Since it’s a favorite topic of mine, I really have to point out this study in PLoS ONE on resveratrol. A large collaboration looked at the gene transcription effects of dosing the compound in mice, compared to a normal diet and to a calorie-restricted one. I can’t do better than the first paragraph of the paper does at setting the scene:
” Caloric restriction (CR) retards several aspects of the aging process in mammals, including age-related mortality, tumorigenesis, physiological decline and the establishment of age-related transcriptional profiles. The wide scope of these actions, and the profound metabolic and hormonal shifts induced by CR has led to efforts at identifying natural or synthetic compounds that mimic the effects of CR in the absence of overt metabolic and endocrine disturbances or reduced caloric intake. Because most age-related diseases are likely to be secondary to the aging process itself, the discovery of such compounds could have a profound public health impact by reducing disease incidence and possibly extending the quality and length of the human lifespan.”
That’s a fine list of things that everyone would like to avoid: cancer, decline, and death. And the last sentence makes a key point, that the age-related diseases are not inevitable, but can be attacked as a group by attacking aging itself. A few years back, that statement might not have made it into a scientific paper at this level, but it can now.
This study had three groups of male mice (in a hybrid strain derived from C57 black): a control group getting 84 kcal/mouse/week of food, a calorie-restricted group getting 25% less chow, and a group getting the first diet plus 4.9 mg/kg of resveratrol, both experimental diets starting at mouse middle age (14 months). This same group had already reported that starting CR at that point in that mouse strain leads to about a 13% increase in lifespan.
As a baseline, they checked the transcriptional changes in young versus old mice on the control diet. There were, for example, about a thousand genes in heart tissue (out of twenty thousand checked) with a highly significant change in their profile. Comparing old heart tissue from the controls to the old tissue from the CR group, 536 genes showed a highly significant difference due to caloric restriction. The resveratrol-treated group, meanwhile, showed the same level of change in 522 genes, from basically the same list.
They also looked at skeletal muscle and brain (neocortex) and found similar but definitely less dramatic effects. In muscle, CR only affected about a quarter of the age-related genes (as opposed to half in the heart), and resveratrol was very similar. In the brain tissue, CR was able to reverse the aging profile in only 19% of age-related genes, and resveratrol lagged with 13%. The take-home message there is that aging can be a different process in different tissues, and attempts to alter it are going to vary across those tissues as well. (Here's the figure covering all these).
Another interesting question is whether CR (or resveratrol) affect other genes that aren’t in the age-related group. The answer is “Oh, yes indeed”, with over seven hundred genes whose profile was altered by CR but are not directly altered by age alone. (Compare that to the thousand age-altered genes, five hundred of which are reversed by CR, and you can see that this could be a source of significant effects). Resveratrol treatment did an extraordinary job of mimicking this profile, affecting 745 of the same 747 genes. The same thing was found in the other tissues – 1164 non-age-related transcription changes were found in skeletal muscle from the CR mice, and resveratrol treatment affected all 1164 of them.
So resveratrol appears to be a pretty close mimic of caloric restriction – but it’s closest in the non-age-related genes, which is interesting. The thing is, there’s no guarantee that all these transcriptional changes are good – presumably a lot of the ones that reverse age-related changes are beneficial (although we don’t know that for sure), but the ones that aren’t involved in aging could be more of a mixed bag. The net effect of CR does seem to be beneficial, but there are a lot of ways to arrive at that end point, and resveratrol could be mimicking the bad as well as the good. Here’s an attempt at figuring out the functions of the various genes involved in each group.
Of course, a much more relevant measure of benefit is how the animals themselves are doing under these treatments. The paper looks at some measures of cardiac function in young and old control mice, as well as in the treatment groups, and find that both CR and resveratrol seem to protect against decline. That chart also shows some other physiological readouts, at least one of which is rather surprising.
There's a huge insulin-signaling component in this whole field of study - the putative target of resveratrol (the sirtuins) are thoroughly tangled up in insulin and insulin-related growth factor (IGF) pathways. Genetic manipulation of several genes in those areas has also shown powerful effects on lifespan and aging in model organisms. So one of the oddities here is that the CR diet showed an effect on IGF-1, but resveratrol didn't. And while both treatment groups showed increased insulin sensitivity, several markers of that (glucose transporters, for example) showed the expected changes in the CR group but not the resveratrol group.
So if resveratrol treatment really does increase lifespan in the same way that caloric restriction does, it could mean that the insulin signaling axis isn't as important in CR as people thought. That's a difficult conclusion to come to, given the other data in the field, so a more reasonable one might be that resveratrol is hitting those same pathways, but not in the same way that CR does. (The similar increase in insulin sensitivity in the two groups argues for that view). Supporting this view, the transcription factor Pgc-1alpha, which is known to be very important for a range of genes in insulin sensitivity, was upregulated in muscle in the CR group but untouched in the resveratrol group.
The weirdest thing about this whole study, though, to my mind was the finding that levels of the prototype sirtuin, SIRT1, were not elevated in either group. That's in direct contrast to results seen in rats and humans under caloric restriction. In fact, in this case SIRT1 levels actually went down in the CR mice. There are several potential conclusions, all of which will keep people busy for a good while: perhaps some (or most?) of the anti-aging benefits of either CR or resveratrol in all species are coming from something else other than the effect on SIRT1. That would sow confusion, for sure. Or perhaps this is only true in mice, but not in rats (or people), and the sirtuin pathway really is the answer in the other species - in that case, you have to wonder what's so special about mice, and just what those different pathways are that kick in with them.
No, this is a very interesting study, and a very hopeful one, but it also points out just how much we don't know. I'm sure many more surprises like this are on the way, both positive and negative ones. But the overall point is made: aging and its effects can be altered. We don't understand quite how it's happening, and there are a lot of things to be worked out, but we can do it. If we can get the details worked out, human history is going to go through the biggest inflection point since fire and agriculture.