobin in red blood cells or be consumed in the formation of S-nitrosoalbumin (Stamler et al., 1992). A portion of the NO that diffuses to arterial myocytes will bind to the heme of guanylate cyclase, resulting in enzyme activation; additionally, some may be consumed in the formation of the S-nitrosylderivative of glutathione (GSH). In cardiac myocytes, which contain considerable amounts of oxyand deoxymyoglobin, the situation may be different, however (Wittenberg and Wittenberg,1989). Kinetic and thermodynamic considerations suggest that here most of the NO will be trapped by myoglobin before it can react with guanylate cyclase, and, thus, NO that diffuses into cardiac myocytes will have a much lesser effect on guanylate cyclase activity and cGMP levels. Experiments in our laboratory with GSH indicate that NO reacts with GSH to yield the S-nitrosyl derivative only in the presence of O,, and that the rate-limiting step inprocess is this

5883

the autoxidation of NO.' Also, the rate of formation of nitrosoamines from NO and amines appears depend on NO autoxito dation (Challis and Kyrtopoulos, 1979). Although unlikely to influence the bioregulatory action of NO, the reaction ofNO reactions of NO with O2 may nevertheless be a key step in other in biological systems, as indicated by these findings. To summarize: based on the kinetic data obtained here and the current estimates of physiological NO concentrations, the reaction of NO with O2 is expected to be very slow at the cellular level and to influence only marginally the diffusion of NO from the site of synthesis to the target enzyme guanylate cyclase.
Acknowledgments-We helpful discussions. thank Drs. D. Magde and M. P. Doyle for

REFERENCES
Cassoly, R., and Gibson, Q . H. (1975)J. Mol. Biol. 91, 301-313 Challis, B. C., and Kyrtopoulos, S. A. (1979)J. Chem. Soc. Perkin I, 299-304 Eguchi, W., Tanigaki, M., Mutoh, K., and Tsuchiya, H. (1989) Kagaku Kogaku Ronbunshu 15, 1115-1119 Kelm, M., and Schrader, J.(1990)Circ. Res. 66, 1561-1575 Malinski, T., Bailey, F., Zhang, 2. G . , and Chopp, M. (1993)J . Cereb. Blood Flow Metab. 13, 355-358 McCall, T., and Vallance, P. (1992)'12ends Physiol. Sci. 13, 1-6 Moncada, S., Palmer, R. M. J., and Higgs,E.A. (1991)Pharmacol. Reu. 43,109-142 Olbregts, J. (1985) Int.J. Chem. Kinet. 17, 835-848 Olson, J . S. (1981)Methods Enzymol. 7 6 , 6 5 2 4 6 7 Palmer, R. M. J., Femge, A. G . , and Moncada, S. (1987) Nature 327, 524-526 Ribeiro, J.M.C., Hazzard, J . M. H., Nussenzveig, R. H., Champagne, D. E., and Walker, F. A. (1993)Science 260, 539-541 Shibuki, K., and Okada,D. (1991) Nature 349,32&328 Stamler, J. S., Jaraki, O., Osborne, J., Simon, D. I., Keaney, J., Vita, J., Singel, D., Valeri, C. R., and Loscalzo, J. (1992)Proc. Natl. Acad. Sci. U. S. A. 89, 76747677 Taha, Z., Kiechle, E , and Malinski,T. (1992)Biochem. Biophys. Res. Commun. 188, 734739 Traylor, T. G . , and Sharma, V. S. (1992)Biochemistry 31,2847-2849 Wink, D. A., Darbyshire,J.F., Nims,R. W., Saavedra, J. E., and Ford, P. C. (1993) Chem. Res. Toxicol. 6 , 23-27 Reu. Physiol. 51, 857-878 Wittenberg, B. A., and Wittenberg, J. B. (1989) Annu.

V. G. Kharitonov, A. R. Sundquist, and V. S . Sharma, unpublished results.