How Cells Age

Parallels between mice and yeast uncover a potentially universal aging mechanism

Elderly mice and aging yeast have more in common than scientists ever suspected. A new study by Harvard Medical School researchers reveals that the biochemical mechanism that makes yeast grow old has a surprising parallel in mice, suggesting it may be a universal cause of aging in all organisms.

"It was very exciting when we made the discovery, because it was so unexpected," says David Sinclair, a Harvard Medical School professor of pathology and senior author of the study, published today in Cell.

In yeast, aging--marked by an inability to continue replicating--is modulated by a protein called Sir2, which has counterparts, called sirtuins, in nearly every known organism. Normally, yeast Sir2 attaches to repeating DNA sequences to keep them stable. It also doubles as a DNA repairer, migrating to damaged spots on the genome and helping to patch them up. When a yeast cell is young, DNA damage is minimal, and Sir2 can keep up both these roles. But as the cell ages and accumulates more and more DNA damage, Sir2 becomes too busy with repairs to consistently stabilize those volatile repeating sequences. Left unsupervised, the repeats recombine into little extrachromosomal loops of DNA that build up and prevent the cell from reproducing.

This mechanism was discovered a decade ago in the MIT lab of Leonard Guarente, where Sinclair was then a postdoctoral researcher. For years, says Sinclair, few scientists suspected it had any relevance for understanding the process of aging in humans or other mammals. Although sirtuins have been linked to aging in a wide variety of organisms, their mechanism of action was understood only in yeast. But now it seems a remarkably similar process may underlie aging in mice as well.

One function of the mouse version of Sir2, called SIRT1, is to regulate how genes are expressed in various tissues. Patterns of expression differ among organs--many genes that need to be active in the liver, for instance, must remain silent in the brain. By binding to regulatory regions alongside certain genes, SIRT1 helps dictate those patterns. Because SIRT1 has also been shown to participate in DNA repair, Sinclair and his colleagues wondered whether increasing DNA damage would compromise the protein's normal regulatory role, as is the case with Sir2 in yeast.

Sure enough, when the researchers treated mouse embryonic stem cells with DNA-damaging hydrogen peroxide, SIRT1 migrated away from regulatory regions of the genome and toward the many areas where DNA strands had broken. As a result, genes that were normally shut off suddenly became active. Gene expression patterns, once exquisitely fine-tuned, went haywire.

"This is something that's eerily parallel to what we know in yeast," says Jan Vijg, chair of genetics at Albert Einstein College of Medicine, who was not involved in the study.

Yeast are the only organism in which the mechanism of aging is well understood, says Sinclair. "We only know for sure why yeast age," he says. "[With] all the other organisms, it's still a black box. But we're hoping that this is an explanation for all organisms."

Guarente agrees that the resemblance to yeast is surprising. "It was interesting to see this commonality," he says. "The degree to which it recapitulates yeast is pretty striking." But he is more skeptical that this particular mechanism will turn out to be universal, cautioning that the process of aging is so chaotic and haphazard that the notion of a universal may not be useful.

Sinclair says the finding also provides a plausible explanation for two well-known phenomena: that DNA damage accelerates aging, and that patterns of gene expression tend to go awry as an animal gets older.

The sirtuins have received considerable attention in recent years for their apparent role in aging. An overabundance of sirtuins extends the life spans of yeast, nematodes, and flies. In addition, molecules that seem to activate sirtuins--such as resveratrol, found in red wine--have a protective effect against some age-related diseases in mice. Sinclair cofounded Sirtris Pharmaceuticals in Cambridge, MA, to investigate the therapeutic possibilities of highly potent resveratrol-like molecules. The company is testing a series of products, including a treatment for treating type 2 diabetes.

The new study adds to this growing body of evidence for the many ways sirtuins contribute to aging and age-related disease. "SIRT1 is reported to do so many different things now; the challenge is going to be figuring out which of those it really does, and which of those are really important for diseases," says Brian Kennedy, another former member of Guarente's lab. Kennedy, now an associate professor of biochemistry at the University of Washington, was not involved in the study.

Guarente also emphasizes the broad importance of sirtuins, beyond the newly discovered SIRT1 mechanism. "The universal in aging we already know is sirtuins; they do so many things," he says. "The best way to approach this is to be able to trigger sirtuins so that you get all of the outputs and all of the benefits that they can bestow," he adds, noting that many of those outputs are unrelated to the new mechanism.

Sinclair and his colleagues also found evidence of a link between the SIRT1 mechanism and cancer, a disease strongly associated with old age. When dosed with resveratrol or beefed up with an extra copy of the SIRT1 gene, mice normally prone to cancer developed fewer tumors. Both of these interventions increased the available amount of SIRT1, likely enhancing the protein's ability to repair the DNA damage that leads to cancer without compromising its function as a gene regulator.

SIRT1 was already known to regulate a handful of mouse genes, but the new study revealed hundreds more. Many of these genes were found to be overexpressed in the brains of aging mice, underlining the potential importance of SIRT1-based gene deregulation in the aging process.

While the striking parallel between mice and yeast suggests that sirtuins' competing dual roles may be relevant in a wide variety of organisms, it remains to be seen just how that mechanism fits into the larger picture of mammalian aging, says Vijg. Nonetheless, Sinclair is confident that his group has uncovered a potentially universal mechanism. "Life, in general, has an Achilles heel," says Sinclair, "and this is it."