If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The superoxide free radical has come to occupy an amazingly central role in a wide variety of diseases. Our metabolic focus on aerobic energy metabolism in all cell types, coupled with some chemical peculiarities of the oxygen molecule itself, contribute to the phenomenon. Superoxide is not, as we once thought, just a toxic but unavoidable byproduct of oxygen metabolism. Rather it appears to be a carefully regulated metabolite capable of signaling and communicating important information to the cell’s genetic machinery. Redox regulation of gene expression by superoxide and other related oxidants and antioxidants is beginning to unfold as a vital mechanism in health and disease.
Free radicals, oxidative stress, and antioxidants have become commonly used terms in modern discussions of disease mechanisms. The purpose of this article is to provide some understanding as to why free radical metabolism seems to occupy a central and remarkably common position in the mechanisms of so many seemingly unrelated types of human disease. In articles to follow in this series, experts in a broad spectrum of clinical specialties will examine more specifically the free radical mechanisms known to be operative in the inflammatory process and in ischemia/reperfusion injury, as they affect a variety of organ systems. In addition, there are some free radical mechanisms that are unique to certain tissue locations, especially those affecting the central nervous system.
In a general sense, substantial insight regarding free radical production can come from the recognition of a few basic facts about the nature of the chemical reactions that support life as we know it, and from an understanding of the quirky structure of molecular oxygen, upon which our metabolism is based. We can gain further insight as we consider the course of our evolution in a redox active milieu. As evolutionary pressures tend to make the best of any bad situation, we will see examples of evolutionary steps to contain, control, or avoid unpleasant toxic species, and other examples of how these reactive species can be turned to constructive purposes contributing to our defense, or even to our internal regulation and communication processes.
A redox primer
At the risk of inducing unpleasant flashbacks to college chemistry courses, let us consider the class of chemical reactions upon which our bodies rely for energy production. Reduction-oxidation (or redox) reactions are at the core of our metabolic machinery. Redox reactions involve the transfer of electrons or hydrogen atoms from one reactant to another. (The process of taking away electrons is called “oxidation,” because oxygen does it so well. The substance receiving electrons becomes “reduced.”) Life on our planet has evolved into two main categories of lifeforms: 1) photosynthesizing plants that capture solar energy and use it to drive thermodynamically unfavorable reactions producing reduced carbonaceous compounds, and 2) the rest of us, who eat these reduced carbonaceous compounds and “burn” them in thermodynamically favorable reactions. The latter reactions allow the electrons (or hydrogen atoms) to return to the molecule that wants them most, molecular oxygen, with the release of large amounts of energy that we can cleverly trap and use to power our lives. This process might be straightforward but for the quirky electronic structure of the oxygen molecule. Oxygen itself is a diradical. (Chemically, a free radical is any molecule containing a single, unpaired electron. Usually, paramagnetic transition metal ions are not considered to be free radicals, although by technical definition they are.) Most molecules contain only pairs of electrons with opposite spins that reside in discrete molecular orbitals and that may or may not participate in bond formation. Oxygen contains two electrons that are not spin paired, and each resides in an orbital of its own. Thermodynamically, oxygen wants to take on additional electrons (two per atom, or four per molecule) to produce water molecules, which have much lower free energy. The unconventional distribution of electrons in the oxygen molecule, however, makes it impossible for oxygen to accept a spin-matched pair of electrons, as badly as it may want them, until one of its unpaired electrons undergoes a spontaneous spin reversal to make pairing possible. At ordinary collisional frequency, the period of contact is too brief for this spin reversal to occur, imposing a kinetic barrier (ie, a large energy of activation) to most oxidative reactions. It is this kinetic barrier that saves us from reacting explosively with an atmosphere of huge thermodynamic oxidizing potential. It is this same kinetic restriction that makes oxygen an ideal terminal electron acceptor for biological systems. Enzymes have binding sites that can hold oxygen in contact with an oxidizable substrate for a much longer time than would occur by simple collision, overcoming the kinetic barrier to reaction. At the same time, enzymes may be designed to trap much of the energy released as the oxidation occurs in a useful high-energy compound, such as ATP.
Oxygen-derived free radicals and oxidants
Our relationship with oxygen is, at best, a difficult one to manage. Occasionally, under normal biological conditions, oxygen does manage to steal away electrons from other molecules by nonenzymatic autoxidations. Because it cannot accommodate a spin-matched pair, it must settle for stealing electrons one at a time. This breaking up of electron pairs results in free radical formation. The one-electron reduction product of oxygen is the superoxide radical, O2•−. If two electrons are transferred, the product is hydrogen peroxide, H2O2, which is not a radical. It is nonetheless still eagerly receptive of two more electrons, causing hydrogen peroxide to be a cytotoxic oxidant. Certain chelates of ferrous iron and cuprous copper are capable of transferring a third electron to hydrogen peroxide, causing lysis of the O-O bond. One fragment is reduced to the state of water; the other fragment is the hydroxyl free radical, HO•, one of the most potent oxidants known. It can initiate lipid peroxidation, cause DNA strand breaks, and indiscriminately oxidize virtually any organic molecule. The fact that it is so indiscriminate actually works in our favor, as most of its targets are expendable. Reactivity and toxicity are not synonymous (
McCord JM. Superoxide production and human disease. In: Jesaitis A, Dratz E, eds. Molecular Basis of Oxidative Damage by Leukocytes. Boca Raton, FL: CRC Press, 1992:225–239.
). Despite the much lesser reactivity of superoxide, its toxicity appears to be substantial precisely because its targets are focused. Cyanide, for example, is not particularly reactive but is extremely toxic, because it effectively strikes one crucial target, cytochrome oxidase.
The family of reactive intermediates resulting from the incomplete reduction of oxygen therefore includes superoxide radical, hydrogen peroxide, and hydroxyl radical. It is not correct to refer to this group as oxygen-derived free radicals, because one member, hydrogen peroxide, is not a radical. Accordingly, several terms are now in use to refer to this family. Reactive oxygen metabolites (ROM) or active oxygen (AO) or variations thereon are the most common. Occasionally, the terms are expanded to include electronically excited oxygen (singlet oxygen), hypochlorous acid (produced from oxygen by the neutrophil enzymes NADPH oxidase and myeloperoxidase), peroxynitrite, and even nitric oxide, as all of these oxidants are derived from molecular oxygen.
The evolution of the aerobic lifestyle
Life on this planet first evolved in a reducing atmosphere. It was not until photosynthetic algae appeared that oxygen began to be introduced into the atmosphere in ever increasing quantities. This shift from a reducing environment to an oxidizing one undoubtedly resulted in some serious evolutionary pressures. One might be surprised when examining modern metabolic pathways to find that very few enzymes actually deal with molecular oxygen, despite the fact that our bioenergetics scheme is completely dependent on the transfer of electrons to this acceptor. In fact, approximately 98% of the oxygen we metabolize is handled by a single enzyme, the cytochrome oxidase in our mitochondria, which transfers four electrons to oxygen in a concerted reaction to produce two molecules of water as the product. The enzyme is structurally quite complex, containing four redox centers (two hemes and two copper ions), each of which can store a single electron. When all centers are reduced, the simultaneous transfer of four electrons to an oxygen molecule occurs with no detectable intermediate steps. One probable reason for this dominance by cytochrome oxidase in the reduction of oxygen is the chemical difficulty of carrying out the reaction in a safe and controlled manner. As discussed above, the reduction of oxygen by anything less than the full complement of four electrons results in the production of active oxygen species. Surely, as primitive metabolic systems were struggling with the shift toward energy-rich oxidative pathways, there must have been many contenders that lost out in the end, because unacceptable levels of active oxygen products were produced. Hence, one evolutionary strategy for survival in an oxidative environment is to restrict opportunities for poorly controlled transfer of electrons to oxygen. Although many enzymes involved in redox pathways exist as reduced intermediates with great thermodynamic potential for the transfer of their electrons to oxygen, the evolutionary pressure has led to kinetic barriers against such reactivity. Instead, most of these high-energy electrons are transferred to NADP+ to produce NADPH, which is itself kinetically resistant to reaction with oxygen. Thus, our electron-conducting metabolic circuits are insulated by evolutionary design to prevent the inadvertent development of short-circuits, much as the electron-conducting circuits in a modern electronic circuit board are insulated to keep electrons flowing in the proper channels. Our insulation is not perfect. At least two sites have been identified in the electron-transport chain (Complex I and ubisemiquinone) where electrons may leak out through breaks in the insulation to waiting oxygen molecules, resulting in the formation of superoxide (
). These sites are diagrammatically represented in Figure 1 as the primary sources of intracellular superoxide generation. It has been estimated that this leakage amounts to 1% to 2% of total electron flux through the mitochondria. Because working myocardium consumes oxygen at approximately 8 mM per minute, rates of superoxide production could exceed 0.1 mM per minute. So, as well designed as we may be, mitochondria are still the major source of accidental free radical production. During exposure to hyperoxia, the rates of leakage from these sites in lung mitochondria are believed to increase in direct proportion to the increased oxygen tension. Healthy adult rats will die within 72 hours if placed in an atmosphere of 100% oxygen, only five times the normal concentration at sea level (
Figure 1A schematic representation of signal transduction pathways for superoxide radical. Mitochondrial respiration accounts for most of the superoxide generated in a cell with leakage sites at Complex I and at ubisemiquinone. The steady-state concentration of superoxide is kept low in all compartments by SODs, not shown. The low levels of the radical remaining may modulate various kinases, or may activate transcription factors directly to effect gene regulation in the nucleus. It is interesting to speculate on the existence of a cell surface receptor for superoxide (R), which might transduce various responses within the cell by means of kinase activation, for example. There is presently no direct evidence for such a receptor.
The evolution of antioxidants and antioxidant enzymes
In addition to evolutionary attempts to avoid the production of reactive byproducts of oxidative metabolism, another very important direction was the ability to synthesize or accumulate antioxidants—molecules that would avidly react with and annihilate active oxygen species before they could inflict oxidative damage to vital components, such as DNA or cell membranes. The result was hundreds of kinds of such antioxidant molecules, especially in plants. Among the most successful of these molecules are the water-soluble antioxidant ascorbic acid (vitamin C) and the lipid-soluble antioxidant α-tocopherol (vitamin E) (
The most efficient way to eliminate undesirable toxic species, of course, is by means of catalysis. Families of antioxidant enzymes have evolved for this purpose, including superoxide dismutases for the elimination of the superoxide radical, and catalases and glutathione peroxidases for the elimination of hydrogen peroxide and organic peroxides. Humans have three genes encoding superoxide dismutases (SOD), which localize in the mitochondria, the cytosol, or the extracellular spaces (
). These genes are derived from two ancestral genes. One gene gave rise to the copper-and-zinc–containing enzymes; the other gave rise to the manganese- or iron-containing enzymes. The genes can be traced back to the most primitive organisms with high degrees of homology around the active sites, indicating that the genes that evolved early as life forms were figuring out how to survive and thrive in the presence of oxygen. The SODs (
) catalyze the reaction: This dismutation or disproportionation reaction makes use of the fact that superoxide is both an oxidant and a reductant, eager to get rid of its extra electron or to take on another. The enzyme uses one superoxide radical to oxidize another. Catalases work in much the same way, because hydrogen peroxide can be a weak reductant as well as a fairly strong oxidant: In higher organisms, glutathione peroxidases appear to have largely supplanted the need for catalase. These enzymes use NADPH as the reducing species for hydrogen peroxide: They can reduce lipid peroxides as well as hydrogen peroxide and are very important enzymes in the prevention of lipid peroxidation to maintain the structure and function of biologic membranes.
The reactivity and toxicity of superoxide radical
Although the chemical reactivity of the superoxide radical is modest, its toxicity is quite easily demonstrated. Escherichia coli contains three genes for SODs: one enzyme uses manganese as its cofactor, one uses iron, one uses copper and zinc. Disruption of the two major genes encoding the manganese and iron enzymes results in a bacterium unable to grow aerobically in minimal medium but still able to grow anaerobically (
). Aerobically, it displays multiple auxotrophies and can grow if all amino acids are added to the minimal medium, indicating that several biosynthetic pathways for amino acids are sensitive to inactivation by the radical. Indeed, certain dehydratases in E. coli have subsequently been shown to be sensitive to inactivation by superoxide radical: the α, β-dihydroxyisovalerate dehydratase (
). More than a dozen other important enzymes have similarly been shown to be inactivated by superoxide, including the following mammalian enzymes: catalase (
McCord JM, Russell WJ. Superoxide inactivates creatine phosphokinase during reperfusion of ischemic heart. In: Cerutti PA, Fridovich I, McCord JM, eds. Oxy-Radicals in Molecular Biology and Pathology. New York: Alan R. Liss, 1988:27–35.
The toxicity of superoxide is seen not only in its ability to inhibit certain enzymes and thereby attenuate vital metabolic pathways, but also in its effects on other major classes of biological molecules. E. coli deficient in SOD activity show increased rates of mutagenesis (
), illustrating the role of the radical, directly or indirectly, in DNA damage. In conditions of ischemia and reperfusion, the most acute problem resulting from the overproduction of superoxide appears to be greatly increased rates of lipid peroxidation. Here, superoxide radical plays paradoxical roles, in that it can both initiate and terminate lipid peroxidation chains (
). This dichotomy of good and bad will be explored further below. The practical result, however, is that a proper balance between oxidants and antioxidants is required. Superoxide radical is not all bad, as we shall see in other ways.
Knockout mice have now been produced for each of the three mammalian SOD genes separately, but not yet in combination. Surprisingly, SOD1 knockouts, missing the cytosolic copper-zinc SOD, get along quite well until they are stressed. They do show increased neurologic and histologic damage after focal cerebral ischema and reperfusion (
). It is the homozygous SOD2 knockouts, however, that dramatically illustrate how toxic superoxide can be. SOD2 encodes the manganese-containing SOD that localizes to the mitochondrion, the site that is responsible by far for most of the cellular production of superoxide. SOD2 knockouts are born small, but alive, but they die within days of birth with a dilated cardiomyopathy (
One of the most fascinating aspects of evolution is the ability to make the best of a bad situation, to make a silk purse from the proverbial sow’s ear. There are clear examples of how active oxygen products, which we generally try to avoid producing at all costs, can actually be put to constructive uses. The best example is the evolution of our phagocytic NADPH oxidase. When first discovered as a biologic metabolite, it appeared that the superoxide radical was simply a noxious cytotoxic byproduct that served no good purpose. That view changed when Babior et al (
) realized that the radical is an important player in our defense against invading microbes. It is now universally accepted that the production of superoxide radical by activated polymorphonuclear leukocytes and other phagocytes is an essential component of their bactericidal armamentarium (
). Precisely because superoxide is cytotoxic, this NADPH oxidase has evolved to circumvent the kinetic stability of NADPH and specifically to allow its oxidation by molecular oxygen with the production of superoxide radical. It equips phagocytes (which have evolved mechanisms to detect and engulf invading microorganisms) with a way to destroy chemically the ingested microbes. In effect, superoxide serves as an extremely broad spectrum antibiotic. The neutrophil is also destroyed in the process, a Kamikaze mission, by its own artillery. In addition, surrounding healthy host cells may be injured or even killed in the crossfire (
). In effect, superoxide is a mediator of inflammation, and SODs display anti-inflammatory activity. It should be noted that our immune defense system tends to overreact to any challenge, as too timid a response may be fatal. The damage associated with the inflammatory process is the price we pay for a vigilant defense system. From a medical therapeutic point of view, our own ingenuity may now have evolved to a position of greater intelligence than our immune system, enabling us to treat ourselves with more targeted synthetic antibiotics (that our bodies have not yet learned to produce) and allowing us to attenuate selectively our inflammatory response to spare ourselves the damage associated with it.
We now believe that the superoxide radical plays additional constructive roles that may be more subtle in nature. When organisms evolved from single-celled creatures to complex, multicellular, multiorgan creatures, a huge paradigm shift took place. For a single-celled organism, the biological imperative is simply to grow and divide without restraint when times are good and food is plentiful. For higher organisms, only epithelial cells (which are continuously being sloughed) are in this grow-and-divide mode. Thus, nearly all the cells in our bodies are under tight constraints that override the biologic imperative that tells our individual DNA molecules to replicate themselves. Certain types of cells are able to escape from the restraints under certain circumstances. For example, fibroblasts are able to proliferate to form scar tissue that is necessary for wound closure and healing. Lymphocytes capable of producing needed antibodies are able to proliferate to create a clone of such cells when appropriately stimulated. In both cases, it appears that superoxide may serve as the signal to override the postmitotic constraints, and both cases may have evolved as secondary responses to the oxidative nature of the primitive immune system’s superoxide-generating machinery.
The boundaries of an open wound become a battlefield for phagocytes versus microbes. The objectives are twofold: to sterilize the wound, and to close the wound. The NADPH oxidase of neutrophils becomes activated, and superoxide and other oxidants derived therefrom, including hydrogen peroxide and hypochlorous acid, accomplish the first objective. It then appears that the phagocyte-generated superoxide serves to stimulate fibroblasts to enter a proliferative mode (
), laying down collagen fibrils and forming scar tissue to close and seal the wound against further infection. The proliferative response of fibroblasts to exposure to superoxide is easily demonstrated in the laboratory (
). Similarly, the clonal expansion of stimulated lymphocytes may be driven largely by superoxide production. B lymphocytes, in particular, contain an NADPH oxidase closely related to the one found in neutrophils and macrophages. When the oxidase is stimulated to produce superoxide by mitogens, the result is cellular proliferation giving rise to a clone of antibody-producing cells.
Redox regulation of gene expression
If oxidative status can signal cells to respond in various ways, we must ask how these signals are transduced, carried, and interpreted, especially by the cell’s genetic machinery. The study of redox regulation of gene expression has exploded in recent years and clearly suggests that oxidants are major determinants of gene expression. Reactive oxygen intermediates have been implicated in the activation of a variety of kinases [such as the Src kinase family (
Cooperativity between oxidants and tumor necrosis factor in the activation of nuclear factor (NF)-kappaB requirement of Ras/mitogen-activated protein kinases in the activation of NF-kappaB by oxidants.
Cooperativity between oxidants and tumor necrosis factor in the activation of nuclear factor (NF)-kappaB requirement of Ras/mitogen-activated protein kinases in the activation of NF-kappaB by oxidants.
). Figure 1 depicts a schematic representation of how reactive oxygen species may regulate gene expression. This extensive interface between oxidants and reductants and the cell’s genetic machinery results in responsiveness to exogenous oxidant exposure and to remarkably effective mechanisms governing redox homeostasis under normal conditions. The unfolding complexity of the system further suggests just how badly things can go wrong when a cell’s redox status is upset.
Oxidative stress and malignancy
Reining in a cell’s biological imperative to proliferate and placing constraints on the natural inclination to replicate DNA and divide is no small feat. Indeed, it may require more sophisticated cellular engineering to squelch the desire to proliferate than to promote it. The connection between mild oxidative stress and cellular growth may date back to the primordial soup. When food is plentiful, metabolism is running at full speed, and there is sufficient energy to support cell division, the rate of superoxide production will also be high (at least in aerobic organisms), producing a state of mild oxidative stress. Conversely, when food supply nears exhaustion, the rate of oxidant production within the cell would drop, possibly signaling insufficient energy production to support the cell’s entry into a vulnerable period of replication.
For a normal postmitotic cell to become malignantly transformed, several conditions may have to be met. It may be necessary to relieve certain evolutionary constraints that tell the cell not to enter the cell cycle leading to mitosis. It may be necessary to provide mild oxidative stress to serve as the driving force for proliferation. It may even be necessary to disable yet another set of evolutionary constraints designed to prevent cells from running amok by triggering apoptosis (
McCord JM, Flores SC. The human immunodeficiency virus and oxidative balance. In: Paoletti R, ed. Oxidative Processes and Antioxidants. New York: Raven Press, 1994;13–23.
). Although wild proliferation may be a mark of success for a bacterium, it is a very dangerous situation in a cell that is part of a human being. The entire organism can be brought down by what begins as a single errant cell that has broken free of its evolutionary constraints. Thus, we have evolved a failsafe system that can detect out-of-control proliferation and can cause programmed self destruction in any cell showing this behavior.
How might a “wannabe” cancer cell achieve and maintain the condition of mild oxidative stress necessary to drive its proliferation? It has been shown that many types of human cancer cells have reduced manganese superoxide dismutase (MnSOD) (
). In most cases, the reduced activity has been assumed to be the result of defective expression of the gene (ie, changes in the promotor region of the gene) (
). Oberley, St. Clair, and others have observed in numerous studies that transfection with the gene for human MnSOD can reverse the malignant phenotype of tumor cells, suggesting that MnSOD functions as a tumor suppressor (
) reported finding a variant sequence containing a cluster of three mutations in the promoter regions of the MnSOD genes from 5 of 14 human cancer cell lines examined. All 5 cell lines were heterozygous for the variant sequence. The mutations change the binding pattern of transcription factor AP-2 and cause a marked diminution in the efficiency of the promoter using a luciferase reporter assay system. Alternatively, mutations in the coding region of MnSOD may adversely affect catalytic efficiency or the stability of the protein.
Oxidative stress and human disease
Perhaps the most noteworthy observation concerning the role of oxidative stress in human disease is the commonality of it. Oxidative stress is now thought to make a significant contribution to all inflammatory diseases [arthritis (
Role of oxidant stress in the adult respiratory distress syndrome evaluation of a novel antioxidant strategy in a porcine model of endotoxin-induced acute lung injury.
Synthetic combined superoxide dismutase/catalase mimetics are protective as a delayed treatment in a rat stroke model a key role for reactive oxygen species in ischemic brain injury.
), and many others. The reason that overproduction of free radicals is a feature of such a broad spectrum of diseases derives from the fact that oxidative metabolism is a necessary part of every cell’s metabolism. If that cell is sick or injured in any way that results in mitochondrial injury (calcium influx, leaky membranes, and so forth), then increased production of superoxide is likely to result. In the series of articles to follow, experts in many of these areas will delineate specific roles for free radicals and oxidative stress in a number of the diseases mentioned above.
References
McCord JM. Superoxide production and human disease. In: Jesaitis A, Dratz E, eds. Molecular Basis of Oxidative Damage by Leukocytes. Boca Raton, FL: CRC Press, 1992:225–239.
McCord JM, Russell WJ. Superoxide inactivates creatine phosphokinase during reperfusion of ischemic heart. In: Cerutti PA, Fridovich I, McCord JM, eds. Oxy-Radicals in Molecular Biology and Pathology. New York: Alan R. Liss, 1988:27–35.
McCord JM, Flores SC. The human immunodeficiency virus and oxidative balance. In: Paoletti R, ed. Oxidative Processes and Antioxidants. New York: Raven Press, 1994;13–23.
This dismutation or disproportionation reaction makes use of the fact that superoxide is both an oxidant and a reductant, eager to get rid of its extra electron or to take on another. The enzyme uses one superoxide radical to oxidize another. Catalases work in much the same way, because hydrogen peroxide can be a weak reductant as well as a fairly strong oxidant: 
In higher organisms, glutathione peroxidases appear to have largely supplanted the need for catalase. These enzymes use NADPH as the reducing species for hydrogen peroxide: 
They can reduce lipid peroxides as well as hydrogen peroxide and are very important enzymes in the prevention of lipid peroxidation to maintain the structure and function of biologic membranes.
00412-5&id=gr1.jpg){kind=link}