"The great secret that all old people share is that you really haven't changed," wrote novelist Doris Lessing. "Your body changes, but you don't change at all." From a genetic point of view, there is a lot of truth in that statement: As we age, the core of our biological being—the sequence of our DNA, which makes up our genes—remains the same. Yet recent research suggests that more subtle chemical changes to our DNA occur as we age. Now, a comparison of the DNA of a newborn baby with that of a centenarian shows that the scope of these changes can be dramatic, and they may help explain why our risk of cancer and other diseases increases as we get older.
DNA is made up of four basic building blocks—adenine, thymine, guanine, and cytosine—and the sequence of these nucleotides within a gene determines what protein it makes. Genes can be switched on and off as needed, and the regulation of genes often involves what are called epigenetic mechanisms in which chemical alterations are made to the DNA. One of the most common of these epigenetic changes involves a methyl group -- one carbon atom and three hydrogen atoms—binding to a nucleotide, usually cytosine. In general, this binding, called methylation, turns off the gene in question.
Recent research suggests that changes in DNA methylation patterns as a person gets older may contribute to human diseases for which risk increases with age, including cancer. To get a better idea of how methylation patterns change with age, a team led by Manel Esteller, an epigenetics researcher at the Bellvitge Biomedical Research Institute in Barcelona, Spain, looked at two extreme cases: A newborn male baby and a man aged 103 years.
The team extracted DNA from white blood cells taken from the blood of the elderly man and from the umbilical cord blood of the baby and determined its methylation pattern using a fairly new technique called whole-genome bisulfite sequencing (WGBS). With WGBS, DNA is exposed to the chemical sodium bisulfite, which has no effect on cytosines with methyl groups bound to them but turns nonmethylated cytosines into another nucleotide called uracil. The result is an epigenetic map that shows exactly which DNA sites are methylated and which are not.
As the team reports online today in the Proceedings of the National Academy of Sciences, it found a significantly higher amount of cytosine methylation in the newborn than in the centenarian: 80.5% of all cytosine nucleotides, compared with 73%. To look at an intermediate case, the team also performed WGBS on the DNA of a 26-year-old male subject; the methylation level was also intermediate, about 78%.
Esteller and his colleagues then took a closer look at the differences between the DNA of the newborn and of the centenarian, but restricted the comparison to regions of the genome where the DNA nucleotide sequences were identical so that only the epigenetic differences would stand out. The team identified nearly 18,000 so-called differentially methylated regions (DMRs) of the genome, covering many types of genes. More than a third of the DMRs occurred in genes that have already been linked with cancer risk. Moreover, in the centenarian, 87% of the DMRs involved the loss of the methyl group, while only 13% involved the gain of one.
Finally, to expand the study, the team looked at the methylation patterns of 19 newborns and 19 people aged between 89 and 100 years old. This analysis confirmed that older people have a lower amount of cytosine methylation than newborns.
The authors conclude that the degree of methylation decreases in a cumulative fashion over time. Moreover, Esteller says, in the centenarian the loss of methyl groups, which turns genes back on, often occurred in genes that increase the risk of infection and diabetes when they are turned on during adulthood. In contrast, the small number of genes in the centenarian that had greater methylation levels were often those that needed to be kept turned on to protect against cancer.
The new work is the first to compare the complete, genome-wide DNA methylation patterns of these two diverse age groups, says Martin Widschwendter, an oncologist at University College London in the United Kingdom who has studied the link between methylation and cancer. Widschwendter, who likens the DNA sequence to the genome's "hardware" and epigenetic changes to its "software," says that the Esteller team's study supports earlier research suggesting that "as a function of age and environmental exposure, this software accumulates defects" that can cause "age-related cancer and degenerative diseases."