Superoxide radical initiates the autoxidation of dihydroxyacetone

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 254, No. 2, May 1, pp. 547-551,1987

Superoxide

Radical Initiates the Autoxidation TADAHIKO

Department

of Biochemistry, Received

August

MASHINO Duke

University

4,1986,

AND Medical

and in revised

IRWIN

FRIDOVICH2

Center, form

of Dihydroxyacetone’

Durham,

North

November

Carolina

27710

l&l986

The aerobic xanthine oxidase reaction causes the cooxidation of dihydroxyacetone in a process which is strongly inhibited by superoxide dismutase but not by catalase, HO. scavengers, or iron-inactivating chelating agents. Several molecules of the sugar can be oxidized per 0; introduced. A free radical chain mechanism, in which 0; acts both as an initiator and as a chain propagator, is proposed. Simple sugars capable of tautomerizing to enediols may now be added to the list of biologically relevant targets for 0;. 0 1987 Academic

Press, Inc.

The toxicity of 0; and the protective effects of SOD3 have been briefly reviewed (1). Because physicochemical studies of 0; have indicated that it is not a very reactive radical (2, 3), many have explained the toxicity of 0; solely in terms of its ability to generate the vastly more reactive hydroxyl radical. For this reason it has seemed important to gather data concerning direct reactions of 0; with biologically relevant compounds (1). Thornalley et al. (4) studied the autoxidation of simple sugars and noted that SOD inhibited partially. This indicates that 0; was, to some degree, functioning as a chain propagating species which, in turn, suggests that 0, might be able to cause the oxidation of such simple sugars. We have chosen to examine this possibility using dihydroxyacetone as the simple sugar and the xanthine oxidase reaction as a source of 0, (5, 6).

MATERIALS

AND

METHODS

Dihydroxyacetone, lithium dihydroxyacetonephosphate, xanthine, DTPA, mannitol, sodium benzoate, cytochrome c type IV, and bovine liver catalase were from the Sigma Chemical Co.; Desferal was from Ciba and EDTA was from Mallinkrodt. Xanthine oxidase was prepared from raw cream by R. D. Wiley (7) and was kindly provided by Dr. K. V. Rajagopalan. Cu,ZnSOD from bovine blood was obtained from Grunenthal, GmbH. The consumption of dioxygen was monitored polarographically with a Clark electrode. The reduction of cytochrome c was followed at 550 nm. RESULTS

Cooxidation of DHA bg the xanthine oxiakse reaction. Addition of DHA to xanthine plus xanthine oxidase in 50 mM potassium phosphate, 0.1 mM EDTA, pH 7.8, caused an augmentation of O2 consumption which exceeded that expected on the basis of a simple additive effect. This is illustrated by the data in Fig. 1. Xanthine in the absence of xanthine oxidase, or xanthine oxidase in the absence of xanthine, did not increase the rate of autoxidation of DHA. The rate of cooxidation of DHA by the xanthine oxidase reaction, was calculated by subtracting from the rate seen with DHA -t- xanthine + xanthine oxidase the rates seen with DHA alone and with xanthine plus xanthine oxidase alone. The

1 This work was supported by research grants from the United States Army Research Office, the Council for Tobacco Research, U.S.A., Inc., and the National Science Foundation. ’ To whom correspondence should be addressed. ’ Abbreviations used: DTPA, diethylenetriaminepentaacetate; DHA, dihydroxyacetone; Cu,ZnSOD, the copperand zinc-containing superoxide fiismutase; SOD, superoxide dismutase; Desferal, desferrioxamine; DHAP, dihydroxyacetonephosphate. 547

0003-9861/S? Copyright All rights

$3.00

0 1987 by Academic Press, Inc. of reproduction in any form reserved.

548

MASHINO

?

,?,

r 2 min

0” 5 =L (D 7

x!o.

?

AND

FRIDOVICH

x.0. ’

\

SOD

(fig/ml)

Minutes

FIG. 1. Cooxidation of DHA by the xanthine oxidase reaction. Reaction mixtures contained 50 PM xanthine, 50 mM DHA, 50 mM potassium phosphate, 0.1 mM EDTA, and 6 nM xanthine oxidase in 2.0 ml at pH 7.8 and at 25°C. The arrows indicate the time of addition of individual components; X denotes xanthine and X0 denotes xanthine oxidase.

rate of this cooxidation, which is presented as the excess O2 consumption, increased as a function of the concentration of DHA, as shown in Fig. 2. The apparent saturation of rate at DHA concentrations exceeding 200 m&i was at least partially due to inhibition of xanthine oxidase by these high concentrations of DHA (data not shown), Raising the pH markedly increased the rate of cooxidation of DHA. Thus, the rate

Dihydoxyactone

FIG. 3. Inhibition bv SOD of the cooxidation of DHA. Reaction mixtures contained 50 PM xanthine, 0.1 mM EDTA, 50 mM potassium phosphate, 6 nM xanthine oxidase, 50 mM DHA, and the indicated concentrations of SOD at pH 7.3 and at 25°C.

achieved in the presence of 100 mM DHA, under the conditions of Fig. 2 (pH 7.8), could be equalled at only 4.0 InM DHA at pH 9.0. The roles of 0; in the cooxidation of DHA. If O,, made by the xanthine oxidase reaction, was responsible for the cooxidation of DHA then SOD should inhibit. Figure 3 demonstrates that SOD did strongly inhibit the cooxidation of DHA. Numerous circumstances have been described in which 0: exerts an effect primarily because of its participation with H202 in the production of a powerful oxidant, possibly HO., by the Haber-Weiss reaction (8-18). In that case H202 is also an essential reactant and catalase should inhibit completely. Figure 4 shows that catalase inhibited the cooxi-

(mM)

FIG. 2. The effect of varying the concentration of DHA. Reaction mixtures contained 50 pM xanthine, 0.1 mM EDTA, 50 mM potassium phosphate, 0.1 mM EDTA, 6 nM xanthine oxidase, and the indicated concentrations of DHA at pH 7.8 and at 25°C. The excess dioxygen consumption, occasioned by the admixture of DHA with xanthine plus xanthine oxidase, is presented here as a function of the concentration of DHA.

Cotolose

(fig/ml)

FIG. 4. Effect of catalase on the cooxidation of DHA. Reaction mixtures contained 50 PM xanthine, 0.1 mM EDTA, 50 mM potassium phosphate, 6 nM xanthine oxidase, 50 mbi DHA. and the indicated concentrations of catalase at pH 7.8 and at 25°C.

OXIDATION

OF DIHYDROXYACETONE

dation of DHA only weakly and to a limit of 65% .The rate of excess Oz uptake in the presence of SOD or catalase was calculated by subtracting the rates seen with DHA + SOD or catalase and with xanthine + xanthine oxidase + SOD or catalase from the rate seen with DHA + xanthine + xanthine oxidase + SOD or catalase. Since the Haber-Weiss reaction depends upon catalysis by Fe (III) or Cu (II), it can be prevented by chelating agents such as DTPA or Desferal, which inactivates these metals. Moreover, if HO. is an essential intermediate in a given process then compounds which react rapidly with HO *, but not with 0; or H202, will also inihibit. Mannitol and benzoate are among the many compounds which have been used to scavenge HO *. The data in Table I demonstrate that the cooxidation of DHA was unaffected by mannitol, or by replacing EDTA by Desferal or DTPA. Benzoate (5.0 mM) inhibited weakly. It follows that most of the cooxidation of DHA by the xanthine oxidase reaction was due to a direct action of 02 on DHA. In the event that 02 was the species directly attacking DHA, then one might expect an electrostatic repulsion to prevent attack upon dihydroxyacetonephosphate

TABLE

I

EFFECTSOF HO - SCAVENGERSAND AGENTSONTHE COOXIDATIONOF

Reaction Complete” Complete Complete Complete Complete Complete Complete Desferal Complete DTPA

+ + -

mixture

xanthine oxidase xanthine 100 mrd mannitol 5 mM benzoate EDTA EDTA + 0.1 mM

CHELATING DHA Relative excess 02 consumption (o/o) 100 0 0 106 80 3 109

- EDTA

+ 0.1 mM 94

“The complete reaction mixture contained 59 &iM xanthine, 0.1 mM EDTA, 59 mM potassium phosphate, and 6 nM xanthine oxidase at pH 7.8 and at 25°C.

549

BY 0,

Dihydroryocetone

(mM1

FIG. 5. Chain length as a function of the concentration of DHA. Reaction conditions were as in Fig. 2 and chain length, expressed as molecules DHA oxidized per 0, introduced, is presented here as a function of the concentration of DHA.

(DHAP). In fact DHAP, unlike DHA, was not detectably cooxidized by the xanthine oxidase reactions, whether or not Mg2+ (10 InM) was present. Chain length. If the cooxidation of DHA was due to an OS-initiated free radical chain reaction then conditions might be found under which several molecules of DHA could be oxidized per 0% produced by the xanthine oxidase reaction. The rate of 02 production was assayed in terms of the rate of reduction of cytochrome c in the presence of saturating levels of ferricytochrome c by applying Em550nm = 21 X 103K’cm-’ (19). The rate of cooxidation of DHA was calculated from the O2 consumption assuming one O2 consumed per DHA oxidized. Figure 5 demonstrates that chain length was a linear function of [DHA] and that chain lengths as high as 6.0 could be achieved. Eflect of chelating agents. Omission of metal chelating agents, such as EDTA, DTPA, or Desferal completely abolished the oxidation of DHA by O,, as shown in Table I. Chelating agents prevent inactivation of xanthine oxidase by trace metal contaminants during its catalytic cycle and that is one possible basis for their effect on the oxidation of DHA. However, under the conditions used in our experiments, xanthine oxidase activity persisted even in the absence of chelating agents (data not shown); so another explanation is required.

550

MASHINO

AND

Thornalley et al. (4) noted that chelating agents suppressed the autoxidation of (Yhydroxycarbonyl compounds and proposed that this reflected catalysis, by trace metals, of the reaction between an enediol oxy radical and 02. We propose that 02 acts both as an initiator and as a chain propagator in the oxidation of DHA (see reaction scheme below). In that case, rapid consumption of 0; by metal-catalyzed reactions [4] and [5] (below) would both shorten chain length and decrease the rate of initiation and thus have the effect of inhibiting the oxidation of DHA. DISCUSSION

DHA was cooxidized by the xanthine oxidase reaction and several lines of evidence indicate that 02, rather than HO * derived from metal-catalyzed Haber-Weiss reacOH0 OH I II Hc-c-cHe I H

H

FRIDOVICH

tion, was the attacking species. Thus, SOD inhibited strongly and completely; catalase inhibited Oz uptake only slightly more than the 50% which would be expected on the basis of the formation of H202 as the stable product of OZ reductions; mannitol, which scavenges HO * did not inhibit; DTPA and desferrioxamine, which make iron unavailable to catalyze the Haber-Weiss reaction, did not inhibit; and finally, the anionic phosphate derivative of DHA, i.e., DHAP, was not cooxidized by the xanthine oxidase reaction, in accordance with the expectations for an anionic attacking species. Since several molecules of DHA were oxidized per 0, introduced we envision a free radical chain oxidation of DHA which can be both initiated and propagated by 02. The following reactions represent a feasible mechanism for such a process:

OHOHOH I I I Hc-c=cH

Dl

H

OHOHOH

I I I HC-CCCH+~,=HO,+HC-CCCH

PI

HH OHOHO [31

HC-C=CH+O HH OH0

OHOHO

0

H~-~=~H+O-(~)H~-~--BH+HO2 H H

r41

2

I

0; + 02 + 2H+ (y) H202 + O2 OHOHO

I

I

I

OH0

I

termination 0

II II

2H;-C=CH+H;-C-CH+H;-C=CH

r51

OH OH OH

I

I

I

El i

Chelating agents, by preventing catalysis of reactions [4] and [5] by trace metals (Me”), would diminish chain termination and thus increase the oxidation of DHA. Chelating agents abound in tivo so we anticipate that this process could occur within living cells.

Complete decomposition of H202, by catalase, should return 50% of the dioxygen consumed in producing that H202. Yet we observed 65% inhibition of excess Oz uptake by catalase (Fig. 4). This can be explained by reactions [2] and [4]. In the absence of

OXIDATION

OF DIHYDROXYACETONE

DHA each 0; made by the xanthine oxidase reaction would give rise to l/2 HzOz, by the dismutation reaction (reaction [5]). In the presence of DHA each 0, can give rise to somewhat more than l/2 H202, with a limit of 1.0 H202 per 02, because of reactions [2] and [4]. Thus, the presence of DHA will increase the amount of H202 produced by the xanthine oxidase reaction and this will not be entirely corrected for by subtracting the sum of the O2 uptakes due to xanthine oxidase + xanthine and that due to DHA autoxidation. The thesis that 0; does find targets within living cells was supported by enumeration of those cases in which 0, has been shown to directly attack biologically relevant molecules (1). We may now add simple sugars, capable of tautomerizing to enediols, to the list of potential targets for 0, and must consider that free radicals of these sugars, generated as a consequence of their interaction with 02, could then cause damage by reacting with other cell components. REFERENCES 1. FRIDOVICH, l-11. 2. BIELSICI, Amer. 3. SAWYER, Chem.

I. (1986)

Arch.

Biochem

B. H. J., AND RICHTER, Chem Sot. 99,3019-3023. D. T., AND VALENTINE, Res. 14.393-400.

Biophys. H. W. (1977) J. S. (1981)

247, J.

ACC.

8. 9. 10. 11.

12.

13.

14.

BY 0;

551

THORNALLEY, P., WOLFF, S., CRABBE, J., ANJJ STERN, A. (1984) B&him Biophys Ada 797,276~287. MCCORD, J. M., AND FRIDOVICH, I. (1968) J. Biol. Chem. 243,5753-5760. FRIDOVICH, I. (1970) J. Biol. Chem 245,4053-4057. WAUD, W. R., BRADY, F. O., WILEY, R. D., AND RAJAGOPALAN, K. V. (1975) Arch. Biochem. Biophys. 169,695-701. BEAUCHAMP, C., AND FRIDOVICH, I. (1970) J. Biol. Chem. 245,4641-4646. GREGORY, E. M., AND FRIDOVICH, I. (1974) J. Batteriol. 117,166-169. ELSTNER E. F., AND KONZ, J. R., (1974) FEBSLett. 45,18-21. LOWN, J. W., BEGLEITER, A., JOHNSON, D., AND MORGAN, A. R. (1976) Cand J. Biockem. 54, 110-119. MIEYAL, J. J., ACKERMAN, R. S., BLUMER, J. L., AND FREEMAN, L. S. (1976) J. BioL Cherry 251,34363441. FONG, K.-L., MCCAY, P. B., POYER, J. L., MISRA, H. P., AND KEELE, B. B. (1976) Chem. Biol. Interact. 15, 77-89. MCCORD, J. M., AND DAY, E. D., JR. (1978) FEBS Mt. 86,139-142.

15. HAJ.UWEL.L, B. (1978) FEBS L&t. 96,238~242. 16. DIGUISEPPI, J., AND FRIDOVICH, I. (1980) Arch. Biochem. Biophys. 205,323-329. 17. SAGONE, A. L., DECKER, M. A., WELLS, R. M., AND DEMOCKO, C. (1980) B&him Biophys. Acta 628, 90-97. 18. FRIDOVICH, S. E., AND PORTER, N. A. (1981) J. Biol. Chem, 256,260-265. 19. MASSEY, V. (1959) B&him Biophys. Acta 34,255256.