Glossarya
Antioxidant: a molecule that protects a biological target against oxidative damage.
Free radical:
The field of free radicals and antioxidants (for definitions of terms see Glossary) is often thought of in terms of the value (or lack of it) of dietary ‘antioxidant’ supplements in keeping us healthy. However, scientists now realize that the field is far more than that – the complex interplay between free radicals, other ‘reactive species’ (RS) (such as H2O2 and peroxynitrite), and antioxidants has been (and still is) a major driver in the evolution and survival of humans, in the way cells communicate and respond to danger, and in age-related diseases of humans and other animals 1, 2, 3, 4.
The purpose of this article, based on a lecture given at the sixteenth World Congress of Basic and Clinical Pharmacology held in Copenhagen from 17–23 July 2010, is to step back a little from the multitudes of recent publications in the field and take stock of fundamental principles. What follows is largely a personal view of where the field of RS/antioxidants or ‘redox biology’ is now, and where it might be heading. I begin by articulating certain key principles.
RS are formed in aerobes both by accidents of chemistry (e.g. the autoxidation of unstable biomolecules such as dopamine) and deliberately, for example by activated neutrophils or nitric oxide synthases (nitric oxide, NO, is a free radical). Mitochondria produce superoxide (O2−) radicals; these are known to be potentially damaging in view of the striking deleterious phenotype of knockout animals lacking its scavenger, the mitochondrial manganese-containing superoxide dismutase [5]). Is mitochondrial O2− production due to simple accidental leakage of electrons to O2, or is it part of an intracellular redox-signalling mechanism, as some have argued 6, 7? I tend towards the former view: there is always an intrinsic potential for electron leakage when electron transport chain components with sufficiently-negative redox potentials to reduce O2 directly to O2− are functioning in the presence of O2 [1]. Strategies to minimize mitochondrial O2− production include the use of low intra-mitochondrial pO2 levels, the arrangement of electron carriers into complexes to facilitate electrons following the ‘correct’ path towards cytochrome oxidase instead of prematurely escaping to O2, and uncoupling proteins 1, 7.
In terms of human survival, probably the most important source of RS is the immune system 1, 4. When humans gathered together, initially in nomadic groups and later congregated into growing populations within (usually dirty) cities, infectious disease was a major threat. It has been stated that the ‘black death’ (bubonic plague) killed one third of the population of Europe – but why not the other two thirds? Some avoided exposure, but in others their immune system dealt with it – they recovered or never became sick, a powerful selection for survival of people with robust and well-coordinated immune systems. Perhaps then the survivors passed on some of the relevant genes, breeding children with similarly powerful immune systems.
How is this relevant to redox biology? RS play an essential role in the immune system. First, they help to kill some infecting organisms. It is essential to mount vigorous RS production and ‘hit microorganisms hard and fast’, because many such pathogens respond to low levels of RS with rapid increases in antioxidant defence systems that render them resistant to much higher RS levels 1, 8. Second, RS help to restrain the immune system (especially T-lymphocytes) from over-activation and prolonging inflammation once the threat has been eliminated 9, 10, 11. Persistent or chronic inflammation is a major determinant of disease later in the human lifespan, and RS play key roles in the origin and pathology of several age-related diseases, particularly cancer and neurodegenerative disorders 1, 12.
There is, I feel, less evidence that RS (apart of course from the well-established case of nitric oxide) play a major role in signalling in the healthy human body; many claims for ‘redox signalling’ (e.g. by H2O2) are based on experiments with cell cultures and tissue explants (about which more later). Nevertheless, evidence that RS play at least some role is steadily accumulating, and the mechanisms that might allow them to be generated at selected sites, persist for a short time, and then be removed, are slowly being elucidated [13]. Evidence that RS are truly important physiological (as opposed to pathological) signalling molecules is presently stronger in plants than in animals (reviewed in [2]). However, RS do seem to be involved in adaptation to ischaemia and to exercise, for example 14, 15, 16, 17. Therefore I list the question of the physiological importance of redox signalling as one of the challenges facing the RS/antioxidant field in the coming years (Table 1).
The human antioxidant defence network is complex and interlocking; it functions to minimize levels of RS while allowing useful roles to continue (Figure 1). Two recent examples: some O2− in muscle appears to be useful, but excess causes harm 14, 17, 19; stem cells need some RS to function properly, but too many RS can impair function [20]. The properties of the peroxiredoxin proteins, which scavenge H2O2 but can be inactivated by it, could be especially important in allowing localized and transient H2O2-dependent signal transduction 13, 21. In humans, the great majority of ‘total antioxidant capacity’ of cells and tissues is contributed by endogenously-synthesized antioxidants such as reduced glutathione (GSH), peroxiredoxins and superoxide dismutase; whereas diet-derived antioxidants are far less important, a point upon which I elaborate below. It follows that, instead of using dietary ‘antioxidants’, a more effective way of strengthening overall antioxidant defence in humans might be mild pro-oxidants, increasing the generation of RS and electrophiles that then trigger increases in the levels of our own antioxidants, such as GSH 22, 23.
Because RS are not completely removed in vivo (Figure 1), it follows that the ability to repair (e.g. DNA repair) or replace (e.g. lipid turnover, degradation of abnormal proteins by the proteasome and by lysosomes) oxidatively-damaged molecules is essential. Failures in some or all of these repair systems could contribute more to ageing and age-related disease than changes in antioxidants or RS production, although the magnitude of the contribution remains unclear 24, 25, 26.
In recent years, better (although still not perfect) biomarkers of oxidative damage in human tissues and body fluids have been developed (Table 2). Note that most are measured in blood or urine, and they could easily fail to detect changes in oxidative damage in small groups of cells or individual organs. For example, rises in oxidative damage levels in the brain are probably not usually reflected in the blood (discussed in [27]). Nevertheless, studies using biomarkers reveal that giving antioxidant supplements to healthy or diseased humans rarely causes much change in systemic levels of oxidative damage 28, 29, 30, 31, 32, 33, 34. The failure of most human intervention trials with antioxidants to modify disease outcome 1, 35, 36 is therefore not surprising. Even if oxidative damage does play a role in the disease, ‘antioxidants’ will be ineffective if they fail to alter the levels of oxidative damage. If RS are involved in the adaptation to tissue injury (as discussed earlier and summarized in Figure 2), then high doses of antioxidants could sometimes be deleterious (legend to Table 1 and 35, 36). We are perhaps fortunate that diet-derived ‘antioxidants’ do not markedly decrease oxidative damage in humans – because otherwise they might sometimes have caused harm rather than good.
If antioxidants generally fail to modulate oxidative damage in humans, what does? Table 3 attempts to summarize our limited knowledge to date. One important point is that laboratory rats and mice appear to be more sensitive to dietary antioxidant levels than are humans, and dietary antioxidant supplementation is therefore more likely to reduce oxidative damage in rodents than in humans 1, 39, 40. Indeed, the antioxidant content of animal feed could account for several discrepancies between results reported from different laboratories [39]. It follows that rodent models of human diseases in which oxidative damage appears to be important in the pathology (such as stroke, atherosclerosis, amyotrophic lateral sclerosis and dementia) are more likely to be responsive to administered antioxidants than are humans. This should be borne in mind when evaluating the potential uses of ‘therapeutic antioxidants’ in humans; such trials could well fail [41].
RS have been suggested to cause everything from baldness to failed pregnancy (perhaps the opposite is true for the latter [36]). Their role in many diseases is likely to be peripheral, but in some it is fundamental (Figure 2). Cancer is probably an example of the latter: RS can cause cancer, contribute to its progression, and sometimes even cure it [42]. Neurodegenerative disease is another example: marked oxidative damage is present in affected brain regions in both Parkinson and Alzheimer diseases; there is some evidence that this damage precedes neurodegeneration and studies on animal models suggest that decreasing oxidative damage can preserve neuronal function 1, 27, 43.
Developing better antioxidants is a major challenge for the redox biology field (Table 1). Many antioxidants are available (Table 4) but most have failed dismally in clinical studies even when there is strong evidence that oxidative damage plays a role in the disease pathology (and the evidence is often not robust). Probable reasons for this failure have been summarized 1, 41, 44. Promising antioxidants under evaluation include low molecular mass catalytic scavengers of RS 45, 46 and targeted
Cell culture studies are used extensively in research and drug development, but a substantial proportion of cell lines are not the cells they are supposed to be [52]. To quote the authors of [52] ‘thousands of misleading studies and potentially erroneous papers have been published’. The scale of the problem is under-appreciated by most scientists [52]. Equally under-appreciated is the fact that the effects of “antioxidants” such as ascorbate, lycopene, epigallocatechin gallate and several other
RS are intimately involved in ageing and age-related disease 54, 55. RS probably do not cause ageing (they could perhaps account for some features of ageing, such as skin wrinkling), nor will dietary antioxidants stop ageing 24, 25, 26, 56, 57. Perhaps mild pro-oxidants might help, by raising endogenous antioxidant defences 57, 58, 59. If ROS really are key components of physiological signalling, perhaps the general signalling failure with ageing [60] could even be due to insufficient RS (or
The biology of ROS and antioxidants is not an esoteric field of study: these species are involved in all aspects of aerobic life. One cannot live without them, nor would one wish to, but ultimately they no doubt contribute to individual mortality. Learning how to stop the latter while preserving the useful functions of RS should be a major research priority. Glossarya Antioxidant: a molecule that protects a biological target against oxidative damage. Free radical: