Prompted by attention given to a recent study claiming to cast doubt on the primary role of damaged mitochondria in aging, here is a lengthy and detailed article from the SENS Research Foundation on what is known of mitochondrial DNA damage and aging. It is worth bearing in mind when reading the scientific literature that any single study, especially if claiming to overthrow the consensus, should always be weighed against the rest of the recent literature in a given field:
The study was of fibroblasts, which are a kind of skin cell. It is interesting and contributes to a long-standing debate in this field about the frequency of specific mitochondrial DNA mutations with age and tissue type, and whether they contribute to specific diseases. It is clear at this point that mitochondrial dysfunction occurs with age and that damage in the form of mutations to mitochondria contributes to the diseases and disabilities of aging. We don't believe that this particular study is actually a challenge to scientists' existing understanding about how changes in mitochondria with age both drive and are driven by cellular and molecular damage, and the diseases and disabilities of aging.
What is actually known about the frequency and impact of specifically age-related mitochondrial mutations? First, in line with the ability of dividing cells to dilute out structural damage, multiple studies in aging rodents and humans report that the mutations in mitochondria that persist in cells and thus accumulate with age are confined almost entirely to cell types that don't divide during adulthood (e.g., brain neurons, heart muscle cells, and skeletal muscle). Second, those mutations are quite surprisingly rare: even in tissues that are actually affected by mitochondrial mutations with age, fewer than 1% - and perhaps as few as 0.1% - of cells are found to be affected.
Still, the evidence suggesting that this damage drives degenerative aging is powerful. The level of oxidative damage to mitochondrial DNA, the rate of accumulation of mitochondrial DNA mutations with age, and the structural vulnerability to such mutations are collectively robustly correlated with species maximum lifespan (the strongest integrative measure of the overall rate of aging in a species). Remarkably, this has recently been demonstrated even in rockfish, whose senescence is nearly negligible: lifespan in rockfish species was found to correlate negatively with the rate of mutation of their mitochondrial, but not nuclear, genomes - a relationship that the investigators' analysis suggested was not likely to be an artifact of tradeoffs with fecundity or the rate of germline DNA replication.
Calorie restriction (the most robust intervention that slows the rate of aging in mammals) lowers the rate of accumulation of mitochondrial deletion mutations with age. And when mice are given a transgene that directs a form of the antioxidant catalase directly to their mitochondria - an enzyme that complements the existing antioxidant machinery in the mitochondria in a way that reduces total mitochondrial DNA oxidative damage, including but not limited to deletion mutations - it extends their mean and maximal lifespan and ameliorates multiple pathologies of aging. Yet no such effects are observed when the same enzyme is directed to sites outside of the mitochondria, or when other antioxidant enzymes are expressed elsewhere in the cell, or even when non-complementary enzymes are sent to the mitochondria.
The apparent paradox in all of this is the strong link between mitochondrial DNA deletions and the rate of degenerative aging in the face of the rarity of such mutations. There are two broad kinds of resolution to this paradox. The first is the tissue-specific one. Although cells overtaken by mitochondria bearing DNA deletions are rare, they can have powerful effects on health in tissues where they are unusually enriched in critical cell types, particularly if relatively few of those cells exist in the first place. Such is the case for the key dopamine-producing neurons in an area of the brain known as the substantia nigra pars compacta (SNc). SNc dopaminergic neurons are much more vulnerable to being overtaken by mitochondria bearing large deletions in their DNA than are other cell types in the brain, and such mutations clearly drive dysfunction, including being tightly liked to Parkinson's disease. The same high regional vulnerability to mitochondrial DNA deletions occurs in people suffering with non-Parkinson movement disorders and even in "normal" aging brains, albeit at a lower rate and yet the finding has no parallel in the smaller and less harmful point mutations.
The other kind of tissue-specific effect relates more to the unique properties of the affected cell type itself, with the cardinal case in this category being skeletal muscle. Unlike most cell types, skeletal muscle "cells" are not isolated from all of their neighbors by a membrane. Instead, the long stretches of skeletal muscle fibers are comprised of multiple segments, each of which contains its own nucleus, which is in turn supported by a local population of mitochondria, with additional mitochondria in the membrane-bound space outside the fiber itself. Mitochondrial DNA deletions not only accumulate with age at a faster pace in skeletal muscle than in many other aging tissues, but because of that structure their effects are much more catastrophic. When a local nucleus' mitochondrial population is overtaken by deletion mutations, the segment first atrophies at that point, and then fails, leading the fiber to split or break locally and ultimately causing the loss of the entire fiber. These processes - loss of energy production and the splitting and loss of fibers - are a key driver of sarcopenia, the age-related loss of skeletal muscle mass and function that occurs even in lifelong master athletes.
Because deletion mutations in mitochondrial DNA are core molecular lesions driving these diseases, repair of these mutations will be central to their prevention, arrest, and reversal. But you can't tell that from a study of skin cells.