DNA methylation is an epigenetic alteration in which genes are decorated with methyl groups. It is one of a range of epigenetic processes that establish a feedback loop linking the pace at which specific proteins are built from genetic blueprints, the activities of those proteins once built, and environmental circumstances in tissues such as nutrient availability, temperature, damage, and disease. All of the switches and dials for molecular machinery inside cells are essentially built on top of the circulating levels of specific proteins, and these are altered via epigenetics: protein levels are in constant flux, as are countless epigenetic modifications to DNA.
In recent years researchers have demonstrated that specific patterns of DNA methylation within this broader tapestry correlate very well with age. Researchers can use these patterns in a tissue sample to identify an individual's age with an accuracy of five years or so. We all age due to the same underlying processes, some of us faster than others largely due to unfortunate lifestyle choices such as lack of exercise, excess weight, and smoking. Small differences in stochastic damage to cells and tissues snowball over the years into comparatively large differences in outcomes: the roots of variability in the mean time to failure in a very complex system. Given that the same forms of damage accumulate in all of us as a side effect of the same metabolic processes, it shouldn't be surprising to find that researchers can pull out patterns in the controlling mechanisms of metabolism - epigenetic alterations - that are tightly coupled to age. These are reactions to the environmental state of being damaged.
Studies that investigate DNA methylation from other perspectives should pick up the same signs of the same underlying processes, and same broad similarities between individuals. This is the case even when looking for signs of differences between old individuals, in search of a better explanation of the genetic contribution to extreme longevity in humans. So far genetic studies have turned up very few associations between genetic variants - meaning actual differences in the structure of specific genes - and longevity. Those that are found in one study rarely show up in others. This suggests that if variants are important in determining survival in extreme old age, then there must be a very large number of such variants with individually small effects, and the patterns of genetic differences must vary widely between regional populations. A very complex picture with little hope of complete understanding or any sort of resulting application in medicine in the near future, in other words. Is this in fact the case, however? These researchers suggest that epigenetic changes are instead where we should look, and that the picture isn't as complex as feared:
Human longevity is believed to be an integrating result of genetic and environmental factors. Although previous studies have shown that genetic variation may explain 20-30% contribution to human longevity, much remains to be known for its underlying genetic mechanism. In the past decade, a number of genes were discovered, in which some specifically genetic alterations may confer advantage in extending the organisms' lifespan, suggesting the existence of longevity genes. These findings however could not fully explain the significantly reduced incidence of age-related diseases in centenarians and their offspring, as it requires a broad effect of longevity genes, including conferring beneficial effects in extending life span as well as suppressing deleterious influence from the disease-associated genes. Alternatively, it is possible that the low prevalence of the age-related diseases in the long-lived people is attributed to a much lower frequency of risk alleles. Unfortunately, the latter notion fails to find support from a recent study in which the long-lived people were shown to carry similar frequencies of risk alleles as did in the young controls. This observation seems to echo with the suggestion that the longevity-related variants may compress the morbidity of long-lived people as these variants were significantly enriched in disease-related genes.
Hitherto, the obtained genetic evidence, based virtually on mutation screening, find no support for the hypothesis that lack of disease-related mutations contributes to healthy aging. However, taking into account the heterogeneity in longevity, in which multiple ways could be adopted to achieve longevity, and the crucial role of epigenetic modification in gene regulation, we hypothesize that suppressing the disease-related genes in the longevity individuals is likely achieved by epigenetic modification, e.g. DNA methylation. A reduction of genome-wide DNA methylation level and locus-specific hyper-methylation has been observed with aging, whereas changes in DNA methylation were reported to be associated with the occurrences of age-related diseases, such as cardiovascular disease, diabetes and cancer.
To test this hypothesis, we investigated the genome-wide methylation profile in 4 Chinese female centenarians and 4 middle-aged controls. 626 differentially methylated regions (DMRs) were observed between both groups. Interestingly, genes with these DMRs were enriched in age-related diseases, including type-2 diabetes, cardiovascular disease, stroke and Alzheimer's disease. This pattern remains rather stable after including methylomes of two white individuals. Further analyses suggest that the observed DMRs likely have functional roles in regulating disease-associated gene expressions. Therefore, our study suggests that suppressing the disease-related genes via epigenetic modification is an important contributor to human longevity.
I'd want to see a much larger study before taking this result at face value, but to find consistencies across populations in this sort of data shouldn't be too surprising given the points made above about the fact that we all age in the same way. Patterns of similarity should be there to be found in many different ways.