Critically discuss the research evidence that oxygen and oxidative stress are responsible for ageing
For decades there have been hundreds of hypotheses regarding theories of ageing (finch). Most of these ideas have been disregarded and several theories now captivate much research on ageing. Random damage secondary to free radicals (FR), notably, reactive oxygen species (ROS), have been implicated in the process of ageing. The damage caused by ROS is derived from oxidative stress theory (Merry). Ageing (or senescence) is defined in this context as the molecular and cellular changes witnessed over a life time that accumulate to cause eventual death (campisi). Research suggests that free radical damage within cells leads to physiological changes indicative of ageing (ashok).
Origins of the oxidative stress theory were first conceptualised by Pearl who observed differences in lifespan between warm and cold blooded animals when hibernating or living in cold environments respectively. He believed that maximum lifespan was inversely proportional to basal metabolic rate (pearl). Decades later Harman proposed a theory of FR damage which proved a suitable mechanism for Pearl’s hypothesis (Harman). ROS such as oxygen ions and peroxides are molecules containing oxygen that are readily reactive. ROS includes FRs, which host a single unpaired electron. This unpaired electron makes the molecule or atom highly unstable until it is again paired, which requires a reaction known as oxidation (Cheeseman). ROS are attained both environmentally and intrinsically. Environmental sources include tobacco smoke and pollution. Intrinsic sources are less avoidable, and are the product of the process of living – ROS are formed by white blood cells to fight infection for example. The most important source intrinsically however is from the mitochondria (lobo et al).
Mitochondria are known as “cellular power houses” given their primary function of ATP production, the universal energy source produced from metabolism. The electron transport chain (ETC) is located within the inner membrane of mitochondria, and is the site of oxidative phosphorylation of ADP. A series of proteins transfer electrons through a series of redox reactions creating a proton gradient which drives ATP synthesis. However, a small proportion of electrons leak from this chain, which can lead to the formation of superoxide. These FRs go on to further damage the ETC, making it leakier, producing further FRs and reducing efficiency.
FRs and ROS cause damage through several mechanisms; firstly proteins containing certain amino acids are readily oxidised, leaving them vulnerable to proteolysis, causing detriment to cellular function. Lipids are also subjected to FR and ROS damage, leading to a number of pathophysiological processes underlying many age-related diseases such as carcinogenesis and atherosclerosis. It has also been reported that DNA, RNA, and in particular, mitochondrial DNA are subject to FR and ROS damage.
Oxidative stress refers to an excess of ROS compared to the physiological antioxidant capacity. Antioxidants (such as superoxide dismutases) are made in the body to counter FR damage, however an imbalance in their concentrations can lead to an individual with either more or less FR damage, which has a direct impact on maximum lifespan. If Oxidative stress theory is true, there are three predictions which will be true that researchers can focus their efforts on trying to prove: Accumulation of oxidised macromolecules with age, manipulation of diet/pharmacological/genetic factors will reduce oxidative stress, and decreasing levels of oxidative stress will increase lifespan.
In recent years there have been studies which would suggest that oxidative stress theory may play a less important role in the process of ageing. The large body of literature correlating macro-molecular free radical damage and ageing are proving less feasible with the emergence of some studies (speakman), for example, those using genetic manipulation to reduce antioxidant capacity which show no difference in maximum life-span (perez). Also, recent studies have suggested that ROS may play an important role in stress response induction, immunity and systemic signalling (poljsak).
So is oxidative stress theory the major hallmark of ageing? We will begin this discussion by exploring the evidence surrounding the aforementioned three predictions that oxidative stress theory would imply. A number of studies have arisen in the past decades that would suggest that age-related increases in oxidative damage happen to a variety of macromolecules (as aforementioned) in a range of organisms. Lipid peroxidation has been subject to much research, given the role of the cell membrane phospholipid bilayer is sensitive to FR damage. In the past, non-specific observations were recorded, however, a new alternative method of using MDA and 4-HNE levels have proven more sensitive (Rikans and Hornbrook). This method shows the non-enzymatic production of compounds that arise from cell membrane attack. It has been shown that age related increases in F2-isoprostane levels occur in male F344 rats (bokov) and Sprague-Dawley rate (Roberts and Reckelhoff, 2001). This method is particularly accurate given there is no diurnal variation in blood levels, it is unaffected by lipid levels in diet, it is detectable in all bodily tissues and it can be measured to a picomolar level.
Proteins are very important with regards to theories of ageing, given their ultimate importance in cellular biology. Age-related changes in oxidative damage are essential to understanding the ageing process. ROS can readily react with almost all amino acids, therefore altering the structure and function of affected proteins. Once altered, proteins must be degraded and reformed to re-establish functioning. Only a few oxidative reactions are reversible. The importance of protein oxidation (carbonyl formation) was realised some decades ago by Levine (Levine), after which, Stadtman discovered that oxidised proteins increase with age in rat liver, human skin fibroblasts and the human lens (stadtman). It has also been reported that global protein carbonyl levels in human tissues increase as a function of age (moskovitz). Measuring carbonyl levels is based on their reactivity with 2,4-dinitrophenylhydrazine (DNPH). Carbonyl levels are measured by the absorbance of the 2,4DNPH at 370nm (stadtmaan and oliver). This method has been less valid in the past due to inter4fernce of free DNPH and the reaction of DNPH with other non-protein molecules. Recently, antibodies have been used to detect specific oxidised residues in proteins (cao and cutler). Using western blots, researchers have found age-related increases in oxidative damage to specific proteins (Agarwal and Sohal, 1993) (Yan and Sohal, 1998).
DNA oxidation has been studied for six decades, with the effects of oxidation on nuclear (nDNA) and mitochondrial DNA (mDNA) being explored. In 1981 Frankel et al discovered thymidine glycol in mouse liver DNA using a thin layer of chromatography. Richter et all detected levels of oxidative damage in rat liver mDNA, some 16 fold times greater than that of nDNA. It was first reported in the 90s the effect of aging on DNA oxidation. It was observed that between 2 and 24 months of age; levels of oxo8dG in male rat’s liver, kidneys and intestines had doubled. This product is commonly used in studies of oxidative damage to DNA as it is commonly a major oxidative lesion, it is easily detectable by HPLC-electrochemical detector system and it is a mutagenic in mammalian and bacterial cells. However, as of recently, there was been much doubt about the validity of these findings. Many research groups have been unable to find changes in oxidative damage with increasing age. It is thought that the varying isolation and analysis of DNA sampling methods account for this difference – Claycamp reported in 1992 that phenol and other organic solvents, the commonly used solvent, can cause oxidative damage itself. Hamilton et al used the NaI method, which dramatically reduced the oxo8dG artefacts during isolation producing a mere tenth to a hundredth of the level obtained versus the classical phenol method. With this method, it was still reported that levels of oxo8dG in nDNA in all tissues of male F344 rats had significant increases with age.
Having explored the effects oxidative damage to macromolecules, let us focus on manipulations to increase life span, and whether or not these have an effect on oxidative damage to macromolecules. McCay demonstrated in 1935 an increase in maximum life-span in rats that were on a calorie restricted (CR) diet (mccay). It was demonstrated some decades later that Caenorhabditis elegans (c.elegans) and Drosophila Melanogaster could be mutated to better maximum life-span (Guarente and Kenyon, 2000).
The effects of CR have been the subject of much research in past decades, given evidence it has been shown to delay most age-related diseases and physiological changes associated with ageing, inferring that the ageing process is retarded (masoro).
It is currently believed that CR alters oxidative damage, which is consistent with the Oxidative stress theory. Deacdes of studies show that CR reduced oxidative damage in rodents, for example, decreases in lipid peroxidation (Davis et al., 1993), protein oxidation (lass et al) and decreases in lipofuscin (a protein and lipid oxidation product) (De et al., 1983). Bokov et al have relatively recently found that CR reduced the aforementioned age-related increases in F2-isoprostane levels in Fischer 344 rats. Measuring levels of oxo8dG shows decreased DNA oxidative damage in mice and rats, using the NaI method (Hamilton et al 2001b). Why does CR reduce oxidative damage? This question has been subject to decades of research, however results seem inconsistent and contradictory. Much research has focused on antioxidant defences, the expression and activity of such, and it has been reported that certain antioxidants are increased in concentration with CR (Rao et al 1990b). However, in some studies, no overall trend was noted (sohal et al). Since these contradictions, investigations are now underway, exploring the association between CR and ROS. Several studies have shown age-related increases in ROS within mitochondria, and that CR reduces this association within rat heart and liver (gredilla et al 2001a+b). Some studies suggest that CR may reduce oxidative damage through the enhancement of repair and production of macromolecules – CR was found to increase the turnover of macromolecules throughout rodent lifespans (Van Remmen et al) and DNA repair is bettered by CR (Lipman et al). To summarise the effects of CR, in rodents, it can reduce damage to macromolecules and reduce the production of ROS. This is consistent with the oxidative stress theory; however, it is not evidential that the anti-ageing effects are secondary to reductions in oxidative damage, given that CR affects several physiological pathways.
Following on from research surrounding life-span manipulation, genetic manipulations have proven promising. In 1987, induction of a mutation in the age-1 gene of C.elegans showed increases in life-span (Johnson). The daf genes in c.elegans have been implicated in data on oxidative stress and life span. Early studies showed; age-1 mutants were resistant to oxidation and expressed higher activities of some antioxidants (copper and zinc superoxide dismutase) (Larsen 93), age-1 mutants were resistant to heat stress and age-1 and daf-2 mutatnts were resistant to UC irradiation (Lithgow 95, murakami and Johnson). It has become generally accepted that increases in c.elegans lifespan were secondary to increased stress and oxidative stress resistance. This would be consistent with oxidative stress theory; however the physiological mechanisms underlying this are unknown. Drosophila, a fruit fly, has also been implicated similar findings, however, like c.elegans, findings suffers the same inconsistencies in the understanding of the underlying mechanisms.
The third prediction that would be observable if oxidative stress theory were true is that of increases in life span with manipulations in ROS production and scavenging. The previously discussed two predictions are implicated with many fruitful studies – however this is not true of ROS manipulation. Generally speaking, it has been found that ingestion of antioxidants or genetic manipulation to increase antioxidant capacity has in most cases shown inconsistent results with regards to levels of oxidatively damaged macromolecules.
The most obvious way to increase antioxidant capacity would be to ingest them and observe differences in life span. This has been done with several antioxidants on a variety of organisms including mice. Some examples include vitamin E, ethoxyquin and thiazolidine carboxylic acid (bokov). Differences in life span were observed however none of the ompunds increased maxmimum life span. Only thiazolidine carboxylic acid was found to do so in fruit flies (bokov, miquel et al). As the effect of antioxidant admission wasn’t measured against oxidative damage, it is concluded generally that concentrations of ingested antioxidants were insufficient to prevent oxidation in cells (bokov).
More recently, there has been interest in the research of longevity potential of superoxide dismutase and catalase mimetics (EUKs). Melov et al 2000 reported that c.elegans inclubated in EUKs had significantsly extended life-spans (medial and maximal) without physical changes. Although this study indicated increases in life span with increased antioxidant defences, later studies have not supported this finding. Bayne and Sohal 2002 reported no effect of EUKs on life span of house flies.
Drosophilia has been subject to P-element mediated transformations to allow mutations of certain genes to overexpress antioxidants. Orr and Sohal 94 found overexpression of copper or zinc superoxide dismutase and catalase increased maximal lifespan, and reduced protein and DNA oxidation. This would suggest that reduced oxidative damage increased survivorship – however, these reults are complicated by the p-element method given it can alter life span alone (Kaiser et al 96). Orr et al 2003 overcame this by using over 90,000 flies – it was found that there was actually a slightly decreased survivorship.
Following this, researchers now use yeast FLP/RT inducible systems to overexpress antioxidant genes to avoid the P-element problem. It was found that overexpression of copper and zinc superoxide dismutase alone increased lifespan 48% (sun and tower 99). Overexpression of MsrA has also been found to increase lifespan (Ruan et al 2002). These findings would suggest that increased life span is correlated with increased resistance to oxidative stress. Whilst these studies are consistent with oxidative stress theory, they do not show that increased life span is correlated to reduced oxidative damage to macromolecules. Since Harman’s work in 1956, much research to date has surrounded the oxidative stress theory. Although most of the data supports the first prediction described for the oxidative stress theory, none of them can prove the theory, as they are at most merely consistent. The second prediction has acquired much research in the form of CR. This is because of the limited models of increased life-span in the past. Most of CR data supports this prediction; however, again it is only consistent with the oxidative theory. Promising developments have been made with the growing number of animal models of increased life-span. Consistently, it has been observed that mutant invertebrates are resistant to stress and oxidative stress. This too is consistent with the oxidative stress theory. However, there is little research on age-related disease in these long lived mutants.
The data given for the third prediction is less clear cut. Although data fails to back up the oxidative stress theory, it does not disprove it either. Only recently has technology given opportunity for testing of oxidative stress. Studies on Drosophila support the third prediction of the oxidative stress theory, however, other studies do not support the theory. Again, a major limitation is that the researchers did not determine if the genetic changes altered accumulation of oxidative damage. Only when proving genetic manipulation alters oxidative damage can we consider evidence for the prediction.
Finch CE. Longevity, senescence and the genome. Chicago IL: University of Chicago Press; 1990.
Pearl, Raymond (1928). The Rate of Living, Being an Account of Some Experimental Studies on the Biology of Life Duration. New York: Alfred A. Knopf
Lobo et al //www.ncbi.nlm.nih.gov/pmc/articles/PMC3249911/
Cheeseman KH, Slater TF. An introduction to free radicals chemistry. Br Med Bull. 1993;49:481–93. [PubMed]
Ashok BT, Ali R. The aging paradox: Free radical theory of aging. Exp Gerontol. 1999;34:293–303.
Perez VI, Bokov A, Van Remmen H, Mele J, et al. 2009. Is the oxidative stress theory of aging dead? Biochim Biophys Acta 1790: 1005–14.
Rikans and Hornbrook (1997), Roberts and Reckelhoff, 2001 //www.sciencedirect.com/science/article/pii/S004763740400171X
Levineï‚· Oxidative modification of glutamine synthetase. I. Inactivation is due to loss of one histidine residue
Stadtmanï‚· Protein oxidation and aging
J. Moskovitz, M.B. Yim, P.B. Chock ï‚· Free radicals and disease
E.R. Stadtman, C.N. Oliverï‚· Metal-catalyzed oxidation of proteinsï‚· J. Biol. Chem., 266 (1991), pp. 2005–2008
H.G. Claycamp ï‚· Phenol sensitization of DNA to subsequent oxidative damage in 8-hydroxyguanine assays
L. Guarente, C. Kenyonï‚· Genetic pathways that regulate ageing in model organisms
E.J. Masoroï‚· Food restriction in rodents: an evaluation of its role in the study of aging
T.E. Johnson ï‚· Aging can be genetically dissected into component processes using long-lived lines of Caenorhabditis elegans