with Fe and the other with Mn, has also been reported. (2). Genes encoding. MnSOD .... selenium was recognized in 1973. Spallholz and Boylan (27) have recently .... response to perpetuate. Cells from adult animals, for example, as opposed.
Regulation
of antioxidant
EDWARD
nzymes1
D. HARRIS
Department
of Biochemistry
and Biophysics,
Texas A&M University,
Free radicals generated by a partial reduction of 02 pose a serious hazard to tissues and vital organs, especially membrane lipids, connective tissues, and the nucleic acids of cells. For protection, enzymes have evolved that specifically attack Of, hydrogen, and organic peroxides, and repair any damage incurred to DNA. With few exceptions, antioxidant enzymes are found in all aerobic and aerotolerant anaerobic organisms. Logic assumes that a basal level of antioxidant enzyme activity is maintained at all times. This may be true. Yet cells must have ways to amplify antioxidant enzyme activity to counter sudden increases in oxygen metabolites. The full details of that regulation are slowly coming to light. Bacteria possess a series of elaborate and interacting genes that can sense specific increases in intracellular H202 and Of. In higher organisms, hormones and metal ion cofactors impose pre- and posttranslational control over the genetic expression of antioxidant enzymes. Furthermore, aging, cellular differentiation, and organ specificity must also be factored into the final equation in higher organisms. This review will discuss some of the more recent findings relevant to antioxidant enzyme regulation in bacteria and higher organisms.Harris, E. D. Regulation of antioxidant enzymes. FASEB J. 6: 2675-2683; 1992. ABSTRACT
Key Wordr: hyperoxia
oxygen catalase
THE TOXICITY ditions. biotics, stimulate
stress oxygen radicals glutathione peroxidase
OF OXYGEN
superoxide
under
IS AMPLIFIED
dismut use
a variety
of con-
Neutrophil activation, hyperoxia, redox-active radiation exposure, and ischemia are events the production of a series of partially reduced
xenothat oxy-
gen molecules that directly or indirectly attack the structural and genetic apparatus of cells. Formation of these molecules in one respect is a natural metabolic occurrence, or depending
on
cell
type,
is part
tect the organism checked, including
of a system
against
through
invading
sacrificial that act
and chromatin. As synthesize quenchers molecules directly
in on
cells
microorganisms.
however, oxygen radicals can the sensitive macromolecules
membranes ability to
which
attack that
redox cycles the oxygen
or as radical
specific itself,
possess either
cell the as
will focus on the their regulation. A multitude of factors are superimposed on higher organisms, including cellular differentiation, aging, cofactor availability, and hormones that regulate expression of antioxidant
preventative
aerobic
most recent It should
and
repair
life forms.
enzymes
has
been
factors
given
and glutathione
peroxidase, massive
The
are
critical
of the attention
to superoxide
this
report.
evidence supporting their input be recognized, however, that both
As most
these
literature
0892-6638/92/0006-2675/$01.50.
© FASEB
survival
of
on antioxidant
dismutase,
enzymes that
to the
catalases,
will be central
is steadily
USA
STRATEGY
AND
OVERALL
PERSPECTIVE
Whereas the control of cellular oxidants in bacteria is regulated by transcription of antioxidant genes, eukaryotes combine antioxidant enzyme function with low molecular weight antioxidant compounds such as a-tocopherol, ascorbic acid, glutathione, and 13-carotene. Several important points emerge. The delicate molecules of life exist in hydrophobic and hydrophilic environments. This creates a need for a family of antioxidant compounds whose solubilities are compatible with the environment of function. With few exceptions, antioxidant enzymes are water soluble and thus function in the
plasma,
the
cytosol,
or
periplasmic
spaces
of cells.
Sec-
spontaneous dismutation of Of by superoxide dismutase generates H202, which itself is a dangerous oxidant in cells. Thus, to respond to an increase in Of cells must not only elevate superoxide dismutase but catalase and glutathione peroxidase as well - the latter to regulate H2O2. Experimental evidence has tended to support the necessity for coordinating antioxidant enzyme expression. The overexpression of CuZnSOD,2 for example, has been correlated with the decomposition of brain neurons (5) and has been considered a major predisposing factor in the radiation sensitivity typical of Down’s syndrome fibroblasts (6). High SOD activity typically is seen with chronic and acute leukemias (7). Abnormal antioxidant enzymes are hallmarks and contributing factors of many tissue patholond,
the
catalytic
or
ogies.
ENZYMES
The three major antioxidant enzymes - superoxide dismutase, glutathione peroxidase, and catalase - are unique in cofactor requirements and cellular locations. Their reactions are shown below
enzymes forming
products that are less offensive. This review antioxidant enzymes. A goal is to understand
enzymes. The will be reviewed.
Texas 77843-2128,
ing precludes a more comprehensive coverage here. The reader should consult excellent and timely reviews on antioxidant enzymes and other facets of oxygen stress that have appeared recently (1-4).
ANTIOXIDANT
targets,
comprise
a defense, cells that intervene
Station,
pro-
Left unother
College
accumulat-
in
‘From the Symposium by the American Institute
of the Federation
Regulation of Antioxidant Enzymes presented of Nutrition at the 75th Annual Meeting
of American
Societies
for Experimental
Biology,
April 24, 1991, Atlanta, Georgia. 2Abbreviations: SOD, superoxide dismutase; MnSOD, manganese-containing superoxide dismutase, CuZnSOD, copper, zinc superoxide dismutase, FeSOD, iron-containing superoxide dismutase, 02, superoxide anion, H2O2, hydrogen peroxide; GSH-Px,
glutathione peroxidase; PH-GPx, phospholipid hydroperoxide glutathione peroxidase; MRI, magnetic resonance imaging; TBH, tertiary-butyl hydroperoxide; TNF, tumor necrosis factor; PQ paraquat.
2675
Of
+
ROOH
+
O2
+
2W
02 (superoxide dismutase) 2H2O (catalase) H20 + GSSG (gJutathione peroxidase)
H202
2H,O2 2GSH ROH
0, +
+
+
The superoxide dismutases (EC 1.15.1.1) convert superoxide anion into H202 and 02, a second-order reaction with a rate constant of 2 x i0 M’ s-i (2). All superoxide dismutases have at least one first transition series metal (Fe, Mn, or Cu) at the active site. Bacteria contain both manganese (MnSOD) and iron (FeSOD) forms of SOD. FeSOD also occurs in some plant species. Eukaryote’s MnSOD is a mitochondrial enzyme. CuZnSOD is found in plants, animals, and fungi. Bacteriocuprein, a membrane-bound CuZnSOD, is present in certain bacteria species (8). Catalase (EC 1.11.1.6) catalyzes the dismutation of H202, forming as neutral products 02 and H20. Glutathione peroxidase (EC 1.11.1.9) catalyzes the reductive destruction of hydrogen and lipid hydroperoxides, using glutathione as an electron donor.
MnSOD and FeSOD
in prokaryotes
coli has no CuZnSOD, but instead relies on constitutive production of FeSOD and induced MnSOD (Table 1). Hybrid SOD, a heterodimer composed of one subunit with Fe and the other with Mn, has also been reported (2). Genes encoding MnSOD (soeM) and FeSOD (sodB) have been cloned, mapped, and sequenced. Early studies showed no detectable MnSOD in E. coli grown under anaerobic conditions, leading to the suggestion that the sodA gene was regulated by 02 or Of (9). Stronger evidence came when it was noted that methyl viologen (paraquat), an in situ Of generator, caused a strong increase in MnSOD in anaerobic cultures (9). The regulation of MnSOD has taken on added complications with the recognition that glucose concentraEscherichia
TABLE
1. Antioxidant
and DNA
repair enzymes
Gene
Enzyme
in bacteria
Regulon
Mode
SODS: MnSOD FeSOD
soxR, jlzr, ?
HPI HPII
katG katE
oxyR
GSH-Px
gorA
oxyR
ahpC
oxyR
ahpF
oxyR
Inducible Inducible Inducible
xth
katF soxR
Inducible Inducible
soxR
Inducible Inducible Inducible Inducible Inducible
Catalases
Constitutive
and peroxidases:
C22 F22 DNA repair enzymes:
Exo III Endo Others:
arcA
Inducible
sodA sodB
IV
G6PD S6 OmpF Acid phosphatase BolA
swf
katF
soxR soxR IcatF katF
Inducible Inducible
To defend against oxidative Stress, bacteria use key enzymes, whose genes in turn are controlled by regulons. At least three regulons regulate MnSOD through control of sodA. Present understanding is that sodB is constitutively expressed although evidence supports transcriptional control through the fir locus (2). E. coil contains two catalases: HPI, an H202inducible enzyme coded by katG and HPII, coded by katE. Both Exo III and HPII increase during the stationary phase of growth. Bo1A is a protein involved in morphogenesis. For a more complete summary of bacterial regulons in oxidative stress, see ref 4.
2676
Vol. 6
June 1992
tions and iron also influence expression of the enzyme (2). Moreover, respiratory growth with nitrate, dimethyl sulfoxide, or ferricyanide acting as terminal electron acceptors (replacing 02) stimulates MnSOD in E. coli (2). The data support a multifaceted regulation scheme, with oxygen metabolites working in concert with other metabolites to determine activity. A second regulon controlling MnSOD expression, the soxR locus, is described below.
MnSOD in eukaryotes MnSOD in eukaryotic cells is strictly a mitochondrial enzyme located in the inner membrane and synthesized by nuclear genes. A larger precursor form is made in the cytosol and transported into the mitochondria by an energydependent process (10, 11). Based on a cDNA isolated from human liver library, the primary translation product contains 198 amino acid residues, including a 24-residue leader peptide. Cell-free translation using a rabbit reticulocyte lysate gives rise to a 26-kDa peptide, which subsequently is cleaved to 24-kDa by mitochondria (11). Like the bacterial enzyme, yeast MnSOD is controlled by oxygen or one of its metabolites (10). Sequences representing the signal peptide must be deleted in order to express the eukaryotic MnSOD gene in E. coli (12). A diminished level of MnSOD activity occurs in a wide variety of tumor cells. The lowered activity is generally a consequence of lower amounts of enzyme protein and its mRNA (13).
CuZnSOD Immunohistochemical and cell fractionation procedures have supported a cytosolic location for CuZnSOD in many cells. In rat hepatocytes, for example, about 70% is cytosolic and 12% is in the nuclei; lysosomes contain the highest content of any organelle, whereas mitochondria, ER, and Golgi barely test positive (14). Endothelial and smooth muscle cells also split CuZnSOD between the nuclear and cytosolic fractions (15). More recent evidence based on immunofluorescence with monoclonal antibody markers now suggests that CuZnSOD in human fibroblasts, hepatoma cells, and yeast cells is predominantly a peroxisomal enzyme. Its cytosolic location apparently arises from rupture of peroxisomes during homogenization (16). The enzyme is found in a granule-rich fraction from beta cells of canine pancreas, which lends support to a possible role in the protection of insulin from oxidation (17). CuZnSOD has been cloned from numerous plant, invertebrate, and vertebrate sources such as maize (18), Drosophilia (19), and rat lung (20). Studies of its regulation must take into account that CuZnSOD exists in multiple isomeric forms, distinguishable by metal ion content, and that the prevailing form at any one time could be a function of the age of the animal (21). Copper is rapidly removed from CuZnSOD in situ by dialysis against isotonic buffers containing 0.1-0.5 M catechol, triethylenetetramine, or tetraethylenepentamine (22). CuZnSODs in wheat, kidney bean, maize, tomato, and spinach show structural homologies, specifically metal content and dimeric structure, but differ slightly in amino acid content and antibody cross-reactivity (23). The multiple forms of the enzyme in plants owe diversity to the specific forms found in the chloroplasts and cytosol, suggesting a evolutionary divergence (24). CuZnSOD from fern and green algae has a molecular weight, subunit structure, absorption spectra, and circular dichroism spectra similar to th mammalian enzyme, including modification to a critical arginine residue (24).
The FASEB Journal
HARRIS
Extracellular
superoxide
dismutase
Extracellular superoxide dismutase (EC 1.15.1.1, EC-SOD), a tetrameric high molecular weight Cu,Zn enzyme, is found in numerous extracellular fluids including plasma, lymph, and synovial fluid (25). EC-SOD from pig, cat, rabbit, guinea pig, and mouse shows at least three forms when resolved on heparin-derivatized affinity columns. Little is known about the factors that regulate this enzyme or enzymes. Its sensitivity to heparin suggests a cell surface origin, possibly from endothelial cells (26). Glutathione
peroxidase
Glutathione peroxidase (GSH-Px) has a critical need for GSH, which itself is a key antioxidant at the hub of numerous competing reactions. A distinct requirement for selenium was recognized in 1973. Spallholz and Boylan (27) have recently provided an excellent contemporary review of functions and properties of glutathione peroxida.ses. Typically, the enzymes from numerous sources have been tetramers, between 76 and 105 kDa, with 4 g atoms of Se per molecule. GSH-Px in plasma is antigenically distinct from its cytosolic counterpart. The plasma enzyme is a glycoprotein, Mr 22,500 on SDS-PAGE. Recently, two overlapping cDNA clones isolated from a placental library and encoding a polypeptide of 226 amino acids showed sequence homology with a 22.5-kDa component in kidney homogenates, implying that the origin of the plasma protein is kidney (28). Two other Se-dependent glutathione peroxidases have been described. One, referred to as phospholipid hydroperoxide glutathione peroxidase (PH-GPx), has broad tissue distribution and is more resistant to selenium depletion. PH-GPx requires higher levels of Se to maintain full activity, which could account for the variable sensitivity of different organs to oxidative stress (29). A second Se-containing enzyme, more closely resembling the plasma enzyme, is found in human milk (30). Catalase Catalase, a heme protein with a single substrate - H202 - is ubiquitously distributed in tissues of all species. Early studies showed that catalase activity in cells was controlled by H202. An early puzzle that took many years to solve was why catalase continue to be synthesized well after the H202 had been broken down. Clarification required the discovery of a specific regulon responsive to H202 (see below).
REGULATION IN BACTERIA
OF
ANTIOXIDANT
ENZYMES
oxyR
ANTIOXIDANT
soxR
Paradoxically, oxyR exerts no control over superoxide dismutases. A dual gene locus, separate and apart from oxyR and designated the soxR regulon (superoxide responsive) controls nine proteins induced in response to superoxide generators (32). Formerly soiR (superoxide inducible), the functional soxR locus consists of the soxR and soxS genes, positioned head to head with an intervening space of 85 bp. The soxR
oxyR (H202 inducible)
/ Catalase
and S. typhimurium adapt to lethal levels of hydrogen peroxide when first given low doses. Low levels elicit a series of H202 metabolizing enzymes as well as DNA repair enzymes (Table 1). All told, some 30 proteins are induced (as identified by two-dimensional polyacrylaxnide gel electrophoresis), only a few of which have been identified with enzymatic activity and genes. Some of these overlap with the proteins induced by heat shock or stress from other oxidants. These data support the existence of a specific genetic locus, the oxyR regulon, that coregulates a group of enzymes, which in E. coli and S. typhimurium function in cellular resistance to H2O2. Figure 1 shows the known antioxidant enzymes controlled by oxyR. It is likely that others
Escherichia
await discovery. Time course studies reveal that 12 of the 30 inducible proteins are seen maximally in 10 mm (31). Adaptation to H202 effects could be a function of the early proteins. The response is sensitive to chloramphenicol, implying de novo synthesis of proteins is a requisite for expression. oxyR deletion mutants that are hypersensitive to H202 fail to express the proteins. These mutants are also prone to an increased frequency of mutagenesis, especially under aerobic conditions. Alkyl hydroperoxidase (AHP), an NADPH-dependent flavoprotein, is elevated 20-fold in the constituent mutant, oxyRl. AHP in S. typhimurium consists of two identical 57-kDa subunits (designated F52 on two-dimensional gels) encoded by aizpF and a 22-kDa protein (C22) encoded by ahpC. Strains that do not manifest the ahpCF gene product are extremely sensitive to organic peroxides. A similarity between AHP and glutathione peroxidase of mammalian cells is evident, with the exception that the bacterial enzyme does not contain selenium. Catalase activity (HPI), a product of the kaiG gene, is also expressed in induced cells. Sensing the H202 is a function of the OxyR protein, whose primary sequence has been deduced from oxyR. Dithiothreitol abolishes activation as does implementing anaerobic conditions (argon gas) when preparing the extracts. Thus, OxyR fits the criteria of a redox sensing protein capable of reversibly altering its conformation in response to the prevailing redox environment of the cell. Although highly reducing conditions negate OxyR activation, the reduced protein continues to bind to DNA strands bearing the oxyR promoter region. DNAse I protection assays, however, suggest that the reduced and oxidized forms of oxyR, while overlapping, do not cover identical residues.
A1kyhydroperoxidase
coli
ENZYMES
Glutathione
Reductase
Figure 1. The oxyR regulon. Three of the better characterized target genes include katG, which encodes HPI catalase; ahpC and ahpF (ahpCF), encoding two components of arylhydrocarbon hydrolase; and gorA, which expresses glutathione peroxidase. Both HPI and arylhydrocarbon hydrolase function in the cellular resistance to H202.
2677
MnSOD (0;
Endonuclease
IV
(DNA repair)
Glucose-6-Phosphate Dehydrogenase (NADPH
production)
s6 (proteinsynthesis) Figure 2. The soxR regulon. Key components include sodA (MnSOD), endo IV (for DNA
(glucose-6-phosphate levels of NADPH ribosomal subunit regulating
absorption
of soxR regulation repair), G-6P DH
dehydrogenase, responsible for maintaining and responsive to low NADPH), S6 (small protein), and ompF (a membrane proteinof antibiotics).
locus exerts positive transcriptional control over at least nine genes, five of which have been characterized (Fig. 2). Demple has recently provided a strong, penetrating analysis of the soxR locus (4). Other than O2, soxR responds to redox cycling agents, decreased levels of cellular NAD(P)H, and antibiotics not necessarily related to redox stress. Control is mediated by the soxR and soxS gene products. A sequential but cooperative mechanism has been proposed.
REGULATION
IN EUKARYOTES
Antioxidant enzyme regulation in tissues of higher animals depends on many diverse factors, including organ specificity, age, developmental stage, prevailing hormone profile, and the availability of active site cofactors. Many tissues show separate and unique responses to factors that regulate antioxidant enzymes. For brevity, only two will be considered.
LUNG
AND
HYPEROXIA
An estimated 5-10% of total 02 consumed by rat lung goes to form reduced oxygen species (33). The lung is the major organ most vulnerable to the peroxidative reactions of oxygen. Studies with lung endothelial cells, pulmonary macrophages, and fetal cells have yielded important insights that tend to support whole animal studies but with greater insights into the finer details of regulation. A pivitol observation that has brought the focus on antioxidant enzymes in hypoxia has been that neonatal rats are more capable than adults in resisting the lethal effects (34). Hyperoxia, or overexposure to oxygen, has been shown to induce catalase, GSH-Px, and SOD activities both in cultured cells and intact animals (35, 36). Enhanced inductive capacity for antioxidant enzymes synthesis, particularly SOD, has been considered a principal reason for the superior resistance of the neonate. A closer examination has shown that of the two SODs, only MnSOD activity is elevated in minced lung of neonatal hypoxic rats and by a mechanism that requires the cells to remain intact (34). But survival could depend on other antioxidant enzymes, and perhaps more important, could require the induced response to perpetuate. Cells from adult animals, for example, as opposed to the neonate show
2678
Vol. 6
June 1992
less capacity to maintain the elevated SOD activity once induced (33). Except for a spontaneous increase in the fetus and late gestation, CuZnSOD activity tends to remain constant in lungs from young (1 month old) and adult (4 to 5 months) rats (20). Hyperoxia, however, is a condition that accelerates the synthesis of CuZnSOD, thus preempting the normal tissue expression of the enzyme, a response not unlike that seen for MnSOD in bacteria stressed with O2. In fact, the apparent parallels in the two systems prompted consideration that O2 could be the inducing factor that creates the elevated levels of enzymes in hyperoxic lung. In support of the hypothesis, bleomycin, a known 02 generator, was found to strongly elevate the total and specific activity of SOD in hamster lung; the particular species of SOD was not identified (37). Another important observation, but one that sparked a controversy, is that rats given injections of endotoxin survive hyperoxia much better than normal animals and have greatly reduced lung injury (38). Endotoxin, a bacterial lipopolysaccharide, induces MnSOD mRNA in pulmonary artery endothelial cells, the site of most hyperoxic damage (39). The controversy stems from a unique experiment by Berg and Smith (40) that found serum from endotoxin-treated rats infused into normal animals afforded excellent protection against hyperoxia, but without raising lung SOD activity in the recipients. Berg’s and Smith’s results (40) cast doubts on MnSOD elevation as the major protective factor in hyperoxia. On the other hand, transgenic mice stand
with higher hyperoxia
basal
levels
of CuZnSOD
activity
with-
(>
99%) better than normal mice and show less evidence of tissue damage (41). Moreover, administering catalase and SOD-containing liposomes to rats also increases the survival in 100% oxygen. Lung catalase and SOD activities are raised 3.1- and 1.7-fold, respectively, in 2 h (42). Again, it is not clear whether any one enzyme can be singled out as the most important protective factor. Prolonged exposure to hyperoxia will gradually raise the GSH-Px, glutathione reductase, and glucose-6-phosphate dehydrogenase activities. Then, too, low molecular weight antioxidants such as GSH cannot be discounted in the overall protection of the lung from hyperoxia. Future studies of lung oxidants will have the advantage of using proton magnetic resonance imaging (MRI), a powerful noninvasive technique, to monitor progressive changes in lung tissues. High-intensity bright spots appear where oxidant damage has occurred. Bray and colleagues (43) have pioneered the use of this technique in lung studies assessing the effect of Zn and Cu deficiencies on lung.
BRAIN The vulnerability of the brain to oxidants has drawn special attention due to the intrinsically low to moderate levels of the catalase, GSH-Px, and superoxide dismutase levels prevalent in normal brain tissue. Increasing the CuZnSOD during development is a necessity for the brain especially when protection of preferentially vulnerable brain neurons to neurodegenerative disease such as Alzheimer’s disease poses a threat at all stages of brain development. CuZnSOD mRNA has been detected in human hippocampus, specifically pyramidal cells of Ammon’s horna, subiculum, and granule cells of dentate gyrus (44). Paradoxically, increments in CuZnSOD also pose a threat to brain by accelerating H202 production. Ceballos-Picot et al. (5) showed that transgenic mice that carried the human CuZnSOD gene and had twice
The FASEB Journal
HARRIS
the CuZnSOD activity in brain neurons had significantly increased peroxidation in the pyramidal cells and the gyrus dentate. The experiments support the hypothesis that trisomy 21 cells associated with the abnormalities of Down’s syndrome work through elevated CuZnSOD.
DIFFERENTIATION, AGING INFLUENCES
DEVELOPMENT,
AND
Differentiation Whereas other tissues may show fluctuations in antioxidant enzymes, erythrocytes maintain a fairly constant and critical level of CuZnSOD throughout the life span of the cell. In immature or undifferentiated blood cells, differentiation is usually accompanied by a change in CuZnSOD activity (3). The changes could result from altered genetic expression or in the case of CuZnSOD, from a deficit of the copper cofactor (45). Differentiation of murine erythroleukemia cells with hexamethylene bisacetimide causes a sharp increase in CuZnSOD whereas antioxidants such as a-tocopherol, f3carotene, or butylated hydroxytoluene cause no response (3). Saito et al. (46) observed a decrease in CuZnSOD activity and CuZnSOD mRNA in both human U937 monocytic leukemia cells and HL-60 promyelotic leukemia cells in response to phorbol and DMSO. MnSOD activity had decreased as much as twofold in HL-60 cells induced with 60 mM dimethylforamide, but CuZnSOD did not change (47). These conflicting reports discourage generalities and suggest that the agent used to cause differentiation could be a determinant of the response observed. Hemin, a complex of iron with porphyrin, lowers CuZnSOD activity in K562 cells but has no apparent affect on CuZnSOD mRNA (45). MnSOD activity is induced rapidly and apparently spontaneously in two undifferentiated tumor cell lines. A resistant line that showed slow stimulation was nondifferentiating (48), suggesting that induction of MnSOD was somehow timed with cellular differentiation. The apparent failure of MnSOD to respond to oxidative insult has been considered a dominant factor in the etiology of oncogenesis (13). An interesting twist is that a 46-fold increase in MnSOD added via liposome incorporation leads to the differentiation of slime mold; nonresponsive strains do not differentiate (3).
Aging Aging effects on SOD activity in tissues and organs has been an area of controversy punctuated by conflicting reports. Part of the reason has been very little uniformity as to the duration of the period examined and the nature of the material (whole cells, whole tissue, etc.). For example, with the exception of a significant decrease in liver, early results supported no change in MnSOD activity in most organs of murine animals with aging (54). However, a later study with select populations of mitochondria isolated from rat brain found significant decreases in the MnSOD activity in light and heavy synaptic mitochondria (55). More recently, De and Darad (56) examined rats at 3, 12, and 24 months and found a strong decrease in CuZnSOD activity in liver, whereas catalase activity became enhanced with age. A safe conclusion is that aging has the potential to disrupt the prevailing antioxidant enzyme profile in tissues. Whether this is genetically programmed or a consequence of the inevitable wear and tear of aging is unknown. In support of regulation at a genetic level, cells from patients afflicted with Bloom’s syndrome, a rare genetic disease characterized by immunological dysfunction and a proness to develop cancer, exhibit elevated levels of superoxide dismutase and O2, which leads to high levels of H202. The inefficient removal of H2O2 has been considered a factor in the disorder (57). A study of the CuZnSOD in older (24 months) rat lung has revealed additional important facets of regulation. Biochemical analyses have shown that in the more mature rat, resistance to a loss in CuZnSOD activity seems correlated with a shift to the most stable isoform of the enzyme (21). That form may also be the one most resistant to metabolic turnover.
REGULATION
BY COFACTORS
The major antioxidant enzymes possess transition metals or selenium at the catalytic site. The availability of the enzyme cofactor, therefore, has the potential to limit the expression of the enzyme activity. Cofactor regulation shows an important departure from regulation in bacteria in that the switch is put to posttranslational events. In yeast, however, the cofactor may also be a transcriptional regulator.
Development
CuZnSOD
Immunohistochemistry has shown lung SOD in preaveolar lung before 23 wk gestation in humans (49). Early in prenatal development of rats, immunoreactive SOD, GSH-Px, and catalase undergo a marked increase in concentration in a kidney and other tissues (50). In liver, CuZnSOD protein increases nearly 10-fold at a time when MnSOD rises only twofold (51). Brain shows a distinct ontogenetic pattern of oxidant enzymes in that rat cortex nuclei have increased levels of CuZnSOD and MnSOD, but maintain low levels of GSH-Px and catalase. Brain mitochondria show increases in MnSOD and GSH-Px but decreased catalase, which is more prevalent in mitochondria from 2-day-old than from 77-dayold rats (52). Of the two SOD enzymes, CuZnSOD is the predominant form in rat brain and liver during development. Whereas brain MnSOD increases steadily throughout life, liver MnSOD plateaus at less than 100 days and remains constant (53). Oxygen tension may have little effect on the quality or quantity of changes observed (3).
Modifications to the diet or the application of specific metal ion chelators limit expression of CuZnSOD in tissues. In human infants and numerous species of animals, including sheep, fowl, and rats, the availability of copper has been shown to be a decisive factor controlling CuZnSOD activity. In one of the better characterized animal species - the rat - a severe copper deficiency lowers CuZnSOD activity in liver, lung, heart, and skeletal muscle. Surprisingly, zinc deficiency has only a marginal effect on the expression of CuZnSOD in erythrocytes (58). Moreover, a copper-free apoenzyme has been observed in tissues of copper-deficient animals and humans. Diethyldithiocarbamate (DDC) and cuprizone, both chelators of ionic copper, suppress SOD levels in rat lung and liver without affecting other enzymes (33). A period of oxidative stress in intact animals and cells in culture tends to enhance the need for copper to sustain CuZnSOD activity (45, 59). In yeast (Saccharomyces cerevisiae), anaerobosis leads to the formation of a CuZnSOD proen-
ANTIOXIDANT
ENZYMES
2679
zyme that requires copper for activation (60). Abundant amounts of active enzyme are achieved only when Cu2 salts in excess of 50 ,LM are added to the growth medium (61). Thus, the evidence is strong that cofactors, especially copper, exert major regulation over CuZnSOD expression. This statement must be guarded, however. Humans suffering from rheumatoid arthritis have lower erythrocyte CuZnSOD activity, yet the copper content of the tissue is normal, which implies that the enzyme protein and not its cofactor limits catalytic activity. The specific activity of erythrocyte CuZnSOD remains constant with aging, suggesting that any decline in activity necessitates a loss in protein as well (62). Similarly, yeast cells (S. cerevisiae) that manifest very low CuZnSOD activity when grown in low copper medium have suppressed levels of both CuZnSOD mRNA and protein (61). When the cells are grown under anaerobic conditions, less enzyme activity is observed; mRNA levels are unchanged, less copper is found in extracts, and an inactive proenzyme accumulates (60).
A dual role for copper
Vol. 6
June 1992
ACE1- [Cu-9
CUP1 ACE1-Cu
#{149}SOD1
site
If apoMT
apoSOD
CuMT
CuZnSOD
Figure 3. Regulation of CuZnSOD in yeast. Coordinate transcriptional control through activation of CUPJ (encoding metallothionein, MT) and SODJ by the ACE1-Cu DNA binding protein is complemented by a posttranslational insertion of copper to activate the apoenzyme product. The ACE1 binding site for SODJ is located at bases -183 to -206 upstream from the start site. Cu in excess will bind to apoMT, a precautionary measure against Cu toxicity.
the coordinated system for regulating antioxidant enzymes. Such a simplistic appraisal, however, does not account for all of the observations. For example, low amounts of copper in the diet of mice and rats not only decrease the CuZnSOD, but also lower the tissue levels of glutathione peroxidase and glutathione transferase (66). The lower selenoperoxidase activity is unexpected but shows that Cu2 affects expression of the selenium-dependent enzyme, possibly by controlling the absorption
or
tissue
retention
of selenium
(67).
Within
the
family of SOD enzymes, lowering CuZnSOD activity leads to a paradoxical increase in MnSOD activity. This response has been viewed as one in which the cell compensates for the loss of one SOD form by increasing the other, thereby attempting to keep the total SOD activity at a near-constant level (68). In other examples, yeast cells (S. cerevisia) cultured in a medium containing less than 50 iM Cu show elevated levels of MnSOD activity and MnSOD mRNA (61). In contrast, chicks starved of copper for 11 days lose both CuZnSOD and MnSOD activities in aortas (69). In E. coli, Fe competes with Mn for binding sites on MnSOD (70), and when given as a supplement to low birth weight infants, iron lowers the CuZnSOD in erythrocytes (71).
HORMONES
INTERACTIONS
Cofactor-dependent regulation must also take into account the potential for biochemical interaction between cofactors. Mechanistically, some of the interactions thus far described have tended to challenge working hypotheses regarding the role of cofactors in regulation. One may be bold to assert that similarities in chemical properties will lead ultimately to substitution and competition between cofactors and affect
2680
[Cu-4-]
in yeast
Regulation of CuZnSOD in yeast (S. cerevisiae) perhaps has provided insight into a unique dual role for copper ions in regulation, specifically in both transcriptional and posttranslational events. Evidence for transcriptional involvement comes from studies of ACE1 proteins that work in conjunction with Cu in inducing biosynthesis of metallothionein, a copper-binding protein, through specific activation of the CUPJ gene for metallothionein. A strain lacking or possessing a defective ACE1 gene fails to induce metallothionein and, unexpectedly, an ability to increase CuZnSOD mRNA synthesis in response to Cu2 (63). The latter observation suggests that CUPJ and SODJ, the CuZnSOD gene, are coordinately regulated through ACE1 protein and copper. The same group (60) that made the initial discovery also reported that without oxygen, less copper entered the cell and an inactive form of CuZnSOD protein accumulated in the cytosol. An active enzyme was obtained when copper ions were added to the extracts, showing that the apoenzyme lacked only copper. This implies that copper is the posttranslational activator. Figure 3 gives the overall scheme of the regulation. The ACE1 binding site has now been shown to encompass bases -183 to -206 upstream from the promoter region of SODJ (64). A strain of yeast cells that carry the scdl allele also fail to synthesize catalytically active CuZnSOD, yet accumulate immunoreactive protein in the cytosol (65). Our best understanding of this is that the scdl allele causes cells to synthesize a modified form of CuZnSOD apoprotein, which for some unknown reason is unable to bind copper andlor assume the conformation of an active enzyme. The mutant provides further evidence that induced enzyme synthesis and the subsequent emergence of an active enzyme are separate metabolic events and that both show a need for copper.
COFACTOR
ACE1
AND
CYTOKINES
Various exogenous and endogenous factors exert major regulation of antioxidant enzymes. Both normal and malignant cells respond to cytolytic effects of tumor necrosis factor (TNF/chacetin) by synthesizing MnSOD (35, 39, 72). TNF increases the MnSOD mRNA in rat lung regardless of the oxygen pressure and without affecting CuZnSOD mRNA or protein (73). Similarly, IL-i induces MnSOD and brings
The FASEB Journal
HARRIS
about a gradual increase in CuZnSOD activity in cultured lung fibroblasts (74, 75). When administered to rats, both TNF and IL-I increase the survival time of rats exposed to hyperoxia (35). Other than SOD, the induction correlates with increased expression of GSH reductase, GSH-Px, and catalase, which remain elevated long after the inducer is withdrawn. Tracheal insuffiation of TNF selectively induces MnSOD mRNA, protein, and activity in normoxiaand hyperoxia-exposed rats; CuZnSOD in unaffected (73). Both superoxide dismutase and catalase activities are elevated in Syrian hamster kidney in response to estrogen treatment (76). Dexamethasone has a similar effect on rat lung catalase and superoxide dismutase (77). The increase correlates with a stimulated transcription rate as opposed to the stability of catalase mRNA, which clearly shows that hormonal effects impinge at the transcriptional level of catalase regulation.
XENOBIOTICS Quinones, viologens, and chemicals of like structure are known to augment O2 production above basal levels. Paraquat (PQ), diquat, menadione, and plumbagin are a few that have been used to induce oxygen stress in otherwise healthy, normal animals and cells. Paraquat, a widely used herbicide, is a potent superoxide generating reagent. MnSOD levels in E. coliare elevated in response to PQ (78). Surprisingly, subcutaneous administration of PQ to rats causes a dose-dependent, linear increase in MT-I in liver and lung (79). The effect requires de novo synthesis of MT-I mRNA. The MT-I increase suggests a potential to redistribute endogenous transition metal ions. Despite the increase in MT-I and its metal ion sequestering activity, lipid peroxidation is still evident in lung and liver of PQ-treated animals. Organic hydroperoxides such as tertiary-butyl hydroperoxide (TBH) weakly induce CuZnSOD and exert little or no effect on GSH reductase or catalase in Chinese hamster V79 cells after brief exposure (80). On the other hand, injections of turpentine into rats fed various levels of copper cause a paradoxical decrease in CuZnSOD in liver (59). The effect has been attributed to a competition between CuZnSOD and ceruloplasmin. The latter protein, which represents a major route for excreting copper from the liver, is elevated in serum from inflamed animals. Rats exposed to ozone for a brief period develop a tolerance to the oxidant that relates to enhanced levels of CuZnSOD, MnSOD, catalase, and GSH-Px in lung tissue (81). Elevating the mRNA concentration for each enzyme seems to be the cause of the increased activities.
to transcriptional regulators that sense sudden changes in oxidant levels. Specific banks of genes with specially designed promoters have evolved to fulfill this function. Whereas E. coli and S. typhimurium use regulons such as oxyR and soxR to modulate SOD, catalase, and peroxidase activities, it is yet to be determined whether a coordinated geneticlevel regulation exists in higher organisms. Survival in an aerobic environment would be expected to depend on a multilevel response system that coordinates the activities of a number of antioxidant enzymes. As exemplified by the ACEI protein in yeast, there would seem to be an important role for metal ion cofactors in regulation of antioxidant enzymes at pre- and posttranslational levels. Because they are known activators of their cognate enzymes, Cu, Fe, Mn, and Se are excellent candidates for coordinating antioxidant enzyme expression. Coordinating the expression of superoxide dismutase with a metallothionein, a metalsequestering protein, is one way of safeguarding cells against protein-free copper ions. Their presence could promote production of the more dangerous hydroxyl radicals through well-defined Fenton chemistry reactions (1). One mystery yet to be resolved is why heme oxygenase, an enzyme that releases the iron atom from a molecule of heme, increases strongly during oxidative stress (82). Clearly, there is still much more to be learned. As exemplified in this review, attempts to understand antioxidant enzyme regulation have focused on single enzymes responding to specific oxidant signals. Because of the more complex interactions taking place in situ, that approach will have to be expanded. Funding for this review was provided in part by National Institutes of Health grant DK41682 and project number H-6621 of the Texas Agricultural Experiment Station. I also express appreciation to numerous colleagues for donating unpublished manuscripts and preprints.
REFERENCES 1. Fridovich,
I. (1989)
Superoxide
dismutases.
An
adaption
to a
paramagnetic gas. j Biol.Chem. 264, 7761-7764 2. Hassan, H. M. (1989) Microbial superoxide dismutases. Adv. Genet. 26, 65-97 3. Allen, R. G. (1991) Oxygen-reactive species and antioxidant responses during development: the metabolic paradox of cellular differentiation. Pmc. Soc. Exp. BioL Med. 196, 117-129 4. Demple, B. (1991) Regulation of bacterial oxidative stress genes. Annu. Rev. Genet. 25, 315-337 5. Ceballos-Picot,
I.,
Nicole,
A.,
Briand,
P.,
Grimber,
G.,
Delacourte, A., Defossez, A., Javoy-Agid, F., Lafon, M., Blouin, J. L., and Sinet, P. M. (1991) Neuronal-specific expression of human copper-zinc superoxide dismutase gene in transgenic
CONCLUSION It may seem ironic that molecules as delicate as enzymes are a first line defense against dangerous and destructive oxygen radicals. In search of a reason, one must consider that enzymes are designed specifically to execute reactions with speed, specificity, and high affinity. These are certainly desirable properties of any antioxidant that must deal with infinitesimally small and often fleeting amounts of radicals that exist in cells. In addition, the major antioxidant enzymes all possess redox elements at their active site, which tends to shield the enzyme protein from the brunt of the chemistry. A third and perhaps most important reason is that enzymes fit into a genetic scheme of regulation in that their concentration in the cell is rapidly elevated in response
ANTIOXIDANT
mice: animal model of gene dosage effects in Down’s synBrain Res. 552, 198-214 6. Schwaiger, H., Weirich, H. G., Brunner, P., Rass, C., HirschKauffmann, M., Groner, Y., and Schweiger, M. (1989) Radia-
drome.
ENZYMES
tion sensitivity of Down’s syndrome fibroblasts might be due to overexpressed Cu/Zn-superoxide dismutase (EC 1.15.1.1), Eur. j Cell Biol.
7. Kokoglu,
48,
79-87
E., Aktuglu,
G., and
superoxide dismutase levels Leukemia Rn. 13, 457-458
Belce, A. (1989) Leucocyte and chronic leukemias.
in acute
8. Steinman, H. M. (1986) Bacteriocuprein superoxide dismutase of photobacterium leiognathi. Isolation and sequence of the gene and evidence for a precursor form. j Biol.Chem. 262, 1882-1887
9. Hassan, H. M., and Fridovich, thesis of superoxide dismutase methyl
viologen.
1. (1977) Regulation of the synin Escherichia coli: induction by J. BioL Chem. 252, 7667-7672
2681
10. Autor, A. P. (1982) Biosynthesis of mitochondrial dismutase in Saccharomyces cerevisiae. Precursor mitochondrial superoxide dismutase made in the J. Biol. Chem. 257, 2713-2718
11. Wisp,
12.
13.
14.
15.
J.
R., Clark,
J.
C., Burhans,
M.S., Kropp,
superoxide form of cytoplasm.
K. E., Korf-
hagen, T. R., and Whitsett, J. A. (1989) Synthesis and processing of the precursor for human mangano-superoxide dismutase. Biochim. Biophys. Acta 994, 30-36 Schrank, I. S. , Sims, P. F. G., and Oliver, S. G. (1988) Functional expression of the yeast Mn-superoxide dismutase gene in Escherichia coli requires deletion of the signal peptide sequence. Gene 73, 121-130 Oberley, L. W., and Oberley, T. D. (1988) Role of antioxidant enzymes in cell immortalization and transformation. Molec. Cell. Biol. 84, 147-153 Chang, L.-Y., Slot, J. W., Geuze, H. J. and Crapo, J. D. (1988) Molecular immunocytochemistry of the CuZn superoxide dismutase in rat hepatocytes. J. Cell BioL 107, 2169-2179 Makita, T, Ishida, T., Kawata, M., and Lee, K. T (1990) Cytochemical and immunohistochemical localization of superoxide dismutase (SOD) in the intima of swine and rat aorta. Ann. N} Acad. Sci. 598, 510-511
16. Keller, G.-A., Warner, T. G., R. A. (1991) Cu,Zn superoxide zyme in human fibroblasts and Sci. USA 88, 7381-7385 17. Crouch, R. K., Patrick, G. J., Simson, dismutase 377-383
J.
18. Cannon,
Steimer, K. S., and Hallewell, dismutase is a peroxisomal enhepatoma cells. Proc. NatI. Aced. Reynolds,
S., Buse, M. G., and
A. V. (1984) Localization of copper-zinc in the endocrine pancreas. Exp. Molec.
R. E., and Scandalios,J.
superoxide Pathol. 41,
during
dismutase
proteins
in
22. Kelner, M. J., Bagnell, R., Hale, B., and Alexander, N. M. (1989) Inactivation of intracellular copper-zinc superoxide dismutase by copper chelating agents without glutathione depletion and methemoglobin formation. Free Red. BioL Med. 6,
Demple,
38.
39.
40.
41.
42.
24. Kanematsu,
44.
K. (1989) CuZn-superoxide
dismu-
45.
46.
47.
VOL. I, pp. 259-291, CRC Press, Boca Raton, Florida 28. Yoshimura, S., Watanabe, K., Suemizu, H., Onozawa, T., Mizoguchi, J., Tsuda, K., Hatta, H., and Moriuchi, T, (1991)
48.
Tissue
Vol.6
June 1992
Biophys.
Ada
Positive
control
of a global
antioxidant
L., Iqbal, protocol
J. Hass,
M., and Massaro, D. (1989) New “rest for inducing tolerance to high 02 exposure in rats. Am. j PhysioL 257, L226-L231
H. P., Chandler,
D. B., and Chen,
Z. (1983)
Increases in lung prolyl hydroxylase and superoxide dismutase activities during bleomycin-induced lung fibrosis in hamsters. Exp. Moles. Pathol. 39, 317-326 Spence, T H., Jr., Jenkinson, S. G., Johnson, K. H., Collins, J. F, and Lawrence, R. A. (1986) Effects of bacterial endotoxin on protecting copper deficient rats from hyperoxia. j AppL PhysioL 61, 982-987 Visner, G. A., Block, E. R., Burr, I. M., and Nick, H. S. (1991) Regulation of manganese superoxide dismutase in porcine pulmonary artery endothelial cells. Am. j PhysioL 260, L444-L449 Berg, J. T., and Smith, R. M. (1988) Protection against hyperoxia by serum from endotoxin treated rats: absence of superoxide dismutase induction. Proc.Soc.Exp. Biol. Med., 187, 117-122 White, C. W., Avraham, K. B., Shanley, P. F, and Groner, Y. (1991) Transgenic mice with expression of elevated levels of copper-zinc superoxide dismutase in the lungs are resistant to pulmonary oxygen toxicity. j Clin. Invest. 87, 2162-2168 Turrens, J. F, Crapo, J. D., and Freeman, B. A. (1984) Protection against oxygen toxicity by intravenous injection of liposome-entrapped catalase and superoxide dismutase. j Clin. Invest. 73, 87-95 Taylor, C. G., Towner, R. A., Janzen, E. G., and Bray, T. M. (1990) MRI detection of hyperoxia-induced lung edema in Zndeficient rats. Free Red. Biol. Med. 9, 229-233 Ceballos, I.,Javoy-Agid, F., Hirsch, E. C., Dumas, S., Kamoun, P. P., Sinet, P. M., and Agid, Y. (1989) Localization of copperzinc superoxide dismutase mRNA in human hippocampus by in situ hybridization. Neurosci. Lett. 105, 41-46 Percival, S. S., and Harris, E. D. (1991) Regulation of Cu,Zn su-
peroxide
27. Spallholz, J. E., and Boylan, L. M. (1991) Gluthathione peroxidase: the two selenium enzymes. In Peroxidases in Chemistry and Biology (Everse, J., Everse, K. E., and Grisham, M. B., eds)
2682
Biochem.
dismutase
with
copper.
Caeruloplasmin
maintains
level
255, 223-228
specific expression of the plasma glutathione peroxidase gene in rat kidney. J. Biochem. (Tokyo) 109, 918-923 29. Weitzel, F, Ursini, F, and Wendel, A. (1990) Phospholipid hydroperoxide glutathione peroxidase in various mouse organs
B. (1990)
adult 37. Gin, S. N., Misra,
43.
S., and Asada,
repletion.
1003-1007 36. Frank, period”
355-360 23. Kanematsu, S., and Asada, K. (1989) CuZn-superoxide dismutases in rice: occurrence of an active, monomeric enzyme and two types of isozyme in leaf and non-photosynthetic tissues. Plant Cell Physiol. 30, 381-391 tases from the fern Equisetum arvense and the green alga Spirogyra sp.: occurrence of chloroplast and cytosol types of enzyme. Plant Cell PhysioL 30, 717-727 25. Marklund, S. L. (1982) Human copper-containing superoxide dismutase of high molecular weight. Proc. NatL Acad. Sci. USA 79, 7634-7638 26. Karlsson, K., and Marklund, S. L. (1988) Extracellular superoxide dismutase in the vascular system of mammals. Biochem. j
and
defense regulon activated by superoxide-generating agents in Escherichia coli. Proc. NatI. Acad. Sci. USA 87, 6181-6185 33. Hass, M. A., and Massoro, D. (1987) Differences in CuZnsuperoxide dismutase induction in lungs of neonatal and adult rats. Am. j Physiol. 253, C66-C70 34. Stevens, J. B., and Autor, A. P., (1977) Induction of superoxide dismutase by oxygen in neonatal rat lung. j Biol. Chem. 252, 3509-3514 35. White, C. W., Ghezzi, P., McMahon, S., Dinarello, C. A., and Repine, J. E. (1989) Cytokines increase rat lung antioxidant enzymes during exposure to hyperoxia. j AppL PhysioL 66,
19. Seto, N. 0. L., Hayashi, S., and Tener, G. M. (1989) Cloning, sequence analysis and chromosomal localization of the Cu-Zn superoxide dismutase gene of Drosophila melanogaster. Gene 75, 85-92 20. Hass, M. A., Iqbal, J., Clerch, L. B., Frank, L., and Massaro, D. (1989) Rat lung Cu,Zn superoxide dismutase. Isolation and sequence of a full-length cDNA and studies of enzyme induction. j Clin. Invest. 83, 1241-1246 21. Ischiropoulos, H., Nadziejko, C. E., and Kikkawa, Y. (1990) Effect of aging on pulmonary superoxide dismutase. Mech. Ageing Dev. 52, 11-26
deficiency
88-94
30. Avissar, N., Slemmon, J. R., Palmer, I. S., and Cohen, H. J. (1991) Partial sequence of human plasma glutathione peroxidase and immunologic identification of milk glutathione peroxidase as the plasma enzyme. j Nutt 121, 1243-1249 31. Christman, M. F., Morgan, R. W., Jacobson, F. S., and Ames, B. N. (1985) Positive control of a regulon for defenses against oxidative stress and some heat-shock proteins in Salmonella typhimurium. Cell 41, 753-762 32. Greenberg, J. T., Monach, P., Chou, J. H., Josephy, P. D., and
G. (1989) Two cDNAs encode
two nearly identical Cu/Zn superoxide maize. Mol. Gen. Genet. 219, 1-8
selenium
1036,
49.
of functional enzyme activity during differentiation of K562 cells.Biochem. j 274, 153-158 Saito, H., Kuroki, T., and Nose, K. (1989) Decrease in Cu-Znsuperoxide dismutase mRNA level during differentiation of human monocytic and promyelotic leukemia cells. FEBS Lett. 249, 253-256 Speier, C., and Newburger, P. F. (1986) Changes in superoxide dismutase, catalase, and the glutathione cycle during induced myeloid differentiation. Arch. Biochem. Biophys. 251, 551-557 Oberley, L. W., Ridnour, L. A., Sierra-Rivera, E., Oberley, T. D., and Guernsey, D. L. (1989) Superoxide dismutase activities of differentiating clones from an immortal cell line. j Cell. Physiol. 138, 50-60 Stange, R. C., Cotton, W., Fryer, A. A., Drew, R., Bradwell,
The FASEB Journal
HARRIS
A. R., Marshall, T, Collins, M. F, Bell, J., and Hume, R. (1988) Studies on the expression of Cu,Zn superoxide dismutase in human tissues during development. Biochem. Biophys. Acta
964,
66.
260-265
50. Hayashibe, H., Asayama, K., Dobashi, K., and Kato, K. (1990) Prenatal development of antioxidant enzymes in rat lung, kidney, and heart: marked increase in immunoreactive superoxide dismutases, glutathione peroxidase, and catalase in the kidney. Pediatr. Ret.27, 472-475 51. Pittschieler, K., Lebenthal, F., Bujanover, Y., and Petell, J. K. (1991) Levels of Cu-Zn and Mn superoxide dismutases in rat liver during development. Gastroenteivlogy 100, 1062-1068 52. Del Maestro, R., and McDonald, W. (1989) Subcellular localization of superoxide dismutases, glutathione peroxidase and catalase in developing rat cerebral cortex. Mech. Ageing Dcv. 48, 15-31 53. Mariucci, G., Ambrosini, M. V., Colanieti, G. (1990) Differntial changes in Cu,Zn and mutase activity in developing rat brain and 753-755
54. Kellogg,
E. W., and Fnidovich,
in the rat and mouse tol. 31, 405-408
L., and Bruschelli, Mn superoxide disliver. Experientia 46,
I. (1976) Superoxide
as a function
dismutase
of age and longevity.j
Geron-
59, 123-128 berg,
T. M.,
Notaro,
J.,
A. A. (1989) Elevated
syndrome: 5239-5243
a genetic
58. Bettger, copper
J.,
W.
superoxide
condition
Fish, T zinc status
and
dismutase
68.
69.
CuZn-supenoxide Biochem.J. 248,
Notaro,
J.,
S., Schumer,
superoxide of oxidative
and
Sand-
dismutase stress.
in Bloom’s Cancer Res. 49,
and O’Dell, B. L. (1978) Effects of rats on erythrocyte stability and activity. Proc. Soc.Exp. Biol.Med. 158,
60. Galiazzo, Marcocci,
61.
62.
63.
64.
65.
F, Ciriolo, M. L., Marmocchi,
75.
Kosman,
ANTIOXIDANT
D.
J.
(1991) Genetic
ENZYMES
76.
R., Carri, M. T, Civitareale, P., F, and Rotilio, G. (1991) Activation
and induction by copper of Cu/Zn superoxide dismutase in Saccharomyces cerevisiae. Presence of an inactive proenzyme in anaerobic yeast. Eur.j Biochem. 196, 545-549 Greco, M. A., Hrab, D. I., Magner, W., and Kosman, D. J. (1990) Cu,Zn supenoxide dismutase and copper deprivation and toxicity in Saccharomyces cerevisiae. j bacteriol. 172, 317-325 Saito, T., Ito, K., Kurasaki, M., Funjimoto, S., Kaji, H., and Saito, K. (1982) Determination of true specific activity of superoxide dismutase in human enythrocytes. Clin.Sci.63, 251-255 Carni, M. T, Galiazzo, F, Ciriolo,M. R., and Rotiio, G. (1991) Evidence for co-regulation of Cu,Zn superoxide dismutase and metallothionein gene expression in yeast through transcriptional control by copper via the ACE 1 factor. FEBS Left. 278, 263-266 Gralla, E. B., Thiele, D. J., Silar, P., and Valentine, J. S. (1991) ACEI, a copper-dependent transcription factor, activates expression of the yeast copper, zinc superoxide dismutase gene. Proc. NatL Acad. Sci. USA 88, 8558-8562 Chang, E. C., Crawford, B. F, Hong, Z., Bilinski, T, and
and biochemical
characterization
erythnocyte
L506-L512 74. Masuda, A., Longo,
279-282 59. DiSilvestro, R. A., and Marten, J. T (1990) Effects of inflammation and copper intake on rat liver and erythrocyte Cu-Zn superoxide dismutase activity levels. J. Nutr. 120, 1223-1227
with
copper.
Effects
in
vivo.
superoxide
dismutase
activity
in low birth
weight infants given iron supplements. Pediatr. Ret. 29, 297-301 72. Asoh, K-I, Watanabe, Y., Mizoguchi, H., Mawatari, M., Ono, M. Kohno, K., and Kuwano, M. (1989) Induction of manganese superoxide dismutase by tumor necrosis factor in human breast cancer MCF-7 cell line and its TNF-resistant variant. Biochein. Biophys. Ret. Commun. 162, 794-801 73. Tsan, M. -F., White, J. E., Treanor, C., and Shaffer, J. B. (1990) Molecular basis for tumor necrosis factor-induced increase in pulmonary superoxide dismutase activities. Am. j PhysioL 259,
J.,
of
dismutase
663-668 70. Beyer, W. F, Jr., and Fnidovich, I. (1991) In vivocompetiton between iron and manganese for occupancy of the active site region of the manganese-superoxide dismutase of Escherichia coli. j Biol. Chem. 266, 303-308 71. Barclay, S. M., Aggett, P. J., Lloyd, D. J., and Duifty, P. (1991)
Reduced
55. Vanella, A., Villa, R. F, Gonini, A., Campisi, A., and Giuffrida-Stella, A. M. (1989) Superoxide dismutase and cytochrome oxidase activities in light and heavy synaptic mitochondria from rat cerebral cortex during aging. j Neurosci. Res. 22, 351-355 56. De, A. K., and Darad, R. (1991) Age-associated changes in antioxidants and antioxidative enzymes in rats. Mech. Ageing Dcv. 57. Nicotera,
67.
of Cu,Zn superoxide dismutase mutants in Saccharomyces cerevisiae. j Biol. Chem. 266, 4417-4424 Prohaska, J. R. (1991) Changes in Cu,Zn-superoxide dismutase, cytochrome c oxidase, glutathione peroxidase and glutathione transferase activities in copper-deficient mice and rats. J. Nutr. 121, 355-363 Jenkinson, S. G., Lawrence, R. A., Bunk, R. F, and Williams, D. M. (1982) Effects of copper deficiency on the activity of the selenoenzyme glutathione peroxidase and on excretion and tissue retention of 75SeO32. j Nutr. 112, 197-204 Chung, K., Romero, N., Tinker, D., Keen, C. L., Amemiya, K., and Rucker, R. B. (1988) Role of copper in the regulation and accumulation of superoxide dismutase and metallothionein in rat liver. j Nutr 118, 859-864 Dameron, C. T., and Harris, E. D. (1987) Regulation of aortic
77.
78.
79.
80.
81.
D. L., Kobayashi,
Y., Appella,
E., Oppen-
heim, J. J., and Matsushima, K. (1988) Induction of mitochondrial manganese superoxide dismutase by interleukin 1. FASEBJ 2, 3087-3091 DiSilvestro, R. A., David, F. A., and Collignon, C. (1991) Interleukin 1 slowly increases lung fibroblast Cu-Zn superoxide dismutase activity levels. Proc. Soc. Exp. BioL Med. 197, 197-200 McCormick, M. L., Oberley, T. D., Elwell, J. H., Oberley, L. W., Sun, Y., and Li, J. J. (1991) Supenoxide dismutase and catalase levels during estrogen-induced renal tumonigenesis, in renal tumors and their autonomous variants in the Syrian hamster. Carcinogenesis 12, 977-983 Clench, L. B., Iqbal, J., and Massaro, D. (1991) Perinatal rat lung catalase gene expression: influence of corticosteroid and hypenoxia. Am. j PhysioL260, L428-L433 Bowen, S. W., and Hassan, H. M. (1988) Induction of the manganese-containing superoxide dismutase in Escherichia coli is independent of the oxidative stress (oxyR-controlled) regulon. j Biol.Chem. 263, 14808-14811 Sato, M. (1991) Dose-dependent increases in metallothionein synthesis in the lung and liver of paraquat-treated rats. Toxicol. AppL Pharmacol. 107, 98-105 Ochi, T. (1990) Effects of an organic hydropenoxide on the activity of antioxidant enzymes in cultened mammalian cells. Toxicology61, 229-239 Rahman, 1.-U., Clench, L. B., and Massaro, D. (1991) Rat lung antioxidant enzyme induction by ozone. Am. j Physiol. 260,
L4l2-L418 82. Saunders, E. L., Maines, M. D., Meredith, M. J., and Freeman, M. L. (1991) Enhancement of hemeoxygenase-1 synthesis by glutathione depletion in chinese hamster ovary cells. Arch. Biochem. Biophys. 288, 368-373
2683
