Superoxide Anion Radical (O·2), Superoxide Dismutases, and Related Matters

Abstract

A field of inquiry may be said to have come of age when conclusions initially viewed as remarkable or even unbelievable are accepted as commonplace. Study of the biology of the superoxide anion radical and of related free radicals, and the defenses thereto, has now reached this happy state of maturity. Superoxide and even hydroxyl radicals are now known to be produced in living systems, and elaborate systems of defense and repair, which minimize the ravages of these reactive species, have been described. New members of the superoxide dismutase, catalase, and peroxidase families of defensive enzymes are being found, as are new targets that are modified by O·̄2. In addition, the involvement of O·̄2 in both physiological and pathological processes is being established. A weighty tome would be needed to encompass a comprehensive coverage of this field of study. This review will describe only aspects of the biology of oxygen radicals that currently engage the interest of the writer. Hopefully they will also be of interest to the reader. Other recent reviews may serve to fill the gaps in this one (1Miller A.F. Sorkin D.L. Comments Mol. Cell. Biophys. 1997; 9: 1-48Google Scholar, 2Gardner, P. R. (1997) BioSci. Rep., in press.Google Scholar, 3McCord J.M. Proc. Soc. Exp. Biol. Med. 1995; 209: 112-117Crossref PubMed Scopus (136) Google Scholar, 4Valentine J.S. Ellerby L.M. Graden J.A. Nichida C.R. Gralla E.B. Kessissoglou D.P. Bioinorganic Chemistry. Kluwer Academic Publishers Group, Dordrecht, Netherlands1995: 77-91Crossref Google Scholar, 5Rosen G.M. Pou S. Ramos C.L. Cohen M.S. Britigan B.E. FASEB J. 1995; 9: 200-209Crossref PubMed Scopus (412) Google Scholar, 6Fridovich I. Annu. Rev. Biochem. 1995; 64: 97-112Crossref PubMed Scopus (2755) Google Scholar). Because of the spin restriction, the univalent reduction of O2 to O·̄2 is a facile process, and O·̄2production by spontaneous as well as enzyme-catalyzed reactions has been demonstrated. The instability of O·̄2 in aqueous solutions is a hindrance to its detection and measurement. This has been circumvented by exploiting its reaction with various “detector” molecules such as ferricytochrome c or spin-trapping agents. These agents are not specific for O·̄2. Thus, reductants other than O·̄2 can reduce ferricytochrome c, and oxidants other than O·̄2 can convert the spin traps to their epr-detectable hydroperoxy derivatives (7Moan J. Wold E. Nature. 1979; 279: 450-451Crossref PubMed Scopus (245) Google Scholar, 8Finkelstein E. Rosen G.M. Rauckman E.J. Arch. Biochem. Biophys. 1980; 200: 1-16Crossref PubMed Scopus (770) Google Scholar). Inhibition by SOD 1The abbreviations used are: SOD, superoxide dismutase; FALS, familial amyotrophic lateral sclerosis; EC-SOD, extracellular SOD. is used to lend specificity to these methods. Detection and measurement of fluxes of O·̄2 within cells is a goal as difficult as it is desirable. Unfortunately enthusiasm for achieving this goal has led many investigators to use flawed methods. An example is the luminescence that can be elicited from lucigenin. Early studies of the lucigenin luminescence elicited by the xanthine oxidase reaction led to the realization that the lucigenin dication must be univalently reduced to the corresponding monocation before reacting with O·̄2 in the process that leads to luminescence (9Greenlee L. Fridovich I. Handler P. Biochemistry. 1962; 1: 779-783Crossref PubMed Scopus (61) Google Scholar,10Allen R.C. Methods Enzymol. 1986; 13: 449-493Crossref Scopus (443) Google Scholar). The chemistry involved was discussed in a more recent review (11Faulkner K. Fridovich I. Free Radical Biol. & Med. 1993; 15: 447-451Crossref PubMed Scopus (329) Google Scholar). Nevertheless the use of lucigenin as a “specific” detector of O·̄2 continues. Quite recently it was shown that the lucigenin monocation radical can autoxidize and thus produce O·̄2, even in cases where no O·̄2 was being produced in the absence of lucigenin (12Liochev S.I. Fridovich I. Arch. Biochem. Biophys. 1997; 337: 115-120Crossref PubMed Scopus (203) Google Scholar). In studies with Escherichia coli, lucigenin was shown to function, much as does paraquat, to increase intracellular O·̄2 production (13Liochev S.I. Fridovich I. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 2891-2896Crossref PubMed Scopus (61) Google Scholar). Hopefully the widespread but inappropriate use of lucigenin luminescence as a measure of O·̄2 will stop. Another luminescent method that is misused to measure O·̄2 is based on luminol. In this case the compound must be univalently oxidized to a luminol radical, which reacts with O·̄2 before the light-emitting pathway is entered upon. The problem in this case is that the luminol radical can spontaneously reduce O2 to O·̄2. Luminol luminescence thus can be caused by a variety of oxidants and in all cases SOD inhibits (14Hodgson E.K. Fridovich I. Photochem. Photobiol. 1973; 18: 451-455Crossref PubMed Scopus (107) Google Scholar, 15Miller E.K. Fridovich I. J. Free Radicals Biol. & Med. 1986; 2: 107-110Crossref PubMed Scopus (25) Google Scholar). Here again the detector is acting as a source of O·̄2. One additional artifactual detector of O·̄2 needs to be mentioned because of its widespread misuse, and that is nitroblue tetrazolium. Many enzymes can cause the reduction of tetrazolium salts to the corresponding formazans. Reduction of nitroblue tetrazolium to the monoformazan requires two electrons and to the diformazan four electrons. When proceeding by a univalent pathway, which is usual, one encounters tetrazoinyl radical intermediates (16Bielski B.H.J. Shive G.G. Bajuk S. J. Phys. Chem. 1980; 84: 830-833Crossref Scopus (364) Google Scholar), which reduce O2 to O·̄2 in a reversible process. SOD by removing O·̄2 displaces this oxidation to the right and thus prevents production of the formazan. For this reason many aerobic tetrazolium reductions are inhibitable by SOD even though O·̄2 was not being produced in the system in the absence of the tetrazolium (17Liochev S.I. Fridovich I. Arch. Biochem. Biophys. 1995; 318: 408-410Crossref PubMed Scopus (58) Google Scholar). Is there any method which can reliably be used as a measure of intracellular O·̄2? There is and it is based on the rapid inactivation of [4Fe-4S]-containing dehydratases, such as aconitase. O·̄2 oxidizes the clusters of these dehydratases resulting in loss of Fe(II), and that is reversible by reduction and reincorporating of Fe(II). The balance between these opposing processes can be used as a measure of O·̄2 and has been so used in E. coli(18Gardner P.R. Fridovich I. J. Biol. Chem. 1992; 267: 8757-8763Abstract Full Text PDF PubMed Google Scholar) and in mammalian cells (19Gardner P.R. Raineri I. Epstein L.B. White C.W. J. Biol. Chem. 1995; 270: 13399-13405Abstract Full Text Full Text PDF PubMed Scopus (436) Google Scholar). The inactivation of aconitase by O·̄2 provides an explanation for previously inexplicable observations. Thus high pO2 (291 mm) increased both glucose utilization and lactate production, by 4–6-fold, in WI38 cells (20Balin A.K. Goodman D.B.P. Rasmussen H. Cristofalo V.J. J. Cell. Physiol. 1976; 89: 235-250Crossref PubMed Scopus (123) Google Scholar). Raising pO2 would increase O·̄2 production, and inactivation of aconitase by O·̄2 would force the cells to rely on fermentation of glucose for energy. In another exampleAspergillus niger was reported (21Kubicek C.P. Röhr M. Appl. Environ. Microbiol. 1985; 50: 1336-1338Crossref PubMed Google Scholar) to accumulate less citrate in the medium when supplied with 0.1 mg/liter Mn(II). In this instance enrichment of the medium with Mn(II) would increase the level of Mn-SOD in the mitochondria, and that in turn would decrease O·̄2 and thus raise aconitase. The final effect would be increased metabolism of citrate via the Krebs cycle and less citrate excretion. Near UV irradiation of E. coli inhibited growth on succinate more than growth on glucose (22Kashket E.R. Brodie A.F. J. Bacteriol. 1962; 83: 1094-1100Crossref PubMed Google Scholar) and inhibited respiration (23Swenson P.A. Schenley R.L. J. Bacteriol. 1974; 117: 551-559Crossref PubMed Google Scholar). Both of these effects can be explained by the inactivation of aconitase by photosensitized production of O·̄2. Aconitase can be inactivated by oxidants other than O·̄2. Of these peroxynitrite is particularly relevant to biology, but in NO-producing cells the level of peroxynitrite is itself dependent upon O·̄2 production. Inactivation by H2O2 is relatively unimportant. Although O·̄2 can initiate and propagate free radical oxidations of leukoflavins, tetrahydropterins, catecholamines, and related compounds and can inactivate [4Fe-4S]-containing dehydratases, it does not significantly attack polyunsaturated lipids or DNA. Yet defects in SODs, which would have the effect of raising intracellular [O·̄2], do lead to cell envelope damage (24Imlay J.A. Fridovich I. J. Bacteriol. 1992; 174: 953-961Crossref PubMed Google Scholar) and to enhanced mutagenesis (25Farr S.B. D'Ari R. Touati D. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 8268-8272Crossref PubMed Scopus (242) Google Scholar). Mechanisms by which O·̄2 can give rise to more potent oxidants could explain these seeming anomalies, and there are several such mechanisms. The simplest of these is protonation to hydroperoxyl radical, whose pK a is 4.8 and which is a much stronger oxidant than is O·̄2. Association of O·̄2 with other cationic centers such as vanadate (26Liochev S. Fridovich I. Arch. Biochem. Biophys. 1986; 250: 139-145Crossref PubMed Scopus (45) Google Scholar) or Mn(II) (27Curnutte J.T. Karnovsky M.L. Babior B.M. J. Clin. Invest. 1976; 57: 1059-1067Crossref PubMed Scopus (52) Google Scholar) also have this effect, but these mechanisms are unlikely to apply generally within cells. There is a mechanism pertinent to living cells, and we may call it thein vivo Haber-Weiss reaction. It is a process in which O·̄2 increases “free” iron by oxidizing the [4Fe-4S] center of dehydrases such as dihydroxy acid dehydrase (28Kuo C.F. Mashino T. Fridovich I. J. Biol. Chem. 1987; 262: 4724-4727Abstract Full Text PDF PubMed Google Scholar), 6-phosphogluconate dehydrase (29Gardner P.R. Fridovich I. J. Biol. Chem. 1991; 266: 1478-1483Abstract Full Text PDF PubMed Google Scholar), fumarases A and B (30Liochev S.I. Fridovich I. Arch. Biochem. Biophys. 1993; 301: 379-384Crossref PubMed Scopus (62) Google Scholar), and aconitase (31Gardner P.R. Fridovich I. J. Biol. Chem. 1991; 266: 19328-19333Abstract Full Text PDF PubMed Google Scholar, 32Flint D.H. Tuminello J.F. Emptage M.H. J. Biol. Chem. 1993; 268: 23369-23376Google Scholar). The released iron is kept reduced by cellular reductants, and the Fe(II) reacts with H2O2, as in the Fenton reaction, to yield Fe(III) + HO⋅ or its formal equivalent, Fe(II)O. This was proposed (33Liochev S.I. Fridovich I. Free Radical Biol. & Med. 1994; 16: 29-33Crossref PubMed Scopus (365) Google Scholar) to provide an explanation for the enhanced O2-dependent mutagenesis exhibited by sodA sodB E. coli, and it has been experimentally verified (34Keyer K. Gort A.S. Imlay J.A. J. Bacteriol. 1995; 177: 6782-6790Crossref PubMed Scopus (188) Google Scholar, 35Keyer K. Imlay J.A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13635-13640Crossref PubMed Scopus (695) Google Scholar). The importance of release of iron by O·̄2 from [4Fe-4S] clusters of dehydrases was underscored by the recent observation of the complementation of sodA sodB E. coli by insertion and overexpression of rubredoxin reductase (36Pianzzola M.J. Soubes M. Touati D. J. Bacteriol. 1996; 178: 6736-6742Crossref PubMed Google Scholar). Rubredoxin reductase may play a role in reconstitution of the oxidatively disassembled [4Fe-4S] clusters of dehydrases and thereby lower the “free” iron in aerobic SOD-null E.coli, or it may somehow scavenge O·̄2 within cells. The primary defense against the damage that can be caused by O·̄2, and by its reactive progeny, is the SODs. The importance of these enzymes has been clarified by the phenotypic deficits of mutants defective in their production and in a number of cases by the complementing effects of homologous or heterologous SODs. These demonstrations have been achieved in bacteria (25Farr S.B. D'Ari R. Touati D. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 8268-8272Crossref PubMed Scopus (242) Google Scholar, 37Purdy D. Park S.F. Microbiol. Rev. 1994; 140: 1203-1208Google Scholar), yeast (38Bilinski T. Krawiec Z. Liczmanski A. Litwinska J. Biochem. Biophys. Res. Commun. 1985; 130: 533-539Crossref PubMed Scopus (134) Google Scholar, 39Bowler C. Van Kaer L. Van Camp M. Van Montagu M. Inzé D. Dhaese P. J. Bacteriol. 1990; 172: 1539-1546Crossref PubMed Google Scholar),Drosophila (40Phillips J.D. Campbell S.D. Michaud D. Charbonneau M. Hilliker A.J. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2761-2765Crossref PubMed Scopus (310) Google Scholar), a nematode (41Ishii N. Takahashi K. Tomita S. Keino T. Honda S. Yoshino K. Suzuki K. Mutat. Res. 1990; 237: 165-171Crossref PubMed Scopus (181) Google Scholar), Neurospora(42Chary P. Dillon D. Schroeder A.L. Natvig D.O. Genetics. 1994; 137: 723-730Crossref PubMed Google Scholar), and even mice (43Reaume A.G. Elliot J.L. et al.Nat. Genet. 1996; 13: 43-47Crossref PubMed Scopus (1051) Google Scholar, 44Li Y. Huang T.-T. et al.Nat. Genet. 1995; 11: 376-381Crossref PubMed Scopus (1462) Google Scholar, 45Lebovitz R.M. et al.Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 9782-9787Crossref PubMed Scopus (843) Google Scholar). The consequences of a lack of both the constitutive Fe-SOD (SOD B) and the inducible Mn-SOD (SOD A) inE. coli include oxygen dependent decrease of growth rate, nutritional auxotrophies, hypersensitivity toward redox cycling compounds such as paraquat and quinones, and an increase in the rate of spontaneous mutagenesis. In yeast similar problems were seen in strains lacking either the cytosolic Cu,Zn-SOD or the mitochondrial Mn-SOD. Envelope damage was made evident by the ability of osmolytes to facilitate the aerobic growth of the sodA sodB E. coli and to partially suppress the amino acid requirements (24Imlay J.A. Fridovich I. J. Bacteriol. 1992; 174: 953-961Crossref PubMed Google Scholar). Support for the free radical theory of senescence was provided by the shortened lifespan of Drosophila with a mutational defect in Cu,Zn-SOD (40Phillips J.D. Campbell S.D. Michaud D. Charbonneau M. Hilliker A.J. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 2761-2765Crossref PubMed Scopus (310) Google Scholar). These flies were also hypersensitive toward paraquat and were sterile. A curtailed lifespan was also evident inCaenorhabditis elegans, which had only half the normal complement of SOD (41Ishii N. Takahashi K. Tomita S. Keino T. Honda S. Yoshino K. Suzuki K. Mutat. Res. 1990; 237: 165-171Crossref PubMed Scopus (181) Google Scholar). Mice lacking Cu,Zn-SOD appeared normal while young but were less able to recover from axonal injury (43Reaume A.G. Elliot J.L. et al.Nat. Genet. 1996; 13: 43-47Crossref PubMed Scopus (1051) Google Scholar) and could not successfully reproduce. They also exhibited a shortened lifespan. 2A. Reaume, personal communication. Lack of Mn-SOD imposed more serious consequences (44Li Y. Huang T.-T. et al.Nat. Genet. 1995; 11: 376-381Crossref PubMed Scopus (1462) Google Scholar, 45Lebovitz R.M. et al.Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 9782-9787Crossref PubMed Scopus (843) Google Scholar). These animals lived only a week or two and exhibited faulty mitochondrial activities in several tissues, especially the heart. The familial form of amyotrophic lateral sclerosis, or FALS, has been shown to be associated with defects in the gene encoding Cu,Zn-SOD (sod 1) (46Rosen D.R. et al.Nature. 1993; 362: 59-62Crossref PubMed Scopus (5568) Google Scholar, 47Denq H.-X. Science. 1993; 261: 1047-1051Crossref PubMed Scopus (1360) Google Scholar, 48Siddique T. Nijhawan D. Hertati A. Neurology. 1996; 47 (suppl.): 27-34Crossref PubMed Google Scholar). That this was to a of of the than to a loss of SOD was by its and was by the ability of the in mice to J. 1994; Google Scholar, P. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: PubMed Scopus Google Scholar). There is for two of One of these is a peroxidase M. S. Gralla E.B. J.A. J.S. Science. 1996; PubMed Scopus Google Scholar), and the other is the ability to of by peroxynitrite J.S. Chem. Res. 1996; 9: PubMed Scopus Google Scholar). Yet another of a Cu,Zn-SOD has been I. L. L. R. A. and I. and that is a peroxidase Cu,Zn-SOD has been reported to the production of H2O2 E.R. Proc. Natl. Acad. Sci. U. S. A. 1990; PubMed Scopus Google Scholar), and the has been shown to be more in this than the E.R. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: PubMed Scopus Google Scholar). for these to the peroxidase of the than to the production of free have been I. Annu. Rev. Biochem. 1995; 64: 97-112Crossref PubMed Scopus (2755) Google Scholar). It may be that all of these to the in amyotrophic lateral is from so a related mechanism be A recent et 1996; PubMed Scopus Google Scholar) that cases of amyotrophic lateral were with to Cu,Zn-SOD this Another that may be to faulty superoxide is in the of the gene for Mn-SOD have been reported J.S. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: PubMed Scopus Google Scholar). This led to the that be to faulty of Mn-SOD. is associated with a of the gene for this is on It has been that this the of this R. D. Y. M. Y. J. Sci. Full Text PDF PubMed Scopus Google Scholar). this Thus a with of Cu,Zn-SOD was associated with of the R.M. H. M.J. Genet. 57: PubMed Scopus Google Scholar). This lack of of with has been R. A. E. D. J. 1996; PubMed Scopus Google Scholar). The SOD has been The cytosolic Cu,Zn-SOD and the mitochondrial Mn-SOD are now by an extracellular which is to as The is a with a It to the cytosolic Cu,Zn-SOD but is for in the extracellular and is by the cells that produce It is and for such as or Thus, in of it the extracellular Proc. Natl. Acad. Sci. U. S. A. PubMed Scopus Google Scholar, J. L. T. J. Biol. Chem. 1992; 267: Full Text PDF PubMed Google Scholar, J.D. Free Radical Biol. & Med. 1996; PubMed Scopus Google Scholar). to O·̄2 released by and other cell It may be for the rapid reaction of O·̄2 with S. Free Radical Res. Commun. 1993; 18: PubMed Scopus Google Scholar) and thus for the of while production of the oxidizing peroxynitrite J. Physiol. 1995; 268: Google Scholar). mice lacking but were more to the effects of of O2 L.M. J. T. Proc. Natl. Acad. Sci. U. S. A. 1995; PubMed Scopus Google Scholar). superoxide has been reported in a variety of and the case of the extracellular SOD has been shown to be a L. 1990; PubMed Google Scholar). An new SOD, with its has been J. Biochem. 1996; PubMed Scopus Google Scholar). This is a whose is It no to known or have been reported in a of bacteria and were such have been with J. Bacteriol. 1990; 172: PubMed Google Scholar, J. Bacteriol. 1994; PubMed Google Scholar, A. 1995; PubMed Scopus Google Scholar, J. Bacteriol. 1993; PubMed Google Scholar, J. Bacteriol. 1996; 178: PubMed Google Scholar), and it is now that they in bacteria where they are to be to the J. Bacteriol. 1990; 172: PubMed Google Scholar, J. Bacteriol. 1994; PubMed Google Scholar). The E. coli Cu,Zn-SOD to be in that it is A. 1995; PubMed Scopus Google Scholar), than being as are other mutants in Cu,Zn-SOD have been exhibited increased toward citrate and toward in or J. Bacteriol. 1993; PubMed Google Scholar), the in J. Bacteriol. 1996; 178: PubMed Google Scholar). and it that Cu,Zn-SOD to the of and Imlay J.A. J. Bacteriol. 1996; 178: PubMed Google Scholar). It may be that the Cu,Zn-SOD as a defense against O·̄2. the many of the defense against more than does the This of is on by any that increases O·̄2 production in E. coli H. E. J. Biol. Chem. 1996; Full Text Full Text PDF PubMed Scopus Google Scholar, 1996; PubMed Scopus Google Scholar). to to a of to oxidatively the c, to the fumarases a and to oxidatively disassembled [4Fe-4S] to decrease the of the aconitase to the aconitase inactivated by and as as the redox which production of which on the a and is a only in its oxidized The is itself only half of the defense against There is another on in to H2O2 and to as the Science. 1990; PubMed Scopus Google Scholar). to be of that which is known and much to be to may because their do not in the of and to the whose are not The of the field and whose it in a may be as another of the rapid growth of in this