Copper and Zinc Containing Superoxide Dismutase (Cu, Zn, SOD) can act as a. Superoxide Reductase (SOR) and as a Superoxide Oxidase (SOO).

Transcription

1 JBC Papers in Press. Published on September 25, 2000 as Manuscript M Copper and Zinc Containing Superoxide Dismutase (Cu, Zn, SOD) can act as a Superoxide Reductase (SOR) and as a Superoxide Oxidase (SOO). By Stefan I. Liochev and Irwin Fridovich + Department of Biochemistry, Duke University Medical Center, Durham North Carolina Running Title: Cu, Zn SOD as a SOR and a SOO

2 + Corresponding Author: Tel: ; Fax: ; Footnotes Abbreviations Used: Fe (II), potassium ferrocyanide; Fe (III), potassium ferricyanide; Cu, Zn SOD, bovine copper and zinc containing superoxide dismutase; Mn SOD, human manganese containing superoxide dismutase cloned into Escherichia coli. Abstract 2

3 The copper and zinc containing superoxide dismutase (Cu, Zn SOD) can catalyze the oxidation of ferrocyanide by as well as the reduction of ferricyanide by. It can thus act as a superoxide dismutase (SOD), a superoxide reductase (SOR), and as a superoxide oxidase (SOO). The Mn SOD does not exert SOR or SOO activities with ferrocyanide or ferricyanide as the redox partners. It is possible that some biological reductants can take the place of ferrocyanide and can also interact with Mn SOD; thus making SOR activity a reality for both SODs. The consequences of this possibility vis á vis H 2 production, the overproduction of SODs, and the role of Cu, Zn SOD mutations in causing familial amyotrophic lateral sclerosis are discussed as well as the likelihood that the biologically effective SOD mimics, described to date, actually function as SORs. Keywords Copper, Zinc Superoxide dismutase, superoxide reductase, superoxide oxidase, ferricyanide/ferrocyanide couple, hydrogen peroxide, amyotrophic lateral sclerosis. 3

4 Introduction 4

5 The family of superoxide dismutases (SODs) encompasses enzymes containing Cu + Zn (1), Mn (2), Fe (3), or Ni (4) at their active sites. They provide a defense against oxidative stress by catalyzing the dismutation of into H 2 plus and do so at close to the limit imposed by diffusional encounters (5, 6). Their mechanisms of action is based upon reduction and reoxidation of the catalytic metal center by ; as illustrated in reactions I and II for the case of Cu, Zn SOD. I. E Cu (II) + E Cu (I) + II. E Cu (I) H + E Cu (II) + H 2. If an electron donor, other than, were to reduce the active site Cu (II) to Cu (I) the SOD would then act as a reductant: oxidoreductase ie. as a superoxide reductase or SOR. On the other hand if an electron acceptor were to replace in reoxidizing the Cu (I) to Cu (II) the enzyme would then act as a superoxide oxidase or SOO. That these ruminations can have biological relevance is shown by the findings that desulfoferrodoxin substitutes for SOD in SODnull E. coli by acting as a SOR (7, 8) and by the similar action of neelaredoxin (9, 10). It had been proposed earlier that a manganic porphyrin studied as a mimic of SOD, was actually acting as a SOR (11). It appeared possible that the Cu, Zn SOD might itself act as a SOR and/or as a SOO under special conditions. If it did so that could provide an explanation for the deleterious effects that have been associated with the overproduction or over administration of this enzyme (1214). 5

6 We chose to investigate the ferrocyanide (Fe (II)/ferricyanide (Fe (III) couple as the electron donor/acceptor couple for studies of the SOR/SOO activity of Cu, Zn SOD for several reasons. Thus: the conversion of Fe (II) to Fe (III) can be followed at 420nm (15); Fe (II) is not autoxidizeable to a noticeable degree; Fe (II) has already been shown to reduce the active site Cu (II) of Cu, Zn SOD (16); and the rate of spontaneous reduction of Fe (III) by is slow with a rate constant at 25 o C of ~ 2.7 x 10 2 M 1 s 1 (17). In what follows we demonstrate that Cu, Zn SOD can act as a Fe (II): oxidoreductase ie. as a SOR and as an : Fe (III) oxidoreductase ie. as a SOO and we discuss the possible consequences of these activities. Materials and Methods K 4 Fe(CN) 6 was from: Fisher Scientific; K 3 Fe(CN) 6, J.T. Baker: MOPS and DETAPAC, Sigma; acetaldehyde, Aldrich; catalase and Cu, Zn SOD, Grunenthal Gmbh; and Mn SOD (bacterial), Diagnostic Data Inc. Bovine cream xanthine oxidase was prepared by Ralph Wiley (18). Acetaldehyde was distilled freshly each day. It was used as the substrate for XO in place of xanthine because urate rapidly reduces Fe(CN) 3 6 (19) and also because acetaldehyde could be used at 20mM allowing relatively large fluxes of to be maintained without significant depletion of substrate. Reactions were performed at 23 o C in 50mM MOPS, 0.1mM DETAPAC, 20mM acetaldehyde, 1.0mM Fe (II) or 0.5mM Fe (III), 0.2mg/ml Cu Zn SOD, and enough XO to cause production of 12nmol /min/ml when Fe (II) oxidation was to be followed, and 15nmol /min/ml when Fe (III) reduction was to be followed. The rate of production was measured by replacing Fe (II), or Fe (III), by 20mM ferricytochrome c and decreasing [XO] by a factor of five. The molar rate of cyt c reduction, which could be 100% inhibited by SOD, was equated with the rate of 6

7 production. Fe (II) oxidation or Fe (III) reduction was followed at 420nm using Em 1 cm 1 = 1, 020 (15). Other components, when present are specified in the figure legends. Results Cu, Zn SOD catalyzes the oxidation of Fe (II) by SOR activity with Fe (II) as the reductant, would entail catalysis of the dependent oxidation of Fe (II). Line 1 in figure 1A shows that Cu, Zn SOD alone caused only a stoichiometric oxidation of Fe (II) to Fe (III) whereas provision of a flux of allowed the enzyme to catalyze that oxidation. Line 2 in figure 1A shows that the flux of did not cause the oxidation of Fe (II) until Cu, Zn SOD was added. Thus Cu, Zn SOD can act as a SOR with Fe (II) serving as the reductant of the active site Cu (II). Under the conditions used the rate of Fe (II) oxidation was ~ 45% of the rate of production. Hence 1.0mM Fe (II) was unable to largely outcompete the much lower steady state concentration of as the reductant for the active site Cu (II) and the enzyme was acting simultaneously as a SOD and as a SOR. Effects of Varying the Concentration of Cu, Zn SOD and the Supply of Raising the [Cu, Zn SOD] at a constant [Fe (II)] should cause a directly proportional increase in the rate of the reduction of the active site Cu (II) by Fe (II), but a less than proportional increase in the reduction of active site Cu (II) by because [ ] falls as [SOD] increases. Thus at constant Fe (II) and constant flux of raising [Cu, Zn SOD] should increase the ratio of SOR to SOD activities. Conversely raising [O 2 ] with other factors held constant should favor SOD activity over SOR activity. Hence increasing the rate of production of does not proportionately increase SOR activity because a larger fraction of the is 7

8 eliminated by the SOD reactions. In keeping with those expectations line 3 in figure 1B shows again that Cu, Zn SOD caused a stoichiometric oxidation of Fe (II) and that subsequent addition of graded amounts of XO caused a less than proportional increase in the rate of Fe (II) oxidation. Thus increasing [XO] fivefold, which would increase the flux fivefold, caused only a 2.3 fold increase in the rate of Fe (II) oxidation. Line 4 shows again that a flux of O 2, per se, was not able to oxidize Fe (II) and that subsequent addition of graded amounts of Cu, Zn SOD caused a less than proportional increase in the rate of Fe (II) oxidation, as did raising the [Fe (II)]. These results are in accord with expectations, on the basis of the effects of these manipulations on the [O 2 ], and on ratio of Cu (II) to Cu (I) at the active site, and will be considered further in the discussion. The Effect of Mn SOD Mn SOD does not catalyze the oxidation for Fe (II) by O 2. Hence it could be used to test the effect of [O 2 ] on the SOD:SOR ratio. Thus 0.15mg/ml Mn SOD slowed the oxidation of Fe (II) by the SOR activity of Cu, Zn SOD (compare lines 5 and 6 in figure 1C); while 0.66mg/ml caused further inhibition; which was less than proportional (line 7) because the SOR:SOD ration increases as [ ] decreases. Such high concentrations of Mn SOD were needed also because it was competing for with the Cu, Zn SOD. Catalase, added to 100u/ml, was without effect on the dependent oxidation of Fe (II) by Cu, Zn SOD. Moreover 0.1mM H 2 did not replace the flux of in facilitating the oxidation of Fe (II) by Cu, Zn SOD (data not shown). It follows that there was no detectable oxidation of Fe (II) by the peroxidase activity of Cu, Zn SOD (20, 21). A Superoxide: Ferrocyanide Oxidoreductase Activity of Cu, Zn SOD 8

9 is known to reduce Fe(III) with a rate constant of 3 x 10 2 M 1 s 1 (17) and as shown by line 1 in figure 2A the flux of produced by the XO reaction (15 nmoles/ml/min) caused the reduction of Fe(III). Cu, Zn SOD added to 0.002mg/ml inhibited; but adding more did not inhibit further and indeed increased the rate of Fe (III) reduction. Line 2 presents a repetition of this experiment but at 1/5 the XO, hence at 1/5 the flux of O 2. Thus at low concentration, Cu, Zn SOD inhibits the reduction of Fe (III) by by catalyzing the dismutative elimination of O 2 ; while higher [Cu, Zn SOD] catalyzes the oxidation of Fe (III) by ie it acts as a superoxide oxidase (SOO). When the concentration of Fe (III) was decreased to 0.1mM its rate of reduction by the flux of was much lower than that flux due to the loss of to the spontaneous dismutation of O 2. Under these conditions the SOO activity of Cu, Zn SOD was evident at lower concentrations of the enzyme. This is illustrated by line 3 in figure 2. Discussion We have seen that Cu, Zn SOD can catalyze: the dismutation of O 2 ; or the reduction of by Fe (II); or the oxidation of by Fe (III). That is to say that it can act as an SOD, a SOR, or a SOO. While these activities were demonstrated utilizing the decidedly unnatural Fe (II)/Fe (III) redox couple; it is possible that, in the reducing environment of the cell, there is some natural redox pair that can interact with the active site of the Cu, Zn SOD. Since a number of otherwise puzzling observations can be explained on the basis of these multiple activities it may be worthwhile to express the component reactions and their relationships more rigorously. 9

10 SOD activity depends upon the sum of reactions I and II above. SOR activity involves reactions III and II; while SOO activity involves reactions IV and I. III. E Cu(II) + Fe(II) E Cu(I) + Fe(III) IV. E Cu(I) + Fe(III) E Cu(II) + Fe(II) The rates of the component reactions can be written: V I = k I [ECu(II)] [ ] V II = k II [E Cu(I)] [ ] V III = k III [E Cu(II)] [Fe(II)] V IV = k IV [E Cu(I)] [Fe(III)] Under steady state conditions the sum of the rates of the reduction of E Cu (II) must equal the sum of the rates of the oxidation of E Cu(I). Hence V I + V III = V II + V IV. In the presence of a constant flux of and when [Fe(II)] >> [Fe(III)] V IV is negligeable and then V. V I + V III = V II, or equivalently VI. K I [E Cu(II)] [ ] + k III [ECu(II)] [Fe(II)] =k II [ECu(I)] [ ] 10

11 When [Cu, Zn SOD] is increased the V III term increases in direct proportion because Fe (II) is in large excess and is effectively constant; but V I and V II will not increase proportionately because [O 2 ] will fall as [Cu, Zn SOD] increases. It follows that increasing [Cu, Zn SOD] will favor SOR activity over SOD activity. This effect was demonstrated in figure 1B. Similarly raising Fe (II) will also increase the SOR:SOD ratio. Anything that decreases the [ ] will favor the SOR reaction over the SOD reaction. This can be made obvious by rearranging equation VI. Thus: VII. k III [ECu(II)] [Fe(II)] = (k II [ECu(I)]k I [ECu(II)]) [ ] The left side of equation VII must fall as [ ] falls, but the SOR:SOD ratio will rise because decreasing [ ] will increase the [ECu(I)]:[ECu(II)] ratio. This is the case because is the only oxidant of ECu(I) under the conditions specified. We have seen that decreasing [ ] by increasing [Cu, Zn SOD], or by adding Mn SOD, or lowering [XO], increased the SOR:SOD ratio in accord with these deductions. The Fe (II)/Fe (III) couple has here been shown to support the SOR and the SOO activities of Cu, Zn SOD, but not of Mn SOD; because this redox couple interacts with the active site Cu, but not with the active site Mn. If the cell contains redox couples competent to interact with both active sites then both Cu, Zn SOD and Mn SOD could act as SORs and as SOOs. Given that cell cytosols are reducing environments, SOR activity is more likely in vivo than is SOO activity. 11

12 Whether SOD enzymes acts only in the SOD mode, or in the SOR or SOO modes, has an effect on the amount of H 2 produced from O 2. Thus, in the SOD reaction 1/2 H 2 is produced per consumed; while in the SOR mode 1.0 H 2 is the yield per and in the SOO mode no H 2 would be made from O 2. Since we have seen that lowering [O 2 ] favors the SOR mode; we deduce that raising [Cu Zn SOD] would increase H 2 production only in the presence of a reductant capable of reducing the active site Cu (II). If were acting to initiate oxidative chain reactions, then the yield of H 2 per could be significantly greater then 1.0 per O 2. In such a situation SODs would decrease H 2 production. If an endogenous reductant, that can act as a SOR substrate, is an essential molecule, or if its oxidized form is toxic, then we can understand the reports that overproduction (13, 14) or over administration (12) of SOD has deleterious effects. Thus increasing [SOD] increases the SOR:SOD ratio because it lowers [ ] and at the same time increases net SOR action. The neurotoxic effect of the mutant forms of Cu, Zn SOD, that have been associated with the familial amyotrophic lateral sclerosis (22) may be due to SOR activity. Thus the mutated Cu, Zn SODs may be able to catalyze the oxidation of essential reductants within motor neurons by O 2 ; a SOR activity not exerted by the wild type enzyme. is scavenged very effectively by desulfoferrodoxin (7, 8) and by neelaredoxin (9, 10) acting as SORs. The SODmimic Mn (III) TMPyP has also been reported to act as a SOR within E. coli (11). Given that low molecular weight SODmimics are certain to be less discriminating than the SODs themselves with regard to interaction with reductants, it seems likely that the biological effects of most of these mimics are due to SOR, rather than to SOD, 12

13 activity. The results of Offer et al (23), who have reported that the nitroxidecatalyzed oxidation of Fe (II) by a flux of could be inhibited by low [Cu, Zn SOD], but not by high [Cu, Zn SOD], can now be understood in terms of the increase in SOR activity with increasing [Cu, Zn SOD] as described above. It should be stressed that any agent acting as a SOR could be beneficial or detrimental depending on: the nature of the reductant consumed and the oxidized product generated therefrom; the availability of that reductant and the possibility of its regeneration; and on the magnitude of the flux of and the availability of critical targets for attack. Thus not only the toxic effects of the nitroxides, as Offer et al suggest, but also their beneficial actions are explicable on the basis of their SOR activity. References 1. McCord, J.M., and Fridovich, I. (1969) J. Biol. Chem. 224, Keele, B.B., McCord, J.M., and Fridovich, I. (1970) J. Biol. Chem. 245, Yost, F.J., Jr., and Fridovich, I. (1973) J. Biol. Chem. 248, Youn, H.D., Kim, E.J., Roe, J.H, Hah, Y.C., and Kang, S.O. (1996) Biochem J. 318, Klug, D., Rabani, J. and Fridovich, I. (1972) J. Biol. Chem. 247,

14 6. Rotilio, G., Bray, R.C., and Fielder, E.M. (1972) Biol. Chem. Biophys. Acta 268, Liochev, S.I., and Fridovich, I. (1997) J. Biol. Chem. 272, Lombard, M., Fontcave, M., Touati, D., and Niviere, V. (2000) J. Biol. Chem. 275, Lombard, M., Touati, D., Fontcave, M., and Niviere, V., (2000) J. Biol. Chem. in press M Jovanovic, T., Ascenso, C., Hazlett, K.R.O., Sikkink, R., Krebs, C., Litwiller, R., Benson, L.M., Moura, I., Moura, T.T.G., Radolf, J.D., Huynh, B.H., Naylor, S., and Rusnak, F. (2000) J. Biol. Chem. in press M Faulkner, K.M., Liochev, S.I., and Fridovich, I. (1994) J. Biol. Chem. 269, Omar, B.A., Gad, N.M., Jordan, N.C., Striplin, S. P., Russel, W. T., Downey, J.M., and McCord, J.M. (1990) Free Rad. Biol. Med 9, Amstad, P., Peskin, A., Shah, G., Mirault, M.E., Moret, R., Zbinden, I., and Cerutti, P (1991) Biochemistry 30, Scott, M.D., Meshnick, J.R., and Eaton, J.W. (1989) J. Bio. Chem. 264,

15 15. Schellenberg, K.A., and Hellerman, L. (1958) J. Biol. Chem. 231, Rotilio, G., Morpurgo, L., Calabrese, L., and Mondovi, B. (1973) Biochem, Biophys. Acta 302, Zehavi, D., and Rabani, J. (1972) J. Phys Chem. 76, Waud, W.R., Brady, F.O., Wiley, R.D., and Rajagopalan, K.V. (1975) Arch. Biochem. Biophys. 169, Fridovich, I., and Handler, P. (1958) J. Biol. Chem. 233, Hodgson, E.K., and Fridovich, I. (1975) Biochemistry 14, Hodgson, E.K., and Fridovich, I. (1975) Biochemistry 14, Estevez, A. Crow., J.P., Sampson, J.B., Reiter, C., Zhuang, Y., Richardson, G.J., Tarpey, M.M., Barbeito, L., and Beckman, J.S. (1999) Science 286, Offer, T., Russo, A., and Samuni, A. (2000) FASE B J. 14,

16 Figure Legends Figure 1A Cu, Zn SOD Catalyzes the O2 Dependent Oxidation of Fe(II) Reaction mixtures contained 20mM acetaldehyde, 0.1mM DETAPAC 0.2mg/ml Cu, Zn SOD, and 50mM MOPS at ph 7.8 and 25 0 C. Line 1: Fe (II) added to 1.0mM at first arrow and XO added at the second. Line 2: all components present excepting Cu, Zn SOD, which was added to 0.2mg/ml at the arrow. Figure 1B Effect of Varying [XO], [Fe (II)], and [Cu, Zn SOD] Line 3: Buffer, DETAPAC, acetaldehyde and Fe (II) were present at the outset. Cu, Zn SOD added to 0.2mg/ml at first arrow, one fifth the usual amount of XO added at the second arrow, four fifth the usual amount of XO added at the third arrow. Line 4: All components excepting Cu, Zn SOD present at outset. Cu, Zn SOD added to 0.02mg/ml at first arrow and to 0.20mg/ml at second arrow. Additional Fe (II) added to 4.0mM at third arrow. Figure 1C Effect of Mn SOD Reaction mixtures as in fig. 1A. Line 5: Cu, Zn SOD added to 0.04mg/ml at arrow. Line 6: as in line 5 but with 0.15mg/ml Mn SOD present at outset. Line 7: as in line 5 but with 0.67mg/ml Mn SOD present at outset. Figure 2 Cu, Zn SOD catalyzes the Dependent Reduction of Fe (III) Reaction mixtures as in Figure 1A except that 0.5 mm Fe (III) replaced the 1.0mM Fe (II). Line 1: XO added at first arrow, Cu, Zn SOD added to 0.002mg/ml at second, to 0.004mg/ml at third, and to 0.2mg/ml at fourth. Line 2: as in line 1 but one fifth the usual amount of XO added at first arrow, Cu, Zn SOD to 0.001mg/ml at second, and to 0.2mg/ml at third. Line 3: 16

17 as in line 1 but with Fe (III) at 0.1mM. XO added at first arrow, Cu, Zn SOD to 0.04mg/ml at second, and to 0.2mg/ml at third. 17

18

19

20

21