In this issue, Kirkwood and Kowald develop a model of mitochondrial DNA damage via free radicals released from the electron transport chain 1. Essentially, the concentration of reactive oxygen species (ROS) in the direct vicinity of the mitochondrial DNA is very high in comparison with that in the mitochondrial matrix on average. The reason: mitochondrial DNA is physically very close to the proteins of the electron transport chain. The mitochondrial genome therefore takes the brunt of the damage meted out by ROS emanating from oxidative phosphorylation. Proteins, lipids and other non-DNA constituents of the cell are recycled, constantly taking oxidised forms out of service; by contrast, DNA damage accumulates.
Though I am simplifying somewhat, the concentration of ROS decays exponentially with distance from the source in the electron transport chain – partly as a result of diffusion, and partly as a result of neutralisation by antioxidants and enzymes that convert ROS – from here on simply called ‘antioxidants’. The authors' assertion is that additional external oxidative stress would have a very small proportional effect in the steep part of the curve near the mitochondrial DNA, and a much larger proportional effect in the shallow part of the curve closer to the centre of the mitochondrial matrix. Via a more complicated analysis involving consideration of the membrane-route of ROS, another consequence is inferred: modulating cellular antioxidant concentrations will have a negligible effect on reducing mitochondrial DNA damage.
Can we justify this model from another perspective, intuitively? I believe so: such antioxidants have, after all, to maintain a redox balance in the larger cell, with nothing more than diffusion to distribute them; they cannot be actively concentrated between the electron transport chain and the mitochondrial DNA just because that is where they would do ‘most good’. Precisely because their function is to maintain a balance in the cell as a whole, they also cannot exist at a monstrously high general concentration in the cell just in order to stave off damage to the mitochondrial DNA: a ‘completely reduced’ cell would not work any better than a ‘completely oxidised’ one.
But there is another likely reason for which it is not in an organism's interest to attempt to quench all ROS close to their mitochondrial source: they act as a stimulus to upregulate the expression of mitochondrially encoded components of the electron transport chain. Essentially, up to a certain threshold a cell interprets free radical leakage merely as suboptimal stoichiometry in the constituents of the electron transport chain, as discussed in 2. Such imbalance causes relatively sluggish electron transfer, and hence free radical leakage. The response is an increase in mitochondrial gene transcription. But mitochondrial gene transcription is linked to mitochondrial genome replication; it hence increases the number of mitochondrial genomes, and by extension, the number of mitochondria in a cell. The process is termed reactive biogenesis 2. As Lane notes, reactive biogenesis is not able to tell the difference between the two causes of free radical leakage, namely (i) stoichiometric imbalance in the protein complexes; and (ii) damaged complexes that no longer fit well, and therefore also have a slower-than-normal rate of electron transfer. Therefore, with increasing age of the organism, the number of mitochondria that leak simply because their electron transport chains are damaged will also be increased via this process. Leakage above a particular species-specific value will induce apoptosis. As Lane discusses, that value is low for long-lived species that produce few offspring (so-called K-strategists), particularly those with a high aerobic demand (e.g. many birds). During embryogenesis in such species, stringent selection for good fit between the proteins of the mitochondrial electron transport chain results in production of relatively few offspring, but with very efficient mitochondria: they leak minimal amounts of free radicals. Conversely, the apoptotic threshold is high for short-lived species that produce many offspring (so-called r-strategists) 2. Put crudely, a rat does not age slowly, rather it burns out, riddled with oxidative damage in all its cells; a pigeon, by contrast, lives much longer because of its much better-fitting electron transport chain. But that goes hand-in-hand with a lower leakage threshold for triggering of apoptosis. A pigeon therefore ages slowly, showing typical tissue atrophy via apoptosis 2: reactive biogenesis inevitably claws back its earlier benefits later in life.
Evidence is mounting that a major part, or even the major part, of age-related degeneration of an organism is, indeed, intimately, and unavoidably, linked to mitochondrial dysfunction; and that dysfunction is mainly caused by an increasingly inefficient electron transport chain that leaks increasing quantities of ROS. The proximal target of ROS is the mitochondrial DNA, hence exacerbating the problem. And if Kirkwood and Kowald are right, that particular process cannot significantly be influenced by the concentration of cellular antioxidants. Where does all this leave the naked mole rat – an unusually long-lived rodent (both in captivity and the wild) 3 that has, paradoxically, relatively high levels of oxidative damage and low antioxidant-based defence 4? Perhaps this creature ‘knows’ that it cannot do much about the damage caused to its mitochondrial DNA by ROS, and it ‘saves’ on the antioxidants. The cellular damage that it accumulates might be mitigated by the animal's doubtless very low metabolic rate: its body temperature differs little with its subterranean surroundings.
Damage to mitochondrial DNA might well be unavoidable, because significantly reducing the concentration of ROS near to the mitochondrial DNA is not possible without fundamentally changing the cell's redox balance (by inference from ref. 1), or sacrificing an important mechanism that optimises mitochondrial efficiency in early life (by inference from ref. 2). The oxidative damage theory of ageing does not appear to be dead; but if damage to mitochondrial DNA is the major factor in ageing, it is hard to see how lifespan can be significantly increased via antioxidant strategies. In this context, lifespan extension would arguably be more likely achievable via mechanisms that mitigate the effects of oxidative damage, rather than seek to reduce oxidative damage per se. But for already long-lived species one has to wonder about the spectre of reactive biogenesis – which tends to expand populations of slightly defective mitochondria. How might we mitigate the effects of such slow, apoptosis-related ageing? …