Tuesday, February 2, 2010

The strange case of the cart pulling the horse: the runner's telomeres

Our culture, and certainly our science culture, firmly believes in the concept of causation that is directed in time order. Only some rather esoteric physicists toy with the idea that time is reversible, or doesn't exist, or is some sort of illusion. In everyday life, at least, cause comes first, then effect. Always. The horse always pulls the cart, never the reverse.

But that may not allow us to identify which comes first, at least if we think sloppily as seems to happen daily in the science press and hence also in the journals the reporters are reporting on. Just-So stories of how something came to be are just too appealing.

So, in the New York Times on Jan 27, there was a story about aging and running....or is it running and aging? It turns out--at least it's reported--that older men who run have bodies like young studs. They may look old on the surface, but if you gaze pruriently at the ends of their....chromosomes, you find that those chromosome ends are capped by structures, called telomeres, that are as long as younger guys' telomeres. Size matters!

Telomeres protect chromosomes from chemical degradation in the cell, so they're good for genetic function. But generally they are reported to shorten with age, and this is argued to be one cause of biological aging and senescence.
In general, telomere loss was reduced by approximately 75 percent in the aging runners. Or, to put it more succinctly, exercise, [the principle investigator] says, ‘‘at the molecular level has an anti-aging effect.’’
If runners' telomeres are longer for their age than non-runners, running must be good for you, right? And then, indeed, we must be able to find an evolutionary explanation for that.

Well, not necessarily. This may be a case of causal order, of carts and horses. Does running inhibit telomere degradation, or do longer telomeres let older men run better? That would be easy to explain: if the whole idea of telomeres and aging has merit, maybe guys with damaged telomeres feel lethargic or in other ways are not inclined to run, or can't run comfortably. One would see the same association: older runners have longer telomeres. In this case, telomeres are the horse, the running man the cart.

Of course it is also possible, though not easy to understand, that running boosts your telomere length, so if older guys run, they maintain their telomeric health. How a cell knows that you're running and how that leads it to keep up telomere maintenance is a critical subject if the man is the horse and the telomere the cart.

But all that assumes there really is a causal connection between running and telomere length, whichever way it may go. Because another widely practiced fault, an obvious one everyone knows about but a temptation few can resist, is to equate correlation with causation. Telomere length may be correlated with running in one's dotage, but there need be no causal connection between them. Older running men probably are (on average) more educated and more into health cultures; they probably also watch more Public Television than slothful older guys, who probably watch more online poker or wrestling. But we doubt that PBS shows lengthen telomeres.....or might they? What if thinking harder has that effect?


Holly Dunsworth said...

This other perspective that you are always able to offer is refreshing!

Anne Buchanan said...

Good to hear that, Holly. I am never quite sure whether having this view of the world is a blessing or a curse!

Ken Weiss said...

Thanks, Holly! That's what we try to do. That's also why we like posts from one Holly Dunsworth, who always has entertaining and clever things to say....and we are awaiting the next one!

Anne Buchanan said...

A hearty second to that!

Holly Dunsworth said...

Decaying fossils... coming right up :).

Michael said...

Nice post. One point I thought might be worth making is that the alternative explanations you suggested do not exactly mean that there is no causal connection.

In the search to explain correlations, we might postulate direct causal connections or, like you alternative suggestions, we might postulate common-cause explanations. The latter are still causal explanations but as you point out, they do not hold that either of the two observed phenomena is the cause of the other.

Ken Weiss said...

That was entirely our intent in this post. Many things might be causally related to both running at older ages and longer telomeres, or one of the latter might be causally related to the former.

Too often, we think, people (including scientists who know better) have favored explanations that are not as critically thought-out as they should be.

Nothing, of course, undermines the finding (if it's true -- a random sample of older non-runners would be worth collecting, as well as different younger people, to confirm telomere length differences. Also, telomeres vary within and among individuals, chromosomes, and cells, so that more data would be needed).

But if the finding is true, its biology could be quite interesting and important to understand.

Ken Weiss said...

I neglected to say, Michael, that there are probably a number of ways to test the idea. One would be to look at mouse or hamster or rat models: compare same-age, same-sex animals as they age, when some have exercise wheels and others don't. Or compare pet horses with race horses. Or apartment dogs with outdoor dogs of the same breed. And it would be easy to think of other ways to look for more clear-cut evidence of causation rather than just correlation.

Michael said...

Yes indeed. Agreed. Distinguishing common-cause relations from cause-effect relations can be done quite easily. As your examples suggest, it is a matter of "wiggling" the putative cause and seeing what happens to the putative effect while holding other variables constant.

If you haven't seen it already, I highly recommend Jim Woodward's book on causation, causal explanation and causal thinking: "Making Things Happen".