December 21, 2006
Evidence That DNA Damage Major Cause Of Aging

Study of the cause of a rare human genetic disease has led to strong evidence that DNA damage accumulation is a major cause of aging.

PITTSBURGH, Dec. 20 – The accumulation of genetic damage in our cells is a major contributor to how we age, according to a study being published today in the journal Nature by an international group of researchers. The study found that mice completely lacking a critical gene for repairing damaged DNA grow old rapidly and have physical, genetic and hormonal profiles very similar to mice that grow old naturally. Furthermore, the premature aging symptoms of the mice led to the discovery of a new type of human progeria, a rare inherited disease in which affected individuals age rapidly and die prematurely.

"These progeroid mice, even though they do not live very long, have remarkably similar characteristics to normal old mice, from their physical symptoms, to their metabolic and hormonal changes and pathology, right down to the level of similar changes in gene expression," said corresponding author Jan Hoeijmakers, Ph.D., head of the department of genetics at the Erasmus Medical Center in Rotterdam, Netherlands. "This provides strong evidence that failure to repair DNA damage promotes aging— a finding that was not entirely unexpected since DNA damage was already known to cause cancer. However, it shows how important it is to repair damage that is constantly inflicted upon our genes, even through the simple act of breathing."

The study found that a key similarity between the progeria-like, or progeroid, mice and naturally old mice is the suppression of genes that control metabolic pathways promoting growth, including those controlled by growth hormone. How growth hormone pathways are suppressed is not known, but this response appears to have evolved to protect against stress caused by DNA damage or the wear-and-tear of normal living. The authors speculate that this stress response allows each of us to live as long and as healthy a life as possible despite the accumulation of genetic damage as we age.

Can we design humans to live longer? Or will we have to constantly repair accumulating damage? How to make the DNA in our cells less prone to accumulation of damage? Will the development of massive computer simulations for computer aided biological engineering allow us to find much better designs for enzymes that protect and repair DNA?

The scientists were set off on the road to make this discovery by their investigations into the causes of a boy's genetic disease.

A German physician had contacted the center about a 15-year old Afghan boy who was highly sensitive to the sun and had other debilitating symptoms including weight loss, muscle wasting, hearing loss, visual impairment, anemia, hypertension and kidney failure.

The boy turned out to have a defect in the DNA repair mechanism called nucleotide excision repair (NER). The scientists were able to trace down the mutation to a particular gene:

When the investigators obtained cells from the boy and tested them for NER activity, they found almost none. Further analysis of the boy’s DNA revealed a mutation in a gene known as XPF, which codes for part of a key enzyme required for the removal of DNA damage. The XPF portion of the enzyme harbors the DNA-cutting activity; whereas a second portion, known as ERCC1, is essential for the enzyme to bind to the damaged DNA. Mutations in either XPF or ERCC1 lead to reduced activity of this key DNA repair enzyme.

"We were completely surprised by the finding that the patient had a mutation in XPF, because mutations in this gene typically cause xeroderma pigmentosum, which is a disease characterized primarily by skin and other cancers rather than accelerated aging," said Dr. Hoeijmakers. "This patient, therefore, has a unique disease, which we named XPF-ERCC1, or XFE-progeroid syndrome."

DNA manipulation technologies have become powerful enough to allow creation of lab mice which have any mutation of interest. So these scientists do what many scientists do when faced with the need to better understand a human genetic variation: They created lab mice that contain the same genetic defect.

To understand why this XPF mutation caused accelerated aging, the investigators compared the expression pattern of all of the genes (approximately 30,000) in the liver of 15-day-old mice that had been generated in the laboratory to harbor a defect in their XPF-ERCC1 enzyme and that had symptoms of rapidly accelerated aging to the genes expressed by normal mice of the same age. This comparison revealed a profound suppression of genes in several important metabolic pathways in the progeroid mice. Most notably, the progeroid mice had a profoundly suppressed somatotroph (growth hormone) axis—a key pathway involved in the promotion of growth and development—compared to normal mice.

Damaged aging bodies probably produce less growth hormone as a way to reduce cancer risk. Aging cells with damaged genomes are at risk of becoming cancerous. Growth hormone exposure would make the cells divide. While a cell is dividing it is at increased risk of further DNA damage. Accumulation of DNA damage eventually causes some cells to mutate in their mechanisms for controlling cell growth. Then they start dividing continuously and you get cancer. Better to turn down the growth hormone and reduce the rate of DNA mutation accumulation than get cancer.

The investigators also found low levels of growth hormones in the progeroid mice and ruled out the possibility that this suppression was due to problems with their hypothalamus or pituitary glands, which regulate growth hormone secretion. Furthermore, they demonstrated that if normal adult mice were exposed to a drug that causes DNA damage, such as a cancer chemotherapy agent, the growth hormone axis was similarly suppressed. In other words, DNA damage somehow triggered hormonal changes that halted growth, while also boosting maintenance and repair.

Turns out these mutant mice get the same pattern of gene expression that normal old mice get. So the more rapid rate of accumulation of DNA damage causes mice to age in the same way as normal mice do but at a faster rate.

Because growth hormone levels go down as we get older, contributing to loss of muscle mass and bone density, the investigators systematically compared the gene expression pattern of their progeroid mice to normal old mice to look for other similarities. What they found was a striking similarity pattern between the progeroid and normal-aged mice in several key pathways.

Indeed, for genes that influence the growth hormone pathway, there was a greater than 95 percent correlation in changes in gene expression between the DNA repair-deficient mice and old mice. And, remarkably, there was a near 90 percent correlation between all other pathways affected in the progeroid mice and the older mice.

These results strongly suggest that most of normal aging is driven by accumulation of damage in DNA.

"Because there were such high correlations between these pathways in progeroid and normal older mice, we are quite confident that DNA damage plays a significant role in promoting the aging process. The bottom line is that avoiding or reducing DNA damage caused by sources such as sunlight and cigarette smoke, as well as by our own metabolism, also could delay aging," explained Dr. Niedernhofer.

We need gene therapies that will repair DNA damage. But if DNA damage involves most of a genome then gene therapy might not be practical. Current gene therapy techniques involve adding just a gene or two. Putting in much larger amounts of DNA is a much harder task. How to get completely new copies of entire genomes into hundreds of billions of cells throughout the body?

For neurons in the brain we have got to find ways to do extensive repair or replacement of chromosomes. We can't replace all our neurons without losing our identities. This result provides additional evidence that brain rejuvenation is our toughest rejuvenation challenge.

Update: In the last 5 years the two reports that have done the most to make full body rejuvenation look harder to me are this report above and another report that showed the blood of young mice improves the regenerative ability of the muscles of old mice. In both cases the upshot is that the scale of the changes needed to do rejuvenation came out looking bigger.

In the case of the young mouse blood, old mice, and muscle regeneration the result indicates that perhaps many cells all over the body excrete compounds into the blood that dampen down stem cells. Even if the compounds that cause this effect come from a few places and are easily blockable the fact that old bodies make stem cell suppressor compounds suggests that old bodies really need to make stem cell suppressor compounds in order to reduce the risk of cancer.

This latest report similarly points in the direction of more extensive changes needed to do rejuvenation. This report is worse news than the mouse blood report because development of ability to deliver full genome gene therapy hundreds of millions of cells strikes me as an incredibly difficult problem to solve. We will probably need some pretty sophisticated nanotechnology to solve it. I hope the nanotech optimists are right about how fast nanotech will advance. To do extensive genome repair we'll probably need nanotechnology.

Share |      Randall Parker, 2006 December 21 10:30 PM  Aging Mechanisms

Lou Pagnucco said at December 22, 2006 7:46 AM:

Here is a bit of cognitive dissonance:

"The long-lived naked mole-rat shows much higher levels of oxidative stress and damage and less robust repair mechanisms than the short-lived mouse, findings that could change the oxidative stress theory of aging
The researchers measured oxidative damage in lipids, DNA and proteins and found that naked mole-rats showed much greater levels of damage to each of these biological molecules, in all tissues assayed, when compared to mice. The study found multiple signs of lipid damage: The level of isoprostanes found in the urine was 10 times higher in the naked mole-rat, the level of malondialdehyde in liver tissue was twice as high and isoprostane levels in heart tissue was two-and-a-half times the level of the mice."

an excerpt from "Naked Mole-Rat Unfazed By Oxidative Stress" at

John said at December 22, 2006 8:45 AM:

In terms of delivering a repaired genome to the cells of the body, I suspect it would be enough to repair/replace
the body's stem cells.

Would that be easier? Hard to say. Perhaps targeted destruction of current stem cells per organ, followed by injecting the appropriate organ with replacement stem cells.

kurt9 said at December 22, 2006 11:18 AM:

Perhaps the resevoirs of stem cells in ones body exists in a sort of hierarchy. Once this hierarchy is worked out, any somatic gene or protein therapy to repair these cells will require that one develop the delivery vector to target the therapy to these particular stem cells only. Once done, the rest of the body will spontaneously self-repair over a period of time.

Perhaps the repair mechanism needed to maintain the functionality of genomic DNA are themselves depended upon the production of sufficient ATP which, in turn, require functional mitochondria in sufficient amounts necessary for this task. If so, perhaps the accumulation of damaged genomic DNA is caused by mitochondrial DNA mutatations. After all, 13 of those mitochondrial genes lie outside the cell nucleus and, thus, do not benefit from the repair enzymes that are prevalent in the nucleus.

Peter said at December 22, 2006 11:27 AM:

It has always seemed to me that DNA damage must be at the root of the aging process. It seems the key to finding out how to stop this from happening resides in our germ cells, which do not age. Since germ cells do not exist in a lead lined chamber, there must be a 100 percent effective repair mechanism available to them.

Randall Parker said at December 22, 2006 4:22 PM:


But the naked mole rat result, if interpreted to argue that the experience lots of DNA damage, would lead one to expect a mutation in DNA repair mechanisms to have little effect on longevity. Yet for that Afghan boy and the mice that is not the case.


Speaking as a software developer I have a high expectation that DNA damage is one of or the major cause of aging. Damaged code means damaged machine. Good code means good machine.


Replenishment of stem cell reservoirs will not fix the 100 billion post-mitotic neurons in the average brain. We can grow replacement organs and replenish stem cell reservoirs in much of the body. I figure between those two we can fix most of what ails us outside of the nervous system. But we have to fix the nerve cells where they are.

We need the ability to deliver whole chromosomes into cells, toss out the chromosomes they replace, and to set the epigenetic state of those chromosomes and cells. That's hard. Just getting the replacement chromosomes in is hard. But we would need to set the epigenetic state differently in different types of cells. So, for example, a dopamine neuron would need different epigenetic state than a GABA neuron. Ditto for glial cells and vascular cells of various types in the brain.

Kurt, we'd be lucky if the mitochondria really were our Achilles Heel. They'll be easier to fix. They have 15k of DNA base pairs. I suspect that the nuclear damage plays a larger role in cellular aging.

rsilvetz said at December 22, 2006 7:42 PM:

As concentration-based computers with a DNA control mechanism, we desperately need to increase the DNA protective mechanisms and the DNA repair mechanisms if we want to have any chance to win the SENS battle.

And yes, I too don't understand why we don't harvest stem cells, expand them viciously, and then freeze 20 5-year doses for future use. It's absurd that we let the majority of stem cells, which almost certainly just sit in the bone marrow and circulate out to tissues, die in the bone marrow.

Kurt said at December 23, 2006 10:41 AM:


The reason why I believe damaged mitochodrial DNA is far more significant than nuclear DNA damage in aging is that the cell nuclei have all of the repair enzymes that the mitochondria do not. Thus, the mitochodrial DNA will accumulate at a much faster rate than nuclear DNA. I'm sure that nuclear DNA damage is a problem, but not one that shows up within current life spans (80-120 years). Also, is it not likely that production of the repair enzymes in the cell nuclei are dependent upon sufficient ATP which, in turn, is produced by the mitochondria? Thus mitochodrial damage will result in damage to nuclear DNA.

Peter said at December 23, 2006 12:44 PM:

In the most recent issue of LEF magazine they are touting a new version of COQ 10 that supposedly retards the aging process substantially in mice. COQ 10 is a major factor in mitchondrial function, of course.

Randall Parker said at December 23, 2006 12:53 PM:


But the mitochondria have 15k base pairs versus 3 billion in the nucleus. That's a 200,000 to 1 ratio of nuclear DNA to mitochondrial DNA on a single mitochondrial strand. Though that's simplisitic. The mitochondria have multiple copies of their mtDNA ring and there are multiple mitochondria in a cell. Still, the nuclear damage problem seems major to me.

I disagree on the time span needed for nuclear mutations to accumulate to a substantial point. I realize that I'm also disagreeing with Aubrey de Grey when I say this (though I haven't asked him lately on this point). But I think this latest study and other evidence point in the direction of nuclear damage really being a big factor. More rapid DNA damage accumulation causes what looks like normal aging (including gene expression profiles) to happen sooner.

As for repair enzymes: They can fix base pair deletions on one half of a double helix strand. But they can't fix deletions that occur on both strands of the double helix at the same location. They also can't fix insertions caused by, for example, viral infections.

Take thymidine dimerization that occurs due to sunlight hitting the skin. Sequential pairs of thymidine bases get popped out by UV hitting them (two T's in a row absorb UV very efficiently and pop out of the double helix as a result). Enzymes go in and do repair. But if, say, a section of the DNA is popped open for transcription when the thymidines get dimerized and if both strands of the double strand have TT pairs then full breaks will occur on both strands and I doubt that enzymes can figure out how to do a correct repair.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©