There's a new paper out in Cell Metabolism on resveratrol and SIRT1, and the press release from Elsevier (Cell Press) is just a tiny bit optimistic. "Study resolves controversy on life-extending red wine ingredient, restores hope for anti-aging pill", says the headline, but believe me, no one paper is going to do that. (This entry has links back to some of the history of the compound and the target, as covered here, but it's a convoluted story indeed). The EmbargoWatch web site calls it a "truly appalling" press release, and while I can't disagree with that, I don't think it particularly stands out: a lot of press releases are appalling.
And I disagree with them when they say that studies like this "probably don't deserve any coverage at all". It's actually a very interesting paper, even if it's not going to resolve any major controversies all by itself. It's from David Sinclair and co-workers (a large international team), and it presents the results of a long-running effort to see if what resveratrol does in animals that don't have the SIRT1 protein at all. That's a good experiment, which cuts right to the question of whether resveratrol's effects are SIRT1-driven or not. Problem is, the traditional knockout mouse model is almost always embryonic lethal in that case, so it's not so simple that generate such animals. The team was able to develop inducible whole-body conditional knockout adult mice, though, and set about dosing them with resveratrol to see what happened then.
Well, quite a few things did. From what I can see, the marquee items are these: normal mice fed a high-fat regimen showed beneficial effects on their mitochondria when given resveratrol, but the knockouts didn't, so that might be a clear connection to SIRT1. Resveratrol's effects on AMPK appear to be SIRT1-dependent (there are several links in this post about that connection, some of which led to papers that hypothesized a SIRT1-independent effect). But resveratrol treatment had good effects on glucose levels in mice, whether or not they had SIRT1 present, so that part seems to be going through some other pathway.
Sinclair's quoted in this Nature News piece as saying that this reflects the nature of resveratrol as a compound. “Resveratrol is a dirty, dirty molecule, very non-specific", he says. I think that's a very fair characterization, which is one of the reasons why I wouldn't take it myself. (That does shed an interesting light on the 2010 controversy when two former Sirtris executives set up their own reveratrol distribution effort, though, doesn't it?)
It would be quite interesting, for the sheer science of it, to take one of the later (apparently cleaner and more targeted) SIRT1 activating compounds that have come out of the GSK/Sirtris work and run it through the same animal model. You might expect the same sorts of SIRT1-driven effects, and perhaps much less of an effect on blood glucose, if that's really some off-target resveratrol thing. But since we're talking about epigenetic enzymes here, prediction is a chancy business. I wonder if this experiment is being done somewhere?