Implications of the New Biology of Aging

Drs. Tomas Prolla and Richard Weindruch are the two principal investigators behind the first reported use of DNA chip technology to fathom the nature of aging and the mechanism by which calorie restriction stops or greatly retards aging.

The preceding interview provides much more, and much more accurate, information than any other source about this revolutionary new development other than the scientists themselves and their actual report in Science, but even so, there is more to the story than what is on the surface. Here is a non-technical discussion of the big picture behind the new findings and the new method.

To set the stage for the following discussion, the following very simple but essentially accurate way of thinking about how your body works will be helpful.

In a way, we are all like digital computers. Humans are basically computers that are programmed using a total of around 80,000 commands per person, each command being a gene. Reproduction allows shuffling of the commands that are put into the program, but the total is always about 80,000 commands (genes) per person, and most of these commands are essentially the same in all people. We are made out of proteins, and each command gives the instructions for making one such protein "part." So, in terms of our genetic endowment, we are in some sense the sum of our 80,000 parts. Other living organisms have different numbers of gene commands, but all living creatures are the product of their genes.

Your programming changes with age

A human baby looks very different than a teenager looks, and looks even more different than an embryo looks. This is because it takes a different set of commands to specify an embryo than it does to specify a baby or a teenager. As you are maturing through different life stages, you use some genes for a while, and then you stop using them and start using other genes instead, because the first set of genes is no longer appropriate for your newer stage of life. By the same token, is it possible that the difference between being an adult and being old is that older people are running a different command set than they once were?

According to the Prolla/Weindruch results, the answer may well be yes. They looked at a big chunk of the command set for old mice and compared that command set to the set of instructions being used to specify what an adult looks like, that is, to specify the adult "phenotype." And what they found was that the two lists of instructions-the list of instructions for being an adult and the list of instructions for being old-were not the same.

So it is apparently the case that there is such a thing as an old age "phenotype," meaning a set of biological characteristics specified by a particular pattern of genes that at least in part defines what old age actually is.

But is it cause or effect?

A central question that remains unanswered is: Is the aging phenotype causal or reactive? In other words, do the changes in gene expression cause old age? Or do these changes merely represent a response to random, uncontrollable damage that is the actual cause of aging?

If the first possibility is correct, then any therapy that could prevent or reverse the switch in the instruction set would prevent or reverse aging. But if the second possibility is correct, then trying to stop the changes in program with age would be beside the point, a bit like putting more gasoline in a car whose problem is a defective engine: a case of treating the symptoms rather than treating the disease itself.

According to The Wall Street Journal, at least one gerontologist, Raj Sohal, has interpreted the Prolla/Weindruch report as a vindication of random damage theories of aging, saying: "The fact that these genes are critical in blocking the damage caused by free radicals is an immensely important discovery. It helps confirm the notion that the formation of free radicals is at the heart of aging, and it gives us ways to seek out medicines that may prevent free radical damage."

The good news is that, in the end, it hardly matters whether the change in the genetic program is the cause or the effect of aging, because in either case, knowing the difference between the adult and the old-age patterns provides such powerful clues to the nature and cause of aging, that remedies should be forthcoming.

Why the DNA chip results are good news

An important consequence of the Prolla/Weindruch paper is that if uncontrollable damage is the cause of aging, it doesn't seem to be a kind of generalized or global damage, since at least 95% of all the genes examined show no change with aging, whereas one would expect most genes to react to widespread, random damage. And this conclusion is in agreement with a vast amount of prior research showing over and over that aging does not affect most of the body's proteins, regardless of how general aging seems on the level of a whole person.

So you can probably forget about the possibility that changes in gene expression with aging are a reflection of free radical or glycosylation damage to every piece of protein and every gene you've got. And that means the damage has to be of a highly non-random, highly specific and, in fact, rather peculiar nature. This is good because anything that is highly specific is a lot easier to discover than something that is generalized and vague.

For example, if aging is caused by mutations (damage to the genes, the program of life), these mutations would have to be universal, not randomized. We can say this because, as the Prolla/Weindruch interview brings to light, you can take any three mice at random and compare them, and you find that, at the gene level, they all look the same: there is nothing random about the changes in gene expression pattern.

What kind of random damage can be imagined that would produce highly selective, non-random effects? Such damage does not fit with any previous concept of random damage except, perhaps, random damage to mitochondrial DNA. If this is the cause of aging, however, remedies have already been proposed, especially by Aubrey de Grey, and solving the problem of aging would be conceptually far simpler than individually adjusting hundreds of genes. In fact, one of the findings of the Prolla/Weindruch report is that a genetic switch for turning on mitochondrial renewal is turned down with aging. Probing that particular change in program could be very informative.

In any case, having the genetic command list or blueprint before us will allow us to determine either that aging is the result of damage, or that it is the result of a more normal change like the change in going from one stage of development to another and, in either case, it now seems apparent that the answer is going to converge on a relatively small set of prime movers of aging.

How many genes?

The paper in Science found more than 100 genes that change two-fold or more with age, and 200 more genes that change less than two-fold, and the authors of the report only got to peak at 10% of the program set for the mouse. Simple extrapolation might therefore suggest that if you were to look at the whole mouse, you could find more than 3,000 changes in program code that might have to be rewritten to address aging-not exactly a trivial undertaking.

Fortunately, this estimate of 3000 age-related genetic instruction changes is wrong. As we already considered, lots of commands are used only during embryological development. You can subtract virtually all of those as age-determining candidates from the total set of genes. You can also subtract all genes required to specify how to make toenails, hair, fingers, uvulas, adenoids, nasal septa, and blue eyes, for example-we really don't have much reason to think that these genes and others like them, which constitute a huge number of genes, have much to do with why we age as a whole and with why we die of old age. So it's more likely that less than 10% of the complete gene set-less than 10% of the genome-represents the possible age-determining genes.

If that's the case, there's no need to multiply the Prolla/Weindruch gene list by 10-the total number of age-governing genes in the whole animal may be close to the total number measured by these investigators in muscle alone: perhaps in the neighborhood of 300 genes. Subtract out two-thirds of these as being addressable by calorie restriction mimicking therapies, and you end up with around 100 genes left. And many of these may be linked to each other in a chain of causation that places the majority under the control of a minority whose changes then dictate the other changes that take place.

It's therefore plausible that we are talking about needing to adjust only 50 core age-determining genes in order to radically extend lifespan. In fact, this conclusion is more than supported by laboratory manipulations of single genes that have produced major changes in both mean and maximum lifespan already.

How can the key genes, however many there may be, be reverted back from the senescent pattern of use to the adult pattern of use? Really, there's no need to worry too much about that. With multi-billion dollar pharmaceutical companies eyeballing multi-trillion dollar profits that would emanate from the control of human age-related diseases, it's pretty clear that answers will emerge, and the dollars involved will encourage the emergence sooner rather than later. But just for fun, consider a bit of contextual background.

The background of biological power

We are standing at the dawn of a new era in biology. Fields like genomics are giving us chips that will soon be able to screen every gene in a human being in an afternoon. In one variation, chips are being created right now that can test for every possible mutation underlying certain kinds of breast cancer for proper diagnosis, and there are more than a dozen methods already in existence for reversing defined genetic mutations in living cells and in mammals (this is called therapeutic DNA repair). Fields like proteomics are giving us tools for scanning all of the proteins being made in the tissues of individuals, and checking for protein "fingerprints" that specify the key differences between health and disease.

And, as just one more example, there are several variations of what is called combinatorial chemistry, in which staggering numbers of variations in drug molecules or in biomolecules can be created and tested in short order, in many cases enabling brilliantly effective therapeutic molecules to be found with no knowledge whatever of the underlying therapeutic target itself. In the case of aging, once you know the pattern of gene usage you want to see, you can use combinatorial approaches to "fish" for molecules that turn on or turn off whole blocks of genes, until, one way or another, you get what you are looking for. And you can do this knowing nothing about what it is that your "fish" may be biting on that brings about aging reversal, so the fishing job is easy.

We are entering a time when we will be able to turn specific genes on and off at will, and with few or no side effects, and when we will be able to add new genes if we want to correct diseases or age-related changes. It is just the right time for the genetics of aging to be laid bare.

Implications

In the past, physics and engineering led to astonishing innovations that tended to leave biologists in the dust. We had airplanes, cars, Tesla coils, refrigerators, and even televisions well before we understood the structure of DNA, let alone how it carries hereditary information. But it has taken less than 50 years to go from the understanding that DNA encodes biological information to the commercial availability of DNA chips that can probe half of the entire set of instructions for making human beings.

As Erik Drexler said, if we are to be conservative and cautious, we must assume that what has been true in the past will continue to be true in the future. And what has been true in the past is an explosive and accelerating pace of revolutionary change and increased technological capability. Biology is catching up, in part because of a merger between biology and computer science, and the implications are truly profound. If we are lucky, we won't have to wait much beyond the dawn of the new millennium to see striking practical results. Let's hope at least a few of these will be in the area of interventive gerontology.

-Gregory Fahy