I posted this in the Skeptics Club, but it's of general interest, so have a read: (From this science website)

Does Exercise Age you more Rapidly?

A great deal of literature suggests that there's an easy way to extend the lifespan of most animals: keep them from eating much. As long as essential nutrients are provided as part of the diet, a sharply limited input of calories can both extend the lifespan of many organisms and stave off some of the less pleasant symptoms of old age, including Alzheimer's-like symptoms. One of the outstanding questions regarding these findings has been one of mechanism: how does having as few as one-third the normal intake of calories feed into the processes that regulate an organism's life span?
One of the ideas that's been floated as an explanation is that there's a relationship between caloric intake and oxidative activities in the cell. As a byproduct, these activities will produce a certain level of reactive oxygen that can cause damage to a cell's components. Given time, the accumulation of damage leads to the aging of the cells, and the organism as a whole. Even as this was proposed, there were a few obvious problems with it. For one, antioxidants in diets should block aging, but don't appear to do so. In contrast, the metabolic activity caused by exercise should age people, but doesn't. A few presentations that are happening at the American Physiological Society's annual conference in Virginia Beach (scientists love those off-season rates) suggest that the proposal is in need of a thorough reworking, at least based on the press releases the APS is putting out.
The work was done in rodents, which have a pretty broad range of life spans. Standard lab mice live for about three years, whereas the naked mole rat is a longevity champion, lasting for nearly 30 years in addition to being eusocial and, well, naked. One study focused exclusively on two strains of mice, one active and the other sedentary. The mice were split into three groups of 100: one active group given a running wheel, one active group without it, and a sedentary group that was given a wheel. Regular dissections of members of each group revealed that energy use correlated with the presence of the running wheel, and none of the mice appeared to have unusual levels of antioxidants. The sedentary group outlived the active strain by about 10 percent of the typical life-span, but the two active groups lived for essentially identical time periods. So, it looks like energy use and longevity don't line up well, at least in these mice.
Meanwhile, others are asking what makes naked mole rats so long-lived? It's not because they don't burn through calories like mice; an examination of markers of oxygen-induced damage shows that the naked mole rats have higher levels of damaged lipids, proteins and DNA than mice. It's still possible that they repair this damage more efficiently than mice, but it's clear that a lower level of oxygen metabolism alone isn't enough to explain this finding. The picture was blurred even further by another study being presented at the meeting, which suggested that lifespan in rodents correlated with levels of a specific isoform of thyroid hormoneórodents with shorter lives had more of the hormone. Thyroid hormone plays a general role in revving up metabolism, which makes sense if lifespan is proportional to metabolic activity. But this finding makes no obvious sense in light of the results that suggest mole rats have a higher metabolism.
Having looked through all of this data, the only thing that's clear is that any relationship between metabolism and aging is going to be very complex.