It wouldn't be too far wrong to regard ourselves as ambulatory chemical processing plants: biology is very complicated and highly organized chemistry, a collection of reactions and products. The operation of our metabolism is as much based on processing oxygen as it is on processing food. Thus you don't don't have to wander all that far into the scientific literature on the biology of aging to find mention of reactive oxygen species (ROS), the mechanisms of oxidative stress, and the various oxidative theories of aging, such as the mitochondrial free radical theory of aging. All sorts of reactive molecules containing oxygen can be found in our biology at any given time, an inevitable byproduct of being an oxygen-processing species.
Cells and their components are intricate assemblies of protein machinery, but all it takes to disrupt a component is for it to react with a passing ROS molecule. It'll quickly be replaced by a cell's repair mechanisms, but in the meanwhile it is broken. Oxidative stress refers to the level of ongoing damage caused by ROS; ambient levels of ROS can rise due to environmental circumstances such as heat or radiation exposure, but we're more interested in what happens during aging. Older theories of aging based on oxidative damage suggest that aging is caused by an accumulation of this damage, and indeed levels of oxidative stress rise with aging, but the relationship isn't that simple.
Evolutionary selection is very ready to use any tool to hand. An individual's biology is a set of interconnected systems that share component molecules, and which are tied together into feedback loops and signal exchanges. Just like every other molecule in our biology reactive oxygen species were long ago co-opted into all sorts of vital mechanisms. This means that it is far from straightforward to talk about ROS and aging, as there are many different roles in metabolism for what at first sight seems to be nothing but a damaging, toxic class of molecule, and these roles are affected by rising levels of ROS in different ways. The specific location in cells and the body and the present circumstances all matter when it comes to what happens when ROS levels increase.
For example, it has been shown that some of the benefits of exercise are based on slightly increased levels of ROS as a signal. Increased use of muscle results in a higher output of ROS from the mitochondrial power plants working away in muscle cells, and cells react to this change with greater housekeeping efforts - an outcome known as hormesis. If tissues and bloodstream are bathed in antioxidants that soak up those ROS, then these benefits of exercise can be blocked. Thus general use of antioxidants in a normal metabolism may potentially do more harm than good.
Similarly, it seems fairly clear at this point, based on work in mice, that targeting antioxidants specifically to mitochondria is a beneficial strategy, and this presumably works by soaking up ROS at source before they can cause harm. How does this impact exercise and its effects on health? As yet unknown. Further, a range of life-extending genetic alterations in nematode worms work by globally increasing or globally reducing ROS output from mitochondria, with either outcome resulting in longer-lived worms. Increased ROS works through hormesis, by increasing repair activities, while reduced ROS output directly reduces damage, or at least that is the present consensus.
Mitochondria are important in aging - that much is worth taking away as a lesson here. I view much of the research into ROS and mitochondria as little more than a confirmation that it is vital to develop the range of envisaged biotechnologies that enable mitochondrial repair and replacement. The mitochondrial free radical theory of aging suggests that aging is in large part caused by the way in which mitochondria damage themselves with their own ROS output. It is that damage that is the important thing, not the ROS, but mitochondrial damage has such a great impact on aging and longevity that even modest changes in either (a) the pace at which they damage themselves or (b) the likelihood of damaged mitochondria being replaced by cellular maintenance mechanisms, both of which are influenced by rates of ROS production, have a measurable effect on longevity in shorter-lived species.
But back to complexity resulting from the uses that ROS are put to in our biology. Here are a couple of papers that illustrate a few more of the ways in which nothing is simple:
The dysfunctional changes of aging are generally believed to be irreversible due to the accumulation of molecular and cellular damage within an organism's somatic cells and tissues. However, the importance of potentially reversible cell signaling and epigenetic changes in causing dysfunction has not been thoroughly investigated. Striking evidence that increased oxidative stress associated with hematopoietic stem cells (HSCs) from aging mice causes dysfunction has been reported. Forced expression of SIRT3, which activates the reactive oxygen species (ROS) scavenger superoxide dismutase 2 (SOD2) [to] reduce oxidative stress, functionally rejuvenates mouse HSCs.
These data, combined with numerous other reports, suggest that ROS act as a signal transducer to play a critical regulatory role in HSCs and at least in some other stem cells. It is likely that ectopic expression of SIRT3 restores homeostasis in gene expression networks sensitive to oxidative stress. This result was surprising because age-associated damage from impaired DNA repair had been thought to be irreversible in old HSCs. These data are consistent with a hypothesis that potentially reversible processes, such as aberrant signaling and epigenetic drift, are relevant to cellular aging. If true, rejuvenation of at least some aged cells may be simpler than generally appreciated.
[The endothelium] cooperates with leukocytes to create openings to provide the infection-fighting cells ready access to their targets. By and large, these ensuing "micro-wounds" are short-lived; as soon as the cells have crossed the endothelium, these pores and gaps quickly heal, restoring the system's efficient barrier function. In cases when these gaps fail to close - and leakage occurs - the results can be devastating, leading to dramatic pathologies including sepsis and acute lung injury.
[Researchers] set up experimental models that mimicked acute, intense inflammation. Using dynamic time-lapse and high-resolution confocal microscopy, the investigators could see the process by which leukocytes were breaching the endothelial cell. In the course of a 10-minute span, they observed that a single endothelial cell tolerated the passage of at least seven leukocytes directly through its body, and that within this brief period, the gaps closed, leaving no sign of the pores.
This response [is] fundamentally dependent on proteins (i.e. NADPH oxidases) that can generate reactive oxygen species (ROS), specifically hydrogen peroxide. ROS are widely implicated in causing cellular, tissue and organ damage when present at excessive levels in the body. But, these findings show that low levels of these molecules - when produced in discrete locations within the cell - are highly protective. "It's tempting to speculate that excess ROS causes vascular breakdown by short-circuiting the recuperative response process and creating 'white noise' that dis-coordinates and disrupts micro-wound healing. It appears that we've got an essential homeostatic self-repair mechanism that is completely dependent on the generation of intracellular ROS, which is opposite to our typical thinking about ROS in cardiovascular health and disease."